河道水量还原计算中的遗传BP人工神经网络方法

将遗传算法GA及人工神经网络BP算法结合起来运用于河道水量的还原计算.结果表明:基于GA BP的河道水量还原优化算法能够克服BP算法自身不可优化的弊病,较好地改善网络全局寻优能力,提高网络速度,防止网络陷入局部最小值.同时该算法较确切地反映出河道水量还原计算中河道水量形成非线性这一本质现象,提高了河道水量还原计算结果的精度.

River water returning calculation based on genetic BP neural network

Returning calculation of river water was carried out on the basis of GA(genetic algorithm) and BP(back propagation) algorithm of artificial neural networks.Results indicate that the optimization algorithm based on GA BP,which does not have the defect that BP algorithm cannot be optimized,can improve the global optimization ability of the net,increase the net speed,and avoid the local minimum value in the net.The algorithm reflects the nonlinear character of water formation in river water returning calculation,and improves the calculation precision.