基于自适应粒子群算法的轨下基础病害识别

Sub-rail Foundation Disease Identification Based on Adaptive Particle Swarm Optimization Algorithm

基于车辆-轨道耦合动力学模型,对不同轨下基础病害情况下的轨枕振动响应进行仿真分析。提出利用支持向量机算法和粒子群算法对轨下基础病害进行识别。为了提高粒子群算法的收敛速度,提出一种自适应粒子群算法,并将所提方法应用于轨下基础病害识别仿真,分析不同病害条件下的轨枕振动特征。研究表明:所提算法的病害识别准确率≥80%,且其算法收敛速度有明显提升。

Based on vehicle-track coupling dynamics model,the simulation analysis of the sleeper vibration response under different sub-rail foundation disease conditions is carried out.It is proposed to adopt SVM(support vector machine)algorithm and PSO(particle swarm optimization)algorithm to identify the sub-rail foundation basic diseases.To improve the convergence speed of PSO,an APSO(adaptive particle swarm optimization)algorithm is proposed,and the proposed method is applied to the identification and simulation of sub-rail foundation basic diseases,so as to analyze the vibration characteristics of sleepers under different disease conditions.The research shows that the disease identification accuracy rate of the proposed algorithm can achieve over 80%,and the convergence speed of the algorithm is significantly improved.