No description available.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23888号 / 工博第4975号 / 新制||工||1777(附属図書館) / 京都大学大学院工学研究科航空宇宙工学専攻 / (主査)教授 藤本 健治, 教授 加納 学, 准教授 丸田 一郎, 教授 松野 文俊 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM

Metamodeling is regarded as a powerful analysis tool to learn the input-output relationship of a system based on a limited amount of data collected when experiments with real systems are costly or impractical. As a popular metamodeling method, Gaussian process regression (GPR), has been successfully applied to analyses of various engineering systems. However, GPR-based metamodeling for time-dependent systems (TDSs) is especially challenging due to three reasons. First, TDSs require an appropriate account for temporal effects, however, standard GPR cannot address temporal effects easily and satisfactorily. Second, TDSs typically require analytics tools with a sufficiently high computational efficiency to support online decision making, but standard GPR may not be adequate for real-time implementation. Lastly, reliable uncertainty quantification is a key to success for operational planning of TDSs in real world, however, research on how to construct adequate error bounds for GPR-based metamodeling is sparse. Inspired by the challenges encountered in GPR-based analyses of two representative stochastic TDSs, i.e., load forecasting in a power system and trajectory prediction for unmanned aerial vehicles (UAVs), this dissertation aims to develop novel modeling, sampling, and statistical analysis techniques for enhancing the computational and statistical efficiencies of GPR-based metamodeling to meet the requirements of practical implementations. Furthermore, an in-depth investigation on building uniform error bounds for stochastic kriging is conducted, which sets up a foundation for developing robust GPR-based metamodeling techniques for analyses of TDSs under the impact of strong heteroscedasticity. / Ph.D. / Metamodeling has been regarded as a powerful analysis tool to learn the input-output relationship of an engineering system with a limited amount of experimental data available. As a popular metamodeling method, Gaussian process regression (GPR) has been widely applied to analyses of various engineering systems whose input-output relationships do not depend on time. However, GPR-based metamodeling for time-dependent systems (TDSs), whose input-output relationships depend on time, is especially challenging due to three reasons. First, standard GPR cannot properly address temporal effects for TDSs. Second, standard GPR is typically not computationally efficient enough for real-time implementations in TDSs. Lastly, research on how to adequately quantify the uncertainty associated with the performance of GPR-based metamodeling is sparse. To fill this knowledge gap, this dissertation aims to develop novel modeling, sampling, and statistical analysis techniques for enhancing standard GPR to meet the requirements of practical implementations for TDSs. Effective solutions are provided to address the challenges encountered in GPR-based analyses of two representative stochastic TDSs, i.e., load forecasting in a power system and trajectory prediction for unmanned aerial vehicles (UAVs). Furthermore, an in-depth investigation on quantifying the uncertainty associated with the performance of stochastic kriging (a variant of standard GPR) is conducted, which sets up a foundation for developing robust GPR-based metamodeling techniques for analyses of more complex TDSs.

Industrial robots are a key component for several industrial applications. Like all mechanical tools, they do not last forever. The solution to extend the life of the machine is to perform maintenance on the degraded components. The optimal approach is called predictive maintenance, which aims to forecast the best moment for performing maintenance on the robot. This minimizes maintenance costs as well as prevents mechanical failure that can lead to unplanned production stops. There already exist methods to perform predictive maintenance on industrial robots, but these methods require additional sensors. This research aims to predict the anomalies by only using data from the sensors that already are used to control the robot. A machine learning approach is proposed for implementing predictive maintenance of industrial robots, using the torque profiles as input data. The algorithms selected are tested on simulated data created using wear and temperature models. The torque profiles from the simulator are used to extract a health index for each joint, which in turn are used to detect anomalous states of the robot. The health index has a fast exponential growth trend which is difficult to predict in advance. A Gaussian process regressor, an Exponentron, and hybrid algorithms are applied for the prediction of the time series of the health state to implement the predictive maintenance. The predictions are evaluated considering the accuracy of the time series prediction and the precision of anomaly forecasting. The investigated methods are shown to be able to predict the development of the wear and to detect the anomalies in advance. The results reveal that the hybrid approach obtained by combining predictions from different algorithms outperforms the other solutions. Eventually, the analysis of the results shows that the algorithms are sensitive to the quality of the data and do not perform well when the data present a low sampling rate or missing samples. / Industrirobotar är en nyckelkomponent för flera industriella applikationer. Likt alla mekaniska verktyg håller de inte för alltid. Lösningen för att förlänga maskinens livslängd är att utföra underhåll på de slitna komponenterna. Det optimala tillvägagångssättet kallas prediktivt underhåll, vilket innebär att förutsäga den bästa tidpunkten för att utföra underhåll på roboten. Detta minimerar både kostnaderna för underhåll samt förebygger mekaniska fel som kan leda till oplanerade produktionsstopp. Det finns redan metoder för att utföra prediktivt underhåll på industriella robotar, men dessa metoder kräver ytterligare sensorer. Denna forskning syftar till att förutsäga avvikelserna genom att endast använda data från de sensorer som redan används för att reglera roboten. En maskininlärningsmetod föreslås för implementering av prediktivt underhåll av industriella robotar, med hjälp av vridmomentprofiler som indata. Metoderna testas på simulerad data som skapats med hjälp av slitage- och temperaturmodeller. Vridmomenten används för att extrahera ett hälsoindex för varje axel, vilket i sin tur används för att upptäcka anomalier hos roboten. Hälsoindexet har en snabb exponentiell tillväxttrend som är svår att förutsäga i förväg. En Gaussisk processregressor, en Exponentron och hybridalgoritmer används för prediktion av tidsserien för hälsoindexet för att implementera det prediktiva underhållet. Förutsägelserna utvärderas baserat på träffsäkerheten av förutsägelsen för tidsserien samt precisionen för förutsagda avvikelser. De undersökta metoderna visar sig kunna förutsäga utvecklingen av slitage och upptäcka avvikelser i förväg. Resultaten uppvisar att hybridmetoden som erhålls genom att kombinera prediktioner från olika algoritmer överträffar de andra lösningarna. I analysen av prestandan visas att algoritmerna är känsliga för kvaliteten av datat och att de inte fungerar bra när datat har låg samplingsfrekvens eller då datapunkter saknas.

