Bayesian Learning-Based Multi-Objective Distribution Power Network Reconfiguration
Tianwei Zhong, Hai‐Tao Zhang, Yuanzheng Li, Lan Liu, Renzhi Lu
Abstract
This article proposes a scheme aiming at solving the reconfiguration problem of distribution power network (DPN) with high wind power penetrations. The virtue of the presented scheme lies in balancing the voltage stability and the absorption rate of wind energy. First, the DPN reconfiguration is formulated as a multi-objective optimization problem, where a curtailment strategy is introduced with the assistance of the secure operations of DPN. Thereby, the absorption rate of the generated wind power is maximized and voltage stability level is improved as well. Meanwhile, a modified multi-objective Bayesian learning-based evolutionary algorithm is applied to yield a Pareto front, which is a tradeoff between absorption rate and voltage stability. Afterwards, A technique for order preference by similarity to an ideal solution (TOPSIS) is adopted to determine the dispatching solution by similarity to an ideal solution. Finally, numerical case studies are conducted on a modified IEEE-33 bus system to verify the effectiveness of the proposed scheme.