Big Data Analytics for Wind Energy Integration: Spatio-Temporal Wind Power Analysis and Stochastic Optimization
Dr. Lei Yang,
Department of Computer Science and Engineering
University of Nevada, Reno
With the increasing penetration of wind into bulk power systems, wind generation has posed a significant challenge to system operators due to the highly variable wind generation. Reliable system operations require accurate wind forecast, especially at the high penetration level of wind generation. In this talk, short-term forecast of wind farm generation is investigated by applying spatio-temporal analysis to extensive measurement data collected from a large wind farm. Specifically, using the data of the wind turbines power outputs recorded across two consecutive years, multiple finite-state Markov chains that take into account the diurnal non-stationarity and the seasonality of wind generation are first developed to capture the fast fluctuations of small amounts of wind generation. To capture the wind ramp dynamics, SVM is employed, based on one key observation from the measurement data that wind ramps always occur with specific patterns. Then, the forecast by the SVM is integrated into each finite-state Markov chain. Based on the SVM enhanced Markov model, short-term distributional forecasts and point forecasts are then derived. The distributional forecast can be utilized to study stochastic unit commitment and economic dispatch problems by using a Markovian approach. Numerical test results, via using realistic wind farm data provided by the National Renewable Energy Laboratory (NREL), demonstrate the significant improved accuracy of the proposed forecast approach.
Lei Yang received the B.S. and M.S. degrees in electrical engineering from Southeast University, Nanjing, China, in 2005 and 2008, respectively, and the Ph.D. degree from the School of Electrical Computer and Energy Engineering, Arizona State University, Tempe, AZ, USA, in 2012. He was a Postdoctoral Scholar with Princeton University, Princeton, NJ, USA, and an Assistant Research Professor with the School of Electrical Computer and Energy Engineering, Arizona State University. He is currently an Assistant Professor with the Department of Computer Science and Engineering, University of Nevada, Reno, NV, USA. His research interests include big data analytics, edge computing and its applications in IoT and 5G, stochastic optimization and modeling in smart cities and cyber-physical systems, data privacy and security in crowdsensing, and optimization and control in mobile social networks. He was a recipient of the Best Paper Award Runner-up at the IEEE INFOCOM 2014. He is currently associate editor for IEEE Access.