Full PaperLloyd George Acoustics
ABSTRACT
Have you ever been asked how you got into acoustics? Are you a recent graduate employed in acoustics and unsure of what the future holds? Do you employ young engineers and acousticians and not know what to do with them half the time or worry if they will stick around? Being the specialised field that it is, getting that first job in acoustics or employing the ideal graduate can be like finding and catching a unicorn. This paper explores what makes the field so unique for those starting out, in an anecdotal telling of one graduate consultant's story, 6 years down the track. The balanced discussion also includes summaries of interviews with employers and educators in the field. Topics explored will interest employers, employees and educators - including insight into retaining young staff, developing knowledge base versus on-the-job training, and shaping the industry to attract new talent and create jobs.
Research Leader Acoustic Signature Management
ABSTRACT
Acoustic signature management is critical for the operational survivability of ships and submarines. This paper discusses some of the research programs at DST focused on developing modelling and prediction capabilities for the understanding and prediction of the acoustic signatures of maritime platforms. The paper considers some of the acoustic signature components for a conventional diesel electric submarine with an open bladed propeller; the development of modelling and prediction capabilities to estimate the acoustic signature for those components; and design procedures and control strategies that may be employed to reduce those signature components.
(1) Maritime Division, Defence Science & Technology Group
ABSTRACT
Measuring acoustic radiation from underwater equipment at sea is expensive. Therefore, it would be highly desirable if the acoustic source level could be obtained accurately and easily in a water tank. Building an anechoic water tank for the audio frequency range is not practicable. In this paper we present a method that can be used to obtain source level from measurements that are conducted in a reverberant water tank. The method is based on room acoustics for air, but only a few underwater applications can be found. The method requires a highly reverberant confined sound space and an accurate measurement of reverberation time (T60). A previous paper applied this to a large water tank successfully by using a broadband T60. The current method utilises the reverberation time in one-third octave bands. The reverberation time indeed was found to be function of frequency. The T60 is used to calculate the critical radius where the direct field and reverberant field of the source are equal. The measured reverberant sound pressure level (SPL) of an object can then be used to give the direct field and therefore the source level at 1 m. To illustrate the accuracy of this method, measurements from an underwater sound source in a reverberant water tank were compared to its known source level. A very good agreement between the two results is obtained.
(1) South University of Science and Technology, No 1088,xueyuan Rd., Xili, Nanshan District,Shenzhen,Guangdong,China 518055 (2) China Aerodynamics Research and Development Center, No.6, South Section,Second Ring Road, Mianyang City, Sichuan Province,P.R. China  621000 (3) China Aerodynamics Research and Development Center
ABSTRACT
Unmanned aerial vehicles (UAVs) can be observed in many applications. In some circumstances, aeroa-coustic noise of UAVs is a big concern and quiet propellers would be more environmentally friendly. Therefore, this research imitates the serrated trailing edge of owl wings which are famous for their low noise emissions. Spectral characteristics, noise reduction and directivity of the noise from a propeller with serrated trailing edge are measured by microphones in an anechoic wind tunnel. The time-average loadings of the serrated and normal propellers are analysed by a six-component balance. The results show that the noise of the propeller is alleviated, and the noise reduction depends on the Strouhal number and flight conditions. During forward flight, the maximum noise reduction is up to 12 dB when the Strouhal number is around 4. The noise reduction on the side of the forward moving blade is larger than that on the other side. During the hovering condition, the maximum noise reduction is up to 7.5 dB when the Strouhal number is around 2, and the noise reduction is axisymmetric. The lift and drag decrease during forward flight condition, but the lift-to-drag ratio remains the same. The aerodynamic performance is not altered during the hovering condition.
(1) School of Mechanical & Mining Engineering, The University of Queensland (2) School of Pharmacy, University of Queensland, Brisbane, Australia.
ABSTRACT
Ultrasonic theranostics using echogenic bubbles allow real-time traceability and site-specific, on de-mand medicine delivery. Ultrasonic manipulation of microbubbles is studied through illustrating the effect of acoustofluidic properties on flow and interaction of bubble populations. Significant displace-ment and clustering of medical microbubbles are observed for excitations resulting in acoustic pres-sures as low as 6 kPa. Microbubble displacements are found to be negligible at 4 kPa. The measured velocities show good agreement with analytical calculations at pressure levels below the threshold of non-linear bubble behaviour. Higher pressures caused faster bubble movement and higher frequencies resulted in faster cluster formation. An extreme effect is also reported where the secondary radiation force overcomes the primary radiation force, which results in bubbles moving towards the acoustic source.
CMST Curtin University
ABSTRACT
The wave properties for a one dimensional periodic structure are related to the eigenvalues and eigenvectors of two pairs of 2x2 matrices M and N, E and G. M is the scattering matrix of coherent complex waves by a single scatterer while N is constructed from M by interchanging eigenvalue and eigenvector parameters. E is the scattering matrix for wave energy fluxes onto and away from the scatterer, while G is constructed from E by interchanging eigenvalue and eigenvector parameters. Equivalently N and G are related to M and E respectively by interchanging forward and backward scattering parameters. This matrix method allows wave solutions for any size structure, including infinite periodic structures where their "cell independent" vectors are just the eigenvectors of M, N, E and G. It is shown that the characteristic equations (CE) for the eigenvalues of E and G can be derived from the CE for the eigenvalues of M and N. Further, all the elements of E and G can be derived from M and N CE parameters. Damped or amplified Bloch-Floquet waves (BFW) are an example of coherent periodic structure waves (PSW) where the difference of backward and forward average phase shifts is +/- pi/2. More generally scatterers may have internal wave modes giving rise to phase sensitive wave-scatterer energy exchanges causing the difference of backward and forward average phase shifts delta to deviate from +/- pi/2. Phase sensitive energy exchanges allow coherent waves to exist in asymmetric periodic structures, unlike previous papers by the author where asymmetric phase shift differences delta and phase insensitive inelastic scattering alone confined PSW to incoherent energy propagation.
AAS
ABSTRACT
This paper presents a discussion on the acoustical effects on absorptive treatments located behind visual screens and perforated facings. All facings will have some acoustic effect. It is desirable to know what effect a facing may have. It may not be significant at the frequencies of interest, that is, acoustically transparent, or it may not be insignificant. This paper discusses some prediction methodologies for both visual screens and perforated facings and compares outcomes with reported measurements. It concludes with some questions which at present (to this author) remain unanswered.
(1) School of Mechanical Engineering, The University of Adelaide, Australia (2) School of Chemical Engineering, The University of Adelaide, Australia (3) Department of Chemistry, School of Physical Sciences, The University of Adelaide, Australia
ABSTRACT
This presentation will provide an overview of resent research on acoustics at the nanoscale including how recent developments in nanotechnology are being utilised to create advanced acoustic materials that incorporate either carbon nanotubes or graphene as components. Methods to overcome the challenges involved in conducting experimental measurements on relatively small samples to characterise the acoustic performance of the developed acoustic materials will be detailed. The acoustic performance of the developed materials will be presented and the hypothesised mechanisms for their enhanced performance described. The simulation tools applicable to modelling acoustics at the nanoscale will be outlined as well as the challenges faced in modelling the acoustic performance of the developed acoustic materials in the audible frequency range and approaches currently being developed to address these challenges.
(1) Southern University of Science and Technology SUSTech, Shenzhen, China, 518055 (2) Key Laboratory of Aerodynamic Noise Control, China Aerodynamics Research and Development Center, Mianyang, China, 621000
ABSTRACT
Cylinders placed in uniform flow produce a vortex shedding tone that can be observed in many commercial and industrial applications, such as chimney stacks, high-speed train pantographs, power lines and aircraft landing gear. Recent advances in passive flow control methods, such as porous coating treatments, have been effective at weakening this vortex shedding tone. The fundamental cause for this noise reduction is not yet fully understood. Adding a porous coating increases the cylinder diameter, leading to an increase in Reynolds number and a subsequent shift in Strouhal number, yet this alone cannot explain the change in measured noise. Furthermore, the true outer diameter of a porous coated cylinder is uncertain, as the outer surface decreases in surface area with increasing porosity, thus making the Strouhal number estimation of the cylinder difficult. Numerical simulations require an approximation method, such as the Darcy profile, to model porous surfaces that do not take into account the different types of porous structures. Despite recent advances in computational technology, direct numerical modelling of porous structures is far too computationally expensive at aerospace-related Reynolds numbers.
An experimental investigation will be presented in this paper. The investigated cylinders are three polyurethane coated cylinders with outer diameters of 60 mm, coating thicknesses of 10 mm with varying porosity (between 85% to 95%), and two bare cylinders with outer diameters of 40 mm and 60 mm. Far-field acoustic measurements conducted in an anechoic wind tunnel were obtained at five flow speeds, between Reynolds numbers of 80,000 to 200,000. Velocity spectral data in the cylinder shear layers were obtained using hot-wire anemometry to help explain the acoustic spectra. The smaller and larger bare cylinders act as references for porosity extremities, being 100% and 0% respectively. From these data, a relationship between an effective porous coated cylinder diameter and porosity will be presented.
(1) UNSW Sydney
ABSTRACT
This paper will describe the anechoic wind tunnel facility located at the University of New South Wales (UNSW) Sydney. The wind tunnel construction, including the inlet contraction and fan system, will be described along with the design of the anechoic chamber and muffler system. The model mounting system is located in an open jet, with integrated force measurement system via a turntable mount. Aeroacoustic measurements are performed with an aeroacoustic microphone array, whose design, construction and signal processing will be explained. Aerodynamic measurements using hot-wire, pitot and PIV systems are used and integrated with acoustic measurements. A remote microphone measurement system has been used with a novel calibration methodology to measure the turbulent wall pressure spectra in a variety of configurations. In addition to an extensive description of the construction and performance of the wind tunnel, selected results from recent airfoil self-noise experiments will be included to demonstrate the acoustic measurement performance of the facility.
(1) Senior Environmental Specialist (2) Airport and Environmental Assurance Team Leader
ABSTRACT
The Integrated Noise Model (INM) was developed by the United States Federal Aviation Administration (FAA) and has been used extensively in Australia for the environmental assessment of aircraft noise, development of Australian Noise Exposure Forecast/Index maps (ANEF/ANEIs), and the generation of number above (Nxx) noise contour maps. The INM software was supported by the FAA until September 2016 and has since been replaced by the Aviation Environmental Design Tool (AEDT). To keep abreast with technology changes, the aviation industry in Australia must migrate away from the use of INM to AEDT. The continual use of the outdated software will eventually affect the accuracy of noise modelling and prediction of noise impacts around air-ports in Australia. This paper aims to informatively discuss the proposed use of AEDT in Australia and to specifically address Airservices' technical endorsement process of ANEFs using the new model. The paper presents preliminary work undertaken to validate the use of AEDT in Australian conditions with data extracted from Airservices' Noise and Flight Path Monitoring System (NFPMS), and presents comparative results of noise con-tours. The paper also aims to review airport noise modelling best practice in Australia and provides insight into modelling with AEDT.
