The aim of this thesis is to design a system for automated accident detection in intersections. The input to the system is a three-second audio signal. The system can be operated in two modes: two-class and multi-class. The output of the two-class system is a label of ?crash? or ?non-crash?. In the multi-class system, the output is the label of ?crash? or various non-crash incidents including ?pile drive?, ?brake?, and ?normal-traffic? sounds. The system designed has three main steps in processing the input audio signal. They are: feature extraction, feature optimization and classification. Five different methods of feature extraction are investigated and compared; they are based on the discrete wavelet transform, fast Fourier transform, discrete cosine transform, real cepstrum transform and Mel frequency cepstral transform. Linear discriminant analysis (LDA) is used to optimize the features obtained in the feature extraction stage by linearly combining the features using different weights. Three types of statistical classifiers are investigated and compared: the nearest neighbor, nearest mean, and maximum likelihood methods. Data collected from Jackson, MS and Starkville, MS and the crash signals obtained from Texas Transportation Institute crash test facility are used to train and test the designed system. The results showed that the wavelet based feature extraction method with LDA and maximum likelihood classifier is the optimum design. This wavelet-based system is computationally inexpensive compared to other methods. The system produced classification accuracies of 95% to 100% when the input signal has a signal-to-noise-ratio of at least 0 decibels. These results show that the system is capable of effectively classifying ?crash? or ?non-crash? on a given input audio signal.

No description available.

No description available.

With the deployment of sensors in hardware equipment and advanced metering infrastructure, system operators have access to unprecedented amounts of data. Simultaneously, grid-connected power electronics technology has had a large impact on the way electrical energy is generated, transmitted, and delivered to consumers. Artificial intelligence and machine learning can help address the new power grid challenges with enhanced computational abilities and access to large amounts of data. This thesis discusses the fundamentals of neural networks and their applications in power systems such as load forecasting, power system stability analysis, and fault diagnosis. It extends application of neural networks to inverter-based resources by studying the implementation and performance of a neural network controller emulator for voltage-sourced converters. It delves into how neural networks could enhance cybersecurity of a component through multiple hardware and software implementations of the same component. This ensures that vulnerabilities inherent in one form of implementation do not affect the system as a whole. The thesis also proposes a comprehensive support vector classifier (SVC)--based submodule open-circuit fault detection and localization method for modular multilevel converters. This method eliminates the need for extra hardware. Its efficacy is discussed through simulation studies in PSCAD/EMTDC software. To ensure efficient usage of neural networks in power system simulation softwares, this thesis entails the step by step implementation of a neural network custom component in PSCAD/EMTDC. The custom component simplifies the process of recreating a neural network in PSCAD/EMTDC by eliminating the manual assembly of predefined library components such as summers, multipliers, comparators, and other miscellaneous blocks. / Master of Science / Data analytics and machine learning play an important role in the power grids of today, which are continuously evolving with the integration of renewable energy resources. It is expected that by 2030 most of the electric power generated will be processed by some form of power electronics, e.g., inverters, from the point of its generation. Machine learning has been applied to various fields of power systems such as load forecasting, stability analysis, and fault diagnosis. This work extends machine learning applications to inverter-based resources by using artificial neural networks to perform controller emulation for an inverter, provide cybersecurity through heterogeneity, and perform submodule fault detection in modular multilevel converters. The thesis also discusses the step by step implementation of a neural network custom component in PSCAD/EMTDC software. This custom component simplifies the process of creating a neural network in PSCAD/EMTDC by eliminating the manual assembly of predefined library components.

Yes / In the last few years, there has been growing interest in near-real-time solar data processing, especially for space weather applications. This is due to space weather impacts on both space-borne and ground-based systems, and industries, which subsequently impacts our lives. In the current study, the deep learning approach is used to establish an automated hybrid computer system for a short-term forecast; it is achieved by using the complexity level of the sunspot group on SDO/HMI Intensitygram images. Furthermore, this suggested system can generate the forecast for solar flare occurrences within the following 24 h. The input data for the proposed system are SDO/HMI full-disk Intensitygram images and SDO/HMI full-disk magnetogram images. System outputs are the “Flare or Non-Flare” of daily flare occurrences (C, M, and X classes). This system integrates an image processing system to automatically detect sunspot groups on SDO/HMI Intensitygram images using active-region data extracted from SDO/HMI magnetogram images (presented by Colak and Qahwaji, 2008) and deep learning to generate these forecasts. Our deep learning-based system is designed to analyze sunspot groups on the solar disk to predict whether this sunspot group is capable of releasing a significant flare or not. Our system introduced in this work is called ASAP_Deep. The deep learning model used in our system is based on the integration of the Convolutional Neural Network (CNN) and Softmax classifier to extract special features from the sunspot group images detected from SDO/HMI (Intensitygram and magnetogram) images. Furthermore, a CNN training scheme based on the integration of a back-propagation algorithm and a mini-batch AdaGrad optimization method is suggested for weight updates and to modify learning rates, respectively. The images of the sunspot regions are cropped automatically by the imaging system and processed using deep learning rules to provide near real-time predictions. The major results of this study are as follows. Firstly, the ASAP_Deep system builds on the ASAP system introduced in Colak and Qahwaji (2009) but improves the system with an updated deep learning-based prediction capability. Secondly, we successfully apply CNN to the sunspot group image without any pre-processing or feature extraction. Thirdly, our system results are considerably better, especially for the false alarm ratio (FAR); this reduces the losses resulting from the protection measures applied by companies. Also, the proposed system achieves a relatively high scores for True Skill Statistics (TSS) and Heidke Skill Score (HSS).

