Thermal Error Modeling Method towards Electric Spindles Based on Digital Twin and Deep Transfer Learning

doi:10.3901/JME.2025.03.052

Abstract

Abstract: Thermal error is one of the main sources of electric spindle system errors, and thermal error modeling is an important means to improve system reliability. In machining scenarios where the electric spindle is loaded with tools, it needs to move in multiple directions, which makes real-time measurement difficult and difficult to collect sufficient thermal error samples. Furthermore, the data distribution under different working conditions presents large discrepancies, and a well-trained model under one working condition failed to obtain satisfactory prediction accuracy when applied to another working condition. To address these issues, a thermal error modeling approach based on digital twins and deep transfer learning is proposed. Firstly, a digital twin model of the thermal behavior of the electric spindle system is established, where the temperature fields and thermal deformation data under different working conditions can be simulated to alleviate the limitation of the scarcity of the thermal error samples in real scenarios. Secondly, a convolutional bidirectional long short-term memory network based on the domain adversarial mechanism is developed. The virtual data generated by the digital twin model is used as the source domain, and the real data is used as the target domain. Convolutional layers of different scales are used as the feature extractor to extract the spatial features of the temperature data from both the source and target domains so as to address the collinearity issue of multi-dimensional temperature features. The bidirectional long short-term memory network is constructed as a predictor to process the time-series relationship between temperature and thermal error and output predictions. Additionally, the adversarial training technique of domain adaptation is employed to confuse the two domain features and minimize the distribution discrepancies between both domains, thereby improving the model's generalization ability. Finally, a multi-source data collaborative collection platform is established to obtain real data under variable working conditions. Different transfer tasks are constructed to validate the proposed method and the results showed the proposed method successfully achieves thermal error modeling in the absence of labeled thermal error samples and exhibits good prediction performance.

Key words: electric spindles, thermal error modeling, digital twin, deep transfer learning, domain adaptation

CLC Number:

TP391
TP399

MA Shuai, LENG Jiewu, CHEN Zhuyun, LI Weihua, LI Bo, LIU Qiang. Thermal Error Modeling Method towards Electric Spindles Based on Digital Twin and Deep Transfer Learning[J]. Journal of Mechanical Engineering, 2025, 61(3): 52-66.