(1) The University of Sydney
ABSTRACT
Acoustically retroreflective surfaces reflect sound back to the source, over a wide range of possible source positions.
This paper explores their potential use in architecture, in the form of an extensive array of square trihedral
corner cubes which could be used for reducing speech distraction in multi-talker environments. The retroreflective
effectiveness of a corner cube array surface is investigated by simulation using the finite-difference time-domain
(FDTD) method. The factors investigated were the distance of an omnidirectional source to the array surface (1,
2 and 3 m) and the size of the individual corner cubes that make up the array surface (0.3, 0.6, 0.9 and 1.2 m
edge lengths), in the 250 Hz - 8 kHz octave bands. Retroreflection performance was assessed by (i) the energy
reflected by the array surface back to the source position, and (ii) the spatial decay rate of reflected energy as a
function of distance from source. Results indicate that a retroreflective array surface comprising corner cubes with
edge lengths of 0.3 m and 0.6 m have most potential as surface treatments in architectural applications over the
range of distances simulated.
Marshall Day Acoustics
ABSTRACT
In each of the decades starting in 1955 advances have been made in understanding the design of rooms for music - in research and in the design of the rooms themselves which often outstrip the limitations of the research. In this talk I shall outline at least one design in each decade and the underlying research. Quite often these were concurrent. A common theme is the search for the "more" than physical science - a "more" which we value in music rooms. 2018 has seen two colloquial on this topic, one on presence in Auckland, and a satellite colloquium in Emmanuel College Cambridge during September. My talk finishes with a map for the production of presence drawn from the experience of 60 years of engagement.
Marshall Day Acoustics
ABSTRACT
Concave surfaces are often avoided by acousticians for fear of focusing and in particular circular room designs are generally avoided for meeting room applications. In 2016 I was approached by Macquarie University to carry out the acoustic design for the refit of their existing circular Council Room. The project was to involve a significant financial investment in a single, high profile chancellery space.
This paper provides a summary of the acoustic effects of surfaces curved in a single dimension and the acoustic requirements for support of speech and the application to circular rooms. The paper then describes the application of these principles to the design of the MQU Council Room project and an examination of the out-comes.
The architectural intent was to highlight the existing circular nature of the space. The acoustic brief noted that speech intelligibility in the existing space was not acceptable requiring that the acoustics in the new fitout space required a significant improvement. The new design needed to provide speech intelligibility between people around the Council Table, the people seated around the periphery of the room. The acoustic design was also required to provide the necessary speech intelligibility for the use of the proposed new video conferencing system to be installed.
The solution was developed using a range of acoustic modelling tools and included 3D impulse response measurements to benchmark the acoustic performance and reflection sequence in the existing space and to perform certification on completion. The most critical element for the rooms acoustic success was the design of the circular ceiling reflectors. These were designed using an iterative process involving collaboration with the Architect, starting with paper cut-outs, before progressing to proprietary 3D modelling and ray tracing.
Rodney Stevens Acoustics
ABSTRACT
Acoustic treatment of internal spaces is vital in controlling the acoustic energy within the room. Precise loca-tion of the materials, selection of the materials and the layout of the room is important in controlling the acoustic environment. This paper investigates 3 different room acoustic models (EASE, I-Simpa and Basic Excel Model) and compares the results with measured on site data. Room modelling includes predicting the performance of the empty space and predicting the performance and location of the acoustic materials. The paper will summarise the limitations of each of the modelling method against the measured data. The scenario to be modelled consists of a large reflective room with a volume greater than 10,000m3.
(1) Norman Disney & Young
ABSTRACT
Cross laminated timber (CLT) is a modern building material which is gaining increasing application in Australian and New Zealand building developments. It is used as a structural wall and floor element, and has certain advantages over traditional building methods utilising concrete floors, particularly relating to the speed with which CLT structures can be constructed on-site. One disadvantage of CLT is that the base timber floor panel has less mass per square meter than concrete. Therefore, CLT typically requires additional layers of material (ceilings and raised floors) to achieve airborne and impact sound insulation performance similar to that of concrete floors and, indeed, to meet Building Code requirements. This paper explores a range of on-site test results obtained from two similar CLT apartment buildings. Standard airborne and impact sound insulation results (DnT,w, FSTC, LnT,w and FIIC) are presented, as well as heavy impact results obtained using a "Japanese ball drop" method (LiA,Fmax) to assess the low-frequency performance of the CLT floors. Various flooring upgrades were tested with the aim of improving the sound insulation performance of the floors. Test results from other apartment buildings with a mixture of concrete floors and timber joist floors are also presented and compared to the CLT floor results.
(1) Embelton
ABSTRACT
Resiliently treated bolted connections are being increasingly utilised as the demand for high density housing is combined with the increasing requirements of acoustic comfort for occupants. The typical design of a bolted connection is for an oversized hole in the equipment bracket or cleat combined with a sandwiched elastomeric pad with resilient sleeve and washer to prevent any metal on metal contact. However, the use of a sleeve and washer, which often contain different stiffness characteristics to the pad, will compromise the theoretical attenuation through the connection and the expected performance loss is difficult to calculate due to a number of variables. In this paper, the vibration transfer through a bolted connection between a steel beam and concrete structure was investigated at low audio frequencies using a number of commonly used isolation configurations, along with correct and incorrect installation. An out of balance motor with variable speed was used to excite the beam at the lowest frequencies, and a large speaker used at the higher frequencies. In order to minimise flanking paths, the end of the beam with the bolted connection was placed on the structure, and the other end supported on a separate structure. Vibration levels were measured both on the steel beam and on the concrete structure directly adjacent to the connection, in order to determine the level of vibration transfer and the frequency domain characteristics relative to a rigid connection.
(1) Isolgomma (2) University of Padua
ABSTRACT
Among the different types of wooden slabs, the CLT one is a solution increasingly used worldwide for both new buildings and restoration, mainly due to its structural, thermal and seismic resistance, and also for its low energy and environmental impact. In this paper the results of the laboratory acoustic characterization of CLT slabs are presented with the application of different possible performance enhancement techniques, according to the ISO 10140 series standards.This specific type of construction element has thicknesses comparable to solid concrete slabs, but it is characterized by modest acoustic performance, mainly due to the low structural mass that distinguishes it. However, significant increases in performance can be achieved both for level of im-pact and airborne sound insulation with adequate intervention techniques. In this paper the results of the acous-tic characterization in laboratory of CLT slabs are presented with the application of different possible tech-niques of performance increase, according to the standards of the ISO 10140 series.
(1) Renzo Tonin & Associates (2) Mechanical Engineering Student at University of NSW
ABSTRACT
In recent years, Sydney has experienced unprecedented levels of infrastructure and building construction activities. Tunnel boring machines, roadheaders, rockbreakers, drilling rigs and vibratory rollers are commonly used plant on underground tunnelling projects and major construction projects which cause significant vibration and ground-borne noise.
While theory surrounding airborne noise propagation is well established worldwide due to the consistent propagation medium of air, ground-borne noise and vibration (GBNV) is not as well defined. Although current theoretical models exist which can predict GBNV levels, such models do not cater to the unique geology of each area and often they provide relatively inaccurate results in comparison to empirical models. Accordingly, GBNV levels are often measured on site from which site-laws and empirical models are established.
Ground-borne vibration measurements were undertaken at different distances from construction machinery and outside and inside buildings surrounding construction activities from a range of recent Sydney projects. Undertaking vibration measurements at buildings allows for the consideration of how these vibrations may affect a building's inhabitants and how ground-borne vibration transforms into regenerated or ground-borne noise inside buildings. The data is presented via regression analyses to produce GBNV versus distance propagation curves, which can assist to describe the extent of areas potentially impacted for various types and sizes of construction machinery.
Vibration measurements were conducted at a range of distances from the source and where the opportunity existed measurements were conducted on the exterior and interior of buildings. Additionally, noise measurements were conducted concurrently, inside subject buildings, to consider the noise produced by the intrusive vibration. The majority of interior vibration and noise measurements were taken inside rooms at basement or ground levels, shielded from external intruding airborne noise.
(1) Octave Acoustics
ABSTRACT
Previous testing and analysis has indicated that standard low-rise lightweight constructed buildings can comply with sound insulation requirements under the building code when framing, plates or columns are used as the load bearing element for simple timber joists, floors and ceilings.
The recent influx of prefabricated high-rise timber development, however, has also brought with it more unorthodox forms of load bearing construction and subsequent design considerations in mitigating structure-borne flanking noise via connected load bearing elements.
This paper presents analysis of the sound insulation properties of prefabricated timber building modules which rely on exposed internal timber cladding as the structural load bearing element in lightweight construction. The analysis applies to modern prefabricated buildings where timber 'slabs', such as Cross Laminated Timber (CLT) and Laminated Veneer Lumber (LVL) and Oriented Strand Board (OSB), can be left exposed as finished cladding and where there is no additional plasterboard framed wall constructed internally.
'Mock up' sound insulation testing was carried out using prefabricated timber cassette floor systems supported on both load bearing timber cladding and the timber framing behind the cladding. A comparison of results is presented between non-isolated and vibration isolated structural and load bearing connections using polyurethane foam to determine the effect on flanking transmission.
(1) CSIRO Infrastructure Technologies (2) RMTI University and CSIRO Infrastructure Technologies
ABSTRACT
Inadequate sealing of edge gaps of building partitions (walls, ceilings etc) can adversely affect the acoustic insulation of a building partition, with consequent loss of amenity to building occupants. Measuring and rating the effectiveness of caulking compounds to protect against such adverse effects is somewhat problematic for a variety of reasons. This paper proposes a method based on ISO 10140-1 Amendment 1 Annex J, and ISO 717-1 to enable laboratory measurement of sound transmission through controlled gaps sealed with the compound under test, and determination of the suitability of such compounds for use in sealing building partitions with specified acoustic insulation requirements. Some examples are provided, from laboratory experiments with test-gaps sealed by different means, demonstrating the outcome of the proposed method in a variety of circumstances.
(1) School of Mechanical and Chemical Engineering, The University of Western Australia
ABSTRACT
The acoustic wave propagator method has been developed and extended since 2000 for investigating the propagation of acoustic and flexural waves. The boundaries are usually introduced by defining the mechanical properties in the wave transmission path. This was found to be highly efficient but slightly inaccurate, especially when a boundary-smoothing technique is used to weaken the discontinuity and maintain the numerical stability. In this paper, an image source method is implemented during modelling, which simulate the boundary reflection by placing virtual sources emits similar signals at the same time. Compared with the results of several previous papers, this is found to provide a more accurate solution for studying wave propagation under different boundary conditions.
