The six-component wheel force load spectrum is a critical input for road simulation tests, CAE analysis, and fatigue life prediction. It plays a vital role in vehicle development and durability testing. To accurately and efficiently predict six-component wheel forces, this paper proposes a prediction method based on Transformer-BiLSTM. Spearman, dCor, and MIC quantify the relationships between operating parameters and six-component wheel force loads from multiple perspectives. An entropy weighting method then weights these parameters objectively and selects input features for the prediction model. The Transformer encoding layer processes the model input and provides more representative time series features for the BiLSTM model, which then predicts the six-component wheel forces. The results show that the predicted six-component wheel forces agree closely with the actual values, with R2 values generally above 0.85. This method accurately and efficiently predicts six-component wheel forces and offers engineering value for virtual load extraction, CAE simulation, and durability design in the automotive industry.
ZouXWangQYuanD, et al. Research on the vibration load spectrum extraction method for electric drive assembly. Eng Res Express2024; 6(4): 045506–045571.
2.
TangCLuoXWuY. Study on multiaxial loading fatigue test of steering knuckle based on road load spectrum. In: Proceedings of the society of automotive engineers (SAE)-China congress, Springer, Singapore, 2017, pp.739–755.
3.
SunCZDuanXLWengY, et al. A study on the evaluation method of vehicle durability performance based on 3D digital road. Automot Eng2017; 39(10): 1211–1216.
4.
StützLKönigTKleyM. Testing of the tire modeling for a vehicle dynamics sim-ulation on a road-to-rig test rig. VDI Ber2022; 2407: 185–196.
5.
LiuWWangRDingR, et al. On-line estimation of road profile in semi-active suspension based on unsprung mass acceleration. Mech Syst Signal Process2020; 135: 1–17.
6.
YangJWangXLiX, et al. Research on dynamic load of Belgian event based on virtual proving ground. SAE World Congress & Exhibition, 2019.
7.
BarbosaBHGXuNAskariH, et al. Lateral force prediction using Gaussian process regression for intelligent tire systems. IEEE Trans Syst Man Cybern Syst2022; 52(8): 5332–5343.
8.
LiuQGGuoHWeiXM, et al. Tire-road force estimation model based on a tire acceleration signal. J Tianjin Univ Sci Technol2023; 56(2): 177–183.
9.
CaiYYuXWangH, et al. Hybrid physics and neural network model for lateral vehicle dynamic state prediction. Proc Inst Mech Eng Part D: J Automobile Eng2024; 238(1): 3–17.
10.
LuoHHuHJYuJH. Prediction of vehicle road load based on deep convolutional neural network-long-short term memory. Automot Eng2021; 7: 46–51.
11.
PanCFengHWangJ, et al. Vehicle trajectory prediction based on spatial-temporal graph transformer networks with LSTM. Proc Inst Mech Eng Part D: J Automobile Eng2025; 0(0). DOI: 10.1177/09544070251398023.
12.
LimBArıkSÖLoeffN, et al. Temporal fusion transformers for interpretable multi-horizon time series forecasting. Int J Forecast2021; 37(4): 1748–1764.
13.
GengMLiJXiaY, et al. A physics-informed transformer model for vehicle trajectory prediction on highways. Transp Res Part C: Emerg Technol2023; 154: 104272.
14.
AminFGharamiKSenB. TrajectoFormer: transformer-based trajectory prediction of autonomous vehicles with spatio-temporal neighborhood considerations. Int J Comput Intell Syst2024; 17(1): 87.
15.
KimMKwakBIHouJU, et al. Robust long-term vehicle trajectory prediction using link projection and a situation-aware transformer. Sensors2024; 24(8): 2398.
ZhangFWuSZhangY, et al. Off-road distributed-drive electric vehicle trajectory tracking control with constrained transformer MPC. Control Eng Pract2025; 165: 106608.
18.
YuanXWangDFengL. A nonlinear decoupling method for wheel force transducer with high-order polynomial model. SAE technical paper, 2024.
19.
JinHJZhouZSZangLG, et al. Simulation and optimization method of vehicle frame structure based on measured load spectrum and its application. J Vib Shock2023; 42(1): 181–189.
20.
WangJLiangYTangJ, et al. Vehicle trajectory reconstruction using lagrange-interpolation-based framework. Appl Sci2024; 14(3): 1173.
21.
EjegwaPAWanzenkeTDOgwucheIO, et al. A robust correlation coefficient for fermatean fuzzy sets based on Spearman’s correlation measure with application to clustering and selection process. J Appl Math Comput2024; 70(2): 1747–1770.
22.
HuJJXiongWFangLM, et al. Forest volume estimation based on distance correlation coefficient and Catboost method. J Central South Univ Technol2023; 43(5): 27–35.
23.
ZhangZWangCGaoY, et al. Short-term passenger flow forecast of rail transit station based on MIC feature selection and ST-LightGBM considering transfer passenger flow. Sci Program2020; 2020: 1–15.
24.
ChenZMYehYLChenTC. Assessment of a regional flood disaster indicator via an entr-opy weighting method. Nat Hazard Rev2018; 19(2): 05018002.
25.
JiangBSBaiYHXuBX, et al. A novel approach for predicting production capacity of tight gas reservoirs based on ensemble learning. China Offshore Oil Gas2024; 36(5): 120–127.
26.
MorozovaABarataTBarlyaevaT, et al. Total electron content PCA-NN prediction model for South-European middle latitudes. Atmosphere2023; 14: 1058.
27.
LiJXiaSZLanHY, et al. Network intrusion detection method based on GRU-RNN. J Harbin Eng Univ2021; 42(6): 879–884.
28.
HochreiterSSchmidhuberJ. Long short-term memory. Neural Comput1997; 9(8): 1735–1780.
29.
GravesASchmidhuberJ. Framewise phoneme classification with bidirectional LSTM networks. In: Proceedings of the 2005 IEEE international joint conference on neural networks, Montreal, QC, Canada, 31 July–4 August 2005, pp.2047–2052. New York, NY: IEEE.
30.
LiJG. Research on fault diagnosis and life prediction of rolling bearings based on machine learning. PhD Thesis, Southwest Jiaotong University, China, 2021.
31.
KrichenM. Convolutional neural networks: a survey. Computers2023; 12(8): 151.
32.
NgaNTTMatosJCNgocSD. Hybrid one-dimensional convolutional neural network—recurrent neural network model for reconstructing missing data in structural health monitoring systems. Machines2025; 13(12): 1101.
33.
VaswaniAShazeerNParmarN, et al. Attention is all you need. Adv Neural Inf Process Syst2017; 30: 6000–6010.
34.
SongXChenKBiZ, et al. Transformer: a survey and application. SSRN. Epub ahead of print 29 November 2024. DOI: 10.2139/ssrn.5211988.
35.
ZhangY. Research on predicting the remaining service life of lithium batteries based on deep learning. PhD Thesis, Anhui University of Science and Technology, China, 2024.
