Matching Items (39)
Description
Due to great challenges from aggressive environmental regulations, increased demand due to new technologies and the integration of renewable energy sources, the energy industry may radically change the way the power system is operated and designed. With the motivation of studying and planning the future power system under these new challenges, the development of the new tools is required. A network equivalent that can be used in such planning tools needs to be generated based on an accurate power flow model and an equivalencing procedure that preserves the key characteristics of the original system. Considering the pervasive use of the dc power flow models, their accuracy is of great concern. The industry seems to be sanguine about the performance of dc power flow models, but recent research has shown that the performance of different formulations is highly variable. In this thesis, several dc power-flow models are analyzed theoretically and evaluated numerically in IEEE 118-bus system and Eastern Interconnection 62,000-bus system. As shown in the numerical example, the alpha-matching dc power flow model performs best in matching the original ac power flow solution. Also, the possibility of applying these dc models in the various applications has been explored and demonstrated. Furthermore, a novel hot-start optimal dc power-flow model based on ac power transfer distribution factors (PTDFs) is proposed, implemented and tested. This optimal-reactance-only dc model not only matches the original ac PF solution well, but also preserves the congestion pattern obtain from the OPF results of the original ac model. Three improved strategies were proposed for applying the bus-aggregation technique to the large-scale systems, like EI and ERCOT, to improve the execution time, and memory requirements when building a reduced equivalent model. Speed improvements of up to a factor of 200 were observed.
ContributorsQi, Yingying (Author) / Tylavsky, Daniel J (Thesis advisor) / Hedman, Kory W (Committee member) / Sankar, Lalitha (Committee member) / Arizona State University (Publisher)
Created2013
Description
The subject of this thesis is distribution level load management using a pricing signal in a Smart Grid infrastructure. The Smart Grid implements advanced meters, sensory devices and near real time communication between the elements of the system, including the distribution operator and the customer. A stated objective of the Smart Grid is to use sensory information to operate the electrical power grid more efficiently and cost effectively. One potential function of the Smart Grid is energy management at the distribution level, namely at the individual customer. The Smart Grid allows control of distribution level devices, including distributed energy storage and distributed generation, in operational real time. One method of load control uses an electric energy price as a control signal. The control is achieved through customer preference as the customer allows loads to respond to a dynamic pricing signal. In this thesis, a pricing signal is used to control loads for energy management at the distribution level. The model for the energy management system is created and analyzed in the z-domain due to the envisioned discrete time implementation. Test cases are used to illustrate stability and performance by analytic calculations using Mathcad and by simulation using Matlab Simulink. The envisioned control strategy is applied to the Future Renewable Electric Energy Distribution Management (FREEDM) system. The FREEDM system implements electronic (semiconductor) controls and therefore makes the proposed energy management feasible. The pricing control strategy is demonstrated to be an effective method of performing energy management in a distribution system. It is also shown that stability and near optimal response can be achieved by controlling the parameters of the system. Addition-ally, the communication bandwidth requirements for a pricing control signal are evaluated.
ContributorsBoyd, Jesse (Author) / Heydt, Gerald T (Thesis advisor) / Datta, Rajib (Committee member) / Sankar, Lalitha (Committee member) / Arizona State University (Publisher)
Created2013
Description
Through the personal experience of volunteering at ASU Project Humanities, an organization that provides resources such as clothing and toiletries to the homeless population in Downtown Phoenix, I noticed efficiently serving the needs of the homeless population is an important endeavor, but the current processes for Phoenix nonprofits to collect data are manual, ad-hoc, and inefficient. This leads to the research question: is it possible to improve this process of collecting statistics on client needs, tracking donations, and managing resources using technology? Background research includes an interview with ASU Project Humanities, articles by analysts, and related work including case studies of current technologies in the nonprofit community. Major findings include i) a lack of centralized communication in nonprofits collecting needs, tracking surplus donations, and sharing resources, ii) privacy assurance is important to homeless individuals, and iii) pre-existing databases and technological solutions have demonstrated that technology has the ability to make an impact in the nonprofit community. To improve the process, standardization, efficiency, and automation need to increase. As a result of my analysis, the thesis proposes a prototype solution which includes two parts: an inventory database and a web application with forms for user input and tables for the user to view. This solution addresses standardization by showing a consistent way of collecting data on need requests and surplus donations while guaranteeing privacy of homeless individuals. This centralized solution also increases efficiency by connecting different agencies that cater to these clients. Lastly, the solution demonstrates the ability for resources to be made available to each organization which can increase automation. In conclusion, this database and web application has the potential to improve nonprofit organizations’ networking capabilities, resource management, and resource distribution. The percentile of homeless individuals connected to these resources is expected to increase substantially with future live testing and large-scale implementation.
