铁道科学与工程学报2025，Vol.22Issue(10)：4489-4501,13.DOI:10.19713/j.cnki.43-1423/u.T20250108
地铁嵌入式轨道结构参数多目标优化研究
Multi-objective optimization research on structural parameters of metro embedded track
摘要
Abstract
To reduce vibration of the vehicle body,rail,track slab,and tunnel walls of the metro embedded track system during operation,and to enhance its overall performance,a parameter optimization method based on the higher-order response surface method and the Non-dominated Sorting Genetic Algorithm Ⅱ(NSGA-Ⅱ)was proposed.The metro vehicle-embedded track-tunnel coupled dynamic analysis model was established by Matlab for dynamics simulation,and the reliability of simulation was verified using measured data.The optimal Latin hypercube sampling method was employed to create a high-dimensional sample parameter space matrix.A fourth-order polynomial response surface model incorporating cross-terms was developed using the least squares regression method,and the validity of different surrogate models was evaluated using the coefficient of determination and the mean absolute percentage error.A multi-objective optimization mathematical model was formulated,leveraging the NSGA-Ⅱ algorithm to obtain the Pareto optimal solution set.Different weight matrices were assigned to each objective,and the optimal solution under different design preferences was determined by using the TOPSIS method.The results of the study show that the dynamic simulation values are basically consistent with the measured values,confirming the accuracy and reliability of the simulation;The higher order response surface determination coefficient is closer to 1,and the average absolute percentage error is smaller than that of BP neural network and LSSVM,indicating superior model precision;By selecting the average weight from the weight matrices as the optimal design,eight key metrics,including the vertical Sperling index,show significant reduction compared to the sample scheme.Specifically,the vertical wheel-rail force,lateral wheel-rail force,vertical rail vibration acceleration,lateral rail vibration acceleration,vertical track slab vibration acceleration,lateral track slab vibration acceleration,and vertical frequency-division vibration level of the tunnel wall decrease by 7.24%,5.39%,18.47%,26.54%,13.22%,28.88%,and 6.51%,respectively.The optimized embedded rail system demonstrates significant vibration reduction across key structural components,including the vehicle body,steel rail,track slab,and tunnel walls,proving that the optimization method proposed in this paper is feasible and effective,and the results of the research can provide technical support and engineering guidance for the subsequent embedded track parameter design and improvement research.关键词
嵌入式轨道/高阶响应面/NSGA-Ⅱ/多目标参数优化/结构减振Key words
embedded track/higher order response surface/NSGA-Ⅱ/multi-objective parameter optimization/structural vibration damping分类
交通运输引用本文复制引用
徐磊,王甲生,杨玲,刘向明,张东风,郭骁,李伟,余志武..地铁嵌入式轨道结构参数多目标优化研究[J].铁道科学与工程学报,2025,22(10):4489-4501,13.基金项目
中南大学创新驱动计划(2023CXQD072) （2023CXQD072）
铁科院集团公司基金资助项目(2022YJ296) （2022YJ296）
中国中铁重大科研计划(2022-重大-14) （2022-重大-14）
中国中铁股份有限公司科技研究开发计划项目(2025-专项-03) （2025-专项-03）
