Peeling Force Prediction of Automobile Seat Slide Rail Based on Iterative Local Weighted Linear Regression

doi:10.3901/JME.2025.17.193

Abstract

Abstract: The peeling force (PF) of the car seat slide rail, an important part of the safety assessment, can comprehensively reflect the safety of the slide rail. The existing PF measuring methods are mainly based on physical tests or numerical simulations, with the problems of high cost, time consumption, incomplete evaluation, and so on. Therefore, we combine numerical simulation and statistical regression to propose a complete PF prediction framework for the automobile seat slide including sensitivity analysis, dataset construction, and force prediction. Firstly, the sensitivity of each geometric variable of the slide rail is obtained by the interpretable linear regression. Secondly, the distance metric is constructed based on the variable sensitivity to evaluate the mutual difference between different working conditions. Thirdly, a reasonable construction scheme of the dataset based on the principle of sparseness is proposed. Finally, iterative local weighted linear regression (ILWLR) is established to achieve an accurate prediction of the slide PF, and the nearest adjacent sample distance is used to quantitatively evaluate the cost of the dataset supplement. Experimental results show that the iterative selection strategy and the weighted distance effectively reduce the model sensitivity to the number of nearest neighbor samples, and the proposed ILWLR method achieves better prediction performance than other local regression methods and statistical learning methods. Besides, considering the cost of data acquisition, an indirect comparison is conducted to verify the superiority of the proposed sparse principle sampling over random sampling in dataset construction. Finally, the mean absolute error (MAE) of the proposed PF prediction framework is 3.3 kN the mean relative error (MRE) is 4.3% in 150 simulated test conditions, and the absolute error is within 2 kN and the relative error is 4% in three physical test conditions, which is excellent for PF prediction task.

Key words: slide rail, peel strength, data driven, sensitivity analysis, locally weighted linear regression

CLC Number:

TH122

CHEN Qian, DONG Xingjian, CHEN Kangkang, LIU Wenbo, YUAN Shun, WU Peigui, NI Hongbin. Peeling Force Prediction of Automobile Seat Slide Rail Based on Iterative Local Weighted Linear Regression[J]. Journal of Mechanical Engineering, 2025, 61(17): 193-204.

