Mechanism-data hybrid modeling method for the drafting section of a sliver machine
Abstract
The rapid development of information technology has enhanced the accuracy and depth of intelligent algorithms. However, significant challenges remain in expanding the application scope of intelligent systems and achieving synergistic coupling between mechanical and digital technologies. Inspired by multiscale dynamic feedback mechanisms in cellular biomechanics, such as cytoskeletal remodeling and mechanotransduction in extracellular matrix fiber networks, this study proposes a bio-inspired hybrid modeling framework that analogizes the integration of mechanical and digital systems to the coordinated molecule-cell-tissue multiscale mechanical responses in biological systems. A high-precision mechanistic model is constructed using Recurdyn multi-body dynamics simulation software to capture the physical characteristics of the drafting section through a strategy analogous to multiscale mechanical modeling of ECM (extracellular matrix) fibrous networks. Simultaneously, a GRU (Gated Recurrent Unit) neural network-based data-driven model is developed to emulate the adaptability of biological neural systems, particularly the feedback regulation of neuronal networks under dynamic mechanical stimuli. By calculating residuals between the mechanistic model, data-driven model, and experimental measurements, a dual-channel SKNet (Selective Kernel Networks) architecture is introduced to mimic the multiscale signal extraction properties of mechanosensitive ion channels in cellular biomechanics. Convolutional kernels of different scales extract residual features, and an LSTM (Long Short-Term Memory) residual model is constructed for compensation. Experimental validation demonstrates that the hybrid model significantly improves prediction accuracy and robustness, with its residual compensation mechanism functionally resembling the dynamic repair processes of cells under mechanical stress. This study provides an efficient solution for drafting section modeling and offers methodological insights for interdisciplinary applications in biomaterial fabrication and tissue engineering bio-inspired design.
References
1. Catanese C, Ayassi R, Pincemin E, et al. A Fully Connected Neural Network Approach to Mitigate Fiber Nonlinear Effects in 200G DP-16-QAM Transmission System. In: Proceedings of the 2020 22nd International Conference on Transparent Optical Networks (ICTON). pp. 1-4. doi: 10.1109/icton51198.2020.9203197
2. Meng J, Xu X, Jiang C, et al. Tensile force field plays a crucial role in local invasion of tumor cells through a mechano-chemical coupling mechanism. Soft Matter. 2024; 20(30): 6002-6015. doi: 10.1039/d4sm00335g.
3. Ye X, Yang Y, Jiao P, et al. Underwater minirobots actuated by hybrid driving method. Journal of Zhejiang University-SCIENCE A. 2023; 24(7): 596-611. doi: 10.1631/jzus.a2300056
4. Siddiqui Q, Abro Z, Yu C. Study of drafting force variability and sliver irregularity at the break draft zone of a draw frame. Textile Research Journal. 2015; 85(14): 1465-1473. doi: 10.1177/0040517514563724
5. Zhang Z, Yu C. Study on drafting force and sliver irregularity on drawing frame. Journal of The Textile Institute. 2012; 103(3): 341-347.
6. Lee J, Wu F, Zhao W, et al. Prognostics and health management design for rotary machinery systems—Reviews, methodology and applications. Mechanical Systems and Signal Processing. 2014; 42(1-2): 314-334. doi: 10.1016/j.ymssp.2013.06.004
7. Du C, Gong F, Zhou Y, et al. A Mechanistic-Based Data-Driven Modelling Framework for Predicting Production of Electric Submersible Pump Wells in Offshore Oilfield. Published online 2024. doi: 10.2139/ssrn.4946667
8. Li Q, Niu F, Yang H, et al. Magnetically Actuated Soft Microrobot with Environmental Adaptative Multimodal Locomotion Towards Targeted Delivery. Advanced Science. 2024; 11(43). doi: 10.1002/advs.202406600
9. Vijay U, Galvez-Aranda DE, Zanotto FM, et al. A hybrid modelling approach coupling physics-based simulation and deep learning for battery electrode manufacturing simulations. Energy Storage Materials. 2025; 75: 103883. doi: 10.1016/j.ensm.2024.103883
10. Michalopoulou F, Papathanasiou MM. An approach to hybrid modelling in chromatographic separation processes. Digital Chemical Engineering. 2025; 14: 100215. doi: 10.1016/j.dche.2024.100215
11. Roland W, Marschik C, Kommenda M, et al. Predicting the Non-Linear Conveying Behavior in Single-Screw Extrusion: A Comparison of Various Data-Based Modeling Approaches used with CFD Simulations. International Polymer Processing. 2021; 36(5): 529-544. doi: 10.1515/ipp-2020-4094
12. Leite WL, Jing Z, Kuang H, et al. Multilevel Mixture Modeling with Propensity Score Weights for Quasi-Experimental Evaluation of Virtual Learning Environments. Structural Equation Modeling: A Multidisciplinary Journal. 2021; 28(6): 964-982. doi: 10.1080/10705511.2021.1919895
13. Linthicum DS. Connecting Fog and Cloud Computing. IEEE Cloud Computing. 2017; 4(2): 18-20. doi: 10.1109/mcc.2017.37
14. Gusev M, Dustdar S. Going Back to the Roots—The Evolution of Edge Computing, An IoT Perspective. IEEE Internet Computing. 2018; 22(2): 5-15. doi: 10.1109/mic.2018.022021657
15. Mitrofanov S, Semenkin E. An Approach to Training Decision Trees with the Relearning of Nodes. In: Proceedings of the 2021 International Conference on Information Technologies (InfoTech). pp. 1-5. doi: 10.1109/infotech52438.2021.9548520
16. Xu W, Peng H, Zeng X, et al. A Hybrid Modeling Method Based on Linear AR and Nonlinear DBN-AR Model for Time Series Forecasting. Neural Processing Letters. 2021; 54(1): 1-20. doi: 10.1007/s11063-021-10651-2
17. Hu X, Shi L, Lin G, et al. Comparison of physical-based, data-driven and hybrid modeling approaches for evapotranspiration estimation. Journal of Hydrology. 2021; 601: 126592. doi: 10.1016/j.jhydrol.2021.126592
18. Panchikkil S, Manikandan VM, Zhang YD. A convolutional neural network model based reversible data hiding scheme in encrypted images with block-wise Arnold transform. Optik. 2022; 250: 168137. doi: 10.1016/j.ijleo.2021.168137
19. Pally RJ, Samadi S. Application of image processing and convolutional neural networks for flood image classification and semantic segmentation. Environmental Modelling & Software. 2021; 105-285.
