Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Meteorology, 1979. / Microfiche copy available in Archives and Science. / Bibliography : leaves 45-48. / by Clare Bertram Billing, Jr. / M.S.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第18699号 / 情博第549号 / 新制||情||97(附属図書館) / 31632 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 阿久津 達也, 教授 鹿島 久嗣, 教授 石井 信 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第19283号 / 工博第4080号 / 新制||工||1629(附属図書館) / 32285 / 京都大学大学院工学研究科社会基盤工学専攻 / (主査)教授 田村 正行, 教授 小池 克明, 准教授 須﨑 純一 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第20705号 / 工博第4402号 / 新制||工||1684(附属図書館) / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 引原 隆士, 教授 土居 伸二, 教授 松尾 哲司 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

No description available.

In today’s world, where air pollution has become a ubiquitous problem, city air is normally monitored. Such monitoring can produce large amounts of data, and this enables the development of statistical and machine learning techniques for modeling and forecasting air quality. However, the complex nature of air pollution makes such data a challenge to fully utilize. To this end, machine learning methods, especially deep neural networks, have in recent years emerged as a promising technology for more accurate predictions of air pollution levels, and the research problem in this work is; To capture and model the complex dynamics of air pollution with machine learning methods, with an emphasis on deep neural networks. Connected to the research problem is the research question; How can machine learning, in particular deep neural networks, be used to forecast air pollution levels and pollution peaks? An emphasis is put on pollution peaks, as these are the episodes when existing forecasting models tend to give the largest prediction errors. In this work, historical data from air monitoring sensors were utilized to train several neural network architectures, as well as a more straightforward multiple linear regression model, for forecasting background levels of nitrogen dioxide in the center of Stockholm. Several evaluation metrics showed that the neural network models outperformed the multiple linear regression model, however, none of the models had the desired structure of the forecast errors, and all models failed to successfully capture sudden pollution peaks. Nevertheless, the results point to an advantage for the more complex neural network models, and further advances in the field of machine learning, together with higher resolution data, have the potential to improve air quality forecasts even more and cross conventional forecasting limits.

No description available.

[pt] Muito freqüentemente, as técnicas utilizadas na identificação de processos estocásticos conduzem a mais de um modelo passível de ser utilizado na caracterização do processo. O problema de escolher entre estes modelos é formulado como um problema de teste de hipóteses, e o teste de razão de verossimilhança é a ele aplicado. Considera-se então a situação particular onde se quer descrever processos de parâmetro discreto (séries temporais) através de modelos ARIMA (autoregressive Integrated Moving Average). O teste de razão de verossimilhança associado ao problema é então deduzido e implementado através do algoritmo de Kalman-Bucy. Comparações com um outro teste usualmente empregado na escolha de modelos para séries temporais mostram a superioridade do teste de razão de verossimilhança. / [en] Very often random process identification techniques lead to several prospective models to characterize the process. The problem of choosing among these models is cast as a hypothesis testing problem, to which a likelihood ratio test is applied. For the special situation in which a choice between two autoregressive integrated moving average models is to made, likelihood ratio is derived and afterwards implemented through the Kalman-Bucy algorithm. Comparisons with another procedure usually connected to time series model choices show likelihood ratio tests are definetely superior.

Sensing and analysis of a structure for the purpose of detecting, tracking, and evaluating damage and deterioration, during both regular operation and extreme events, is referred to as Structural Health Monitoring (SHM). SHM is a multi-disciplinary field, with a complete system incorporating sensing technology, hardware, signal processing, networking, data analysis, and management for interpretation and decision making. However, many of these processes and subsequent integration into a practical SHM framework are in need of development. In this study, various components of an SHM system will be investigated. A particular focus is paid to the investigation of a previously developed damage detection methodology for global condition assessment of a laboratory structure with a decking system. First, a review of some of the current SHM applications, which relate to a current UCF Structures SHM study monitoring a full-scale movable bridge, will be presented in conjunction with a summary of the critical components for that project. Studies for structural condition assessment of a 4-span bridge-type steel structure using the SHM data collected from laboratory based experiments will then be presented. For this purpose, a time series analysis method using ARX models (Auto-Regressive models with eXogeneous input) for damage detection with free response vibration data will be expanded upon using both wired and wireless acceleration data. Analysis using wireless accelerometers will implement a sensor roaming technique to maintain a dense sensor field, yet require fewer sensors. Using both data types, this ARX based time series analysis method was shown to be effective for damage detection and localization for this relatively complex laboratory structure. Finally, application of the proposed methodologies on a real-life structure will be discussed, along with conclusions and recommendations for future work

Many methods have been used to forecast real-time electricity prices in various regions around the world. The problem is difficult because of market volatility affected by a wide range of exogenous variables from weather to natural gas prices, and accurate price forecasting could help both suppliers and consumers plan effective business strategies. Statistical analysis with autoregressive moving average methods and computational intelligence approaches using artificial neural networks dominate the landscape. With the rise in popularity of convolutional neural networks to handle problems with large numbers of inputs, and convolutional neural networks conspicuously lacking from current literature in this field, convolutional neural networks are used for this time series forecasting problem and show some promising results. This document fulfills both MSEE Master's Thesis and BSCPE Senior Project requirements.