Noise emission is a major drawback of the positive displacement machine. The noise source can be divided into structure borne noise source (SBNS) and fluid borne noise source (FBNS). Passive techniques such as valve plate optimization have been used for noise reduction of axial piston machines. However, passive techniques are only effective for limited operating conditions or at least need compromises in design. In this paper, active vibration control of swash plate is investigated for vibration and noise reduction over a wide range of operating conditions as an additional method to passive noise reduction techniques. A 75cc pump has been modified for implementation of active vibration control using the swash plate. One tri-axial acceleration sensor and one angle sensor are installed on the swash plate and a high speed servovalve is used for the swash plate actuation. The multi-frequency two-weight least mean square (LMS) filter synthesizes the servovalve input signal to generate a destructive interference force which minimizes the swash plate vibration. An experimental test setup has been realized using Labview field-programmable gate array (FPGA) via cRIO. Simulation and experimental studies are conducted to investigate the possibility of active vibration control.

The field of coupled resonant systems is a rich research area with enumerable real-world applications, including the fields of neural computing and pattern recognition, energy harvesting, and even modeling the behavior of certain types of biological systems. This work is primarily focused on the study of the behaviors of two subsets of this field: large networks of globally coupled resonators (which, in this work, refers to passive, damped resonant elements which require external stimulus) and smaller networks of oscillators (referring to active devices capable of self-sustained motion), which are coupled through a network of light-sensitive resistive elements. In the case of the former, we begin by developing an analytical and experimental framework to examine the behaviors of this system under various conditions, such as different coupling modalities and element-level parametric mistunings. Once a proper understanding of the dynamics of these systems has been established, we go on to develop the system into a single-input, single-output, multi-analyte volatile organic compound sensor. For the study of oscillator networks, we begin by building a device which utilizes a network of Colpitts oscillators, coupled through a series of color-filtered CdSe photocells. We then establish that through the analysis of particular emergent behaviors (most notably, frequency locking within the network), this type of system may show promise as a threshold color sensor. By exploiting these behaviors, this type of system may find applications in neuromorphic computing (particularly in optical pattern recognition).

<p> </p> <p>Gas sensors have many applications. Volatile organic compound (VOC) sensors are used for monitoring air quality in homes and office spaces, as well as monitoring manufacturing environments where a wide variety of VOCs can be produced. These gases can include formaldehyde, which can be toxic to humans at concentrations as low as 1 ppm. Other applications for gas sensors include flammable refrigerant detection. With the move towards developing more environmentally friendly appliances, many companies have started to use refrigerants such as R600a (isobutane) and R32 (difluoromethane), which have a much lower global warming potential (GWP) than their predecessors, such as R134a and R410a. While this move is beneficial for the environment, steps to ensure their safe usage have not been widely implemented to date. Therefore, sensors to detect VOCs at or below exposure limits, as well as flammable refrigerants at or below lower flammability limits (LFL), should be developed to ensure undue hazards are identified and mitigated. </p>

A common choice of bearing for industrial applications such as turbomachinery and rotating compressors is the Tilting-Pad Journal Bearing (TPJB) due to its excellent stability properties. TPJB's are however limited by the reduction of damping in the fluid film at high velocities. In order to overcome this, the Active Tilting-Pad Journal Bearing (ATPJB) has been developed. By adding the possibility of high-pressure radial oil injection through servo-valves which can be controlled via a feedback-loop control system, the classically purely mechanical TPJB becomes a mechatronic device called ATPJB.  The objective of this project is to conduct an experimental evaluation of the dynamical behavior of the ATPJB test rig located at the Technical University of Denmark, use the experimental results to modify the previously developed dynamical model which is used for the calculation of a model-based control system. The control system is to be implemented and experimentally validated at high velocities. Improvements made to the test rig in order to achieve high velocities have been documented and described in this work. The mathematical modeling of the individual components, reduction methods, and the global system assembly is covered with an extensive overview. Parameters of the model have been made frequency dependant in order to have an accurate model, resulting in good agreement with experimental data over a wider operational range. With the implemented Linear Quadratic Gaussian controller it is shown that ATPJB has extended operational range compared to TPJB and shows reduction of vibrations over rotational speeds spanning from 1000 RPM to 10,000 RPM. The ATPJB-technology, as it is implemented in this project, does not improve frictional losses in the system. It is argued that the added sensing and actuating systems inherited in the ATPJB technology make the technology highly suitable for the ideas of Industry 4.0 and also allows for the implementation of Early Fault Diagnosis which gives an economical incitement to invest in ATPJB-technology.