20. Chen Z. Machine remaining useful life prediction method based on global-local attention compensation network. Reliability Engineering & System Safety. 2025; 255: 110652. doi: 10.1016/j.ress.2024.110652
21. Peng G, Qiu H, Li Y, et al. A Sub-Channel Spatial Homogeneity-Based Channel Estimation Method for Underwater Optical Densely Arrayed MIMO Systems. Journal of Marine Science and Engineering. 2024; 12(11): 2030. doi: 10.3390/jmse12112030
22. Cao Y, Ren X, Guo Q, et al. Model Reference Adaptive System of Permanent Magnet Synchronous Motor Based on Current Residual Compensation Without Position Measurement. Actuators. 2024; 13(11): 446. doi: 10.3390/act13110446
23. Li Y, Chen X. Attributed graph subspace clustering with residual compensation guided by adaptive dual manifold regularization. Expert Systems with Applications. 2024; 255: 124699. doi: 10.1016/j.eswa.2024.124699
24. Zhang X, Huang Z, Li Q, et al. Legged robot-aided 3D tunnel mapping via residual compensation and anomaly detection. ISPRS Journal of Photogrammetry and Remote Sensing. 2024; 214: 33-47. doi: 10.1016/j.isprsjprs.2024.05.025
25. Liu J, Yang X. Remote sensing image super-resolution based on cross residual compensation mechanism and attention mechanism. Multimedia Tools and Applications. 2023; 83(18): 55641-55657. doi: 10.1007/s11042-023-17645-3
26. Yang F, Chen X, Chai L. Optimal graph edge weights driven nlms with multi-layer residual compensation. EURASIP Journal on Advances in Signal Processing. 2021; 2021(1). doi: 10.1186/s13634-021-00800-z
27. Sustainability Research-Sustainable Development; Study Findings on Sustainable Development Detailed by Researchers at Memorial University of Newfoundland (Goal-Driven or Data-Driven Inventory of Sustainability Indicator Initiatives in Rural Canada). Information Technology Newsweekly. 2020; 994-998. Available: https://xueshu.baidu.com/usercenter/paper/show?paperid=130r0gx0ky120ed0f32v009078272700&site=xueshu_se (accessed on 10 March 2025)
28. Nekoonam A, Montazeri‐Gh M. Noise‐robust gas path fault detection and isolation for a power generation gas turbine based on deep residual compensation extreme learning machine. Energy Science & Engineering. 2023; 11(11): 4001-4018. doi: 10.1002/ese3.1576
29. Chen H, Hu L, Wang H, et al. Comparison of Active Full Compensation Control Algorithms for Single-phase Grounding Residual Current in Distribution Networks. Journal of Physics: Conference Series. 2023; 2530(1): 012011. doi: 10.1088/1742-6596/2530/1/012011
30. Information Technology; New Data from Russian Academy of Sciences Illuminate Findings in Information Technology (Data-driven Modeling of Stomatal, Mesophyll, and Biochemical Regulation of Scots Pine Photosynthesis). Information Technology Newsweekly. 2020; 350-360. Available: https://xueshu.baidu.com/usercenter/paper/show?paperid=1f0a0v2053160mg0ym180tn0rn477618&site=xueshu_se&hitarticle=1 (accessed on 10 March 2025)
31. Mo L, Xiao X, Chen Y, et al. A Rapid Decompression Failure Mechanism Model Coupling Diffusion-Deformation of O-Ring Sealing Rubber Exposed to High-Pressure Carbon Dioxide. Published online 2024. doi: 10.2139/ssrn.5021481
32. Guo F, Zhang H, Song C, et al. Identification and Precise Optimization of Key Assembly Error Links for Complex Aviation Components Driven by Mechanism and Data Fusion Model. Published online 2024. doi: 10.2139/ssrn.4944504
33. Xie N, Xu W. Analysis of the Evaluation Data of College Classroom Teaching Quality based on Fuzzy Mathematics Model. Frontiers in Computing and Intelligent Systems. 2024; 10(3): 86-94. doi: 10.54097/h7fd6821
34. Tuly SS, Joardder MUH, Welsh ZG, et al. A novel mechanistic model for predicting shrinkage kinetics in plant-based foods by integrating solid matrix mobility and viscoelasticity. Journal of Food Engineering. 2025; 387: 112346. doi: 10.1016/j.jfoodeng.2024.112346
Copyright (c) 2025 Author(s)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright on all articles published in this journal is retained by the author(s), while the author(s) grant the publisher as the original publisher to publish the article.
Articles published in this journal are licensed under a Creative Commons Attribution 4.0 International, which means they can be shared, adapted and distributed provided that the original published version is cited.
Submit a Paper