References

[1] BRYAN J. International status of thermal error research (1990)[J]. CIRP Annals，1990，39(2):645-656.
[2] MAYR J，JEDRZEJEWSKI J，UHLMANN E，et al. Thermal issues in machine tools[J]. CIRP Annals，2012，61(2):771-791.
[3] MAREŠ M，HOREJŠ O，HAVLÍK L. Thermal error compensation of a 5-axis machine tool using indigenous temperature sensors and CNC integrated Python code validated with a machined test piece[J]. Precision Engineering，2020，66:21-30.
[4] LI Z，WANG Q，ZHU B，et al. Thermal error modeling of high-speed electric spindle based on Aquila Optimizer optimized least squares support vector machine[J]. Case Studies in Thermal Engineering，2022，39:102432.
[5] WEI X，YE H，MIAO E，et al. Thermal error modeling and compensation based on Gaussian process regression for CNC machine tools[J]. Precision Engineering，2022，77:65-76.
[6] TAN F，YIN G，ZHENG K，et al. Thermal error prediction of machine tool spindle using segment fusion LSSVM[J]. The International Journal of Advanced Manufacturing Technology，2021，116(1):99-114.
[7] LIU Y，MIAO E，LIU H，et al. Robust machine tool thermal error compensation modelling based on temperature-sensitive interval segmentation modelling technology[J]. The International Journal of Advanced Manufacturing Technology，2020，106(1):655-669.
[8] ZHANG X，LIU Y，LOU P，et al. Thermal error modeling for heavy duty CNC machine tool based on convolution neural network[C]//2019 IEEE 3rd Information Technology，Networking，Electronic and Automation Control Conference (ITNEC). IEEE，2019:665-669.
[9] CHEN Y，CHEN J，XU G. A data-driven model for thermal error prediction considering thermoelasticity with gated recurrent unit attention[J]. Measurement，2021，184:109891.
[10] LIU J，MA C，GUI H，et al. Thermally-induced error compensation of spindle system based on long short term memory neural networks[J]. Applied Soft Computing，2021，102:107094.
[11] GUO J，XIONG Q，CHEN J，et al. Study of static thermal deformation modeling based on a hybrid CNN-LSTM model with spatiotemporal correlation[J]. The International Journal of Advanced Manufacturing Technology，2022:1-13.
[12] LIU P，DU Z，LI H，et al. Thermal error modeling based on BiLSTM deep learning for CNC machine tool[J]. Advances in Manufacturing，2021，9(2):235-249.
[13] CHEN Z，XIA J，LI J，et al. Generalized open-set domain adaptation in mechanical fault diagnosis using multiple metric weighting learning network[J]. Advanced Engineering Informatics，2023，57:102033.
[14] 赵树煊，张洁，汪俊亮，等. 基于两阶段深度迁移学习的面料疵点检测算法[J]. 机械工程学报，2021，57(17):86-97. ZHAO Shuxuan，ZHANG Jie，WANG Junliang，et al. Fabric defect detection algorithm based on two-stage deep transfer learning[J]. Journal of Mechanical Engineering，2021，57(17):86-97
[15] 邓柯楠，高栋，马守东，等. 基于迁移学习的铣削机器人定位误差补偿方法[J]. 机械工程学报，2022，58(14):170-180. DENG Kenan，GAO Dong，MA Shoudong，et al. An efficient error compensation method for milling robot based on transfer learning[J]. Journal of Mechanical Engineering，2022，58(14):170-180
[16] CHEN Z，LIAO Y，LI J，et al. A multi-source weighted deep transfer network for open-set fault diagnosis of rotary machinery[J]. IEEE Transactions on Cybernetics，2023，53(3):1982-1993.
[17] 陈祝云，钟琪，黄如意，等. 基于增强迁移卷积神经网络的机械智能故障诊断[J]. 机械工程学报，2021，57(21):96-105. CHEN Zhuyun，ZHONG Qi，HUANG Ruyi，et al. Intelligent fault diagnosis for machinery based on enhanced transfer convolutional neural network[J]. Journal of Mechanical Engineering，2021，57(21):96-105
[18] LI P，LOU P，YAN J，et al. The thermal error modeling with deep transfer learning[J]. Journal of Physics:Conference Series，2020，1576(1):12003.
[19] ZHANG X，WANG C，LOU P，et al. Machine Tools Thermal Error Modeling with Imbalanced Data Based on Transfer Learning[C]//2020 IEEE 5th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA). IEEE，2020:214-218.
[20] GRIEVES M，VICKERS J. Digital twin:Mitigating unpredictable，undesirable emergent behavior in complex systems[J]. Transdisciplinary Perspectives on Complex Systems:New Findings and Approaches，2017:85-113.
[21] TAO F，ZHANG H，LIU A，et al. Digital twin in industry:State-of-the-art[J]. IEEE Transactions on Industrial Informatics，2018，15(4):2405-2415.
[22] LENG J，WANG D，SHEN W，et al. Digital twins-based smart manufacturing system design in Industry 4.0:A review[J]. Journal of Manufacturing Systems，2021，60:119-137.
[23] LIU J，GUI H，MA C. Digital twin system of thermal error control for a large-size gear profile grinder enabled by gated recurrent unit[J]. Journal of Ambient Intelligence and Humanized Computing，2023，14(2):1269-1295.
[24] LU Q，ZHU D，WANG M，et al. Digital twin-driven thermal error prediction for CNC machine tool spindle[J]. Lubricants，2023，11(5):219.
[25] GANIN Y，USTINOVA E，AJAKAN H，et al. Domain-adversarial training of neural networks[J]. The Journal of Machine Learning Research，2016，17(1):2096-2030.
[26] JIAN B，GUO Y，HU C，et al. Prediction of spindle thermal deformation and displacement using back propagation neural network[J]. Sensors & Materials，2020，32:431-445.
[27] CHENG Y，ZHANG X，ZHANG G，et al. Thermal error analysis and modeling for high-speed motorized spindles based on LSTM-CNN[J]. The International Journal of Advanced Manufacturing Technology，2022，121(5):3243-3257.
[28] WU C，XIANG S，XIANG W. Spindle thermal error prediction approach based on thermal infrared images:A deep learning method[J]. Journal of Manufacturing Systems，2021，59:67-80.
[29] ZOU Z，YAN W，MA W，et al. Development of thermal error mapping model for the dry gear hobbing machine based on CNN-DAE integrated structure and its application[J]. The International Journal of Advanced Manufacturing Technology，2021，113(7):2343-2354.
[30] CHEN Y，ZHOU H，CHEN J，et al. Spindle thermal error modeling method considering the operating condition based on long short-term memory[J]. Engineering Research Express，2021，3(3):35019.
[31] LIANG Y C，LI W D，LOU P，et al. Thermal error prediction for heavy-duty CNC machines enabled by long short-term memory networks and fog-cloud architecture[J]. Journal of Manufacturing Systems，2022，62:950-963.
[32] LIU Y，LI K，TSAI Y. Spindle thermal error prediction based on LSTM deep learning for a CNC machine tool[J] Applied Sciences，2021，11(12):5444.