基坑支护结构水平变形预测的遗传神经网络方法

采用遗传算法和误差反向传播算法相结合的混合算法来训练前馈人工神经网络,先用遗传学习算法进行全局训练,再用BP算法进行精确训练。就遗传算法过程中的选择、变异进行了探索,提出了用BP网络训练产生变异的遗传算法。作为实例,将该方法应用于预测基坑支护结构水平变形中。结果表明,该方法有收敛速度较快、预测精度高等优点。

APPLICATION OF GENETIC NEURAL NETWORK METHOD TO FORECASTING HORIZONTAL DEFORMATION OF SUPPORT STRUCTURE AT FOUNDATION PIT

A new method for training the artificial neural network is presented. In this method, the genetic algorithm is used to train network with updating the weights and the valves to minimize the error between the network output and the desired output. Then the back propagation(BP) algorithm is used to further train the artificial neural network. A new method of using the course of training BP network to get mutation is expounded. For the mutation of the genetic algorithm, the choice and the mutation of the genetic algorithm is studied. As an example, the method was used to predict horizontal displacement of deformation of support structure to foundation pit . The result suggests that the method is feasible.