摘要
机组组合是一个大规模、非线性混合整数优化问题,求解比较困难,为了提高粒子群算法的全局和局部搜索能力,提出一种惯性权值自适应调整的粒子群算法。该算法按照适应度的大小将粒子群分成两个子群,然后根据适应度的进化速度和进化停滞系数动态调整惯性权值。通过对典型函数的测试以及10台机组24小时的优化调度,计算结果表明该方法收敛精度较高。
Unit commitment is a large-scale and mixed-integer non-linear programming problem. An adaptive inertia weight of particle swarm optimization algorithm (AWPSO) is presented to increase the global and local search. The population is divided into two sub-populations according to the value of fitness, and the inertia weight is formulated as a function of evolution speed and stagnate state. The algorithm is tested with well-known benchmark functions and the simulation results with systems of up to 10 units and 24-h scheduling horizon are presented. The experiments show that the convergence accuracy is increased.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2009年第15期15-18,共4页
Power System Protection and Control
关键词
粒子群算法
惯性权值
自适应
机组组合
particle swarm optimization
inertia weight
adaptability
unit commitment