Power Loss Prediction of Integrated Hydro-mechanical Transmission Device in Tracked Vehicle Based on Stacking Ensemble Learning

doi:10.12382/bgxb.2025.0175

Abstract

Abstract:

Power loss is one of the critical parameters for evaluating the performance of integrated hydro-mechanical transmission device (IHMTD) in tracked vehicle.The efficiency of IHMTD is comprehensively assessed by predicting its power losses under different operating conditions,thereby ensuring the good mobility of tracked vehicles.This paper proposes a method based on stacked ensemble learning for predicting the power losses of IHMTDs under various working conditions.The efficiency data from 75 tracked vehicles IHMTD under multiple working conditions is used for power loss prediction,and the Random Forest,LightGBM,AdaBoost,CatBoost and XgBoost algorithms are integrated together through stacking to effectively predict the power losses of IHMTD.Additionally,SHapley Additive exPlanations (SHAP) values are employed to analyze the impacts of various factors and models on power loss prediction.The experimental results show that the root mean square error (RMSE) of power loss prediction is 6.6,and the goodness of fit (R²) reaches 0.976 on the training set; while on the test set,these metrics are 8.92 and 0.961,respectively.Further analysis reveals that the input torque is the primary factor affecting power losses,and within the stacked ensemble framework,the Random Forest algorithm contributes the most to improve the prediction accuracy.

Key words: integrated hydro-mechanical transmission device, power loss prediction, stacking ensemble learning, contribution explanation

CLC Number:

U463.22

LI Shenlong, ZHANG Jinbao, WANG Liyong, ZHANG Jinle, WANG Min. Power Loss Prediction of Integrated Hydro-mechanical Transmission Device in Tracked Vehicle Based on Stacking Ensemble Learning[J]. Acta Armamentarii, 2025, 46(11): 250175-.

Figures/Tables 12

Fig.1 Schematic diagram of IHMTD

Fig.2 Power loss test bench of IHMTD

Table 1 Design of power loss test scheme for IHMTD

速度/ (r·min^-1)	挡位
速度/ (r·min^-1)	N	F1	F2	F3	F4	F5	F6	R1	R2
1200	0	0	0	0	0	0	0	0	0
1400	0	0	0	0	0	0	0	0	0
1600	0	0/50	0/50	0/50/100	0/50/100	0/50/100	0/50/100	0/50	0/50
1800	0	0/50	0/50	0/50/100	0/50/100	0/50/100	0/50/100	0/50	0/50
2000	0	0/50/100	0/50/100	0/50/100	0/50/100	0/50/100	0/50/100	0/50	0/50
2200	0	0/50/100	0/50/100	0/50/100	0/50/100	0/50/100	0/50/100	0/50	0/50

Fig.3 Statistical analysis of power loss data under different working conditions

Fig.4 The variation of power loss with load under different gears and speeds

Fig.5 Framework of the stacking ensemble learning

Fig.6 Fitting and prediction evaluation indicators of the stacking ensemble learning

Fig.7 Fitting and prediction performance of the stacking ensemble learning

Fig.8 Comparison radar chart of the power losses of random test samples at various operating points

