Dynamic Reactive Power Optimization Of Distribution System Based On PSO-DP Algorithm

Reactive power optimization is an important method for safe and economic operation of distribution system. Transmission of reactive power in electric network can be reduced by adjusting the tap changer of transformer and switching on or off some reactive power compensation devices. It can reduce system losses and maintain acceptable voltage levels. Reactive power/voltage control equipments operate frequently due to load fluctuations. It is difficult to make control schemes for different time. Many practical issues lead to major difficulties in solving the problem of reactive power optimization. Dynamic reactive power optimization is hot in the world today.Dynamic reactive power optimization is a nonlinear integer programming problem which considers the conditions of load changes with time and the limitation of equipment's regulation times. The objective function of the mathematical model is neither differentiable nor continuous. Both equality constraints and inequality constrains should be considered with discrete and continuous variables. It is difficult to solve the problem with one of existing algorithms. Through various algorithms complementary advantages, forming mixture algorithms to solve reactive power optimization problem is the development direction of the research.The different conditions of load curve division influence control equipment's regulation times and action time. The more load curve sectionalized, the better optimal result was, and the more adjusted times of on-load tap changing and capacitor. Therefore, reasonable load curve dividing contribute to obtain optimal control strategy of control equipment. The paper put forward the method of dynamic load dividing strategy according to the rate of reactive power load.Through the research of dynamic reactive power optimization and its algorithm, the paper chose on-load tap changer and switched compensators as the control variables and established two-phase optimization model. The first phase optimization is to minimize the power losses and voltage deviation without considering the limitation of equipment's regulation times. Static reactive power optimization problem is solved with particle swarm optimization (PSO) algorithm. The second phase optimization is to optimize equipment's regulation times to extend its service life. Optimization objective is to minimize all the equipments' regulation times. The optimal solution and several suboptimum solutions of the first phase optimization are saved as state variables of the second objective function. The second phase optimization is solved with dynamic programming (DP) algorithm.According to the studies of dynamic load division strategy, established mathematic model and two algorithms, this paper proposed complete ideas to solve the problem of dynamic reactive power optimization. By constantly modifying the rate of reactive powerε, the PSO algorithm and DP algorithm are used repeatedly to solve the established mathematical model of dynamic reactive power optimization until they meet all of the constraints.Simulation results in Qing Zhou distribution system show that the proposed method based on PSO-DP algorithm can obtain good convergence property and speed. The optimal result meets requirements of secure and economic operation of distribution system and reduces each equipment's regulation times. The PSO-DP method is suitable for solving dynamic reactive power optimization problem.