This research contributes to an empirical design analysis framework for informing the implementation of a type of microsound synthesis called Particle Synthesis (PS). Microsounds are sound particles with a duration typically lasting from thethreshold of perception up to ~100ms (Roads, 2001a). Microsound synthesis software is popular in music production and is used for processing audio. PS, however, still remains relatively unknown to composers and only recently have independent developers made PS available on popular computer platforms. Variations in implementation, however, disguise similar underlying techniques and produce inconsistencies. Perhaps this is because independent developers (often individual researchers/hobbyists) lack the development formalities of large software corporations. Although different sound classes are generated by PS systems, they share very similar parametric features. However, parameters, which are key in one system,are commonly implemented partially or omitted in another. In order to obtain a wider more flexible range of microsound sound classes, users have to operate several of the systems simultaneously across incompatible operating systems and environments. To address this situation, this thesis examined the questions as to what specific functional and usability criteria might inform new PS designs and whether this criteria and consequent design framework would be successful in informing new PS artefacts. Key theoretical foundations of microsound techniques are identified, and their use in 20th century music composition examined. PS techniques lack a specific design framework. However, Jaffe (1995) proposed criteria to assess general synthesiser technologies, and design analysis principles called Cognitive Dimensions (CDs) were revealed which help evaluate the design of information artefacts including software synthesisers. Many elements of Jaffe's criteria may be viewed as industrial design principles, which may be generalised into microsound synthesis criteria. By combining the CDs framework with the new microsound artefact criteria and operationalising them as variables, an empirical analysis framework for studying usability factors of PS software was made possible. Furthermore, these can be applied to the functional elements of design. Subsequently, empirical studies were conducted in which seven identified PS implementations were quantitatively assessed against the new microsound criteria in order to ascertain the weight of individual usability and functional characteristics across the systems. The collated results from the studies confirmed that many common usability and functional features exist. They further revealed uneven and incomplete implementation of features for composition of sound particles over multiple timescales, and seamless manipulation of continuous and discrete particle sound streams. The results were collated in order to establish specific design goals and used to implement and adapt a prototype PS compositional system called the Elementary Signal Engine (ESE), which includes the common and disparate features of the systems studied. This is the ancillary contribution to this work. The success of the new MS design analysis framework is subsequently evaluated by studying the ESE artefact using the same analysis framework against the design goals established from the original seven PS artefact analyses.

An apparatus has been constructed with which it is possible to produce, in a tithe, constant and uniform foams with controllable density and bubble size. The sound source is situated at one end of the tube and a probe microphone is used as detector. It has been found that owing to a series of transverse mote: occurring in the frequency range used, measurements of the bulk properties of the foam were difficult to make except at low frequencies. A hypothesis to explain both the existence of these modes and the discrepancy between the measured and theoretical values of the compressional sound velocity is proposed in terms of the existence of a foam shear modulus. Theoretical analysis of the propagation of waves in an elastic medium which is confined in a rigid cylinder and of the thermal damping in a foam are contained in appendices.

The implementation of the Acoustic Sub-seabed Interrogator (ASI)—from its initial experimental concept research phase, which the author pursued at the University of Bath in the form of his PhD thesis (early 1980s), through to full-scale prototyping (late 1980s, early 1990s) and finally to commercialization (mid-1990s onward)—has had the underpinning support and funding of major granting agencies and industrial sponsors/investors. The present work, which forms the content of this Doctor of Science thesis, is the manifestation of over twenty-five years of research and development costing tens of millions of dollars. Therefore, I wish to mention the sponsorship of the agencies, granting institutions, and investors who backed the concept from its early formation stage and pursued the engineering developments leading to today’s commercial market acceptance and utilization. From 1982 to 1986, the National Research Council of Canada (NRC), under an Industrial Research Application Program (IRAP), supported this University of Bath PhD research project. This led to the support of major research granting agencies, Oil & Gas industrial sponsorships, and research centres to help back the development of a scientific prototype and trials in 1990 and continuing on to 1995. This early development of the ASI was made possible by the Centre for Cold Ocean Resources Engineering (C-CORE), the Natural Sciences and Engineering Research Council of Canada (NSERC), the Atlantic Canada Opportunities Agency (ACOA), the Atlantic Geoscience Centre, and the NRC Institute for Marine Dynamics (now the Institute of Ocean Technology [IOT]). The industrial sector was instrumental in providing major operational funds for sea trials around the world; the first sponsors were Mobil Oil Canada Limited, Gulf Canada Resources Limited, Petro-Canada Resources Limited, and Esso Resources Canada Limited. From 1995 to 2005, various research grants by NSERC and NRC-IRAP were issued to Guigné to conduct scientific laboratory and near-shore studies into the physics of the ASI’s acoustics as it applied to imaging the seabed. This was complemented with major contracts by the Canadian Department of Fisheries and Oceans under their Northern Cod Science Program. The acoustic work related to quantifying the effects of otter trawling on benthic habitats. This work was supplemented by the financial support of the National Water Research Institute (NWRI) for investigations into the use of broadband acoustics for mapping subtle discontinuous lakebed features.

