基于GA-BP神经网络的结构损伤位置识别

针对传统BP神经网络训练中存在的一些问题，提出了一种基于遗传算法(GA)-BP神经网络混合技术识别结构损伤位置的方法。该方法利用基因实数编码的遗传算法优化BP网络的结构及初始参数，从而大大提高了神经网络的训练精度。运用GA-BP网络与传统BP网络技术分别对两个算例进行了结构损伤定位的识别仿真，结果表明遗传BP稳定性好，精度高，对噪声有很好的鲁棒性，便于工程应用。

Structural Damage Localization Based on GA-BP Neural Network

In order to improve the limitation which often occurs in the training process of BP neural network, a method for damage localization based on genetic algorithm(GA)-BP neural network(BPNN) combined technology is presented in the paper. The genetic algorithm coding in the real number is used to optimize the structural and original parameters of BP neural network so that the network can obtain more accurate results by learning the training patterns. Two numerical simulations are studied using GA-BP and traditional BP neural network respectively. Results show that GA-BP neural network is more stability, precision and robustness in localizing the structural damage than traditional BP neural network.