提出了一种应用遗传神经网络判定大坝变形预报因子重要度的方法,分析了如何从学习后的网络权重中提取预报因子重要度信息,应用数理统计方法消除了网络学习初始权重对判定结果的影响。用一实例对提出的方法进行了验证,表明该方法具有很强的操作性和较高的准确性。
A new method of using genetic neural network to determine the factor weight of dam deformation is put forward. The way to extract the information of the factor weight from the trained network link weight is analyzed. The impact of primary network weight on final determination result can be eliminated by symbolic statistics. Experiment shows that the' method is highly operational with high accuracy.