Energy harvesting from ambient vibration has become an attractive topic in the recent years. Initial studies aimed to maximise the performance of small linear device for different excitation scenarios. These devices were assumed to be located in hostile and inaccessible environments and be able to provide energy for low powered sensors. Due to the limited size of the energy harvesters, however, the amount of power produced was small. More recently, many researchers have considered using nonlinear stiffness to improve the performance of these devices. This thesis, however, focuses on the use of nonlinear damping in energy harvesters. Nonlinear damping may be unwanted and introduced as the mechanism of the harvester, but can also be deliberately introduced to improve the dynamic range of the harvester. Typically ambient vibration generates a relative displacement between the suspended mass and the base in an energy harvester, which induces an electromotive force (EMF) in a circuit that is used to harvest electrical energy. It is possible to introduce nonlinear mechanical damping by having a circuit with a nonlinear resistance. Specifically, a load, in which the current is a third-power function of the voltage is compared with an equivalent linear load for three kinds of excitation such as harmonic, random white noise and random bandlimited noise. According to the numerical and analytical results, the cubic load provides more harvested power at resonance at low levels when compared to an equivalent linear load at the same level of excitation. As the frequency bandwidth of a random excitation becomes wider, to the limit of white noise, as the power generated by the cubic converges to the linear case. Electromagnetic transducer energy harvesters usually adopt a conversion mechanism of motion, such as ball screw or rack and pinion, which introduce a source of loss such as friction. Static friction is then added to the model and this is shown to affect the harvested power at low input levels. Another proposed strategy consists of adjusting the electric load according to the input level, which can also enlarge the dynamic range of performance of energy harvested compared to a device with constant load. To demonstrate the effectiveness of the level-dependent load, an energy harvesting device was designed and manufactured, which comprised of an oscillating beam sprung to the base, and attached to a generator. Across the terminals of the generator, an electric resistance is mounted and the voltage measured is used to compute the harvested power. Experiments are conducted by exciting the harvested with a harmonic input at resonance via a shaker. A level-dependent load and a constant load were separately tested; with results that are in good agreement with the simulations, it is shown that by adjusting the load according to the input level, the harvested power is increased compared to a linear constant load.

Steel reinforcements bars (rebars) are vital to reinforced concrete (RC) structures and their damage leads to catastrophic failure. Most damage occurs due to corrosion and delamination, and an early detection is necessary. Wave based methods are popular for detecting the damage remotely. However, most of these techniques require direct contact with the rebars. The aim of this project is to exploit guided waves to detect damage of the rebars via measurements only on the concrete surface. The Wave Finite Element (WFE) is well suited to predict the wave characteristics of RC waveguides. It requires knowledge of the mass and stiffness matrices of only a segment of the waveguide which can be obtained from conventional FE analysis. A new RC model approach using embedded reinforcements is suggested and compared to conventional FE models. Next, the WFE methodology is discussed including associated numerical and ill conditioning errors. Wave solutions for RC beams with and without prestress, in the form of dispersion curves and mode shapes, were found to be similar. Having established free wave solutions for an undamaged RC waveguide, one can couple it to a damaged segment that can be modelled in FE. Alternatively, if the damage is modelled as a reduction in rebar diameter over a certain length, then this permits the WFE coupling approach to couple damaged and undamaged RC models for better computational efficiency. High magnitudes for the reflection coefficients due to damage are found, which are associated with evanescent waves at their cut-on frequencies. Based on these findings, a new damage identification algorithm is proposed in which the amplitudes of left and right propagating waves are estimated from surface measured forced responses. The experimental based methodology was successful in detecting rebar reduction without any prior knowledge of the dispersion relations. Experimental validation of the algorithm is found to be successful and in good agreement with simulations. The potential and limitations of the algorithm for practical structures are discussed.

