Similar articles

Abstract: The wheel wear of light rail trains is difficult to predict due to poor information and small data samples. However, the amount of wear gradually increases with the running mileage. The grey future prediction model is supposed to deal with this problem effectively. In this study, we propose an improved non-equidistant grey model GM(1,1) with background values optimized by a genetic algorithm (GA). While the grey model is not good enough to track data series with features of randomness and nonlinearity, the residual error series of the GA-GM(1,1) model is corrected through a back propagation neural network (BPNN). To further improve the performance of the GA-GM(1,1)-BPNN model, a particle swarm optimization (PSO) algorithm is implemented to train the weight and bias in the neural network. The traditional non-equidistant GM(1,1) model and the proposed GA-GM(1,1), GA-GM(1,1)-BPNN, and GA-GM(1,1)-PSO-BPNN models were used to predict the wheel diameter and wheel flange wear of the Changchun light rail train and their validity and rationality were verified. Benefitting from the optimization effects of the GA, neural network, and PSO algorithm, the performance ranking of the four methods from highest to lowest was GA-GM(1,1)-PSO-BPNN>GA-GM(1,1)-BPNN>GA-GM(1,1)>GM(1,1) in both the fitting and prediction zones. The GA-GM(1,1)-PSO-BPNN model performed best, with the lowest fitting and forecasting maximum relative error, mean absolute error, mean absolute percentage error, and mean squared error of all four models. Therefore, it is the most effective and stable model in field application of light rail train wheel wear prediction.

基于改进灰色GM(1,1)模型的轻轨车辆车轮磨耗预测

[1]CaiH, WangYL, SongCW, et al., 2022. Prediction of surface subsidence based on PSO-BP neural network. Journal of Physics: Conference Series, 2400(1):012046.

[2]CaiL, LinJD, LiaoXY, 2022. Life prediction of ship CXF cable using a non-equidistant grey model with small samples. IEEE Transactions on Power Delivery, 37(6):5094-5101.

[3]ChangJH, ZhangJZ, LiWG, et al., 2022. Big data dressing recommendation based on ant colony algorithm to optimize BP network. Proceedings of the Second International Conference on Advanced Algorithms and Signal Image Processing (AASIP 2022), article 124751W.

[4]ChenZW, 2019. Research on Wheel Size Prediction and Re-Profiling Strategy for the Wheel-Set of High-Speed EMU. MS Thesis, Southwest Jiaotong University, Chengdu, China (in Chinese).

[5]ChiZX, LinJ, ChenRR, et al., 2020. Data-driven approach to study the polygonization of high-speed railway train wheel-sets using field data of China’s HSR train. Measurement, 149:107022.

[6]ChudzikiewiczA, KorzebJ, 2018. Simulation study of wheels wear in low-floor tram with independently rotating wheels. Archive of Applied Mechanics, 88(1-2):175-192.

[7]CorreaN, VadilloEG, SantamariaJ, et al., 2016. A versatile method in the space domain to study short-wave rail undulatory wear caused by rail surface defects. Wear, 352-353:196-208.

[8]DengYQ, LiuL, LiMY, et al., 2023. A data-driven wheel wear prediction model for rail train based on LM-OMP-NARXNN. Journal of Computing and Information Science in Engineering, 23(2):021012.

[9]FanN, WangSW, LiuCX, et al., 2017. Wheel wear prediction of high-speed train using NAR and BP neural networks. IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), p.126-130.

[10]HanP, ZhangWH, 2015. A new binary wheel wear prediction model based on statistical method and the demonstration. Wear, 324-325:90-99.

[11]HuangGB, ZhuQY, SiewCK, 2006. Extreme learning machine: theory and applications. Neurocomputing, 70(1-3):489-501.

[12]JiangYZ, ZhongWS, WuPB, et al., 2019. Prediction of wheel wear of different types of articulated monorail based on co-simulation of MATLAB and UM software. Advances in Mechanical Engineering, 11(6):1-13.

[13]JiangZQ, BanjevicD, E MC, et al., 2017. Optimizing the re-profiling policy regarding metropolitan train wheels based on a semi-Markov decision process. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 231(5):495-507.

[14]LiY, TangMA, GuBH, et al., 2015. Forecast research on trucks wheel tread wear based on grey changeable weight combination model. Journal of Railway Science and Engineering, 12(1):160-165 (in Chinese).

