基于粒子群优化反向传播神经网络的铁路路基沉降量预测

铁路路基沉降规律复杂、沉降量难以预测，因此提出一种粒子群优化反向传播神经网络的路基沉降量预测模型。传统神经网络建模时存在收敛速度慢、易陷入局部极小值等不足。因此，通过粒子群算法修正网络的初始权/阈值，提高全局收敛性，建立基于粒子群优化的反向传播神经网络预测模型。通过对宝中线实测数据进行仿真实验，结果显示：经粒子群优化的神经网络可避免局部极小问题，加快网络收敛速度，提高了对铁路路基沉降量的预测精度。

Prediction of Railway Subgrade Settlement Based on Back Propagation Neural Network Optimized by Particle Swarm

In view of the fact that the settlement deformation law of railway subgrade is complex and the settlement is difficult to predict, a back propagation neural network optimized by particle swarm optimization was proposed to predict the settlement of railway subgrade. Considering the shortcomings of using traditional back propagation neural network to establish settlement prediction model, such as slow convergence speed and easy to fall into local minimum. Therefore,a particle swarm optimization algorithm was used to modify the initial weight and threshold of back propagation neural network, the global convergence of the network was optimized, and the prediction model based on back propagation neural network optimized by particle swarm was established. Through the simulation experiment on the measured data of Baozhong Railway, the results show that the back propagation neural network optimized by particle swarm optimization can avoid the local minimum problem, accelerate the convergence speed of the network, and improve the prediction accuracy of railway subgrade settlement.