节点文献
基于BP神经网络和遗传算法的高速铁路弓网受流质量优化
Optimization of Pantograph-catenary Current Collection Quality of High Speed Railway Based on BP Neural Network and Genetic Algorithm
【作者】 卢琪;
【导师】 张继旺;
【作者基本信息】 西南交通大学 , 载运工具运用工程(专业学位), 2022, 硕士
【摘要】 弓网受流质量是制约和影响高速列车供电质量和运行速度的重要因素,其中弓网之间的接触力标准差是评价受流质量的重要指标。本文以优化弓网受流质量为目标,结合弓网系统建模、试验设计方法和优化算法对高速铁路受电弓及接触网参数进行了优化。主要完成了以下工作:(1)基于有限元法,本文分别建立了简单链型悬挂和弹性链型悬挂接触网与三质量块受电弓的二维弓网耦合模型。通过EN 50318标准和实际线路结果对所建立的弓网仿真模型进行了验证。(2)将中心复合试验设计方法以及BP神经网络和遗传算法结合,建立了一种基于BP神经网络和遗传算法的弓网受流质量优化方法。优化结果表明,该方法较传统的响应面法对高速铁路受电弓和接触网的优化均能取得理想的效果。(3)基于建立的弓网受流质量优化方法对京津线和京沪线的受电弓与接触网参数进行优化。分别构建了以京津线受电弓(SSS400+)、京沪线受电弓(DSA380)前弓和后弓、京津线简链接触网初始形状、京沪线弹链接触网的吊弦至定位点距离、吊弦初始长度和线索张力为输入,弓网接触力标准差为输出的BP神经网络模型。通过遗传算法搜寻极值,获取了各优化参数的最优值,结果使得弓网接触力标准差分别降低了30.05%、29.30%、20.86%、37.68%、27.23%、8.65%和9.47%。同时选取定位点最大抬升量、弓网接触点最大抬升垂直位移、不同运行速度下接触力标准差以及吊弦张力对优化结果进行验证。结果表明,除了线索张力的优化,其他优化的结果均可采用。(4)采用基于神经网络权值矩阵和Garson方程的评估方法量化了所建神经网络模型输入变量的相对重要程度。结果表明,对于SSS400+型受电弓,接触力标准差对各参数的变化均比较敏感;对于DSA380型受电弓,上框架质量对结果的影响更大。而对于接触网的计算结果表明:简链接触网参数对弓网接触力标准差的影响程度:第二根吊弦距离定位点距离>第一根吊弦距离定位点距离>接触线预弛度;弹链接触网参数对弓网接触力标准差的影响程度:第二根吊弦距离定位点距离>第一根吊弦距离定位点距离>第三根吊弦距离定位点距离;第三根吊弦初始长度>第二根吊弦初始长度>第一根吊弦初始长度;接触线张力>弹性吊索张力>承力索张力。
【Abstract】 Pantograph-catenary current collection quality is a significant factor that restricts and affects the power supply quality and operation speed of high-speed train.The standard deviation of contact force(CFSD)between pantograph and catenary is an important index to evaluate the current collection quality.Aiming at optimizing the current collection quality of pantograph and catenary,combined with pantograph-catenary system modeling,experimental design method and optimization algorithm,the parameters of pantograph and catenary of high-speed railway were optimized.Main work of this thesis can be summarized as follow:(1)Based on the finite element method,2D pantograph-catenary models which coupled simple and elastic chain suspension catenary with three mass pantograph was established in this thesis.The established pantograph catenary simulation model was verified by EN 50318 standard and the results of actual line.(2)Combining the central composite test design method,BP neural network and genetic algorithm,a kind of pantograph-catenary current collection quality optimization approach based on BP neural network and genetic algorithm was proposed.The optimization results show that this approach can achieve ideal results in the optimization of pantograph and catenary of high-speed railway compared with the traditional response surface method.(3)Based on the established optimization approach,the parameters of pantograph and catenary of Beijing-Tianjin line and Beijing-Shanghai line was optimized.The BP neural network model was constructed with the input of Beijing Tianjin line pantograph(SSS400 +),the leading and trailing pantograph of Beijing Shanghai line pantograph(DSA380),the initial shape of Beijing Tianjin line simple chain suspension catenary,the distance from the droppers to the positioning point,the initial length of droppers and the tension of the contact wire,messenger wire and stitch wire of Beijing Shanghai line elastic chain suspension catenary,as well as the output of the CFSD of pantograph and catenary.By searching the extreme value through genetic algorithm,the optimal value of each optimization parameter was obtained.The results showed that the CFSD of pantograph and catenary was reduced by30.05%,29.30%,20.86%,37.68%,27.23%,8.65% and 9.47% respectively.The optimization results were verified by the maximum lifting amount of the positioning point,the maximum lifting vertical displacement of the pantograph catenary contact point,the standard deviation of the contact force under different operating speeds and the tension of droppers.The results showed that the optimization results are effective except for the optimization of the tension of contact wire,messenger wire and stitch wire.(4)The evaluation method based on neural network weight matrix and Garson equation was used to quantify the relative importance of the input variables of the neural network model.The results show that the standard deviation of contact force of SSS400+ pantograph is sensitive to the changes of various parameters.For DSA380 pantograph,the quality of upper frame has a greater impact on the results.The results of catenary show that: The influence weight of catenary parameters on the CFSD of pantograph and simple chain suspension catenary: the distance between the second dropper and the positioning point > the distance between the first dropper and the positioning point > the contact wire pre-sag;The influence weight of catenary parameters on CFSD pantograph and elastic chain suspension catenary: the distance between the second dropper and the positioning point > the distance between the first dropper and the positioning point > the distance between the third dropper and the positioning point;The initial length of the third dropper > the initial length of the second dropper > the initial length of the first dropper;Contact wire tension > stitch wire tension > messenger wire tension.
- 【网络出版投稿人】 西南交通大学 【网络出版年期】2024年 02期
- 【分类号】U225;U264.34