This work is on Gaussian-process based approximation of a code which can be run at different levels of accuracy. The goal is to improve the predictions of a surrogate model of a complex computer code using fast approximations of it. A new formulation of a co-kriging based method has been proposed. In particular this formulation allows for fast implementation and for closed-form expressions for the predictive mean and variance for universal co-kriging in the multi-fidelity framework, which is a breakthrough as it really allows for the practical application of such a method in real cases. Furthermore, fast cross validation, sequential experimental design and sensitivity analysis methods have been extended to the multi-fidelity co-kriging framework. This thesis also deals with a conjecture about the dependence of the learning curve (ie the decay rate of the mean square error) with respect to the smoothness of the underlying function. A proof in a fairly general situation (which includes the classical models of Gaussian-process based metamodels with stationary covariance functions) has been obtained while the previous proofs hold only for degenerate kernels (ie when the process is in fact finite-dimensional). This result allows for addressing rigorously practical questions such as the optimal allocation of the budget between different levels of codes in the multi-fidelity framework.

Despite the alarming incidence rate and potential for debilitating outcomes of sports-related concussion, the underlying mechanisms of injury remain to be expounded. Since as early as 1950, researchers have aimed to characterize head impact biomechanics through in-lab and in-game investigations. The ever-growing body of literature within this area has supported the inherent connection between head kinematics during impact and injury outcomes. Even so, traditional metrics of peak acceleration, time window, and HIC have outlived their potential. More sophisticated analysis techniques are required to advance the understanding of concussive vs subconcussive impacts. The work presented within this thesis was motivated by the exploration of advanced approaches to 1) experimental theory and design of impact reconstructions and 2) characterization of kinematic profiles for model building. These two areas of investigation resulted in the presentation of refined, systematic approaches to head impact analysis that should begin to replace outdated standards and metrics.