<p>Despite the increasing demand for the hydraulic pump noise reduction, there is yet to be an established straightforward solution to reduce noise emissions. This is primarily due to a lack of understanding of the complete mechanism underlying noise generation and propagation, which involves complex interactions between three domains. Study of the physical phenomena of the hydraulic pump noise is typically separated into three categories, namely fluid-borne noise (FBN), structure-borne noise (SBN), and air-borne noise (ABN). In this light, this study examines the noise generation and propagation of hydraulic pumps in all three domains numerically and experimentally, taking external gear pumps (EGPs) as a reference. </p> <p>In conventional pump noise studies, the outlet pressure ripple in the fluid domain, which typically refers to has been the key focus to minimize, and FBN typically refers to the outlet pressure ripple. Fortunately, attempts to minimize ripples have resulted in some promising solutions that are now on the market (e.g., dual-flank gear pumps). However, since the noise generated by gear pumps involves several other significant and coherent noise sources, this approach has some limitations. In view of this, the current study describes FBN in a wider context to include all potential noise sources in the fluid domain, and their mutual effects on noise are investigated.</p> <p>Another aspect of the vibration and noise of the pump that is not often investigated is its “field” behaviors. Many significant works in vibroacoustic analysis or noise solutions rely on the simple measurements of acceleration or sound pressure at a single or few local points. Since vibration and noise are functions of not only time but also "space", this practice has also served as one of the obstacles to a comprehensive understanding of noise generation. Therefore, this study contributes to topic of the vibroacoustic field behaviors.</p> <p>Furthermore, when prototyping or designing new pumps, inefficient trial-and-error methods are often used, and it demonstrates the necessity of the acoustic model of the pumps for virtual prototyping. The major limiting factor towards the development of this type of models is high computational costs. Another technical challenge is that most of vibroacoustic analysis commercial software usually requires the user’s manual works for the simulation setup. In this regard, another aim of this study is to develop a computationally inexpensive and automated acoustic model that does not need manual inputs of users, so that the model can be used as a virtual prototyping tool with various design parameters.</p> <p>To sum up, the primary goal of this research is to numerically and experimentally investigate the vibroacoustic field behaviors and formulate the acoustic model to be used as a virtual prototyping tool with the experimental validation. To achieve these objectives, this research employs the well-established computational and experimental methods of vibro-acoustic analysis.</p> <p>The analysis of FBN makes use of the HYGESim tool, which has been developed to study EGMs at Maha Fluid Power Research Center. This tool solves the main flow based on the lumped parameter approach in conjunction with different solution schemes for lubricating interfaces and body dynamics. From the HYGESim results, all potential noise sources within the working fluid, such as inlet and outlet pressure ripple and dynamic pressure at the tooth space volumes, hydrodynamic journal bearings, and the lateral lubricating interface, are properly mapped to the structure using appropriate simplifications. </p> <p>When it comes to SBN, the modal superposition approach is exploited for the fast prediction of vibration fields. Therefore, considerable efforts are expended both numerically and experimentally to obtain accurate modal information. Particular attention is paid to the modeling of the mechanical connections between components and modeling of constraints in numerical modal analysis using the finite element method (FEM). Moreover, the vibration mode shapes are categorized according to the dominant motions that the pump body exhibits. Then, two different approaches, namely the full numerical model and the hybrid model, are introduced for the estimation of the vibration field during the operation; for the modal expansion, the former uses numerical modal information, while the latter uses experimentally determined modal information. Finally, the numerical model results are compared to the operational deflection shape (ODS) measured during pump operation, and a good agreement is observed.</p> <p>For the ABN prediction, the boundary element method (BEM) is used by taking the predicted vibration information as an input. The BEM solver development is elaborated to numerically replicate the acoustic environments where the noise measurement is conducted. With the developed BEM solver, two units that have the different gear and groove designs that fit into the same casing are tested, and as the key outcome, their sound power level, sound power spectrum, sound pressure distributions are presented. For model validation, the noise measurements are performed according to the ISO standard in the semi-anechoic chamber at Maha using a custom-designed robot arm. These validations demonstrate the ability of the developed model to predict the overall sound power levels with an averaged error of 1.87 dB and capture the general trends of measured sound power spectrum and sound pressure level distribution under various operating conditions. Furthermore, the developed model provides the reasonably fast computation time.</p> <p>Finally, using the developed acoustic model, a parametric study is performed with the backflow groove as a design variable. It is discussed how the volumetric efficiency and noise performance vary with the design changes, which demonstrates the model potential as a virtual prototyping tool.</p>

