改进拟态物理学算法在交直流系统无功优化中的应用

Reactive optimization of hybrid AC-DC power system based on improved APO

拟态物理学算法(APO)具有较好的全局搜索能力,且能避免粒子群等算法常出现的易早熟收敛现象,具有较好的稳定性、快速收敛和鲁棒性。针对交直流混联系统无功优化问题,在矢量模型的拟态物理学算法的基础上,提出了一种基于改进拟态物理学算法的无功优化算法。为便于处理约束问题,算法中通过定义不可行度函数判断不可行解,并采用收缩因子将不可行解拉回可行域内,再通过多维搜索方法寻求最优解。通过IEEE30节点系统仿真测试,并与其他优化算法进行比较,结果表明该算法具有较好的求解效率和准确性。

The APO is a superior algorithm with global search capability, and it can avoid premature convergence that easily occurs in Particle Swarm Optimization（OPS） and other algorithms. Besides, it has better stability, rapidity convergence and robustness. Aimed at reactive optimization problem of hybrid AC-DC system, and according to the APO of the vector model, a reactive optimization algorithm is proposed based on a kind of improved APO. In order to solve the restricted problem, the algorithm defines infeasibility degree function to judge infeasible solutions, haul the infeasible solution back to the feasible zone by use of the constriction factor, and search the optimal solution based on multidimensional search algorithm. According to the simulation test of IEEE 30-bus system, and compared it with other algorithms, it turns out that the improved APO is accurate and efficient in searching the solutions.