With advancements in technology, electric vehicles are dominating the world making Internal Combustion engines less relevant, and hence vehicles are becoming quieter than ever before. But noise levels remain a significant concern for both passengers and automotive manufacturers. The vehicle's interior noise can affect the overall driving experience and even the safety of the driver and the passengers. The two main sources of vehicle interior noise are attributed to air-borne noises and structure-borne noises. A modern automobile is a complicated vibration system with several excitation sources like the engine, transmission system, tire/road interface excitation, and wind noise. With electric vehicles on the rise, the engine and transmission noise is practically eliminated, and effective preventive measures and control systems are already in place to reduce the aerodynamic-based noise, vibrations, and harshness (NVH) in modern automobiles making the structure-borne noise the most crucial of the noise sources. Tire/road interaction noise being the most dominant among the structure-borne noise is the main concern of the vehicle interior noise. The two main sources of vehicle interior noise induced by the tire pavement interaction noise are structure-borne noise induced by the low-frequency excitation and air-borne noises produced by the mid and high-frequency excitation. The present study tested an all-season tire over varying operational conditions such as different speeds, normal loads, and inflation pressures on an asphalt surface. Two tri-axial accelerometers attached 1800 apart from each other on the inner liner of the tire of a Volkswagen Jetta were used to measure the circumferential, lateral, and radial acceleration data. An Inertial Measurement Unit (IMU) and velocity box (VBOX) were instrumented in the vehicle to measure the acceleration at the center of gravity (COG) position of the vehicle and the longitudinal velocity of the vehicle respectively. The vehicle was also equipped with a modified hybrid of Close Proximity Testing (CPX) and On-Board Sound Intensity (OBSI) sound measurement systems which were designed and manufactured in-house to measure the tire/road interaction noise at the leading and trailing edges of the tire/road contact patch. Another microphone was instrumented inside the passenger compartment of the vehicle at the passenger's seat right ear position over the tire mounted with the sound measurement system to measure the vehicle interior noise as interpreted by the passengers in the vehicle. Two data acquisition systems coupled with a real-time Simulink model were used to collect all the measured data, one for the noise signals and the other for velocity and acceleration signals. The focus of the current study is to review different generation and amplification mechanisms of the structure-induced tire/road interaction noise and find the relevant dominant frequency ranges of the vehicle interior noise induced by the structure-borne noises using already established physics-based models and correlation techniques. It also aims to find correlations between tire acceleration, vehicle interior noise, and tire pavement interaction noise and their effect on different operational conditions like load, inflation pressure, and velocity. All the signals are studied in the time, frequency, and spectral domain and insights have been drawn on different tire/road noise generation and amplification mechanisms. / Master of Science / Structure-induced vehicle interior noise is one of the main concerns surrounding the automotive NVH industry and tire/road interaction noise being the most dominant source among the structure-borne noises affecting the vehicle interior noise is a major problem to the tire and automotive manufacturers nowadays. It leads to discomfort for the driver and the passengers in the vehicle and can cause fatigue, which in turn can directly affect the vehicle's safety. Several attempts have been made to reduce vehicle interior noise using statistical, physics-based, and hybrid models, but the research is still nowhere near completion. The current study aims to identify the frequency ranges affecting the structure-borne noise-induced vehicle interior noise and uses data-driven approaches in estimating the vehicle interior noise using only the acceleration of the tire. A test setup was designed and developed in-house where a tri-axial accelerometer embedded inside the inner liner of the tire measures the X, Y, and Z acceleration signals. Several microphones are instrumented at the tire/road contact surface and inside the passenger cabin to measure the tire/road interaction noise and the vehicle interior noise. The longitudinal velocity of the vehicle and the accelerations at the center of gravity of the vehicle have also been measured. Multiple data-driven models have been developed to directly predict the vehicle interior noise and tire/road interaction noise using the accelerometer data. This research is directly helpful for the automotive and tire industries by giving them insights on designing and developing quieter tires by using data-driven approaches and further using these with active control systems can mask the vehicle interior noise to acceptable levels in real-time.

<p>Bridging the accuracy of ab initio (AI) QM/MM with the efficiency of semi-empirical<br> (SE) QM/MM methods has long been a goal in computational chemistry. This dissertation<br> presents four ∆-Machine learning schemes aimed at achieving this objective. Firstly, the in-<br> corporation of negative force observations into the Gaussian process regression (GPR) model,<br> resulting in GPR with derivative observations, demonstrates the remarkable capability to<br> attain high-quality potential energy surfaces, accurate Cartesian force descriptions, and reli-<br> able free energy profiles using a training set of just 80 points. Secondly, the adaptation of the<br> sparse streaming GPR algorithm showcases the potential of memory retention from previous<br> phasespace, enabling energy-only models to converge using simple descriptors while faith-<br> fully reproducing high-quality potential energy surfaces and accurate free energy profiles.<br> Thirdly, the utilization of GPR with atomic environmental vectors as input features proves<br> effective in enhancing both potential energy surface and free energy description. Further-<br> more, incorporating derivative information on solute atoms further improves the accuracy<br> of force predictions on molecular mechanical (MM) atoms, addressing discrepancies arising<br> from QM/MM interaction energies between the target and base levels of theory. Finally, a<br> comprehensive comparison of three distinct GPR schemes, namely GAP, GPR with an aver-<br> age kernel, and GPR with a system-specific sum kernel, is conducted to evaluate the impact<br> of permutational invariance and atomistic learning on the model’s quality. Additionally, this<br> dissertation introduces the adaptation of the GAP method to be compatible with the sparse<br> variational Gaussian processes scheme and the streaming sparse GPR scheme, enhancing<br> their efficiency and applicability. Through these four ∆-Machine learning schemes, this dis-<br> sertation makes significant contributions to the field of computational chemistry, advancing<br> the quest for accurate potential energy surfaces, reliable force descriptions, and informative<br> free energy profiles in QM/MM simulations.<br> </p>