(1) Day Design Pty Ltd
ABSTRACT
The National Construction Code (NCC) 2016 for Class 2, 3 and 9c buildings requires services to be acoustically separated from habitable and non-habitable rooms by a specified performance criterion outlined in Part F5.6. Pipe systems consisting of lagged or acoustic pipes provide an acoustic benefit over traditional PVC pipes. Since neither lagged or acoustic pipes contribute to the separating construction between the services and room, rating their acoustic performance in accordance with the NCC Deemed-to-Satisfy Provisions in Part F5.6 is impossible. This paper presents and describes an established method to test and a unique calculation meth-od to rate these acoustic pipe systems to demonstrate that the construction of these systems provides an acoustic benefit that may be compared against the NCC requirement. Results of tested lagged and acoustic pipes that are in the public domain are summarised. Various issues with the testing method, calculation proce-dure and NCC criteria are discussed. This paper promotes discussion directed towards the establishment of an Industry Code of Practice for Australian acousticians to enable standardised acoustic rating of innovative pipe systems in accordance with the NCC.
ABSTRACT
Mechanical Services have become an essential part of modern building design. The acoustic design requirements for the building mechanical services form a significant portion of the overall building acoustic design package and is often a main driver for end-user amenity. This paper examines some of the issues encountered and options to assist in the mitigation of the risks to acoustic professionals.
ABSTRACT
Surf wave parks are an emerging commercial development that are increasingly being proposed to be built. When considering noise control they present large areas of water where substantial inertial masses are displaced in order to create surfable waves of heights, currently up to 2 m, that continuously break in sections of the water body area with a high frequency and long duration. Patron, traffic, plant and machinery noise are often misperceived by the public to be the main contributing noise sources, where long durational noise from resonance of air in the tube of the wave or cavitation of the bubbles created in the spilling or breaking process are dominant. Airborne generation of noise from breaking waves has been shown to be complex, containing tonal, modulating and broadband components, which are all additive when assessing noise dose. A case study is presented of a wave park proposal in Tompkins Park, Alfred Cove, Western Australia alongside the Swan River. This proposal has been controversial due to its placement next to a protected migratory water bird sanctuary, and the large number of noise sensitive receivers in the surrounding residential neighbourhood. Characteristics of wave noise are examined indicating placement and assessment problems relative to the location.
(1) School of Mechanical and Mining Engineering, University of Queensland (2) AECOM
ABSTRACT
Strategies for the removal of extraneous noise from Leq and Lp samples cannot necessarily be applied to exceedance levels (LN) without the introduction of significant error. This thesis aims to quantify the error associated with the filtration and logarithmic combination of spectral LN to estimate overall LN without extraneous noise. The problem is primarily viewed within the context of L90 mine noise measurements corrupted by insect or bird noise.
Digital PCM samples of mine noise were combined with various samples of extraneous bird, insect and frog noise. Spectral LN in one-third octave bands were calculated for the combined signal, and filtered based on the spectral profile of the extraneous noise. The filtered LN were logarithmically summed to estimate the overall LN. This estimate was compared to the overall LN of the control mine noise sample.
The amount of error introduced using this process was calculated for a range of LN (L1, L10, L50, L90, L99), and over a range a measurement intervals (0.1 s to 1 hour). The results were correlated with the nature of the samples. The results were used to inform a set of recommendations regarding the conditions under which filtering and summation of spectral LN can be performed with acceptable error.
Dr
ABSTRACT
Birds are an important group of animal that ecologist monitor using autonomous recordings units as an crucial indicator of health of an environment. There is not yet an adequate method for automated bird call recognition in acoustic recordings due to high variations in bird calls and the challenges associated with bird call recogni-tion. In this paper, we use ResNet-50, a deep convolutional neural network architecture for automated bird call recognition. We used a publicly available dataset consisting of calls from 46 different bird species. Spectro-grams (visual features) extracted from the bird calls were used as input for ResNet-50. We were able to achieve 60%-72% accuracy of bird call recognition using ResNet-50.
(1) University of Adelaide
ABSTRACT
The common assumption when predicting community noise levels from exhaust stacks is that the sound radiation is axisymmetric, which is appropriate for low Mach number, low temperature flows, and in the absence of cross-winds. However, this assumption is inaccurate when the conditions do not hold. This paper investigates experimentally the effects of a cooler cross-flow on the sound directivity of turbulent, heated exhaust gas from a stack. A small-scale experimental rig (1:100) has been built that reproduces the flow, temperature and acoustic conditions seen in large-scale open-cycle gas turbines. The University of Adelaide's Wind Tunnel was used to create cross-flows that resemble the cross-wind conditions observed near open-cycle gas turbines. The small-scale experimental rig and the test conditions were scaled using relevant non-dimensional fluid dynamics and acoustic parameters. Directivity measurements from the experiments have shown that the hot exhaust plume leads to an asymmetric acoustic radiation pattern from the exhaust stack. The experimental results presented in this paper shows that the presence of a hot exhaust plume and cross-flow alters sound directivity in the far-field downstream by increasing levels up to 11 dB, which has the potential to impact nearby communities.
(1) GHD Ltd (2) SA Environment Protection Authority
ABSTRACT
Advancements in acoustic instrumentation and data acquisition enables the analysis of substantial amount of data collected during long term noise monitoring programmes. Generally, variations of environmental noise at particular location are random. However, recent research and analysis based on substantial amount of data collected in urban areas shows the presence of patterns or major cycles in time histories of the noise levels. This paves a way to exploring the concept of a generic time history and predictions of future noise.
Neural networks represent an attractive tool to explore and predict future noise levels based on the previous measured values. The paper considers opportunities for predicting noise levels using neural networks. The networks were trained on noise data acquired in variety of urban and suburban noise environments, such as road, rail and industrial noise sources. Different modelling approaches are considered for attempts to forecast a noise level time history. Potential use of the technique in environmental noise monitoring opens opportunities for expecting or preventing non- compliance.
(1) NSW Dept of Planning and Environment
ABSTRACT
Unwanted music from outdoor events is considered a form of noise pollution which presents a unique set of challenges for regulators when compared to other environmental noise sources. Unlike noise generated by sources such as transport or industry where lower levels are always desirable, there is a minimum level of music below which patron experience will be unacceptable. The challenge for regulators therefore lies in balancing the need for entertainment against the impacts of outdoor music on the surrounding population. Regulators and organisers of outdoor music festivals in rural environments are generally required to comply with receiver-based noise limits in noise catchments which range from very low backgrounds to those which may have dominant natural or transportation noise. With this in mind, this paper describes the approach undertaken to develop a practical and realistic set of noise objectives for a music festival site on the north coast of NSW, Australia.
(1) Griffith University School of Environment and Science
ABSTRACT
Over the last ten years there has been a significant increase in the number and type of remote monitoring systems used to measure a range of natural and man-made environmental parameters. Initially developed as stand-alone systems that required a field technician to down load the data, these monitoring units are now web enabled allowing for the centralised storage of data in repositories. The data is as diverse as it is detailed and the quantity of the data is both challenging to manage and challenging to interpret. The increase in the number of monitoring systems has coincided with an increase in the requirement for the timely notification of changes to environmental parameters. The current trend is to use 'dashboard' and 'traffic light' systems that incorporate alarms using email and SMS notifications which result in reactive management strategies. This paper provides an overview of the research behind Situation Awareness and Data Visualisation that can assist in the integration of monitoring data from a range of different data streams. The research indicates that the user's perception and awareness of the situation is enhanced when they are provided with appropriate data visualisation. Furthermore, the increased situation awareness results in decision-making that can be pre-emptive rather than reactive.
(1) Griffith University School of Environment and Science
ABSTRACT
This paper shows how the management of sound emissions from the constantly changing environs of a large open-cut mine can be modelled as a closed-loop control system and how using this approach has already been shown to decrease the noise immision levels from an open cut mine. In traditionally control theory, a closed-loop control system (also called a feedback control system) uses a 'set point' as a reference value, to characterise the desired output response of a process. The actual measured 'output response', is compared to the set point and the difference between the two, the error, becomes the input signal to the 'control device'. The measured difference between the set point and the output response of the process can be due to a change in the set point or due to a disturbance to the system. A well-tuned control system is characterised by its stability, its ability to respond to external disturbances and its reliable and repeatable performance.
For an open-cut mine the desired output response is to not exceed a specific noise limit while maximising pro-duction and minimising any periods of shutdown. To achieve this, the disturbances to the system must be managed. Disturbances to the system include changes to the mine plan, changes in equipment location and changes in the weather. The current control strategy for open cut mines can be insensitive to the error signal and the response to disturbances can be too slow resulting in difficulty bringing the system back under control without enforcing shutdown measures. In this paper it is shown how the management of sound immissions from an open-cut mine can be representated as a closed-loop control system. This has enabled the elements of the control system to be examined using traditional control theory and facilitated the development of a new management tool that is being implemented in a case study in the Hunter Valley.
Retired Engineer
ABSTRACT
The paper deals with airborne and underground acoustic models and electronic modelling developed for real life simulations. It examines the translation of the physical understanding of sound in the air and underground transmission into mathematical models that can simulate acoustical propagation, noise reception and reverberation in the air and underground. Electronic models apply complex numbers and electrical theory to equivalent circuits to determine characteristic impedances and ultimately the voltage gain and frequency response of the equivalent electroacoustic model. These models programed on Excel spreadsheets can be used in a variety of research and operational applications to predict and diagnose the performance of acoustical systems operating in air and underground environments and electronic equivalent circuits. Descriptions of and guidelines for selecting and using available propagation, noise reception, reverberation and electronic models are highlighted. Instructive case studies demonstrate applications in airborne, underground and electronic simulation.
The paper will provide guidance in the selection and application of airborne, underground acoustic and electronic models. Simulation is fast becoming an accurate, efficient and economical alternative to actual field testing and in-situ training.
Real life examples cover explosive sounds (airblast and ground vibration) and re-radiated noise and low frequency ground vibration from underground railway operations. There are further applications for construction noise and vibration, shooting noise, and motor racing noise; space does not permit them to be included here. Modelling for electronic and electro-acoustic applications will be presented, along with examples of electrical characteristics of passive bandpass filters, transfer characteristics of audio transmission lines and impedance matching.