ContributorsKhurana, Baani Kaur (Author) / Bazzi, Rida (Thesis director) / Sankar, Lalitha (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Department of Information Systems (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
As the field of machine learning increasingly provides real value to power system operations, the availability of rich measurement datasets has become crucial for the development of new applications and technologies. This dissertation focuses on the use of time-series load data for the design of novel data-driven algorithms. Loads are one of the main factors driving the behavior of a power system and they depend on external phenomena which are not captured by traditional simulation tools. Thus, accurate models that capture the fundamental characteristics of time-series load dataare necessary.
In the first part of this dissertation, an example of successful application of machine learning algorithms that leverage load data is presented. Prior work has shown that power systems energy management systems are vulnerable to false data injection attacks against state estimation. Here, a data-driven approach for the detection and localization of such attacks is proposed. The detector uses historical data to learn the normal behavior of the loads in a system and subsequently identify if any of the real-time observed measurements are being manipulated by an attacker.
The second part of this work focuses on the design of generative models for time-series load data. Two separate techniques are used to learn load behaviors from real datasets and exploiting them to generate realistic synthetic data. The first approach is based on principal component analysis (PCA), which is used to extract common temporal patterns from real data. The second method leverages conditional generative adversarial networks (cGANs) and it overcomes the limitations of the PCA-based model while providing greater and more nuanced control on the generation of specific
types of load profiles. Finally, these two classes of models are combined in a multi-resolution generative scheme which is capable of producing any amount of time-series load data at any sampling resolution, for lengths ranging from a few seconds to years.
ContributorsPinceti, Andrea (Author) / Sankar, Lalitha (Thesis advisor) / Kosut, Oliver (Committee member) / Pal, Anamitra (Committee member) / Weng, Yang (Committee member) / Arizona State University (Publisher)
Created2021
Description
Infectious diseases spread at a rapid rate, due to the increasing mobility of the human population. It is important to have a variety of containment and assessment strategies to prevent and limit their spread. In the on-going COVID-19 pandemic, telehealth services including daily health surveys are used to study the prevalence and severity of the disease. Daily health surveys can also help to study the progression and fluctuation of symptoms as recalling, tracking, and explaining symptoms to doctors can often be challenging for patients. Data aggregates collected from the daily health surveys can be used to identify the surge of a disease in a community. This thesis enhances a well-known boosting algorithm, XGBoost, to predict COVID-19 from the anonymized self-reported survey responses provided by Carnegie Mellon University (CMU) - Delphi research group in collaboration with Facebook. Despite the tremendous COVID-19 surge in the United States, this survey dataset is highly imbalanced with 84% negative COVID-19 cases and 16% positive cases. It is tedious to learn from an imbalanced dataset, especially when the dataset could also be noisy, as seen commonly in self-reported surveys. This thesis addresses these challenges by enhancing XGBoost with a tunable loss function, ?-loss, that interpolates between the exponential loss (? = 1/2), the log-loss (? = 1), and the 0-1 loss (? = ∞). Results show that tuning XGBoost with ?-loss can enhance performance over the standard XGBoost with log-loss (? = 1).
ContributorsVikash Babu, Gokulan (Author) / Sankar, Lalitha (Thesis advisor) / Berisha, Visar (Committee member) / Zhao, Ming (Committee member) / Trieu, Ni (Committee member) / Arizona State University (Publisher)
Created2021
Description
The introduction of parameterized loss functions for robustness in machine learning has led to questions as to how hyperparameter(s) of the loss functions can be tuned. This thesis explores how Bayesian methods can be leveraged to tune such hyperparameters. Specifically, a modified Gibbs sampling scheme is used to generate a distribution of loss parameters of tunable loss functions. The modified Gibbs sampler is a two-block sampler that alternates between sampling the loss parameter and optimizing the other model parameters. The sampling step is performed using slice sampling, while the optimization step is performed using gradient descent. This thesis explores the application of the modified Gibbs sampler to alpha-loss, a tunable loss function with a single parameter $\alpha \in (0,\infty]$, that is designed for the classification setting. Theoretically, it is shown that the Markov chain generated by a modified Gibbs sampling scheme is ergodic; that is, the chain has, and converges to, a unique stationary (posterior) distribution. Further, the modified Gibbs sampler is implemented in two experiments: a synthetic dataset and a canonical image dataset. The results show that the modified Gibbs sampler performs well under label noise, generating a distribution indicating preference for larger values of alpha, matching the outcomes of previous experiments.
