Prediction Model for the Wall Thickness Reduction Rate in Tube Bending Based on the Random Forest and the Gradient Boosting Tree

ZHU Yingxia; YUAN Chen; WANG Lei; CHEN Wei; XU Jiangping

doi:10.3969/j.issn.1674-6457.2026.03.020

PDF(2222 KB)

Journal of Netshape Forming Engineering ›› 2026, Vol. 18 ›› Issue (3) : 184-193. DOI: 10.3969/j.issn.1674-6457.2026.03.020

Composites Forming

Prediction Model for the Wall Thickness Reduction Rate in Tube Bending Based on the Random Forest and the Gradient Boosting Tree

ZHU Yingxia^*, YUAN Chen, WANG Lei, CHEN Wei, XU Jiangping

Author information +

History +

Abstract

The work aims to construct a mapping relationship between process parameters and the wall thickness reduction rate with machine learning methods and establish a high-precision prediction model for the wall thickness reduction rate in tube bending, thereby providing a data-driven solution for effective defect prediction. Firstly, a simulation scheme was designed with six process parameters, such as diameter and bending degree, etc. as variables, and experimental validation was conducted to establish a database. Secondly, two machine learning models—Random Forest (RF) and Gradient Boosting Regression Tree (GBRT)—were developed. Finally, the effect of training sample size (ranging from N=100 to 808) on the generalization performance of the two models was compared and analyzed. Under a medium sample size (N=400), both machine learning models achieved optimal prediction performance. Among them, the GBRT model exhibited a coefficient of determination (R²) as high as 0.976 9 on the test set, a variance accounted for (VAF) of 97.69%, and a root mean square error (RMSE) and a mean absolute error (MAE) as low as 0.347 0 and 0.175 9, respectively. All performance metrics were significantly superior to those of the RF model under the same conditions (R²=0.922 9, VAF=92.36, RMSE=0.633 5, MAE=0.363 8). Beyond a sample size of 400, both models showed a trend of performance saturation and eventual degradation. Using process parameters outside the training range for GBRT model validation revealed an average prediction error of less than 7%. The GBRT prediction model constructed in this study can accurately and efficiently map the complex nonlinear relationship between the bending process parameters of bimetallic composite tubes and the wall thickness reduction rate, demonstrating the feasibility of machine learning methods as alternatives or supplements to traditional finite element analysis in this field.

Key words

bimetallic composite tube / bending / wall thickness reduction / machine learning / prediction model

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

ZHU Yingxia, YUAN Chen, WANG Lei, CHEN Wei, XU Jiangping. Prediction Model for the Wall Thickness Reduction Rate in Tube Bending Based on the Random Forest and the Gradient Boosting Tree[J]. Journal of Netshape Forming Engineering. 2026, 18(3): 184-193 https://doi.org/10.3969/j.issn.1674-6457.2026.03.020