(1) Cogent Acoustics
ABSTRACT
Urban design in recent years has been embracing "compact city" design concepts. With such trends in urban design and developments of mid to high-rise mixed-use buildings, associated noise sources are often found in close proximity to our living environment. In satisfying the need for increased population density by integrating various land uses within a mixed development zone, planners, designers and policy makers often find challenges in controlling noise in the built environment. Whilst environmental planners and the policy makers aim to reduce noise within the urban built environment by setting noise limits, acoustic consultants are typically tasked with providing cost-effective solutions to achieve the environmental noise limits with almost no tolerance. The resulting condition is "buildup" of ambient noise level cumulatively over the years though satisfying the environmental noise limits for individual development. As a result, noise annoyance is increasingly perceived in urban areas which needs to be addressed through appropriate design strategies to enhance our living environment to acoustically comfortable.
Acoustic comfort in the built urban environment, beyond traditional approach of noise control and management, has often received less attention in urban design. The quality of a "place" is highly influenced by our perception of sound in the surrounding environment. It is therefore of utmost importance that sounds in our built environment are perceived positively. In aspiring to achieve acoustic comfort, it is imperative that innovative design strategies are integrated into urban design to promote the positive aspects of sound, in addition to reducing noise level through traditional noise control techniques.
This paper presents a review of relevant literature on urban soundscape and discusses the theoretical background of acoustic comfort in outdoor urban environments. A number of strategies that can be incorporated into urban design to promote acoustic comfort are also proposed.
(1) SA EPA (2) GHD Pty Ltd
ABSTRACT
Modern instrumentation utilises internet connectivity and cloud based technology to allow for large scale and long term noise monitoring programs making them easier and simpler to perform. The data acquired during monitoring projects can be utilised for acoustic modelling of different areas to produce strategic noise maps for a wide range of applications. Accurate assessment of noise for large urban zones may require substantial computational resources or processing time. Acoustic modelling can be performed within a shorter time frame if spatial resolution for the area is reduced. This may be a detriment to the accuracy of the modelling but reduces the required computational resources.
This paper is intended to provide recommendations on modelling resolution that can be used for assessing noise in an urban environment. These recommendations are derived from modelling performed for large scale strategic noise monitoring programs. The paper details assessment of the modelling accuracy for different sized modelling grids in suburban and multi-story building environment. Results of the research can be used to assist in choosing acoustic modelling parameters for different urban areas.
Advitech Pty Ltd
ABSTRACT
Machine Learning is often cited as an effective means of analysing complex datasets. This paper presents a case study that considers the ability of Machine Learnt models to identify common sources of environmental and industrial noise in rural receiving environments. Preliminary assessment indicates that Machine Learnt models are at least as effective as a human listener at identifying certain sources of environmental noise. Fur-thermore, the results of source contribution assessment using classified models are consistent with manually derived contribution results. These classification frameworks may present a means for automatically screening extraneous noise contributions from environmental monitoring data.
(1) Ambient
ABSTRACT
Many levels of government have strategic planning schemes for the growth of cities and regional areas. Regarding transport noise, planning schemes often define zones to either restrict development, or to ensure new or upgraded developments are constructed in a manner that protects the inhabitants by limiting transportation noise intrusion.
The California Building Code defines external noise criteria, which are used to trigger the need for an acoustic assessment and potential building treatments to limit noise ingress. To assist the public in identifying which properties are impacted, local governments provide maps, developed using basic noise modelling, or simple set-back distances away from major transport routes. This approach ignores the value of terrain and topographical features in screening transport noise, potentially placing many properties incorrectly within impacted zones and adding unwarranted acoustic assessment costs.
Two road traffic noise models of the California Bay Area have been built. The first holds road sources only to replicate noise categories published by local authorities, while the second model adds buildings and terrain throughout the 21,000 km2 (13,000 mi2) area. A comparison between the two methods demonstrates an almost 50% reduction in the number of properties incorrectly zoned, when stratetic noise modelling is used. This removes the need for an acoustic assessment at these properties, thus lowering planning and/or construction costs within the region.
(1) Senior Environmental Specialist (2) Airport and Environmental Assurance Team Leader
ABSTRACT
The Integrated Noise Model (INM) was developed by the United States Federal Aviation Administration (FAA) and has been used extensively in Australia for the environmental assessment of aircraft noise, development of Australian Noise Exposure Forecast/Index maps (ANEF/ANEIs), and the generation of number above (Nxx) noise contour maps. The INM software was supported by the FAA until September 2016 and has since been replaced by the Aviation Environmental Design Tool (AEDT). To keep abreast with technology changes, the aviation industry in Australia must migrate away from the use of INM to AEDT. The continual use of the outdated software will eventually affect the accuracy of noise modelling and prediction of noise impacts around air-ports in Australia. This paper aims to informatively discuss the proposed use of AEDT in Australia and to specifically address Airservices' technical endorsement process of ANEFs using the new model. The paper presents preliminary work undertaken to validate the use of AEDT in Australian conditions with data extracted from Airservices' Noise and Flight Path Monitoring System (NFPMS), and presents comparative results of noise con-tours. The paper also aims to review airport noise modelling best practice in Australia and provides insight into modelling with AEDT.
(1) Southern University of Science and Technology SUSTech, Shenzhen, China, 518055
ABSTRACT
Acoustic beamforming is an experimental method that is used to locate and quantify acoustic sources and is becoming increasingly popular as a research tool in academia and a competitive edge for acoustic consultants. The design of the beamforming array depends on the allowable size of the array, the frequency range of interest, the expected distance of the acoustic sources to the array and the number of microphones. To date, logarithmic spiral arrays and their variations, such as the Underbrink array, have been widely used in commercial applications and academia, yet no array design has been conclusively proven to be optimal, bound by the aforementioned design factors. Furthermore, large arrays require many microphones and expensive data acquisition systems that may be cost prohibitive, yet a smaller, well-designed adjustable array may be a more attractive option.
A microphone array design method, namely the iterative microphone removal method, has been previously presented by the current authors that shows improved beamforming outputs relative to a range of logarithmic spiral and randomised array patterns with the same set of constraints. Based on minimum main lobe width and maximum sidelobe criteria, an array is designed such that microphones are removed from a large array stencil to arrive at a least-compromised array design.
This paper will present new arrays possessing 48-channels installed within a 169-channel stencil, designed using monopole sources at various locations, rather than at the scanning grid centre only. To date, the iterative microphone removal method has only been used in numerical simulations using monopole sources located at the centre of the beamforming scanning grid. An experimental verification of these new arrays will be presented. A small speaker emitting white noise will be moved across several scanning grid locations to determine the performance of the array designs. They will be compared with the expected numerical results of logarithmic spiral and randomised array patterns.
University of Technology Sydney
ABSTRACT
Accelerometers are traditionally used in acoustics for the measurement of vibration. Higher derivatives of motion are rarely considered when measuring vibration. This paper presents a novel usage of a triaxial accelerometer to measure acceleration and use these data to explain jerk and higher derivatives of motion. We are all familiar with the terms displacement, velocity and acceleration but few of us are familiar with the term jerk and even fewer of us are familiar with the terms snap and crackle. We experience velocity when we are displaced with respect to time and acceleration when we change our velocity. We do not feel velocity, but rather the change of velocity i.e. acceleration which is brought about by the force exerted on our body. Similarly, we feel jerk and higher derivatives when the force on our body experiences changes abruptly with respect to time. The results from a gymnastic trampolinist where the acceleration data were collected at 100 Hz using a device attached to the chest are presented and discussed. In particular the higher derivates of the Force total equation are extended and discussed, namely: jerk∙d3x/dt3, snap∙d4x/dt4, crackle∙d5x/dt5, pop∙d6x/dt6 etc.
(1) SLR Consulting Australia Pty Ltd
ABSTRACT
Accelerometers can be mounted to structures in a variety of ways: using studs, magnets or adhesives (including wax) are the most widely documented. It is well known that the chosen mounting method will effect the usable frequency range of the accelerometer. However, while accelerometer manufacturers specify the usable frequency range of accelerometers, these specifications usually reflect test results using stud mounting in idealised laboratory conditions. Specifications of the usable frequency range for magnetic mounting methods are limited and are often generic; typically only one magnet and accelerometer combination are presented with no commentary regarding the potential effects of the magnets size, strength, or mass relative to the mass of the accelerometer. This paper compares the frequency response of four different accelerometers, which have been mounted to a dynamic mass shaker using stud mounts, a magnet with a flat base and a dual rail magnet.
(1) College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, China, 310027 (2) Department of Mechanical Engineering, The University of Western Australia, 35 Stirling Highway, CRAWLEY, WA, 6009
ABSTRACT
In practice, a measured vibration signal is often mixed with the inherent thermal noise existing in the measurement system. It is difficult to use traditional frequency-domain filtering techniques as the desired vibration signal and the unwanted thermal noise may have components in the same frequency band. In this research, a dual-sensor vibration measurement system and a Kalman filter based on a linear prediction model are developed to reduce the thermal noise in the measured data. This paper presents a mathematical analysis of the linear-prediction-based Kalman filter and examines the effects of the prediction error and measurement error on the filtering performance. The results show that the linear-prediction-based Kalman filter can reduce the prediction error compared to the traditional random-walk model. The effect of unsteady measurement error on filtering performance is also investigated. A simulation example is used for illustration. The simulation result shows that the linear-prediction-based Kalman filter achieves a better anti-drift performance than the conventional low-pass filter, and the delay of the linear-prediction-based Kalman filter is smaller than that of the conventional low-pass filter.
(1) Wood and Grieve Engineers (2) Faculty of Architecture, Design and Planning, The University of Sydney
ABSTRACT
The annoyance caused by unpleasant low-frequency sound is known to be influenced by many factors, but studies of the influence of differences in such signals reaching a listener's two separate ears are not common. This is despite the fact that when such sound intrudes into indoor environments, the received binaurally incoherent low-frequency sound can exhibit values of Interaural Cross Correlation (IACC) that reduced well below that measured in the free field, and are associated with clearly audible differences relative to similar signals that are more binaurally coherent. In this study, a selection of three unpleasant low-frequency sounds were submitted to chaotic amplitude modulation with three different peak modulation rates (2 Hz, 5 Hz and 10 Hz), and preented as both coherent and incoherent signals via a laterally positioned pair of subwoofers. The results clearly revealed that binaural incoherence influences the unpleasantness of low frequency sound perceived by human listeners. The study also revealed that of the three peak modulation rates tested, the lowest rate (2 Hz) was associated with greater reported annoyance than that reported at the higher rate (5 Hz and 10 Hz).
