轨道交通车辆踏面制动闸调器螺杆连接优化研究  

Research on screw connection optimization of rail transit vehicle tread brake adjuster

作　　者：巨子琪 兰宏伟[2] 宰晨光 JU Ziqi;LAN Hongwei;ZAI Chenguang(Xi'an Traffic Engineering lnstitute,Xi'an 710300,China;Lanzhou Jiaotong University,Lanzhou Gansu 730070,China)

机构地区：[1]西安交通工程学院,西安710300 [2]兰州交通大学,兰州730070

出　　处：《自动化与仪器仪表》2020年第12期70-74,共5页Automation & Instrumentation

基　　金：西安交通工程学院中青年项目:轨道交通车辆闸调器的优化设计(No.19KY-02)。

摘　　要：针对轨道交通车辆踏面制动闸调器螺杆连接优化问题,提出一种结合人工神经网络和遗传算法的优化方法。通过MATLAB提供的神经网络工具箱,将S型正切函数当成输入层神经元传递函数,将S型对数函数当成输出层神经元传递函数,将正交实验中的样本数据当成训练样本,通过三层人工神经网络对螺杆连接有限元分析的实验样本进行学习与训练,为后续轨道交通车辆踏面制动闸调器螺杆连接优化提供基础。对轨道交通车辆踏面制动闸调器螺杆连接设计进行调整,将最大应力与构件质量当成优化目标,通过加权组合的方式把两个目标函数转换成一个,对解空间的解数据进行编码处理,描述成遗传空间的基因串结构数据,通过适应度函数完成解的优劣性的评价,在获取最高适应度个体的情况下结束计算,将其当成最优解输出。经验证,所提方法可有效增强轨道交通车辆踏面制动闸调器螺杆连接强度,同时减少重量,增强了结构的整体性能。Aiming at the optimization of screw connection of rail transit vehicle tread brake adjuster,an optimization method combining artificial neural network and genetic algorithm is proposed.Through the neural network toolbox provided by MATLAB,the S-type tangent function is regarded as the input layer neuron transfer function,the S-type logarithm function is regarded as the output layer neuron transfer function,and the sample data in the orthogonal experiment is regarded as the training sample.Through the three-layer artificial neural network,the experimental sample of screw connection finite element analysis is studied and trained for the following rail transit vehicle tread optimization to provid of screw connection of brake adjuster provides.Adjust the screw connection design of rail transit vehicle tread brake adjuster,take the maximum stress and component quality as the optimization objective,transform the two objective functions into one by the way of weighted combination,code and process the solution spatial solution data,describe the gene string structure data of genetic space,complete the evaluation of the solution quality through the fitness function,and obtain In the case of the individual with the highest fitness,the calculation is ended and the output is regarded as the optimal solution.The screw connection design of the tread brake adjuster of rail transit vehicles is adjusted.The maximum stress and component quality are regarded as the optimization objectives.The two objective functions are converted into one by weighted combination.The solution spatial solution data is coded and described as genetic spatial gene string structure data In the case of the highest fitness individual,the calculation is finished as the output of the optimal solution.It is verified that the method can effectively enhance the screw connection strength of the rail transit vehicle tread brake adjuster,reduce the weight and enhance the overall performance of the structure.

关 键 词：交通车辆 闸调器 螺杆连接 优化 

分 类 号：TH132[机械工程—机械制造及自动化]