[15]LiY, LuanYZ, ZhangST, 2016. Application of the optimization of background value in non-equidistant GM(1,1) model in the settlement prediction. Beijing Surveying and Mapping, (5):48-52 (in Chinese).

[16]LiaoGL, 2014. Research on the Security State Prediction and Lathing Strategy Optimization for the Wheelset of Urban Rail Train. MS Thesis, Beijing Jiaotong University, Beijing, China (in Chinese).

[17]LiuBL, ZhangYF, PanDB, et al., 2024. Amphibious vehicle’s resistance optimization through neural networks and genetic algorithms. Physics of Fluids, 36(6):065129.

[18]LiuF, YangXW, ChenDW, et al., 2022. Wheel wear characteristics analysis of Changchun 70% low-floor light rail vehicles. Urban Mass Transit, 25(9):11-15 (in Chinese).

[19]ŁuczakB, FirlikB, StaśkiewiczT, et al., 2020. Numerical algorithm for predicting wheel flange wear in trams–validation in a curved track. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 234(10):1156-1169.

[20]NingY, JinYP, PengYD, et al., 2022. Small obstacle size prediction based on a GA-BP neural network. Applied Optics, 61(1):177-187.

[21]PradhanS, SamantarayA, BhattacharyyaR, 2019. Multi-step wear evolution simulation method for the prediction of rail wheel wear and vehicle dynamic performance. Simulation, 95(5):441-459.

[22]ShebaniA, IwnickiS, 2018. Prediction of wheel and rail wear under different contact conditions using artificial neural networks. Wear, 406-407:173-184.

[23]SuKX, ZhangJW, ZhangJW, et al., 2023. Optimisation of current collection quality of high-speed pantograph-catenary system using the combination of artificial neural network and genetic algorithm. Vehicle System Dynamics, 61(1):260-285.

[24]WangMQ, WangY, ChenEL, et al., 2022a. High-speed train tread wear prediction model based on I-ML-ELM. Chinese Journal of Theoretical and Applied Mechanics, 54(6):1720-1731 (in Chinese).

[25]WangMQ, JiaSX, ChenEL, et al., 2022b. Research and application of neural network for tread wear prediction and optimization. Mechanical Systems and Signal Processing, 162:108070.

[26]WangSW, GuoH, ZhangSY, et al., 2022. Analysis and prediction of double-carriage train wheel wear based on SIMPACK and neural networks. Advances in Mechanical Engineering, 14(3):1-12.

[27]WojciechowskiŁ, GapińskiB, FirlikB, et al., 2020. Characteristics of tram wheel wear: focus on mechanism identification and surface topography. Tribology International, 150:106365.

[28]XingZY, MaoLL, LiaoGL, et al., 2014. Forecasting of wheelset size of urban rail train based on PSO-SVM model. Journal of Shenyang University of Technology, 36(4):411-415 (in Chinese).

[29]XuXP, 2021. Research on Wheelset Parameter Detection Technology and Wear Prediction. MS Thesis, Nanjing University of Science & Technology, Nanjing, China (in Chinese).

[30]YangZ, XingZY, WangL, et al., 2018. Optimization of wheel re-profiling strategy based on statistical wear model. Railway Standard Design, 62(1):142-148 (in Chinese).

[31]YuanK, XuSA, FuYQ, et al., 2024. Research on weigh-in-motion algorithm of vehicles based on WOSA-BP. Chinese Journal of Sensors and Actuators, 37(1):50-57 (in Chinese).

[32]ZhangSR, WangBT, LiXE, et al., 2016. Research and application of improved gas concentration prediction model based on grey theory and BP neural network in digital mine. Procedia CIRP, 56:471-475.

[33]ZhangY, ZhangJW, LuoL, et al., 2019. Optimization of LMBP high-speed railway wheel size prediction algorithm based on improved adaptive differential evolution algorithm. International Journal of Distributed Sensor Networks, 15(10):1-9.

[34]ZhongLS, ChenLY, GongJH, et al., 2015. Prediction of wheel tread wear volume based on least squares support vector machine optimized by coupled simulated annealing. Application Research of Computers, 32(2):397-402 (in Chinese).

[35]ZhuAH, YangS, LiQ, et al., 2019. Research on prediction of metro wheel wear based on integrated data-model-driven approach. IEEE Access, 7:178153-178166.

[36]ZhuY, WangWJ, LewisR, et al., 2019. A review on wear between railway wheels and rails under environmental conditions. Journal of Tribology, 141(12):120801.