University of Sydney and Defence Science and Technology Group
ABSTRACT
Ambient noise in the ocean has been studied for around 80 years and many hundreds of papers have been published. It is called noise in the context of the operation of sonar, since it is the main noise against which sonar signals must be detected, even though it has included many studies from mechanisms of sound generation to the use of sound by marine animals. Generally ambient noise refers to the background noise from all sources, excluding sounds from individual identifiable sources. Recently, there has been increasing reference to soundscapes for what appears to be much the same phenomenon. This has coincided with the development of interest by biologists in ambient noise, mainly in the context of the effects of noise on marine animals and their use of sound. Is soundscape simply a trendy new name for ambient noise or is it something different? Judging by what has been published on ocean soundscapes, it seems to be little more than a trendy new name. However, the term soundscapes has been widely used for much longer in many diverse applications from music, to architecture to terrestrial ecology and covers more than just ambient noise. This paper examines the wider use of the term soundscapes in other applications. It considers how these compare with ambient noise in the ocean and how ocean soundscapes might be defined to be more than simply a trendy name for ambient noise as well as being consistent with usage in other areas. The concept of ambient noise as the background noise against which signals of interest must be detected means that ambient noise is both unwanted sound and a necessary component of the sonar equation. This applies equally to our sonars and to the use of sound by marine animals. Soundscape is a broader concept, including all sounds of the environment, including not just the ambient noise but also signals of individual sources, whether the signals of interest to the sonar operator or to a marine animal, or signals that interfere with the capability of the sonar or the animal to make use of sound. For the ocean context, probably the most useful analogy is with soundscapes in terrestrial ecology.
retired
ABSTRACT
Piledriving involves a ram striking an anvil, below which is a cushion, helmet, and pile. For a steel pile, the cushion is usually an assembly of Micarta and aluminium layers. Following impact, the waveform of the cushion strain and hence the stress on the pile head are derived by solving the nonlinear equations of cushion motion using the Runga-Kutta method, based on two reported compressive stress-strain curves (CSSC) for Micarta: one for an annular cushion and the other for a solid disc. The vibration of the immersed pile and sound radiated into the water are modelled theoretically. The model is applied to an acoustic measurement described in the literature, and the initial positive and negative peaks of the predicted pressure waveforms are compared with data measured at nine depths. When the cushion is neglected, the errors over the nine depths are -1 to 5 dB for the positive peaks and -1 to 9 dB for the negative peaks. For the annular cushion those errors are -7 to -5 dB for the positive peaks and -5 to -1 dB for the negative peaks. The corresponding errors for the solid cushion are only -2 to 0 dB and -1 to 3 dB.
(1) School of Mechanical and Chemical Engineering, The University of Western Australia (2) Department of Instrument Science and Engineering, Zhejiang University
ABSTRACT
Correct and efficient vibration model of the laminated structure is an cornerstone for predicting noise and vibration from power transformers. The complexity in the transformer core, as laminated structure, makes the modelling of the core vibration difficult. In this paper, the viscosity and friction between two adjacent layers of the laminated beam are included in the derivation of the bending moment of the transverse vibration of the beam. As a result, equivalent Young's modulus and damping ratio of the beam are obtained as functions of the visco-elastic constant and friction factor between the layers, and used to determine the frequency response functions of laminated beams. The modelled and measured frequency response functions of laminated beams are compared and in good agreement. The effect of the clamping force and bonding treatment between the layers are also discussed.
(1) School of Mechanical Engineering, University of Adelaide, South Australia 5005, Australia (2) Department of Aeronautics, Imperial College London, London SW7 2AZ, UK
ABSTRACT
In the present article, a modified nonlinear continuum model is presented for the large-amplitude vibration of ultrasmall tubes taking into account the effect of a geometric imperfection. The higher-order size-dependent theoretical model is developed via help of a nonlocal strain gradient theory of elasticity in conjunction with the Euler-Bernoulli beam theory. The roles of the stress nonlocality together with the effects of the strain gradient are modelled via the developed nonlocal strain gradient model. The presented scale-dependent model is able to better capture the effect of the size of ultrasmall tubes on the vibration behaviour. According to the work/energy principle, size-dependent relations are presented for the elastic energy and motion energy of the micro/nanoscale tube as well as the work performed by external forces. Using these relations, the size-dependent differential equations of the described problem are then derived. Lastly, the Galerkin decomposition scheme and the continuation-based scheme are employed to extract the nonlinear vibration characteristics of nonlocal strain gradient ultrasmall tubes with a geometric imperfection. In addition to the effect of the imperfection, the influence of stress nonlocality as well as strain gradients on the nonlinear vibration of ultrasmall tubes are investigated.
(1) School of Mechanical Engineering, University of Adelaide, South Australia 5005, Australia (2) Department of Aeronautics, Imperial College London, London SW7 2AZ, UK
ABSTRACT
In the present paper, the role of the flow pulsation in the scale-dependent complex vibration of ultrasmall tubes under the action of an axial pretention is investigated. The main focus of the present study is on the chaotic vibration response of the ultrasmall system. The effect of the internal energy friction is also considered via help of the Kelvin-Voigt model. In practical applications, tubes and pipes are usually surrounded by an elastic foundation. Therefore, in the current study, it is assumed that the ultrasmall tube is embedded in a nonlinear elastic medium. The theory of couple stress as well as the theory of an Euler-Bernoulli beam are utilised to develop a modified scale-dependent model. An energy/work principle is also utilised for the derivation of the energy potential, motion energy and the external work as well as the equations of motions. The influences of displacement components along both the length and thickness of the ultrasmall tube on the complex vibration are considered. Employing Galerkin's procedure gives the discretised version of the differential equations of the pulsatile fluid-conveying ultrasmall tube embedded in a nonlinear elastic foundation. Finally, application of a time-integration scheme gives the nonlinear characteristics of the complex vibration.
(1) SoundBASE Consulting Engineers (2) The University of Queensland (3) ACRAN
ABSTRACT
Acoustic louvres are regularly used to control noise emissions from mechanical plant and equipment rooms, building services and industrial equipment. The sound insulation performance of acoustic louvres is quantified in a number of different ways by louvre manufacturers. While performance indicators such as insertion loss, static transmission loss, weighted sound reduction index and noise reduction are frequently used by manufacturers, do they accurately represent the in-situ sound insulation performance of acoustic louvres? In this paper a critical review of standard and non-standard test methodologies is presented. An overview of each test methodology is presented with a discussion of the advantages and disadvantages of each method. A comparison is also made of the test method results to determine whether any of them are able to accurately quantify the sound insulation performance of acoustic louvres.
(1) The University of Adelaide
ABSTRACT
This paper presents a theoretical analysis of a planar quasi-zero stiffness magnetic levitation system with improved rotational passive stability for the application of vibration isolation. Through proper choices of equilibrium position and magnet geometry (horizontal gap parameter), the system is made stable in the vertical and rotational degrees of freedom, thus requiring active stability control only in the horizontal degree of freedom. It is found that the control effort is inversely proportional to the magnitude of the rotational passive stability. The effect of these parameters on the stability and vibration response of the system is examined. The nonlinearity of the system is also investigated by comparing the vibration response to its linear representation, in which it is shown that the system becomes increasingly nonlinear as it is perturbed further away from its equilibrium position.
(1) Naval Group Research, FRANCE & School of Mechanical and Manufacturing Engineering, UNSW, AUSTRALIA (2) Naval Group Research, FRANCE (3) IEMN, ISEN Department, UMR 8520 CNRS, FRANCE (4) School of Mechanical and Manufacturing Engineering, UNSW
ABSTRACT
Metamaterials can be engineered for noise and vibration control due to their ability to manipulate waves. A common metamaterial design comprises periodic inclusions embedded in a host medium, for example, as a viscoelastic coating applied to the outer hull of a marine vessel. Depending on geometrical arrangement and on the physical properties of the materials, band gaps may appear in such structures, i.e. waves are evanescent. Mechanisms governing the generation of band gaps arise from destructive wave interference associated with periodicity of the scatterers as well as local resonances of the scatterers. In order to study the features of a metamaterial, the effective parameters of an equivalent homogeneous medium are commonly derived. This work presents a comparison between different inverse approaches to derive the frequency-dependent effective properties of a metamaterial, corresponding to the effective complex wavenumber, effective speed of sound and effective density. These effective properties are shown to strongly vary with frequency ranges associated with the band gaps in which wave propagation does not occur.
(1) Department of Instrument Science and Engineering, Zhejiang University, Hangzhou, 310027, China (2) Department of Mechanical Engineering, The University of Western Australia, Perth, WA 6009, Australia
ABSTRACT
Winding vibration is one of the most important noise sources in loaded power transformers. However, accurate and effective modelling of the winding vibration response is often hindered by the complicated conductor and insulation material distributions in the winding structure. In this paper, the winding is described by a coupled-ring model, which are applied on three experimental winding structures. The vibration response of the winding to a point force excitation is then analysed using the finite element method. This simplified model allows efficient prediction and understanding of the vibration characteristics of complex winding structures.
Naval Group Research
ABSTRACT
Acoustic radiated noise in water is an important requirement for some types of ships, and one of the main tasks in the design process is to specify the vibratory levels of each noisy equipment item in order to comply with the overall objective. The design process generally uses predictive models based on transfer function chains, with the assumption that there is a weak coupling between the different stages of the chain. That assumption is relevant in the classical case where a machinery item is elastically mounted, but no more when it is rigidly mounted on an intermediate structure. Here, there is a strong coupling between the vibratory source and the supporting structure. Starting from the characteristics of both decoupled subsystem, the method of mobility is used to describe exactly the vibratory response of the whole assembly. Whereas the method in the direct way is already known, it is not straightforward to apply in practice for naval structures in the inverse way. Therefore, this paper will focus on the derivation of the inverse method, and a numerical case study will be presented.
Pyrotek
ABSTRACT
Rainfall on metal roofing can often result in undesired noise, caused by raindrops impacting the highly resonant metal sheeting. An investigation was conducted on the effectiveness of applying Pyrotek Decidamp® SP80, a sprayable water-based damping compound, to the underside of corrugated metal roofing in order to reduce rainfall noise. A common insulation material typically used for reducing rainfall noise was also tested in combination with the damping material. Testing was conducted to ISO 10140-5 using heavy rainfall on standard residential corrugated metal roofing, along with thicker industrial corrugated metal roofing. Testing was also conducted to ISO 10140-2 in order to measure the improvement in transmission loss. The test results revealed that the damping compound was highly effective in reducing rainfall noise and increasing transmission loss on metal roofing, concluding that it could be a viable solution on its own or as an element of a more comprehensive system.