Fig.9 The degree of influence of various factors on power loss

Fig.10 Prediction of power loss with input torque under different loads

Fig.11 The contributions of various algorithms in power loss prediction

References 22

[1]	高蕾, 刘志浩, 高钦和, 等. 多轴特种车辆传动系统机械摩擦阻力损失与传动效率研究综述[J]. 振动与冲击, 2023, 42(18):138-154.
	GAO L, LIU Z H, GAO Q H, et al. Review on the mechanical friction resistance loss and transmission efficiency of a multi-axle special vehicle transmission system[J]. Journal of Vibration and Shock, 2023, 42(18):138-154. (in Chinese)
[2]	王斌, 宁斌, 陈辛波, 等. 齿轮传动搅油功率损失的研究进展[J]. 机械工程学报, 2020, 56(23):1-20. doi: 10.3901/JME.2020.23.001
	WANG B, NING B, CHEN X B, et al. Research progress in churning losses of gear transmission[J]. Journal of Mechanical Engineering, 2020, 56(23):1-20. (in Chinese) doi: 10.3901/JME.2020.23.001
[3]	徐保荣, 张金豹, 姚李刚, 等. 液力机械综合传动装置低温阻力矩特性研究与验证[J]. 兵工学报, 2023, 44(6):1829-1836. doi: 10.12382/bgxb.2022.0182
	XU B R, ZHANG J B, YAO L G, et al. Study and validation of resistance torque characteristics of hydro-mechanical comprehensive transmission device under low temperature conditions[J]. Acta Armamentarii, 2023, 44(6):1829-1836. (in Chinese) doi: 10.12382/bgxb.2022.0182
[4]	李义军, 毕小平, 马宁, 等. 车辆传动装置产热建模与计算分析[J]. 中国工程机械学报, 2013, 11(3):216-221.
	LI Y J, BI X P, MA N, et al. Modeling and analysis on the transmission heat generation of vehicles[J]. Chinese Journal of Construction Machinery, 2013, 11(3):216-221. (in Chinese)
[5]	SHEN Y, RINDERKNECHT S. A method on modelling and analyzing the power losses in vehicle transmission[J]. Forschung im Ingenieurwesen, 2018, 82(7):261-270. doi: 10.1007/s10010-018-0276-y
[6]	高蕾, 刘志浩, 高钦和, 等. 特种车辆传动系统效率优化与性能评估方法研究综述[J]. 火炮发射与控制学报, 2024, 45(3):6-16.
	GAO L, LIU Z H, GAO Q H, et al. A review of research methods for efficiency optimization and performance evaluation of special vehicle transmission systems[J]. Journal of Gun Launch & Control, 2024, 45(3):6-16. (in Chinese)
[7]	张金豹, 邹天刚, 高秀才, 等. 基于中位秩加权的车辆传动装置直驶模式递效率性能评估[J]. 兵工学报, 2022, 43(增刊1):54-59.
	ZHANG J B, ZOU T G, GAO X C, et al. Straight driving efficiency assessment of vehicle transmission device based on the median-rank weighting[J]. Acta Armamentarii, 2022, 43(S1):54-59. (in Chinese) doi: 10.12382/bgxb.2022.A018
[8]	张金豹, 盖江涛, 安媛媛, 等. 基于智能优化相关向量机的综合传动变工况直驶传递效率预测[J]. 兵工学报, 2023, 44(1):270-278. doi: 10.12382/bgxb.2022.0728
	ZHANG J B, GAI J T, AN Y Y, et al. Straight driving transmission efficiency prediction of integrated transmission under variable working conditions based on relevance vector machine with intelligent optimization[J]. Acta Armamentarii, 2023, 44(1):270-278. (in Chinese)
[9]	KHATTI J, GROVER K S. Prediction of compaction parameters of compacted soil using LSSVM,LSTM,LSBoostRF,and ANN[J]. Innovative Infrastructure Solutions, 2023, 8(2):76. doi: 10.1007/s41062-023-01048-2
[10]	QIU S H, CUI X P, PING Z W, et al. Deep learning techniques in intelligent fault diagnosis and prognosis for industrial systems:a review[J]. Sensors, 2023, 23:1305. doi: 10.3390/s23031305 URL
[11]	LI W X, LAW K L E. Deep learning models for time series forecasting:a review[J]. IEEE Access, 2024, 12:92306-92327. doi: 10.1109/ACCESS.2024.3422528 URL
[12]	FAWAGREH K, GABER M M, ELYAN E. Random forests:from early developments to recent advancements[J]. Systems Science & Control Engineering:An Open Access Journal, 2014, 2:602-609.
[13]	WANG R H. AdaBoost for feature selection,classification and its relation with SVM,a review[J]. Physics Procedia, 2012, 25:800-807. doi: 10.1016/j.phpro.2012.03.160 URL
[14]	CHEN T Q, GUESTRIN C. Xgboost:a scalable tree boosting system[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York,NY,US:ACM, 2016:785-794.
[15]	KE G L, MENG Q, FINLEY T, et al. LightGBM:a highly efficient gradient boosting decision tree[J]. Advances in Neural Information Processing Systems, 2017, 30:3149-3157.
[16]	HANCOCK J T, KHOSHGOFTAAR T M. CatBoost for big data:an interdisciplinary review[J]. Journal of Big Data, 2020, 7:94. doi: 10.1186/s40537-020-00369-8
[17]	MIENYE I D, SUN Y. A survey of ensemble learning:Concepts,algorithms,applications,and prospects[J]. IEEE Access, 2022, 10:99129-99149. doi: 10.1109/ACCESS.2022.3207287 URL
[18]	李驰运, 缪建明, 沈丙振. 基于改进Stacking集成学习方法的武器装备体系作战效能预测[J]. 兵工学报, 2023, 44 (11):3455-3464. doi: 10.12382/bgxb.2022.0797
	LI C Y, MIAO J M, SHEN B Z. Operational effectiveness prediction of weapon equipment system based on improved stacking ensemble learning method[J]. Acta Armamentarii, 2023, 44 (11):3455-3464. (in Chinese) doi: 10.12382/bgxb.2022.0797
[19]	WANG Y W, SUN Y Y, DAN Y Q, et al. Online load-loss risk assessment based on stacking ensemble learning for power systems[J]. Frontiers in Energy Research, 2023, 11:1281368. doi: 10.3389/fenrg.2023.1281368 URL
[20]	GUO X F, GAO Y, ZHENG D, et al. Study on short-term photovoltaic power prediction model based on the Stacking ensemble learning[J]. Energy Reports, 2020, 6:1424-1431. doi: 10.1016/j.egyr.2020.11.006 URL
[21]	SANTOS M R, GUEDES A, SANCHEZ-GENDRIZ I. SHapley additive exPlanations (SHAP) for efficient feature selection in rolling bearing fault diagnosis[J]. Machine Learning and Knowledge Extraction, 2024, 6:316-341. doi: 10.3390/make6010016 URL
[22]	WU L Y, LI J H, ZHANG J W, et al. Prediction model for the compressive strength of rock based on stacking ensemble learning and shapley additive explanations[J]. Bulletin of Engineering Geology and the Environment, 2024, 83:439. doi: 10.1007/s10064-024-03896-3