Application of Sequential Quadratic Programming Algorithm based on region search method in reactive power optimization

In order to improve the convergence of reactive power optimization and the accuracy of optimization result, this paper proposes the sequential quadratic programming (SQP) algorithm based on region search method. The algorithm combines global SQP with trust-region search method. Through adding the trust-region constraint into SQP sub-question, the linear search method is improved to region search method and step length can be determined while search direction is solved. Moreover, calculating reactive power optimization in IEEE30 bus system and comparing with other methods, results show that the improved algorithm has great convergence and optimization effect