Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Control and Monitoring Strategies of Smart Grids Using Artificial Intelligent Methods
Several new challenges have arisen recently in the operation of power systems. First, the high penetration of renewable resources in distribution systems in the form of microgrids add uncertainty and complexity to the systems. Variability of renewable resources can violate the robustness of microgrids and cause system instability. The second challenge stems from increasing amount of power trading in the restructured power system; thus, systems are being pushed closer to their operations boundaries. In this situation, power system security can become a critical concern, since occurring contingencies in the system can lead to system blackout. The 2008 California blackout serves as an example. The research to follow has a twofold-objective: First, to address the robust and optimal operation of renewable resources in the microgrid by designing microgrid energy management; and Second, to alleviate the higher risk of operating in restructured power markets by developing online models for real-time system monitoring. In the first part, we propose distributed energy management in a two-staged energy market, with considering probability of distributed energy resources outages. The uncertainties involved in the nature of microgrids due to the variability in renewable generation is modeled using iterative Monte Carlo simulations and infused into energy management framework. A reinforcement learning algorithm is created as an AI algorithm to allow generation resources, distributed storages, and customers to develop optimal strategies for energy management and load scheduling in the microgrid system. In the second part, we develop an online power system monitoring module utilizing a state-of-the-art machine learning and a convolutional neural network based on the AI approach. In the online intelligent pattern recognition module, the data streaming of system phase angle, driven from phasor measurement units or a state estimator, is processed to assess power system stress conditions. Power system stress is an indicator about transmission lines overloading. The proposed module can reveal the hidden patterns between phase angles of buses and system stress conditions to provide fast and accurate stress status and predict the severity of system stress. The models proposed can be used to improve microgrid energy management and bulk power system security.
Electrical engineering|Computer Engineering|Artificial intelligence
Foruzan, Elham, "Control and Monitoring Strategies of Smart Grids Using Artificial Intelligent Methods" (2019). ETD collection for University of Nebraska - Lincoln. AAI27667466.