abstract: The students of Arizona State University, under the mentorship of Dr George Karady, have been collaborating with Salt River Project (SRP), a major power utility in the state of Arizona, trying to study and optimize a battery-supported grid-tied rooftop Photovoltaic (PV) system, sold by a commercial vendor. SRP believes this system has the potential to satisfy the needs of its customers, who opt for utilizing solar power to partially satisfy their power needs. An important part of this elaborate project is the development of a new load forecasting algorithm and a better control strategy for the optimized utilization of the storage system. The built-in algorithm of this commercial unit uses simple forecasting and battery control strategies. With the recent improvement in Machine Learning (ML) techniques, development of a more sophisticated model of the problem in hand was possible. This research is aimed at achieving the goal by utilizing the appropriate ML techniques to better model the problem, which will essentially result in a better solution. In this research, a set of six unique features are used to model the load forecasting problem and different ML algorithms are simulated on the developed model. A similar approach is taken to solve the PV prediction problem. Finally, a very effective battery control strategy is built (utilizing the results of the load and PV forecasting), with the aim of ensuring a reduction in the amount of energy consumed from the grid during the “on-peak” hours. Apart from the reduction in the energy consumption, this battery control algorithm decelerates the “cycling aging” or the aging of the battery owing to the charge/dis-charges cycles endured by selectively charging/dis-charging the battery based on need. ii The results of this proposed strategy are verified using a hardware implementation (the PV system was coupled with a custom-built load bank and this setup was used to simulate a house). The results pertaining to the performances of the built-in algorithm and the ML algorithm are compared and the economic analysis is performed. The findings of this research have in the process of being published in a reputed journal. / Dissertation/Thesis / Masters Thesis Electrical Engineering 2018

Electrical distribution networks face many operational challenges as various renewable distributed generation (DG), such as solar photovoltaic (PV) systems and wind, become part of their structure. Unlike conventional distribution systems, where the only unpredictable aspect is the load level, the intermittent nature of DG poses additional uncertainty levels for distribution system operators (DSO). The voltage quality problem considers the most restrictive issue that hinders high DG integration into distribution grids. Voltage deviates from the nominal grid voltage limits due to the excess power from the DG. DSOs are accustomed to improving the voltage profile by optimal adjustments of the on-load tap changers, voltage regulator taps and capacitor banks. Nevertheless, due to the frequent variability of the output energy from DG, these devices may fail in doing the needful. Battery energy storage systems (BESS) and smart PV inverter functionalities are regarded as promising solutions to promote the seamless integration of renewable resources into distribution networks. BESS are utilized to store the surplus energy during the high penetration of renewable DG that causes high voltage levels and discharge the stored energy when the distribution grid is heavily loaded, which leads to the low voltage levels. Smart PV inverters regulate the network voltage by controlling the reactive power injection or absorption at the inverter end. This dissertation proposes a management strategy that coordinates BESS and smart PV inverter reactive power capability to improve voltage quality in the distribution systems with high PV and wind penetrations. The proposed management method is based on a bi-level optimization algorithm consisting of upper and lower optimization levels. The proposed method determines the optimal location, capacity, numbers and BESS charging and discharging rates to support the distribution system voltage and to ensure optimal deployment of BESS. Case studies are conducted to evaluate the proposed voltage control method. The large size PV system and wind turbine impacts are studied and simulated on the modified IEEE-34 bus test feeder. In addition, the proposed method is applied to the modified IEEE low voltage test feeder to investigate the effectiveness of installing residential rooftop PV systems on the distribution system's voltage. Experimental results show promising outcomes of the proposed method in controlling the distribution networks' voltage. In addition, a day-ahead forecast of PV power output is developed in this dissertation to assist the DSOs to accurately predict the future amounts of PV energy available and reinforcing the decision-making process of batteries operation. Hybrid forecasting models are proposed based on machine learning algorithms, which utilize support vector regression and backpropagation neural network, optimized with three metaheuristic optimization algorithms, namely Social Spider Optimization (SSO), Particle Swarm Optimization (PSO) and Cuckoo Search Optimization (CSO). These algorithms are used to improve the predictive efficacy of the selected algorithms, where the optimal selection of their hyperparameters and architectures plays a significant role in yielding precise forecasting outcomes. / Doctor of Philosophy / The need for more renewable energy has grown significantly, and many countries are embracing these technologies. However, the integration of distributed generation (DG), such as PV systems and wind turbines, poses several operational problems to the distribution system. The voltage problem represents the most significant issue that needs to be addressed. The traditional voltage control equipment may not cope with the rapid fluctuation and may impact their service life. The continuous developments in the battery energy storage systems (BESS) and the smart PV inverter technologies result in increasing the hosting capacity of DG. BESS can store the excess power from the distributed generators and supply this energy to the grid for different operational objectives. On the other hand, the advanced PV inverter's reactive power capability can be exploited from which the grid can attain many benefits. This dissertation aims at providing a reliable control method to the voltage profile in distribution networks embedded with high PV and wind energy by optimal coordination between the operation of the BESS and the smart PV inverter. In addition, the solar forecasting can mitigate the uncertainty associated with PV system generation. In this dissertation, the PV power forecasting application is applied in the distribution system to control the voltage. Through utilizing PV power forecasting, the decision-making for battery operation can be upheld and reinforced. The BESS can store the surplus energy from the PV system as needed and supply it back in low PV power incidents. Experimental results indicate that proper coordination between the BESS and smart PV inverter is beneficial for distribution system operation that can seamlessly integrate PV and wind energy.