Abstract: A study is conducted on the reactive power optimization in distribution network of the wind farm. The reactive power characteristics of the double-fed wind turbine are analyzed, and the wind farm is used as a continuous reactive power adjustment means to participate in the reactive power optimization of the power system. A multi-objective reactive power optimization method is proposed based on fuzzy quotient and chaotic sequence dynamic particle swarm optimization algorithm. The fuzzy entropy weight method is used to solve the defects of subjective weight and objective weight, and a multi-objective reactive power optimization decision model with wind power is proposed based on static voltage stability margin, voltage deviation and active power loss. In order to solve the local convergence with traditional particle swarm algorithm, combined with Chebyshev mapping and Logistic mapping, a combinatorial chaotic mapping is used to enhance the homogeneity of the initial particles in the particle initialization, and Logistic chaos optimization is introduced to the process of algorithm optimization, strengthening the global optimization ability of the algorithm. The IEEE14-node system is used as a test case, and the result proves the feasibility and validity of the proposed method.

Key words: entropy weight method, wind farm, combinatorial chaotic sequence, reactive power optimization, power system