ContributorsCole, Erika Lingo (Author) / Sankar, Lalitha (Thesis advisor) / Lan, Shiwei (Thesis advisor) / Pedrielli, Giulia (Committee member) / Hahn, Paul (Committee member) / Arizona State University (Publisher)
Created2022
Description
Modern physical systems are experiencing tremendous evolutions with growing size, more and more complex structures, and the incorporation of new devices. This calls for better planning, monitoring, and control. However, achieving these goals is challenging since the system knowledge (e.g., system structures and edge parameters) may be unavailable for a normal system, let alone some dynamic changes like maintenance, reconfigurations, and events, etc. Therefore, extracting system knowledge becomes a central topic. Luckily, advanced metering techniques bring numerous data, leading to the emergence of Machine Learning (ML) methods with efficient learning and fast inference. This work tries to propose a systematic framework of ML-based methods to learn system knowledge under three what-if scenarios: (i) What if the system is normally operated? (ii) What if the system suffers dynamic interventions? (iii) What if the system is new with limited data? For each case, this thesis proposes principled solutions with extensive experiments. Chapter 2 tackles scenario (i) and the golden rule is to learn an ML model that maintains physical consistency, bringing high extrapolation capacity for changing operational conditions. The key finding is that physical consistency can be linked to convexity, a central concept in optimization. Therefore, convexified ML designs are proposed and the global optimality implies faithfulness to the underlying physics. Chapter 3 handles scenario (ii) and the goal is to identify the event time, type, and locations. The problem is formalized as multi-class classification with special attention to accuracy and speed. Subsequently, Chapter 3 builds an ensemble learning framework to aggregate different ML models for better prediction. Next, to tackle high-volume data quickly, a tensor as the multi-dimensional array is used to store and process data, yielding compact and informative vectors for fast inference. Finally, if no labels exist, Chapter 3 uses physical properties to generate labels for learning. Chapter 4 deals with scenario (iii) and a doable process is to transfer knowledge from similar systems, under the framework of Transfer Learning (TL). Chapter 4 proposes cutting-edge system-level TL by considering the network structure, complex spatial-temporal correlations, and different physical information.
ContributorsLi, Haoran (Author) / Weng, Yang (Thesis advisor) / Tong, Hanghang (Committee member) / Dasarathy, Gautam (Committee member) / Sankar, Lalitha (Committee member) / Arizona State University (Publisher)
Created2022
Description
Event identification is increasingly recognized as crucial for enhancing the reliability, security, and stability of the electric power system. With the growing deployment of Phasor Measurement Units (PMUs) and advancements in data science, there are promising opportunities to explore data-driven event identification via machine learning classification techniques. This dissertation explores the potential of data-driven event identification through machine learning classification techniques. In the first part of this dissertation, using measurements from multiple PMUs, I propose to identify events by extracting features based on modal dynamics. I combine such traditional physics-based feature extraction methods with machine learning to distinguish different event types.Using the obtained set of features, I investigate the performance of two well-known classification models, namely, logistic regression (LR) and support vector machines (SVM) to identify generation loss and line trip events in two datasets. The first dataset is obtained from simulated events in the Texas 2000-bus synthetic grid. The second is a proprietary dataset with labeled events obtained from a large utility in the USA.
My results indicate that the proposed framework is promising for identifying the two types of events in the supervised setting. In the second part of the dissertation, I use semi-supervised learning techniques, which make use of both labeled and unlabeled samples.I evaluate three categories of classical semi-supervised approaches: (i) self-training, (ii) transductive support vector machines (TSVM), and (iii) graph-based label spreading (LS) method. In particular, I focus on the identification of four event classes i.e., load loss, generation loss, line trip, and bus fault. I have developed and publicly shared a comprehensive Event Identification package which consists of three aspects: data generation, feature extraction, and event identification with limited labels using semi-supervised methodologies. Using this package, I generate eventful PMU data for the South Carolina 500-Bus synthetic network. My evaluation confirms that the integration of additional unlabeled samples and the utilization of LS for pseudo labeling surpasses the outcomes achieved by the self-training and TSVM approaches. Moreover, the LS algorithm consistently enhances the performance of all classifiers more robustly.