References

[1] 工业和信息化部, 中国国家标准化管理委员会. GB 15083-2019汽车座椅、座椅固定装置及头枕强度要求和试验方法[S]. 北京: 中国标准出版社, 2019. Ministry of Industry and Information Technology, Standardization Administration of the People`s Republic of China. GB 15083-2019 Strength requirement and test method of automobile seats, their anchorages and any head restraints[S]. Beijing: Standards Press of China, 2019.
[2] 中国汽车工业协会. T/CAAMTB 46-2021 乘用车座椅用长滑轨技术要求和试验方法[S]. 北京: 中国标准出版社, 2021. China Association of Automobile Manufacturers. T/CAAMTB 46-2021 Technical requirements and test methods of long slide for passenger car seat[S]. Beijing: Standards Press of China, 2021.
[3] XIANG X, ZHANG H, ZHANG F, et al. Failure analysis of the conductive slide support rod of section insulator applied in urban rail transport system[J]. Engineering Failure Analysis, 2023, 148: 107217.
[4] OH M, LEE M K, KIM N. Robust design of roll-formed slide rail using response surface method[J]. Journal of Mechanical Science and Technology, 2010, 24(12): 2545-2553.
[5] LIU F P, SHANG J Z, LUO Z R, et al. Design of space sliding rails and mechanical analysis[J]. Applied Mechanics and Materials, 2014, 624: 207-212.
[6] 马雅丽, 叶志明, 孙守林, 等. 基于机床刚度的滑轨联接面性能分析[J]. 组合机床与自动化加工技术, 2013(11): 1-4. MA Yali, YE Zhiming, SUN Shoulin, et al. Research on slideways characteristics based on machine tool stiffness[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2013(11): 1-4.
[7] BAN H, ZHANG Y, FENG S. A data-driven approach for real-time prediction of thermal gradient in engineered structures[J]. Journal of Mechanical Science and Technology, 2022, 36(3): 1243-1249.
[8] WANG L, LIU G, ZHANG C, et al. FEM simulation-based adversarial domain adaptation for fatigue crack detection using lamb wave[J]. Sensors, 2023, 23(4): 1943.
[9] LIU Y, ZHAO Y, LIN Q, et al. DeviationGAN: A generative end-to-end approach for the deviation prediction of sheet metal assembly[J]. Mechanical Systems and Signal Processing, 2023, 204: 110822.
[10] 吴少杰, 刘怀举, 张仁华, 等. 基于正交实验和数据驱动的喷丸表面完整性参数预测[J]. 表面技术, 2021, 50(4): 86-95. WU Shaojie, LIU Huaiju, ZHANG Renhua, et al. Prediction of surface integrity parameters of shot peening based on orthogonal experiment and data-driven[J]. Surface Technology, 2021, 50(4): 86-95.
[11] 陈豪龙, 柳占立. 基于数据驱动模型求解热传导反问题[J]. 计算力学学报, 2021, 38(3): 272-279. CHEN Haolong, LIU Zhanli. Solving inverse heat conduction problems based on data-driven models[J]. Chinese Journal of Computational Mechanics, 2021, 38(3): 272-279.
[12] CHEN J, LI Y, LIU X, et al. A data-driven minimum stiffness prediction method for machining regions of aircraft structural parts[J]. The International Journal of Advanced Manufacturing Technology, 2022, 120(5-6): 3609-3623.
[13] HUI Y, BAI X, YANG Y, et al. A data-driven CUF-based beam model based on the tree-search algorithm[J]. Composite Structures, 2022, 300: 116123.
[14] 蒋茂飞, 许可, 刘亚龙, 等. 一种改进的局部线性回归估计器及其在雷达高度计海况偏差估计中的应用[J]. 电子与信息学报, 2016, 38(9): 2314-2320. JIANG Maofei, XU Ke, LIU Yalong, et al. Improved local linear regression estimator and its application to estimation for radar altimeter sea state bias[J]. Journal of Electronics & Information Technology, 2016, 38(9): 2314-2320.
[15] LIANG J, XIAO K, POINTER M R, et al. Spectra estimation from raw camera responses based on adaptive local-weighted linear regression[J]. Optics Express, 2019, 27(4): 5165-5180.
[16] KISI O, OZKAN C. A new approach for modeling sediment-discharge relationship: Local weighted linear regression[J]. Water Resources Management, 2017, 31(1): 1-23.
[17] GELER Z, KURBALIJA V, RADOVANOVIĆ M, et al. Comparison of different weighting schemes for the kNN classifier on time-series data[J]. Knowledge and Information Systems, 2016, 48(2): 331-378.
[18] HUANG M, LIN R, HUANG S, et al. A novel approach for precipitation forecast via improved K-nearest neighbor algorithm[J]. Advanced Engineering Informatics, 2017, 33: 89-95.
[19] ZHANG S, LI X, ZONG M, et al. Learning k for kNN classification[J]. ACM Transactions on Intelligent Systems and Technology, 2017, 8(3): 1-19.
[20] ZHANG S, LI X, ZONG M, et al. Efficient kNN classification with different numbers of nearest neighbors[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(5): 1774-1785.
[21] RODRIGUES É O. Combining Minkowski and Chebyshev: New distance proposal and survey of distance metrics using k-nearest neighbours classifier[J]. Pattern Recognition Letters, 2018, 110: 66-71.
[22] YE H J, ZHAN D C, JIANG Y. Fast generalization rates for distance metric learning: Improved theoretical analysis for smooth strongly convex distance metric learning[J]. Machine Learning, 2019, 108(2): 267-295.
[23] 杨德. 试验设计与分析[M]. 北京: 中国农业出版社, 2002. YANG De. Experimental design and analysis[M]. Beijing: China Agriculture Press, 2002.
[24] XU G, WEI H, WANG J, et al. A local weighted linear regression (LWLR) ensemble of surrogate models based on stacking strategy: Application to hydrodynamic response prediction for submerged floating tunnel (SFT)[J]. Applied Ocean Research, 2022, 125: 103228.
[25] CHEN W, XIE X, WANG J, et al. A comparative study of logistic model tree, random forest, and classification and regression tree models for spatial prediction of landslide susceptibility[J]. Catena, 2017, 151: 147-160.
[26] YILMAZ I, KAYNAR O. Multiple regression, ANN (RBF, MLP) and ANFIS models for prediction of swell potential of clayey soils[J]. Expert Systems with Applications, 2011, 38(5): 5958-5966.