References

[1] ZHU Y X, CHEN W, TU W B, et al.Three-Dimensional Finite Element Modeling of Rotary-Draw Bending of Copper-Titanium Composite Tube[J]. The International Journal of Advanced Manufacturing Technology, 2020, 106(5/6): 2377-2389.
[2] LIU X, LIN Y, FU H, et al.Preparation of the Capillary Copper/Titanium Composite Pipe by Floating-Plug Drawing Processing and Its Microstructure and Properties[J]. Chinese Journal of Engineering, 2017, 39(3): 417-425.
[3] PUGACHEVA N, KRYUCHKOV D, BYKOVA T, et al.Studying the Plastic Deformation of Cu-Ti-C-B Composites in a Favorable Stress State[J]. Materials, 2023, 16(8): 3204.
[4] MOSHIR S K, VAN HOA S, SHADMEHRI F.Structural Analysis of Composite Tubes Using a Meshless Analytical Dimensional Reduction Method[J]. International Journal for Numerical Methods in Engineering, 2021, 122(13): 3191-3217.
[5] JIN K, YUAN Q W, TAO J, et al.Analysis of the Forming Characteristics for Cu/Al Bimetal Tubes Produced by the Spinning Process[J]. The International Journal of Advanced Manufacturing Technology, 2019, 101(1/2/3/ 4): 147-155.
[6] LI H, SHI K P, YANG H, et al. Significance Analysis of Processing Parameters on Wall Thinning in Tube Bending[J]. Advanced Materials Research, 2012, 622/623: 437-441.
[7] LI H P, LIU Y L, ZHU Y X, et al.Global Sensitivity Analysis and Coupling Effects of Forming Parameters on Wall Thinning and Cross-Sectional Distortion of Rotary Draw Bending of Thin-Walled Rectangular Tube with Small Bending Radius[J]. The International Journal of Advanced Manufacturing Technology, 2014, 74(5): 581-589.
[8] KARTHIKEYAN M, JENARTHANAN M P.Experimental Study of Wall Thinning Behaviour in Boiler Pipes during Bending Process[J]. Australian Journal of Mechanical Engineering, 2020, 18(sup1): S88-S94.
[9] XIONG J Y, LIANG W, DING Y, et al.Quantitative Analysis of Wall Thinning of Bimetallic Clad Steel Tube Based on Pulsed Eddy Current[J]. Process Safety Progress, 2022, 41: 118-128.
[10] HU S H, CHENG C, ABD E A, et al.Forming Characteristics of Thin-Walled Tubes Manufactured by Free Bending Process-Based Nontangential Rotation Bending Die[J]. Thin-Walled Structures, 2024, 194: 111313.
[11] FANG J, OUYANG F, LU S Q, et al.Wall Thinning Behaviors of High Strength 0Cr21Ni6Mn9N Tube in Numerical Control Bending Considering Variation of Elastic Modulus[J]. Advances in Mechanical Engineering, 2021,13(5): 16878140211021241.
[12] NATH D, ANKIT, NEOG D R, et al. Application of Machine Learning and Deep Learning in Finite Element Analysis: A Comprehensive Review[J]. Archives of Computational Methods in Engineering, 2024, 31(5): 2945-2984.
[13] VOLKAN GRÜ, BAH M M, EVIK M.Machine Learning for the Prediction of Problems in Steel Tube Bending Process[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 108584.
[14] YAZDANI A, KNOCHE J, ENGEL B, et al.Tube Geometry Prediction in Rotary Draw Bending Process Using Random Forest Regression[J]. Automatisierungstechnik, 2025, 73(4): 223-231.
[15] AHMAD BHAT S, HUSSAIN I, HUANG N F.Soil Suitability Classification for Crop Selection in Precision Agriculture Using GBRT-Based Hybrid DNN Surrogate Models[J]. Ecological Informatics, 2023, 75: 102109.
[16] ZHU Z Y, ZHENG Y H, WANG X Y, et al.Forecasting China’s Precious Metal Futures Volatility: GBRT Models and Time-Model Dimension Combination of Tree SHAP[J]. International Review of Financial Analysis, 2025, 104: 104249.
[17] HE C.Enterprise Financial Risk Warning Based on Random Forest Algorithm[J]. Procedia Computer Science, 2025, 261: 1229-1237.
[18] TIRUMANADHAM N S K M K, PRIYADARSHINI V, PRAVEEN S P, et al. Optimizing Lung Cancer Prediction Models: A Hybrid Methodology Using GWO and Random Forest[M]. Cham: Springer Nature Switzerland, 2025: 59-77.
[19] SEGHIER M E A B, PLEVRIS V, SOLORZANO G. Random Forest-Based Algorithms for Accurate Evaluation of Ultimate Bending Capacity of Steel Tubes[J]. Structures, 2022, 44: 261-273.
[20] WU T Y, HUNG Y C.Tube Wrinkle State Estimation in Bending Machine Using Servo Signal Analysis and Random Forests[J]. The International Journal of Advanced Manufacturing Technology, 2025, 136(7): 3329-3344.
[21] 岳峰丽, 郭威, 刘明华, 等. 基于机器学习算法的超声波预测TP2铜材晶粒度模型优化研究[J]. 精密成形工程, 2025, 17(9): 185-194.
YUE F L, GUO W, LIU M H, et al.Optimization of Ultrasonic Prediction of TP2 Copper Grain Size Model Based on Machine Learning Algorithm[J]. Journal of Netshape Forming Engineering, 2025, 17(2): 185-194.
[22] MALASHIN I, TYNCHENKO V, GANTIMUROV A, et al.Boosting-Based Machine Learning Applications in Polymer Science: A Review[J]. Polymers, 2025, 17(4): 499.
[23] BARTON N A, HALLETT S H, JUDE S R, et al.Predicting the Risk of Pipe Failure Using Gradient Boosted Decision Trees and Weighted Risk Analysis[J]. NPJ Clean Water, 2022, 5(1): 22.
[24] ZHAO Y J, LAI M, WU Y Q, et al.Fatigue Life Prediction of Aluminum-Steel Magnetic Pulse Crimped Joints Based on Point Cloud Measurement and Gradient Boosting Regression Trees[J]. International Journal of Fatigue, 2025: 109020.
[25] WANG G Q, RUAN Y L, WANG H X, et al.Tribological Performance Study and Prediction of Copper Coated by MoS₂ Based on GBRT Method[J]. Tribology International, 2023, 179: 108149.
[26] LIANG W Z, LUO S Z, ZHAO G Y, et al.Predicting Hard Rock Pillar Stability Using GBDT, XGBoost, and LightGBM Algorithms[J]. Mathematics, 2020, 8(5): 765.
[27] ZHU Y X, WAN M M, WANG Y, et al.Size Effect Mechanism of Cross-Section Deformation and Section Hollow Coefficient-Bending Degree of the Thin-Walled Composite Bending Tube[J]. Materials & Design, 2021, 212: 110274.
[28] MCKAY M D, BECKMAN R J, CONOVER W J.A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code[J]. Technometrics, 2000, 42(1): 55-61.