Research Of The Power System Dynamic Reactive Power Optimization Based On The Differential Annealing Algorithm

Reactive power optimization is an important mean to ensure the power system security, economic, and stable operation. Under the conditions of satisfying secure operation of the power system, the power loss of power system is reduced by reasonably distributing the generator output, controlling the on-load tap changer and adjustable reactive power compensation device, reasonably optimizing of reactive power flow, and reducing the reactive power flow between the regions. Currently, there are many literatures about the research of a single time section of power system static reactive optimization, but the actual power system load changes over time. So, considering the combination of the reactive power optimization and the dynamic changes of load is of more practical application value.In this paper, considering the dynamic changes of the load characteristic and the constraints of reactive power compensation equipment action number, dynamic reactive power optimization model is established. Considering the trait of the differential evolution algorithm, simulated annealing strategy is introduced to the differential evolution algorithm with the combination of the advantage of simulated annealing algorithm, which effectively solves the problem that the differential evolution algorithm easily fall into local stagnation and poor snap action ability, and the improved differential evolution algorithm is applied to the solution of dynamic reactive power optimization model; according to the prediction, the load is obtained and divided into24periods, the use of the based heuristic decoupling strategy to decouple the operation time of control equipment from the time-space effectively solves the problem that the control device is constrained by action number; the comprehensive sensitivity analysis is conducted to the control devices which obtain the action permission in each period, then the size of sensitivity is used to reasonably arrange the switching sequence of control device.At last, the IEEE-30standard test system and a real system are used analyzed the simulation of the established model in this paper. The simulation results show that the model in this paper is effective, and it has practical significance to the off-line calculation about the power system dynamic reactive power optimization.