In this thesis, I present a decentralized sparse Gaussian process regression (DSGPR) model with event-triggered, adaptive inducing points. This DSGPR model brings the advantages of sparse Gaussian process regression to a decentralized implementation. Being decentralized and sparse provides advantages that are ideal for multi-agent systems (MASs) performing environmental modeling. In this case, MASs need to model large amounts of information while having potential intermittent communication connections. Additionally, the model needs to correctly perform uncertainty propagation between autonomous agents and ensure high accuracy on the prediction. For the model to meet these requirements, a bounded and efficient real-time sparse Gaussian process regression (SGPR) model is needed. I improve real-time SGPR models in these regards by introducing an adaptation of the mean shift and fixed-width clustering algorithms called radial clustering. Radial clustering enables real-time SGPR models to have an adaptive number of inducing points through an efficient inducing point selection process. I show how this clustering approach scales better than other seminal Gaussian process regression (GPR) and SGPR models for real-time purposes while attaining similar prediction accuracy and uncertainty reduction performance. Furthermore, this thesis addresses common issues inherent in decentralized frameworks such as high computation costs, inter-agent message bandwidth restrictions, and data fusion integrity. These challenges are addressed in part through performing maximum consensus between local agent models which enables the MAS to gain the advantages of decentral- ization while keeping data fusion integrity. The inter-agent communication restrictions are addressed through the contribution of two message passing heuristics called the covariance reduction heuristic and the Bhattacharyya distance heuristic. These heuristics enable user to reduce message passing frequency and message size through the Bhattacharyya distance and properties of spatial kernels. The entire DSGPR framework is evaluated on multiple simulated random vector fields. The results show that this framework effectively estimates vector fields using multiple autonomous agents. This vector field is assumed to be a wind field; however, this framework may be applied to the estimation of other scalar or vector fields (e.g., fluids, magnetic fields, electricity, etc.).

Work in this dissertation is motivated by reducing the design cost at the early design stage while maintaining high design accuracy throughout all design stages. It presents four key design methods to improve the performance of Efficient Global Optimization for multidisciplinary problems. First, a fidelity-calibration method is developed and applied to lower-fidelity samples. Function values analyzed by lower fidelity analysis methods are updated to have equivalent accuracy to that of the highest fidelity samples, and these calibrated data sets are used to construct a variable-fidelity Kriging model. For the design of experiment (DOE), a dynamic sampling method is developed and includes filtering and infilling data based on mathematical criteria on the model accuracy. In the sample infilling process, multi-objective optimization for exploitation and exploration of design space is carried out. To indicate the fidelity of function analysis for additional samples in the variable-fidelity Kriging model, a dynamic fidelity indicator with the overlapping coefficient is proposed. For the multidisciplinary design problems, where multiple physics are tightly coupled with different coupling strengths, multi-response Kriging model is introduced and utilizes the method of iterative Maximum Likelihood Estimation (iMLE). Through the iMLE process, a large number of hyper-parameters in multi-response Kriging can be calculated with great accuracy and improved numerical stability. The optimization methods developed in the study are validated with analytic functions and showed considerable performance improvement. Consequentially, three practical design optimization problems of NACA0012 airfoil, Multi-element NLR 7301 airfoil, and all-moving-wingtip control surface of tailless aircraft are performed, respectively. The results are compared with those of existing methods, and it is concluded that these methods guarantee the equivalent design accuracy at computational cost reduced significantly. / Doctor of Philosophy / In recent years, as the cost of aircraft design is growing rapidly, and aviation industry is interested in saving time and cost for the design, an accurate design result during the early design stages is particularly important to reduce overall life cycle cost. The purpose of the work to reducing the design cost at the early design stage with design accuracy as high as that of the detailed design. The method of an efficient global optimization (EGO) with variable-fidelity analysis and multidisciplinary design is proposed. Using the variable-fidelity analysis for the function evaluation, high fidelity function evaluations can be replaced by low-fidelity analyses of equivalent accuracy, which leads to considerable cost reduction. As the aircraft system has sub-disciplines coupled by multiple physics, including aerodynamics, structures, and thermodynamics, the accuracy of an individual discipline affects that of all others, and thus the design accuracy during in the early design states. Four distinctive design methods are developed and implemented into the standard Efficient Global Optimization (EGO) framework: 1) the variable-fidelity analysis based on error approximation and calibration of low-fidelity samples, 2) dynamic sampling criteria for both filtering and infilling samples, 3) a dynamic fidelity indicator (DFI) for the selection of analysis fidelity for infilled samples, and 4) Multi-response Kriging model with an iterative Maximum Likelihood estimation (iMLE). The methods are validated with analytic functions, and the improvement in cost efficiency through the overall design process is observed, while maintaining the design accuracy, by a comparison with existing design methods. For the practical applications, the methods are applied to the design optimization of airfoil and complete aircraft configuration, respectively. The design results are compared with those by existing methods, and it is found the method results design results of accuracies equivalent to or higher than high-fidelity analysis-alone design at cost reduced by orders of magnitude.