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第21091号 / 工博第4455号 / 新制||工||1692(附属図書館) / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 五十嵐 晃, 教授 八木 知己, 准教授 古川 愛子 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

The ability to rapidly assess the condition of a structure in a manner which enables the accurate prediction of its remaining capacity has long been viewed as a crucial step in allowing communities to make safe and efficient use of their public infrastructure. This objective has become even more relevant in recent years as both the interdependency and state of deterioration in infrastructure systems throughout the world have increased. Current practice for structural condition assessment emphasizes visual inspection, in which trained professionals will routinely survey a structure to estimate its remaining capacity. Though these methods have the ability to monitor gross structural changes, their ability to rapidly and cost-effectively assess the detailed condition of the structure with respect to its future behavior is limited.<div>Vibration-based monitoring techniques offer a promising alternative to this approach. As opposed to visually observing the surface of the structure, these methods judge its condition and infer its future performance by generating and updating models calibrated to its dynamic behavior. Bayesian inference approaches are particularly well suited to this model updating problem as they are able to identify the structure using sparse observations while simultaneously assessing the uncertainty in the identified parameters. However, a lack of consensus on efficient methods for their implementation to full-scale structural systems has led to a diverse set of Bayesian approaches, from which no clear method can be selected for full-scale implementation. The objective of this work is therefore to assess and enhance those techniques currently used for structural identification and make strides toward developing unified strategies for robustly implementing them on full-scale structures. This is accomplished by addressing several key research questions regarding the ability of these methods to overcome issues in identifiability, sensitivity to uncertain experimental conditions, and scalability. These questions are investigated by applying novel adaptations of several prominent Bayesian identification strategies to small-scale experimental systems equipped with nonlinear devices. Through these illustrative examples I explore the robustness and practicality of these algorithms, while also considering their extensibility to higher-dimensional systems. Addressing these core concerns underlying full-scale structural identification will enable the practical application of Bayesian inference techniques and thereby enhance the ability of communities to detect and respond to the condition of their infrastructure.<br></div>

We are seeing a revolution in computing and artificial intelligence; intelligent machines have become ingrained in and improved every aspect of our lives. Despite the increasing number of intelligent devices and breakthroughs in artificial intelligence, we have yet to achieve truly intelligent environments. Audio is one of the most common sensing and actuation modalities used in intelligent devices. In this thesis, we focus on how we can more robustly integrate audio intelligence into a wide array of resource-constrained platforms that enable more intelligent environments. We present systems and methods for adaptive audio filtering that enables us to more robustly embed acoustic intelligence into a wide range of real time and resource-constrained mobile, embedded, and cyber-physical systems that are adaptable to a wide range of different applications, environments, and scenarios. First, we introduce methods for embedding audio intelligence into wearables, like headsets and helmets, to improve pedestrian safety in urban environments by using sound to detect vehicles, localize vehicles, and alert pedestrians well in advance to give them enough time to avoid a collision. We create a segmented architecture and data processing pipeline that partitions computation between embedded front-end platform and the smartphone platform. The embedded front-end hardware platform consists of a microcontroller and commercial-off-the shelf (COTS) components embedded into a headset and samples audio from an array of four MEMS microphones. Our embedded front-end platform computes a series of spatiotemporal features used to localize vehicles: relative delay, relative power, and zero crossing rate. These features are computed in the embedded front-end headset platform and transmitted wirelessly to the smartphone platform because there is not enough bandwidth to transmit more than two channels of raw audio with low latency using standard wireless communication protocols, like Bluetooth Low-Energy. The smartphone platform runs machine learning algorithms to detect vehicles, localize vehicles, and alert pedestrians. To help reduce power consumption, we integrate an application specific integrated circuit into our embedded front-end platform and create a new localization algorithm called angle via polygonal regression (AvPR) that combines the physics of audio waves, the geometry of a microphone array, and a data driven training and calibration process that enables us to estimate the high resolution direction of the vehicle while being robust to noise resulting from movements in the microphone array as we walk the streets. Second, we explore the challenges in adapting our platforms for pedestrian safety to more general and noisier scenarios, namely construction worker safety sounds of nearby power tools and machinery that are orders of magnitude greater than that of a distant vehicle. We introduce an adaptive noise filtering architecture that allows workers to filter out construction tool sounds and reveal low-energy vehicle sounds to better detect them. Our architecture combines the strengths of both the physics of audio waves and data-driven methods to more robustly filter out construction sounds while being able to run on a resource-limited mobile and embedded platform. In our adaptive filtering architecture, we introduce and incorporate a data-driven filtering algorithm, called probabilistic template matching (PTM), that leverages pre-trained statistical models of construction tools to perform content-based filtering. We demonstrate improvements that our adaptive filtering architecture brings to our audio-based urban safety wearable in real construction site scenarios and against state-of-art audio filtering algorithms, while having a minimal impact on the power consumption and latency of the overall system. We also explore how these methods can be used to improve audio privacy and remove privacy-sensitive speech from applications that have no need to detect and analyze speech. Finally, we introduce a common selective audio filtering platform that builds upon our adaptive filtering architecture for a wide range of real-time mobile, embedded, and cyber-physical applications. Our architecture can account for a wide range of different sounds, model types, and signal representations by integrating an algorithm we present called content-informed beamforming (CIBF). CIBF combines traditional beamforming (spatial filtering using the physics of audio waves) with data driven machine learning sound detectors and models that developers may already create for their own applications to enhance and filter out specified sounds and noises. Alternatively, developers can also select sounds and models from a library we provide. We demonstrate how our selective filtering architecture can improve the detection of specific target sounds and filter out noises in a wide range of application scenarios. Additionally, through two case studies, we demonstrate how our selective filtering architecture can easily integrate into and improve the performance of real mobile and embedded applications over existing state-of-art solutions, while having minimal impact on latency and power consumption. Ultimately, this selective filtering architecture enables developers and engineers to more easily embed robust audio intelligence into common objects found around us and resource-constrained systems to create more intelligent environments.

