基于小波神经网络的非线性系统工程安全性评价研究

RESEARCH ON THE SAFETY EVALUATION OF NONLINEARITY SYSTEM ENGINEERING BASED ON WAVELET NEURAL NETWORK

针对现有非线性系统工程安全性评价方法中存在的不足之处,本文将小波理论与神经网络相结合,提出了一种可应用于系统工程安全性评价的小波神经网络模型。最后将该网络应用于某大跨度悬索桥的安全评价中。研究结果表明:训练后的WNN模型收敛速度优于BP神经网络,有效减少了平均训练误差。研究成果可应用于其他系统工程的安全性评价中。

In view of the existing deficiency in the safety evaluation of nonlinearity system engineering, we combine wavelet theory with neural network and propose a wavelet neural network model which can be used in safety evaluation of system engineering. Finally, we apply the proposed neural network in the safety evaluation of a large span suspension bridge. The results show that the convergence speed of the WNN model is better than BP neural network and can effectively reduce average trained errors. The proposed model can be applied to the safety evaluation of the other system engineering.