(1) College of Power and Energy Engineering, Harbin Engineering University, 145 Nantong Street, Harbin, Heilongjiang 150001, China (2) School of Mechanical and Chemical Engineering, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia
ABSTRACT
In this paper, a nonminimum-phase secondary path measured inside a car is analysed and its influence on noise reduction is discussed. To improve control performance, a new nonlinear algorithm using a configuration with a bilinear filter and trigonometric expansions is proposed for a control system with a nonminimum-phase secondary path. Within the controller of an adaptive feedforward control system, the proposed algorithm achieves high-order nonlinearities and has a better ability to predict a non-Gaussian signal. The application of this algorithm in re-ducing the interior road noise of a car is illustrated through simulations. Since road noise is complicated, six acceleration signals from a car's body are adopted as reference signals and one loudspeaker is used as an actuator to reduce noise near one microphone. The simulation results show that the proposed algorithm achieves better results than a traditional linear algorithm.
(1) The HEARing CRC, University of Melbourne (2) Murdoch University (3) The HEARing CRC, National Acoustic Laboratories
ABSTRACT
Risks to the hearing health of patrons and staff in the music industry have been well documented and in recent years, regulations have been introduced in a number of countries in Europe as a means of limiting sound exposure for attendees at live music events such as gigs, concerts and festivals. In Australia, sound levels at live music venues are generally stipulated in planning permits and/or liquor licenses, but are largely focussed on preventing the emission of sound to neighbouring premises rather than reducing risks to hearing of patrons or staff. In this study, we investigated the use of a sound level management system (10EaZy) as a way of reducing exposure levels for patrons and staff, without interfering with patrons' enjoyment of the musical performances. More than 200 sound level measurements were taken in six inner-city Melbourne venues during live music performances before and after installation of the 10EaZy system at each venue. Measurements from the before- and after-periods were compared in terms of overall level (LAeq) and also the proportion of time spent at high volumes. Preliminary results suggest that use of 10EaZy resulted in significantly less time spent at higher volumes, particularly for loud performances. Results of subjective questionnaires indicated that patron satisfaction was unaffected when the 10EaZy system was in use.
(1) Isolgomma
ABSTRACT
The isolation of vibrations in the railway industry nowadays plays a primary role in the design, construction and testing of rail track system for city application. Over the years the awareness of the importance of a design that considers and protects the structures that add to the railway line from the effect of the vibrations caused by the transit of the trains has matured. Isolgomma has long been engaged in the study of technical solutions for vi-bration isolation in various fields of application including the railway sector. The experience gained has allowed us to produce and propose this article the case study of tramway of Palermo with the presentation of the de-sign process and the evaluation of in situ measurements.
(1) WSP
ABSTRACT
The Level Crossing Removal Project - Caulfield to Dandenong (LCRPCTD) was established to remove nine level crossings along the Cranbourne / Pakenham railway line in the south-eastern Suburbs of Melbourne. As part of the management of construction noise, WSP assessed the noise impacts from specialised construction techniques required to install bridge spans in a narrow rail corridor that backed onto residential properties on both sides.
Traditional methods of bridge span installation were not feasible due to the proximity of residential dwellings and lack of space immediately outside the corridor. The viaduct segments were therefore delivered to a central compound in the Melbourne suburb of Murrumbeena, where the segments were assembled to form the bridge spans. The equipment consisted of large gantry cranes in the central compound which acted as a home base for the Straddle Carrier, a large transportation rig that anchored itself to the bridge piers and placed the bridge spans into place without the need for ground clearance. It was the first time this machinery has been used in residential Australia.
Straddle Carriers are seldom used for construction, and have limited documentation on the associated noise emissions - the information was insufficient for any predictive noise modelling exercise, therefore requiring detailed measurements of the equipment.
This paper provides insight on the construction activities carried out within the compound, and the use of predictive modelling, and real-time noise monitoring to influence construction methods and scheduling to actively manage noise impacts. More importantly, the paper describes the construction technique using this Straddle Carrier, along with the details of the carrier components that emit noise. The paper also provides the noise measurement techniques for the Straddle Carrier adopted in the near / far field to establish data suitable for future predictive modelling. It was found that the Straddle Carrier's Hydraulic Unit, already housed in an acoustic enclosure and fitted with silencers was the dominant noise emitting plant.
The paper establishes Sound Power Levels for the Straddle Carrier derived from measurements to enable the development of predictive models and prediction of noise impacts. The model was calibrated by +4 dB to align with the actual measurements taken at the boundary of the residential properties, taking suspected diffraction, shielding, and distance losses into account.
The paper elaborates a case study with the use of field measurements, real-time monitoring and predictive modelling of unique construction techniques in a challenging environment to successfully manage noise effects.
(1) Griffith University (2) RIVM, The Netherlands
ABSTRACT
This paper reports the final outcomes of a systematic review of literature (1980 -2014) of changes in health re-sulting from transport noise control measures. This was part of the evidence-base for the planned revision of the World Health Organization's Environmental Noise Guidelines for the European Region. While there are many studies of transport noise interventions which report a consequent change in emissions, or change in noise lev-els near sources, only a limited number have reported resultant health outcomes of receivers. These health out-comes were mostly of annoyance, with a few examining sleep disturbance, cardiovascular effects, and cogni-tive development in children. The majority of studies included were of road traffic noise; fewer for aircraft noise and rail traffic noise. Because of diversity in the studies, a meta-analysis of association between changes in level and changes in outcome was not feasible. However, it was possible to conclude, though within a limited evidence base, that transport noise interventions do change the health outcomes reported by those who expe-rience the intervention, irrespective of the source type, the outcome, or the intervention type. The minimum magnitude of the change in annoyance outcomes can be predicted from the interventions using a relevant ex-posure-response function.
(1) Renzo Tonin & Associates NSW Pty Ltd (2) Roads and Maritime Services
ABSTRACT
RONDA (ROad Noise Data Acquirer) is a CPX trailer conforming to ISO/CD 11819-2 intended for measuring tyre-pavement noise. The trailer is of the open frame type without an enclosure. Tyre-pavement noise measurements along the recently constructed Hunter Expressway were conducted using the RONDA CPX trailer to quantify noise levels for various pavement types along the expressway, which include a variety of concrete and asphalt pavement types. Results from measurements were analysed to determine the tyre-pavement noise levels for each specific pavement type.
WSP
ABSTRACT
New Zealand's longest road tunnel was opened to traffic in July 2017. At a length of 2.4km, Waterview Tunnel completes the Western Ring Route, an alternative north-south motorway link across Auckland. The tunnel includes multiple technical systems to manage public safety, including a deluge fire protection system, tunnel ventilation and jet fans, emergency lighting, emergency telephones, variable message signage, CCTV cameras, radio rebroadcast system, and a public address (PA) system. Initial tuning and commissioning measurements of the PA system raised concerns that the system may not meet design requirements for speech intelligibility in the main tunnel bores. A review of the installed PA was conducted, and options considered for improvement. This paper covers the review process, options considered, testing, and resultant performance of the PA system after implementation of upgrades and re-tuning. Some general recommendations for PA system design and commissioning are presented, to inform future implementation of systems in similarly challenging environments.
(1) UWA (2) DST (3) DST/UWA
ABSTRACT
The marine environment consists of many different sound sources covering a wide frequency range. Accurately identifying and analysing these sound sources is difficult and time consuming. This is complicated by effects such as variable ambient noise, multi-pathing and multiple sources. One promising technique for analysing such complex data sets is machine learning. This has been successfully used in many other applications. In this paper we will use it to detect snapping shrimp impulses. These are a dominant noise source in shallow tropical waters and ideal for testing new algorithms. The logistics regression method is used as the main algorithm. A snapping shrimp acoustics matrix (SSAM) is constructed from features such as the band energy ratio, frequency centroid, spectrum flatness, etc. It has been ensured that the extraction speed of the SSAM is sufficiently fast such that it is suitable for real time processing. A number of data sets for different locations covering a range of conditions will be analysed and compared.
University of Technology Sydney
ABSTRACT
Accelerometers are traditionally used in acoustics for the measurement of vibration. Higher derivatives of motion are rarely considered when measuring vibration. This paper presents a novel usage of a triaxial accelerometer to measure acceleration and use these data to explain jerk and higher derivatives of motion. We are all familiar with the terms displacement, velocity and acceleration but few of us are familiar with the term jerk and even fewer of us are familiar with the terms snap and crackle. We experience velocity when we are displaced with respect to time and acceleration when we change our velocity. We do not feel velocity, but rather the change of velocity i.e. acceleration which is brought about by the force exerted on our body. Similarly, we feel jerk and higher derivatives when the force on our body experiences changes abruptly with respect to time. The results from a gymnastic trampolinist where the acceleration data were collected at 100 Hz using a device attached to the chest are presented and discussed. In particular the higher derivates of the Force total equation are extended and discussed, namely: jerk∙d3x/dt3, snap∙d4x/dt4, crackle∙d5x/dt5, pop∙d6x/dt6 etc.
(1) The University of Adelaide
ABSTRACT
The responses of birds to overflying remotely piloted aircraft (or drones) are poorly understood. While direct observations of behaviour can provide an understanding of wildlife response, this approach is unable to quantify the potential physiological responses of individuals. To do this, a dummy bird egg containing a microphone was substituted into the nests of incubating adult seabirds. Audio recordings were made to capture the sounds of the birds' heart-beat. The recordings were used to calculate a resting heart rate, which was to be compared with the heart rate when a remotely piloted aircraft flew overhead. The focus of this paper is the signal processing methods that were evaluated to process the audio signals to extract the instantaneous heart rate of the animal.
Dr
ABSTRACT
Birds are an important group of animal that ecologist monitor using autonomous recordings units as an crucial indicator of health of an environment. There is not yet an adequate method for automated bird call recognition in acoustic recordings due to high variations in bird calls and the challenges associated with bird call recogni-tion. In this paper, we use ResNet-50, a deep convolutional neural network architecture for automated bird call recognition. We used a publicly available dataset consisting of calls from 46 different bird species. Spectro-grams (visual features) extracted from the bird calls were used as input for ResNet-50. We were able to achieve 60%-72% accuracy of bird call recognition using ResNet-50.
(1) University of Adelaide (2) DST Group
ABSTRACT
Rolling element bearings eventually become worn and fail by developing surface defects, such as spalls, dents, and pits. Previous researchers have tested bearings with defects that have sharp 90° rectangular edges that were used to develop analytical models of a defective bearing and defect size estimation methods. These models have limitations that require smooth surfaces and constant curvature of the bearing components; as well as assuming the defect profile. An analytical model has been developed for a rolling element bearing that uses a measured defect profile and removes the limitations of previous analytical models that use analytical expressions for contact area and force. The predicted vibration response of a bearing with a defect on the outer raceway was compared with experimental results. It was found that the new analytical model was able to predict the vibration response of a defective bearing. Current defect size estimation methods that use time-series data to estimate the size; these methods do have an aliasing issue when the defect is larger than the separation angle of the rolling elements. In this paper, a method for determining if the length of a spall defect is greater than the separation angle of the rolling elements using the varying stiffness of the bearing assembly is presented. The developed model and experimental data have been made publicly available.