<p>  </p> <p>The development of an increasingly interconnected infrastructure and its rapid evolution demands engineering testing solutions capable of investigating realistically and with high accuracy the interactions among the different components of the problem to study. The examination of any of these components without losing the interaction of the other surroundings components is not only realistic, but also desirable. The more interconnected the whole system is, the greater the dependencies. Real-time Hybrid Simulation (RTHS) is a disruptive technology that has the potential to address this type of complex interactions or internal couplings by partitioning the system into numerical (better understood) substructures and experimental (unknown) substructures, which are built physically in the laboratory. These two types of substructures are connected through a transfer system (e.g., hydraulic actuators) to enforce boundary conditions in their common interfaces creating a synchronized cyber-physical system. However, despite the RTHS community has been improving these hybrid techniques, there are still important barriers in their core methodologies. Current control approaches developed for RTHS were validated mainly for linear applications with limited capabilities to deal with high uncertainties, hard nonlinearities, or extensive damage of structural elements due to plasticity. Furthermore, capturing the realistic dynamics of a structural system requires the description of the motion using more than one degree of freedom, which increases the number of hydraulic actuators needed to enforce additional degrees of freedom at boundary condition interface. As these requirements escalate for larger or more complex problems, the computational cost can turn into a prohibitive constraint. </p> <p>In this dissertation, the main research goal is to develop and validate a nonlinear controller with capabilities to control highly uncertain nonlinear physical substructures with complex boundary conditions and its parallel computational implementation for accurate and realistic RTHS. The validation of the proposed control system is achieved through a set of real-time tracking control and RTHS experiments that explore robustness, accuracy performance, and their trade-off </p>

Active Structural Acoustic Control (ASAC) is a subcategory of the more widely known field of Active Noise control (ANC). ASAC is different from traditional ANC methods because it seeks to attenuate noise by altering the noise producing structure instead of altering the acoustic waves traveling through the air. The greatest challenge currently facing ASAC researchers is that a suitable parameter has not yet been discovered which can be easily implemented as the minimization quantity in the control algorithms. Many parameters have been tried but none effectively attenuate the sound radiation in a way that can be easily implemented. A new parameter was recently developed which showed great potential for use as a minimization quantity. This parameter has been termed the "weighted sum of spatial gradients" (WSSG) and was shown by previous researchers to significantly reduce noise emissions from a vibrating simply supported plate in computer simulations. The computer simulations indicate that WSSG-based control provides as good or better control than volume velocity and does so with a single point measurement which is relatively insensitive to placement location. This thesis presents the experimental validation of the WSSG computer simulations. This validation consists of four major components. First, additional research was needed in to extend the use of WSSG from computer simulations to experimental setups. Second, the WSSG-based control method was performed on simply supported plates to validate the computer simulations. Third, the WSSG-based control method on was used on clamped plates to validate the computer simulations, and fourth, the WSSG-based control method was validated on plates with ribs. The important results are discussed and conclusions summarized for each of these sections. Recommendations are made for future work on the WSSG parameter.