论文部分内容阅读
针对大坝安全监控中小波神经网络模型(WNN)训练时间较长且易陷入局部极小值的缺陷,提出采用粒子群优化算法(PSO)取代传统的梯度下降法对小波神经网络中的各参数进行优化,建立了PSO-WNN模型并用于大坝安全监测的拟合和预报。实例结果表明,PSO-WNN模型收敛速度快、预测精度及稳定性高,为大坝变形监测分析提供了一种有效的新型建模方法。
In view of the shortcoming of long training time and easily falling into local minimum in the wavelet neural network model (WNN) of dam safety monitoring, the particle swarm optimization algorithm (PSO) is adopted to replace the traditional gradient descent method to measure the parameters of wavelet neural network Optimized, the PSO-WNN model was established and used for fitting and forecasting dam safety monitoring. The experimental results show that the PSO-WNN model has the advantages of fast convergence, high prediction accuracy and high stability, which provides an effective new modeling method for dam deformation monitoring and analysis.