基于自调节粒子群算法的电力系统经济负荷分配

带阀点效应的经济负荷分配问题具有不连续、不可导、非凸、非线性的目标函数,同时还受到电力平衡和运行约束的制约,很难应用经典数学算法求解。针对忽略网损的经济负荷分配问题,文章提出了一种自调节粒子群算法,通过可行化调整机制保证解的可行性,同时采用自适应变异算子提高解的多样性,防止算法早熟收敛,提高算法的寻优速度。为验证算法的有效性,文中对多个经济负荷分配问题进行了测试,与其它智能算法的比较结果证明该算法可以有效找到可行解,避免陷入局部最优,能实现问题的快速求解。

Self-Adjusted Particle Swarm Optimization Algorithm Based Economic Load Dispatch of Power System

The objective function of economic load dispatch (ELD) with valve-point effect possesses its own features such as nonlinear, nonconvex, discontinuous and unable to derive; furthermore, ELD is restricted by electric power balance and operation constraints, so it is very difficult to solve ELD by classic mathematical algorithm. To solve ELD problem with network losses neglected, the authors propose a self-adjusted particle swarm optimization (SPSO) algorithm which can ensure feasibility of the solution by means of feasible regulation scheme; at the same time to prevent precocious convergence of the algorithm and to quicken the search speed, a kind of adaptive mutation operator is adopted to improve the diversity of the solution. To verify the effectiveness of the proposed algorithm, this algorithm is tested by three ELD cases and the obtained results are compared with those obtained by other intelligent algorithms. Comparison results show that with the proposed algorithm feasible solutions can be effectively found, local optima can be avoided and the fast solution of ELD can be implemented.