摘要
在遗传算法优化BP神经网络的电梯群控系统基础上,利用遗传算法对候梯时间、乘梯时间、舒适度和运行能耗等评价函数进行搜索具有全局性的进化解,通过BP神经网络按照进化解权值进行寻优,从而获得全局最优解。为提高遗传算法的效率,在遗传操作过程中采用最优个体保存策略,同时采用了交叉率和变异率能够随适应度自动改变的自适应算法,提高了达到最优解的收敛速度。得出最优派梯方案,实现电梯的多目标优化调度。
Based on genetic algorithm to optimize the BP neural network in elevator group control system, evalation functions concerning the waiting time, riding time, comfort and energy consumption are used to search for an overall solution, and at the same time, evolutionary solution right value optimization in BP neural network is applied to do the search to get a global optimal solution. In order to improve the efficiency of GA, the best individual preservation strategy in the process of genetic manipulation is used. At the same time the adaptive algorithm, in which the crossover rate and mutation rate can be self-adapted, is used to improve the convergence speed of the optimal solution. Thus, an optimal, multipurpose elevator dispatching programme is derived.
出处
《陕西理工学院学报（自然科学版）》
2015年第4期36-39,58,共5页
Journal of Shananxi University of Technology:Natural Science Edition
关键词
电梯群控
多目标优化
BP神经网络
遗传算法
elevator group control
multi-purpose optimization
BP neural network
genetic algorithm
