约束自适应粒子群优化算法及水厂调度

针对粒子群优化算法应用于约束优化问题时易陷入局部极小值的问题,提出了一种改进的粒子群优化算法. 该算法综合了约束优化问题的目标函数值和约束函数的违反度值作为粒子群优化算法的双适应度值, 采用了双适应值动态判断粒子群优化算法中粒子的优劣. 违反度值的计算引入了自适应加权系数,相应地提出了调整各权系数的自适应策略, 并改进了粒子群优化算法的粒子竞争选择策略,拓展了粒子群优化算法的单适应值的应用范围.应用约束自适应粒子群优化算法实现了城市水厂的节能优化调度. 结果表明, 该算法收敛速度快且结果可靠. 粒子群优化算法为解决工程约束优化问题提供了一条可行途径.

Adaptive constrained particle swarm optimization algorithm and urban water supply scheduling

Considering that the particle swarm optimization(PSO) algorithm can be easily trapped into the local minimal value in constrained optimization problems,a modified constrained particle swarm optimization algorithm was proposed.The objective function value and the violation value of constraint functions were effectively combined to form two fitnesses, and the fitnesses were adopted to estimate if the particle was superior or not in a dynamic way.The adaptive weight function was adopted in the calculation of the violation value.The strategy of keeping an adaptive relation of weight coefficients was proposed,and the strategy of swarm tournament selection was improved.The application localizations of the single fitness of PSO were widened as well.The modified constrained PSO algorithm was applied to solve energy optimization problems of the urban water supply process,which showed that the convergent speed of the algorithm is fast and the result is valid.A feasible approach to solve the industrial constraint optimization problems with PSO was provided.