Defence Science & Technology Group
ABSTRACT
Acoustic sensors can be used to localize moving sound sources. The trajectory of a sound source moving in air or under water at constant velocity along a horizontal linear path is fully specified by a set of five motion parameters: velocity, altitude or depth, time of closest point of approach (CPA) to a sensor, horizontal range at CPA, and azimuth angle at CPA. A single (ground-based) microphone is able to estimate the speed, CPA time, and CPA slant range of a narrowband airborne (sound) source in transit using instantaneous frequency (IF) measurements from the sensor. A single microphone is also able to estimate the speed, CPA time, and altitude of a broadband airborne source in transit using multipath delay measurements from the sensor. The localization capability is enhanced when two microphones are used instead of one. By exploiting the inter-sensor multipath delays, a pair of microphones is able to estimate the speed, altitude, CPA time, and CPA horizontal range of a broadband airborne source in transit. If the left-right ambiguity can be resolved by some means or if it is known a priori on which side (left or right) of the sensor pair the source's CPA position is located, then the entire set of source motion parameters can be estimated. The same technique can be applied to the underwater domain where a pair of hydrophones is used to localize a broadband underwater (or surface) sound source. The first part of the paper describes this source motion parameter estimation method and evaluates its performances in the air domain using real data recorded from a ground-based microphone array in a field experiment for ten different transits of a jet aircraft, and in the underwater domain using real data recorded from a bottom-mounted hydrophone array in a shallow water experiment for four different transits of a small vessel and a single transit of a rigid-hulled inflatable boat. The second part of this paper considers estimating the entire set of motion parameters of a transiting broadband airborne sound source that contains a narrowband tone using both IF and time delay measurements from a pair of underwater acoustic sensors. The source motion parameter estimation method is described and its performance is evaluated using real data recorded from a towed hydrophone array in a deep water experiment for four different transits of a turbo-prop aircraft.
UNSW Sydney
ABSTRACT
Vibration-based machine diagnostics extracts information about machine health on the basis of response signals, which are always a combination of forcing function and transfer function effects. In contrast to structural health monitoring, where information on health is conveyed almost entirely by changes in the structural dynamic (ie transfer function) properties, and where forcing functions (eg from wind, waves, traffic etc) contain almost no information about health, almost the opposite is the case for machine faults. Probably about 70% of fault features from various components such as gears, bearings, etc, are obtained from changes in the forcing functions, though certain faults, such as growth of a crack in a machine casing, might still be indicated by changes in the modal properties. This paper discusses the ways in which blind separation of forcing function and transfer function effects can be achieved in machine response signals, so as to indicate the most likely faults. This is particularly important for variable speed machines, such as wind turbines, where many forcing functions, such as gearmesh frequencies, vary directly with the speed, but responses are affected by passage through fixed frequency resonances. On the other hand, bearing fault signals are often carried by fixed frequency resonances, but the spacing of the impulse responses corresponding to the faults varies with the speed. Ideally, all machine vibration signals should be divided into their forcing and transfer function components, and this may be essential in certain complicated situations. Operational modal analysis, or determination of the modal properties of structures in operation, has developed rapidly in recent years, and is widely used in structural health monitoring of bridges, buildings, wind turbine structures etc, but the effects of ambient excitation are then more easily dealt with or discarded. Recent developments in cepstral methods give much promise for achieving a better separation of forcing function and modal properties in machine signals, with both being retained, even in the case of complicated forcing functions more typical of machines such as wind turbine drive trains as opposed to their structures. The separation of forcing and transfer functions at different speeds should also have application in the design and development of machines, for example at the prototype stage, allowing updating of analytical simulation models.
WSP
ABSTRACT
New Zealand's longest road tunnel was opened to traffic in July 2017. At a length of 2.4km, Waterview Tunnel completes the Western Ring Route, an alternative north-south motorway link across Auckland. The tunnel includes multiple technical systems to manage public safety, including a deluge fire protection system, tunnel ventilation and jet fans, emergency lighting, emergency telephones, variable message signage, CCTV cameras, radio rebroadcast system, and a public address (PA) system. Initial tuning and commissioning measurements of the PA system raised concerns that the system may not meet design requirements for speech intelligibility in the main tunnel bores. A review of the installed PA was conducted, and options considered for improvement. This paper covers the review process, options considered, testing, and resultant performance of the PA system after implementation of upgrades and re-tuning. Some general recommendations for PA system design and commissioning are presented, to inform future implementation of systems in similarly challenging environments.
WSP
ABSTRACT
New Zealand's longest road tunnel was opened to traffic in July 2017. At a length of 2.4km, Waterview Tunnel completes the Western Ring Route, an alternative north-south motorway link across Auckland. The tunnel includes multiple technical systems to manage public safety, including a deluge fire protection system, tunnel ventilation and jet fans, emergency lighting, emergency telephones, variable message signage, CCTV cameras, radio rebroadcast system, and a public address (PA) system. Initial tuning and commissioning measurements of the PA system raised concerns that the system may not meet design requirements for speech intelligibility in the main tunnel bores. A review of the installed PA was conducted, and options considered for improvement. This paper covers the review process, options considered, testing, and resultant performance of the PA system after implementation of upgrades and re-tuning. Some general recommendations for PA system design and commissioning are presented, to inform future implementation of systems in similarly challenging environments.
Tampere University of Technology
ABSTRACT
Sound carries lots of information about the environment, which can be automatically extracted by computational analysis methods. Modern machine learning methods enable automatic pattern recognition of sounds, by for example detecting when specific target sounds of interest are active. This has several applications, for example in acoustic surveillance and monitoring, context-aware devices, and multimedia information retrieval.
This talk will given an overview of the general methodology for environmental sound recognition. We will first discuss technical and scientific problems related to the task. Since the methods are heavily based on machine learning which aims automatically obtaining models for sounds from training data consisting of audio recordings and metadata about sources, we will discuss different ways to obtain such data, and how they affect the performance of the methods.
We will then describe general machine learning methodology based on deep neural networks. We present a convolutional recurrent neural network based approach that has been shown to produce state of the of art results in a wide range of different sound event detection tasks. The use of convolutional layers allows learning automatically a suitable high-level representation to recognize different types of sound sources, and the use of recurrent layers enables modeling long-term temporal context which is required for reliable sound recognition.
We present results from various evaluations, also including the recent DCASE (Detection and Classification of Acoustic Scenes and Events) public evaluation campaign. We will present a case study of environmental noise monitoring, where we detected continuously the activity of a rock crushing station based on sound over the duration of 50 days. We will also analyze and discuss the performance of humans vs. machines in environmental sound recognition.
(1) School of Engineering, RMIT University (2) Centre for Audio, Acoustics and Vibration, Faculty of Engineering and Information (3) Sound Intuition Pty Ltd, Melbourne, Australia
ABSTRACT
Gas Metal Arc Welding (GMAW) is an arc welding process that forms an electric arc between a consumable electrode and the base metal with a shielding gas to protect the arc. In GMAW, there are various metal transfer modes such as the short circuit mode, the globular, spray mode, and the rotational transfer mode, which show different arc stabilities, weld pool penetrations and spatter production. Identifying the metal transfer modes is critical for process monitoring and quality control of GMAW. In this paper, a method for metal transfer mode identification from the welding sound is presented. A microphone mounted on the welding helmet is used to record the sound signals generated by GMAW under different metal transfer modes, which are analysed in both time and frequency domains. New psychoacoustic parameters based on the subjective perception of an expert welder are extracted to distinguish the metal transfer modes. The Gaussian Mixture Model (GMM) classifier is utilised to identify the metal transfer mode from welding sound signals and a 10-fold cross validation results show that the classification accuracy is above 90 percent. This work is the first systematic study to identify metal transfer modes from welding sound signals, which is advantageous to the conventional plasma theories because it is non-invasive, robust and easy to use.
(1) Defence Science and Technology, Australia (2) Department of Mechanical Engineering, University of Western Australia, WA, Australia
ABSTRACT
The idea of applying the microphone array technology for machine condition monitoring of a mine site has been investigated at the University of Western Australia. As a key part of the investigation, this paper focuses on the sound source localization and signal extraction using the beamforming technique. Instead of a single large array, multiple small arrays are used, as practically, they can be easily deployed. In terms of array processing, they are normally processed as independent arrays. However, with synchronized sampling they can also be processed as a single combined array. These two processing approaches are termed the independent array approach (IAA) and the synchronized array approach (SAA), respectively. The main advantage of SAA is its greater array gain and higher beam resolution. However, the arrays need to be placed close to one another to avoid producing aliasing lobes. The IAA on the other hand has no such restrictions and the arrays can be placed independently in any position. This may provide gains in signal to interference and noise ratio (SINR) for individual sound sources in certain circumstances. In order to compare the performance of the two approaches under various conditions, numerical simulation is carried out. The performances are evaluated in terms of the accuracy of source localization and the fidelity of signal extraction. The general guideline for using the two approaches is provided.
(1) UWA (2) DST (3) DST/UWA
ABSTRACT
The marine environment consists of many different sound sources covering a wide frequency range. Accurately identifying and analysing these sound sources is difficult and time consuming. This is complicated by effects such as variable ambient noise, multi-pathing and multiple sources. One promising technique for analysing such complex data sets is machine learning. This has been successfully used in many other applications. In this paper we will use it to detect snapping shrimp impulses. These are a dominant noise source in shallow tropical waters and ideal for testing new algorithms. The logistics regression method is used as the main algorithm. A snapping shrimp acoustics matrix (SSAM) is constructed from features such as the band energy ratio, frequency centroid, spectrum flatness, etc. It has been ensured that the extraction speed of the SSAM is sufficiently fast such that it is suitable for real time processing. A number of data sets for different locations covering a range of conditions will be analysed and compared.
(1) CMST, Curtin University (2) DST Group
ABSTRACT
Naval operations rely heavily on a knowledge of the undersea environment, particularly for predicting ship and submarine sonar performance, a requirement of tactical planning. This is particularly challenging in regions with highly dynamic oceanography, such as Australia's Northwest Shelf, home to much of the country's offshore oil and gas industry, and therefore an area of considerable strategic importance.
This paper presents the results of fully 3D modelling of acoustic propagation parallel to the crests of nonlinear internal waves at a frequency of 7 kHz. Measured data from the Integrated Marine Observing System was used to derive a three-dimensional sound velocity field that was moved past the acoustic propagation path at a speed typical of nonlinear internal wave group velocities in this area. Acoustic propagation modelling was carried out using the latest version of Bellhop3D, a Gaussian beam tracing model.