ContributorsTaghipourbazargani, Nima (Author) / Kosut, Oliver (Thesis advisor) / Sankar, Lalitha (Committee member) / Pal, Anamitra (Committee member) / Dasarathy, Gautam (Committee member) / Arizona State University (Publisher)
Created2023
Description
In recent years, the proliferation of deep neural networks (DNNs) has revolutionized the field of artificial intelligence, enabling advancements in various domains. With the emergence of efficient learning techniques such as quantization and distributed learning, DNN systems have become increasingly accessible for deployment on edge devices. This accessibility brings significant benefits, including real-time inference on the edge, which mitigates communication latency, and on-device learning, which addresses privacy concerns and enables continuous improvement. However, the resource limitations of edge devices pose challenges in equipping them with robust safety protocols, making them vulnerable to various attacks. Two notable attacks that affect edge DNN systems are Bit-Flip Attacks (BFA) and architecture stealing attacks. BFA compromises the integrity of DNN models, while architecture stealing attacks aim to extract valuable intellectual property by reverse engineering the model's architecture. Furthermore, in Split Federated Learning (SFL) scenarios, where training occurs on distributed edge devices, Model Inversion (MI) attacks can reconstruct clients' data, and Model Extraction (ME) attacks can extract sensitive model parameters. This thesis aims to address these four attack scenarios and develop effective defense mechanisms. To defend against BFA, both passive and active defensive strategies are discussed. Furthermore, for both model inference and training, architecture stealing attacks are mitigated through novel defense techniques, ensuring the integrity and confidentiality of edge DNN systems. In the context of SFL, the thesis showcases defense mechanisms against MI attacks for both supervised and self-supervised learning applications. Additionally, the research investigates ME attacks in SFL and proposes countermeasures to enhance resistance against potential ME attackers. By examining and addressing these attack scenarios, this research contributes to the security and privacy enhancement of edge DNN systems. The proposed defense mechanisms enable safer deployment of DNN models on resource-constrained edge devices, facilitating the advancement of real-time applications, preserving data privacy, and fostering the widespread adoption of edge computing technologies.
ContributorsLi, Jingtao (Author) / Chakrabarti, Chaitali (Thesis advisor) / Fan, Deliang (Committee member) / Cao, Yu (Committee member) / Trieu, Ni (Committee member) / Arizona State University (Publisher)
Created2023
Description
Generative Adversarial Networks (GANs) have emerged as a powerful framework for generating realistic and high-quality data. In the original ``vanilla'' GAN formulation, two models -- the generator and discriminator -- are engaged in a min-max game and optimize the same value function. Despite offering an intuitive approach, vanilla GANs often face stability challenges such as vanishing gradients and mode collapse. Addressing these common failures, recent work has proposed the use of tunable classification losses in place of traditional value functions. Although parameterized robust loss families, e.g. $\alpha$-loss, have shown promising characteristics as value functions, this thesis argues that the generator and discriminator require separate objective functions to achieve their different goals. As a result, this thesis introduces the $(\alpha_{D}, \alpha_{G})$-GAN, a parameterized class of dual-objective GANs, as an alternative approach to the standard vanilla GAN. The $(\alpha_{D}, \alpha_{G})$-GAN formulation, inspired by $\alpha$-loss, allows practitioners to tune the parameters $(\alpha_{D}, \alpha_{G}) \in [0,\infty)^{2}$ to provide a more stable training process. The objectives for the generator and discriminator in $(\alpha_{D}, \alpha_{G})$-GAN are derived, and the advantages of using these objectives are investigated. In particular, the optimization trajectory of the generator is found to be influenced by the choice of $\alpha_{D}$ and $\alpha_{G}$. Empirical evidence is presented through experiments conducted on various datasets, including the 2D Gaussian Mixture Ring, Celeb-A image dataset, and LSUN Classroom image dataset. Performance metrics such as mode coverage and Fréchet Inception Distance (FID) are used to evaluate the effectiveness of the $(\alpha_{D}, \alpha_{G})$-GAN compared to the vanilla GAN and state-of-the-art Least Squares GAN (LSGAN). The experimental results demonstrate that tuning $\alpha_{D} < 1$ leads to improved stability, robustness to hyperparameter choice, and competitive performance compared to LSGAN.
ContributorsOtstot, Kyle (Author) / Sankar, Lalitha (Thesis advisor) / Kosut, Oliver (Committee member) / Pedrielli, Giulia (Committee member) / Arizona State University (Publisher)
Created2023