On-Line Energy Management of Microgrid via Parametric Cost Function Approximation

• Published in: IEEE transactions on power systems Vol. 34; no. 4; pp. 3300 - 3302
• Main Authors: Shuai, Hang, Fang, Jiakun, Ai, Xiaomeng, Yao, Wei, Wen, Jinyu, He, Haibo
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-07-2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Approximation; Batteries; Computational geometry; Convexity; Cost function; Distributed generation; Electric power grids; Energy management; gradient descent; Microgrids; on-line optimization; Operating costs; Optimization; Parameters; Parametric cost function approximation (CFA); Stochastic processes; Uncertainty; uncertainty in microgrid; Wind power generation; Wind turbines
• ISSN: 0885-8950; 1558-0679
• DOI: 10.1109/TPWRS.2019.2912491

This letter proposes a parametric cost function approximation (CFA)-based on-line operation strategy for microgrids, considering the uncertainty from distributed renewable energy. It aims to obtain on-line scheduling solutions with the minimal expectation of the operational cost. For the non-convex optimization problem in this letter, the stochastic gradient descent-based parameter optimization method is proposed to find the optimal parameter used in the parametric CFA policy. The proposed algorithm is demonstrated to be highly effective by using numerical experiments, indicating the high potential for practical applications.