The results indicate that during periods of high internal wave activity, changes in incoherent transmission loss of tens of dB could occur at a range of 15 km over a time scale of approximately ten minutes. This paper presents more detailed results and discusses the advantages and limitations of this type of modelling.
(1) National Centre for Maritime Engineering and Hydrodynamics, Australian Maritime College, University of Tasmania, Launceston, TAS 7250, Australia (2) Maritime Division, Defence Science and Technology Group, Melbourne, VIC 3207, Australia
ABSTRACT
Accurate prediction of unsteady propeller loads for propellers operating in non-uniform flow (wake of a ship hull) is vital for assessing the suitability of candidate propellers in the context of noise and onboard vibration. This requires an assessment of the flow around the hull, the response of the propeller to the incident flow and the propeller interactions with the flow field and other appendages/surfaces. This paper examines a subset of these by evaluating the suitability of several computational codes for assessing the unsteady response of a model propeller to wake inhomogeneity. Unsteady thrust and torque predictions from the Boundary Element Method (BEM) code PROCAL and Unsteady-RANS solver Star-CCM+ are compared with a prediction from the unsteady lifting surface theory code PUF-2 and experimental measurements carried out at the David Taylor Research Centre (DTRC). To reproduce the experimental conditions with URANS, a consistent methodology for representing the wake field is introduced. It is found that the BEM method can provide predictions with accuracy close to that of the URANS calculations for low frequency harmonics.
(1) Defence Science and Technology Group
ABSTRACT
The authors previously simplified their analytically complex model "JBZ" of coherent acoustic reflection loss at the ocean surface, and demonstrated that that simplified JBZ model gave loss results similar to those obtained from Monte Carlo parabolic equation (PE) modelling of equivalent scenarios. These models obtained a surface roughness loss inclusive of the refractive effects of a uniformly stratified distribution of near-surface, wind-driven bubbles. Further approximation of the simplified JBZ model has since been carried out. The approximated model, which is the subject of this paper, combines descriptions of roughness loss and bubble-refraction described for the earlier model versions, however, by restricting the ranges of wind speeds, frequencies and grazing angles, achieves an even simpler form. It is now shown that, for a limited, but useful, range of scenarios, for grazing angles as much as about 10 degrees in some cases, the coherent reflection loss at the wind-driven ocean surface, in dB, is well approximated by a function which is linear in grazing angle, where that angle is taken as the angle of acoustic incidence at the bottom of the bubbly region.
(1) University of Technology Sydney (2) Brunel University London
ABSTRACT
Modelling the propagation of sound waves in the ocean is challenging because one must account for spatial variation in the fluid properties and the ocean geometry, as well as couple the fluid to a seabed that supports both shear and compressional waves. To simplify this problem it is convenient to treat the ocean as a uniform waveguide in the axial direction and then expand the acoustic pressure over a set of normal eigenmodes. This forms the basis the normal mode approaches that typically use analytic methods to solve the governing dispersion relations, see for example the well-known ORCA algorithm. However, analytic methods find it difficult to accommodate depth dependent properties and it is common to partition up the ocean into multiple uniform waveguides, and then to use certain assumptions about the way in which the waveguides interact with one another. This inevitably introduces approximations and becomes very complex as the number of propagating modes increases. Furthermore, coupling the fluid to a physically correct representation of the ocean floor is very challenging for an analytic approach. This article presents an alternative approach that is based on the finite element method. It is shown that finite elements can be used to solve the normal mode problem using a fast and efficient approach that does not suffer from the traditional problems associated with the finite element method, such as excessive problem size associated with discretising the entire ocean. The advantage of using a numerical approach is that one may then properly account for depth dependent properties and, crucially, the fluid can readily be coupled to a full elastodynamic representation of the seabed, which supports both shear and compressional waves and permits the implementation of the physically correct transverse boundary conditions. This paper introduces this finite element based approach and then provides predictions of sound pressure level for a point source located at an arbitrary depth in the ocean. The solution time for the finite element approach is discussed and it is demonstrated that axial sound pressure distributions for the waveguide can be computed relatively quickly using modest computational resources.
DST Group
ABSTRACT
In various sonar applications it is important to model the acoustic response of a bubble cloud to a sonar pulse of different types. Previously, a high-fidelity Monte Carlo model of scattering of the acoustic pulse from a bubble cloud was developed in the single scattering approximation [1]. It is of interest to know how well the single scattering approximation is applicable to practical cases of bubble clouds generated by surface ship turbulent wakes, breaking waves, underwater explosions etc. In the current paper we describe the implementation of the acoustic multiple scattering between bubbles in a bubble cloud to model the cloud response to an active sonar pulse. The implementation is based on the self-consistent approach of Foldy [2]. The bubble cloud is modelled as an aggregation of identical omni-directional point scatterers, which is a good approximation in the case of small bubbles. The dependence of the multiple scattering effect on bubble volume fraction and sonar frequency will be presented in the paper.
1. Kouzoubov, A. Monte Carlo model of acoustic response from a bubble cloud. in ACOUSTICS 2017. 2017. Perth, Australia: Australian Acoustical Society.
2. Foldy, L.L., The multiple scattering of waves. I. General theory of isotropic scattering by randomly distributed scatterers. Physical Review, 1945. 67: p. 107-119.
(1) Maritime Division, Defence Science and Technology Group, Victoria, Australia
ABSTRACT
This paper models and analyses the structural behaviour and radiated sound due to arbitrary force excitations of a cylindrical enclosure totally submerged in a heavy fluid. The enclosure consists of a cylindrical shell with internal bulkheads and ring stiffeners. The cylindrical shell is closed by circular plates. The structural responses of the shell and plates are calculated by solving the shell and plate equations of motions using a wave approach. The far-field radiated sound pressure from the shell and plates is then calculated by using Helmholtz integral. The effect on the vibration and far-field sound pressure of the whole enclosure resulting from the arbitrary force excitations is investigated. The influence of the bulkheads and ring stiffeners on the structure is discussed. Good agreement is obtained between the analytical results and results from numerical finite element / boundary element models.
(1) DST Group, Australia
ABSTRACT
There are currently over a billion miniature microphones produced every year to feed the global market for smartphones. There is a growing trend to develop smartphones with water-resistant capabilities. One of the significant points of weakness for a water-resistant smartphone is the microphone. A capacitive microphone can be very susceptible to both dust and water egress. This has motivated the development of cheap piezo MEMS microphones, which are less susceptible to dust and water egress. As a selling point, the manufacturer of the Vesper piezo MEMS microphone indicates that the device can work underwater as a hydrophone. Small, cheap, low-power and reliable MEMS hydrophones could prove incredibly useful in the underwater domain, enabling a variety of IoT type applications. We test the Vesper piezo MEMS microphone in an underwater environment to assess the manufacturer's claims with regard to its use as a hydrophone. We compare both the sensitivity and the directionality of the microphone to a standard hydrophone.
(1) Defence Science & Technology Group, PO Box 1500 Edinburgh, SA 5111, Australia
ABSTRACT
The effect of surface reflections, such as the Lloyd's mirror effects, is well known. Earlier International Standards for measurements of underwater radiated noise from marine platforms in deep water, such as NATO STANAG 1136:1995, ANSI/ASA S12.64-2009, and ISO 17208-1:2016, acknowledge the importance of surface reflections, but do not provide methods to account for their effects. The recent Standard ISO 17208-2 (2017) attempts to correct for the coherent Lloyd's mirror effects at low frequencies, but does not account for non-coherent energies diffusely scattered from rough surfaces. In this paper, we model the effect of reflections from both smooth and rough sea surfaces, taking into account both coherently reflected and incoherently scattered energies.
Dr
ABSTRACT
The present author showed previously that, if the sound speed profile (SSP) in a vertically stratified fluid half-space is described by a specific formula, the exact solution of the depth-separated Helmholtz equation is a Bessel function with the argument exponentially depending on depth. As, according to this formula, the square of the refraction index "n" changes exponentially with depth, it is suggested here that this SSP be called the "n^2-exponential SSP". It is shown that the n^2-exponential SSP in a limiting case can approximate the linear SSP in the isothermal surface duct. This approximation is compared with a commonly used approximation by means of the n^2-linear SSP. It is shown that the equation for the horizontal wavenumbers in a medium with the n^2-exponential SSP in its approximate form coincides with the corresponding approximate equation for the n^2-linear SSP. Green's function for the acoustic field trapped in the duct with the n^2-exponential SSP is obtained from a known equation. The acoustic field calculated with the use of the Green's function is compared with the acoustic field obtained by means of numerical models BELLHOP and RAMGeo. It is shown that the spatial distribution of the acoustic field described by the Green's function is close to the one calculated by these models.
(1) College of Science and Engineering, Flinders University, Clovelly Park, Adelaide, SA 5042, Australia (2) Adelaide Institute for Sleep Health, College of Medicine and Public Health, Flinders Uni-versity, Bedford Park, Adelaide, SA 5001, Australia
ABSTRACT
Despite the significance of listening tests in identifying the human response to wind farm noise (WFN), little attention has been paid to methodological approaches relevant to WFN listening tests to date. Moreover, evi-dence on the potential adverse effects of WFN is still not well established. This paper thereby sheds light on the different quantification approaches of human response to WFN characteristics. There is also a discussion on the quality of current evidence regarding the effect of WFN on annoyance and sleep disruption. In the con-text of listening tests, separating WFN characteristics can be beneficial in many ways. Firstly, acceptable threshold levels for each component of WFN for daytime annoyance and night-time sleep disturbance can be quantified. Second, the most annoying characteristic of WFN can be identified. Third, the most annoying char-acteristics of WFN can be identified by combining two or more components of WFN. Finally, this helps deter-mine the sufficiency of current penalties applied to WFN characteristics.
(1) Flinders University
ABSTRACT
In addition to the overall noise level, periodic variations in the loudness of wind turbine noise, known as amplitude modulation (AM), also significantly contribute to the annoyance experienced by residents living near wind farms. Due to the high dependence of AM on meteorological conditions and the type of wind turbines, the level and duration of AM are hard to predict. These characteristics have an important impact on the annoyance response of residents. The level of annoyance is expected to depend on the AM depth, the number of AM occurrences and the AM continuity. The aim of this paper is to investigate AM characteristics in the vicinity of two wind farms in South Australia. It has been found that to successfully quantify tonal AM based on the "Reference Method" proposed by the UK Institute of Acoustics, removing the A-weighting, changing the range of band-pass filter frequency and reducing the prominence ratio are also necessary. AM density at night-time is much higher than at daytime (25% versus 15%). However, there is not significant difference between AM depth at night-time and daytime. Furthermore, AM is more likely to occur when the wind turbines are operating significantly below their maximum rated power.