针对BP神经网络收敛速度慢和易于陷入极小值的问题,采用将遗传算法全局寻优和BP神经网络局部寻优相结合的方法,提高了BP神经网络的计算精度和收敛速度.应用该模型对空调水系统进行了辨识,并以冷水机组和水泵能耗最小为目标进行优化,取得了满意的效果.%The convergence speed of BP neural network is slow, and it easily immerges in partial mini-mum value. Aiming at this problem, the accuracy and convergence speed of BP neural network were im-proved by combining genetic algorithm based on global optimization with the BP neural network based on local optimization. The method was used to to identify the cold water chiller. The minimum energy con-sumption of cold water chiller and water pump was taken as optimization goal. The obtained result is sat-isfactory.
展开▼