This thesis covers the dynamic modelling of un-cracked and cracked rotating beams. Accordingly, a new model of a rotating cracked beam is developed using the finite element and the Rayleigh-Ritz method to characterise and analyse its dynamic behaviour. The effect of various parameters are investigated, such as rotational speed, hub ratio and slenderness ratio. In addition, the critical speed, buckling speed and veering phenomena are identified. The numerical results produced are shown to be in good agreement with models based on finite element representations. In addition to the theoretical investigations, experimental validation is presented. A test rig was designed and manufactured with a changeable rotating hub mount for different test requirements. Moreover, the rig was conceived to incorporate capabilities such as applying variable rotational speed using a variable frequency driver and provide vertical base excitation input to the centre of rotation of the hub. The tests were performed using random excitation at the root of the rotating cantilever beam to excite the flapwise modes of the beam. The responses were then measured optically using a high-speed camera, and the images were post-processed using a digital image correlation (DIC) method. This non-invasive optical method was used to extract the temporal deflection of the beam. The frequency response functions are then obtained from the measured responses. The estimated modal frequencies were compared with numerical simulations to validate the Rayleigh-Ritz and FE numerical models at various rotational speeds. Furthermore, an experimental crack detection was implemented and the results showed a good match to the introduced actual crack location and depth. The crack detection approach on the rotating cracked beam uses the fundamental and second natural frequencies. For vibration control of the rotating beam, a real-time velocity feedback control was applied using a remote single optical high-speed camera. An electromagnetic actuator was designed and mounted on the rotating hub to apply a feedback force on the rotating beam. The results for vibration control of the rotating beam show significant active damping and reduction in the amplitude of the first resonance over a wide range of rotational speeds.

No description available.

Broadband interaction noise is a major source of noise in turbofan engines and will become more dominant with the increase of the bypass ratio. Generated by the interaction of turbulent fan wakes with outlet guide vanes, it can be numerically predicted by restricting the configuration to the stator only and introducing turbulent fan wakes with stochastic methods. In this work, a new method to inject turbulence using multiple sources of vorticity is proposed. It can generate either one-component or two-component frozen turbulence while being easy to implement and has no influence on the parallelization of an already existing solver. This method is successfully applied to an isolated 2D aerofoil and compared to experiments. In complex 2D geometries, the distortion of turbulence upstream of an aerofoil plays an important role in interaction noise, yet little is known regarding the mechanisms involved. Thus, the second part of this work focuses on turbulence distortion near the leading edge. To provide more physical insights in these mechanisms, a numerical vorticity approach in the frequency domain is developed. It allows the decomposition of the vorticity field into the incoming vorticity which is distorted close to the leading edge, and the bound vortices at the solid surfaces and wakes which represents the vorticity response of the aerofoil to respect a non-penetration condition at its boundaries. The size of the stagnation region at the leading edge is found to be a key factor in understanding turbulence distortion. Indeed, the larger this region, the more skewed the incoming vorticity field is. It results in attenuated transverse velocities retrieved from this vorticity field as well as reduced velocity perturbations from the bound vortices, which decreases the incoming turbulence levels, and therefore leads to noise reductions. However, at low wavenumbers, as the wavelength is large with respect to the stagnation region, no turbulence alteration is observed. To maximize the effect on the distortion of turbulence, the geometric changes need to be located as close as possible to the leading edge, as it will have more influence on the size of the stagnation region. Indeed, in this work, the geometry after the maximum thickness is found to have no effect on turbulence distortion.

