论文部分内容阅读
针对大型水电站厂内经济运行中最优负荷分配问题的高维性、复杂非线性和实时性,提出一种收敛性全面改善的改进退火粒子群算法.改进算法采用了整体改进策略:初始种群生成方面,采用初始种群解空间生成法避开机组空蚀振动区;适应度函数设计方面,加入惩罚项提高算法搜索效率;进化操作方面,加入自适应惯性权重、交叉运算、变异运算,改善算法的全局与局部收敛性能;引进模拟退火算法,提高算法的局部收敛性,保证算法以较大概率收敛于全局最优解.以三峡水电站厂内经济运行为实例,与现有算法进行了对比,结果表明:改进的退火粒子群算法在收敛速度与收敛精度方面均有一定的优势,适用于求解水电站负荷分配优化问题.
Aiming at the high dimensionality, complex nonlinearity and real-time of the optimal load distribution problem in large-scale hydropower plant, this paper proposes an improved annealing particle swarm optimization with overall improved convergence. The improved algorithm adopts the overall improvement strategy: initial population generation In terms of fitness function design, the penalty term is added to improve the search efficiency of the algorithm. In terms of evolutionary operations, adaptive inertia weight, cross-over operation, mutation operation and improved algorithm are used Global convergence and local convergence performance.At the same time, the simulated annealing algorithm is introduced to improve the local convergence of the algorithm and ensure that the algorithm converges to the global optimal solution with a high probability.Comparing with the existing algorithms, the economic operation of the Three Gorges Hydropower Station It shows that the improved annealing particle swarm optimization algorithm has some advantages in terms of convergence speed and convergence accuracy, and is suitable for solving the load distribution optimization problem of hydropower station.