This project involved the design and testing of a prototype in-line chip quality monitor for gathering process control information for the manufacturers of wood chips. This monitor specifically addresses three common complaints with current chip sampling procedures. Chip sampling occurs too late in the process. It is inadequate. It is too infrequent to develop management information. The monitor is composed of a double screen drum separator to divide chips into oversize, accepts, and pins/fines. Counterbalanced tip buckets are used to weigh each size class. Tip bucket cycles are recorded by a computer via magnetic proximity switches attached to each bucket. This information is then used to chart production of chip size classes, updated continuously over the sorting period. This monitor is capable of sorting one ton of chips per hour. Two trials were conducted to test the monitor. One in a lab environment, and one on site at a chip mill. Both trials compared monitor output with independent samples classified using a Williams classifier. The trials showed that outputs were consistent with Williams output. This monitor can effectively chart chip distribution information. This process control information provides the manufacturer with immediate knowledge of chipper performance. / Master of Science

Edge intelligence can reduce power dissipation to enable power-hungry long-range wireless applications. This work applies edge intelligence to quantify the reduction in power dissipation. We designed a wireless sensor node with a LoRa radio and implemented a decision tree classifier, in situ, to classify behaviors of cattle. We estimate that employing edge intelligence on our wireless sensor node reduces its average power dissipation by up to a factor of 50, from 20.10 mW to 0.41 mW. We also observe that edge intelligence increases the link budget without significantly affecting average power dissipation. / Master of Science / Battery powered sensor nodes have access to a limited amount of energy. However, many applications of sensor nodes such as animal monitoring require energy intensive, long range data transmissions. In this work, we used machine learning to process motion data within our sensor node to classify cattle behaviors. We estimate that transmitting processed data dissipates up to 50 times less power when compared to transmitting raw data. Due to the properties of our transmission protocol, we also observe that transmitting processed data increases the range of transmissions without impacting power dissipation.

We present the results of applying a Fast Boosted Decision Tree (FBDT) algorithm to the task of distinguishing muons from pions in K-Long and Muon (KLM) detector of the Belle II experiment. Performance was evaluated over a momentum range of 0.6 < p < 5.0 GeV/c by plotting Receiver Operating Characteristic (ROC) curves for 0.1 GeV/c intervals. The FBDT model was worse than the benchmark likelihood ratio test model for the whole momentum range during testing on Monte Carlo (MC) simulated data. This is seen in the lower Area Under the Curve (AUC) values for the FBDT ROC curves, achieving peak AUC values around 0.82, while the likelihood ratio ROC curves achieve peak AUC values around 0.98. Performance of the FBDT model in muon identification may be improved in the future by adding a pre-processing routine for the MC data and input variables. / Master of Science / An important task of a high-energy physics experiment is taking the input information provided by detectors, such as the distance a particle travels through a detector, the momentum, and energy deposits it makes, and using that information to identify the particle's type. In this study we test a machine learning model that sorts the particles observed into two categories—muons and pions—by comparing the particle's input values to a threshold value at multiple stages, then assigns a final identity to the particle at the last stage. This is compared to a benchmark model that uses the probabilities that these input variables would be seen from a particle of each type to determine which particle type is most likely. The ability of both models to distinguish muons and pions were tested on simulated data from the Belle II detector, and the benchmark model outperformed the machine learning model.

Mit seiner fokussierten Funktionsweise hat der Mikrobloggingdienst Twitter im Laufe des vergangenen Jahrzehnts eine beachtliche Präsenz als Kommunikationsmedium in diversen Bereichen des Lebens erreicht. Eine besondere Weise, auf die sich die gestiegene Sichtbarkeit Twitters in der täglichen Kommunikation häufig manifestiert, ist die gezielte Verwendung von Hashtags. So nutzen Unternehmen Hashtags um die auf Twitter stattfindenden Diskussionen über ihre Produkte zu bündeln, während Organisatoren von Großveranstaltungen und Fernsehsendungen durch Bekanntgabe ihrer eigenen, offiziellen Hashtags Zuschauer dazu ermutigen, den Dienst parallel zum eigentlichen Event als Diskussionsplattform zu nutzen. [... aus der Einleitung]

Recently computational topology had an important development in data analysis giving birth to the field of Topological Data Analysis. Persistent homology appears as a fundamental tool based on the topology of data that can be represented as points in metric space. In this work, we apply techniques of Topological Data Analysis, more precisely, we use persistent homology to calculate topological features more persistent in data. In this sense, the persistence diagrams are processed as feature vectors for applying Machine Learning algorithms. In order to classification, we used the following classifiers: Partial Least Squares-Discriminant Analysis, Support Vector Machine, and Naive Bayes. For regression, we used Support Vector Regression and KNeighbors. Finally, we will give a certain statistical approach to analyze the accuracy of each classifier and regressor. / Recentemente a topologia computacional teve um importante desenvolvimento na análise de dados dando origem ao campo da Análise Topológica de Dados. A homologia persistente aparece como uma ferramenta fundamental baseada na topologia de dados que possam ser representados como pontos num espaço métrico. Neste trabalho, aplicamos técnicas da Análise Topológica de Dados, mais precisamente, usamos homologia persistente para calcular características topológicas mais persistentes em dados. Nesse sentido, os diagramas de persistencia são processados como vetores de características para posteriormente aplicar algoritmos de Aprendizado de Máquina. Para classificação, foram utilizados os seguintes classificadores: Análise de Discriminantes de Minimos Quadrados Parciais, Máquina de Vetores de Suporte, e Naive Bayes. Para a regressão, usamos a Regressão de Vetores de Suporte e KNeighbors. Finalmente, daremos uma certa abordagem estatística para analisar a precisão de cada classificador e regressor.