Many-Objective Robust Optimization for Dynamic VAR Planning to Enhance Voltage Stability of a Wind-Energy Power System
Yuan Chi,Yan Xu,Rui Zhang
DOI: https://doi.org/10.1109/tpwrd.2020.2982471
IF: 4.825
2021-02-01
IEEE Transactions on Power Delivery
Abstract:Growing integration of wind power and increasing use of induction motor loads dramatically affect a power system's dynamic performance. To address the challenges, this paper proposes a robust dynamic VAR planning method to enhance the voltage stability of wind energy power systems under uncertainty. Firstly, two indices are developed to evaluate the steady-state and the short-term voltage stability of the system. Then, a robustness index is proposed to quantitatively evaluate the robust optimality of the solutions, considering the uncertainties from wind power output and STATCOMs initial state before a contingency. Four objectives are optimized simultaneously: 1) total planning cost, 2) short-term voltage stability index, 3) steady-state voltage stability index, and 4) robustness of the solutions. Finally, an Adaptive Non-dominated Sorting Genetic Algorithm-III based on Latin Hypercube Sampling (A-NSGA-III-LHS)is designed to solve this many-objective optimization problem with 3 improvements over the standard NSGA-III: 1) adaptive mutation rate, 2) Differential Evolution operator, 3) Latin Hypercube Sampling based initial population generation. The proposed method is tested on the New England 39-bus system with an industry-standard complex load model, showing high robust optimality and computational efficiency over conventional methods.
engineering, electrical & electronic