大型水泵机组的遗传BP神经网络算法

Large pump sets of genetic-BP neural network algorithm

利用BP神经网络在处理非线形问题上的优势来处理大型水泵机组的传递函数的非线形。并结合遗传算法寻找适合机组传递函数的最佳网络拓扑结构,为了控制的实时性要求网络的结点尽可能的少,因此在遗传算法的适应度函数中将时间的因素考虑进去。结果是,考虑时间的算法得到的网络结构要优于不考虑时间的算法所得到的网络结构。

The neural network has better capacity for dealing with nonlinear matter in order that the large pump sets has nonlinear transmission function. To find the best the topological structure of the network fitting the transmission function of the pump sets with the Genetic Algorithm.Fewer network nodes can improve the realtime control so that the time factor includes in the fitness function of the Genetic Algorithm. As a result, the Genetic Algorithm including time factor are superior to the Genetic Algorithm excluding time factor. 