Dynamic force has been measured previously at the seat interface beneath the sitting human body exposed to vertical vibration. However, how the distribution of forces over the interface contributes to the total force has not been identified. This study seeks to understand biodynamic responses to vertical vibration at the ischial tuberosities, the middle thighs, and the front thighs, so as to allow the development of a biodynamic model representing the overall and localised responses. In this thesis, for convenience, the transfer function between the dynamic force at a location and the acceleration at the same location is referred to as the ‘localised apparent mass’. How localised apparent masses at the ischial tuberosities, the middle thighs, and the front thighs contribute to the overall apparent mass of the body, and how thigh contact affects the distribution, were investigated initially. The vertical apparent mass at the ischial tuberosities dominated the overall apparent mass around 5 Hz for all postures, but that at the thighs dominated the apparent mass around 8 Hz when the feet were not supported. For all postures, the fore-and-aft cross-axis forces showed a slightly lower principal resonance frequency at the middle thighs and front thighs than at the ischial tuberosities. It is suggested that more than one mode contributes to the resonance in the fore-and-aft cross-axis apparent mass. When the feet were unsupported, the nonlinearities in the biodynamic responses were greater at the thighs than that at the ischial tuberosities. How support from both rigid and soft backrests affect biodynamic responses measured at the seat pan were then investigated. For both rigid and soft backrests inclined from the vertical by 30 degrees, the vertical forces were greater at the ischial tuberosities than at the thighs at all frequencies less than 15 Hz. With increasing inclination of the rigid backrest, the resonance in the overall vertical in-line apparent mass at the seat pan broadened, and the frequency of the principal resonance in the overall fore-and-aft cross-axis apparent mass at the seat pan decreased. Irrespective of the backrest stiffness and backrest inclination, the frequency of the resonance in the overall vertical apparent mass at the seat pan was not correlated with the frequency of the resonance in the forces measured normal to the backrest. The effect of thigh contact on the apparent mass of the body sitting on a foam cushion was then investigated. The frequency of the resonance in the vertical transmissibility of the foam was around 4 Hz at the ischial tuberosities but around 6 to 8 Hz at the front thighs. When sitting on the foam, the localised apparent masses at the middle thighs and the front thighs showed a resonance around 4 Hz, correlated with the principal resonance frequency in the vertical transmissibility of the foam measured at the ischial tuberosities. It is suggested that the forces at the thighs were affected by motions of the whole body at the ischial tuberosities. Differences between the overall vertical apparent mass of the body measured with the foam and with a rigid seat decreased with decreasing thigh contact. A finite-element model was developed to reproduce the overall vertical in-line apparent mass and the overall fore-and-aft cross-axis apparent mass of the body sitting on a rigid seat while exposed to vertical vibration. The model has the upper-body represented by rigid bodies interconnected by revolute joints, and the buttocks and thighs represented by deformable parts covering the pelvis and femur bones. The shape of the thighs and buttocks was adjusted to the anthropometry of a subject, and viscoelastic material with different properties was assigned to the soft tissues of the buttocks, middle thighs, and front thighs. The model is shown to represent both the static pressure and the localised biodynamic responses over a rigid seat with different thigh contact conditions.

In modern aircraft configurations, the interaction between the exhausted jet with the wing and high-lift devices is set to increase substantially the total aircraft noise generated during take-off. Regarding ultra-high bypass ratio turbofan engines mounted underneath the wing, the jet plume is expected to interact strongly with rigid surfaces. The interaction between the high-speed flow and aircraft structures modifies the 'pure' jet mixing noise and enhance the scattered hydrodynamic field. Thus, understanding the changes to the jet turbulence flow field caused by the presence of a solid boundary is vital to predict and mitigate the engine noise produced by commercial aircraft which will be introduced imminently. In this thesis, the statistics of the streamwise component of the velocity of subsonic jets is investigated in detail. In close-coupled installed jet configurations, the interaction between the jet and a rigid surface leads to a local flow acceleration and to lower turbulence levels near the solid boundary. These effects are consistent with the conservation of momentum and the 'beak-down' of the eddies in that region. The characteristic length scales of these installed jet configurations are smaller than the isolated jet scales. The effects of forward flight upon the turbulence field is also considered. A stretching factor with flight velocity is obtained by considering the jet virtual origin. Models for the coherence decay, time and length scales are proposed based on experimental evidence and the assumption of frozen turbulence in the region of maximum turbulence kinetic energy. For the first time, two-point statistic models are proposed for high-subsonic and installed jet configurations. Finally, analysis of the far field of installed jets show that a strong interaction between the jet turbulence field and a solid boundary generates an additional high-frequency noise. Thus, the existence of a noise source mechanism related to the flow-structure interplay is demonstrated. It is hoped that the experimental data and analysis presented in this work provide feedback for jet noise source modelling and also future numerical and large-scale laboratory experiments.