Manufacturing Technology Today

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Thermal Error Modeling of Machine Tool Spindle Through an Ensemble Approach


Affiliations
1 Indian Institute of Technology Madras, Chennai, India
     

   Subscribe/Renew Journal


Thermal error compensation of machine tool is cost-effective than other methods. Towards this, data-driven machine learning (ML) algorithms have been used to produce accurate prediction models. However, ML models have limitations, such as overfitting, requiring a large data etc. In present work, a hybrid model is proposed by exploiting the linear regression (LR), support vector machine (SVM), neural network (NN), and decision tree (DT) models. For this purpose, the optimum weights to each constituent model is identified by cosine similarity maximization. The developed models are validated against the experimental data. The prediction results with optimized weight are compared with equal weights and the root means square error (RMSE) for both methods are 1.8879 and 2.8978, respectively. The RMSE shows that the hybrid model produces good accuracy for both small and large data sets compared to individual models.


Keywords

Hybrid Model, Cosine Maximization, Thermal Error, Support Vector Machine, Linear Regression, Neural Network
User
Subscription Login to verify subscription
Notifications
Font Size

  • Kou, G., & Lin, C. (2014). A cosine maximization method for the priority vector derivation in AHP. European Journal of Operational Research, 235(1), 225-232. https://doi.org/10.1016/j.ejor. 2013.10.019

  • Li, Z., Li, G., Xu, K., Tang, X., & Dong, X. (2021). Temperature-sensitive point selection and thermal error modeling of spindle based on synthetical temperature information. Intl. Journal of Advanced Manufacturing Technology, 113(3-4), 1029-1043. https://doi. org/10.1007/s00170-021-06680-9

  • Lin, C.-J., Su, X.-Y., Hu, C.-H., Jian, B.-L., Wu, L.-W., & Yau, H.-T. (2020). A linear regression thermal displacement lathe spindle model. Energies, 13(4), 949. https://doi.org/10.3390/ en13040949

  • Lin, W., & Fu, J. (2010). Support vector machine and neural network united system for NC machine tool thermal error modeling. Proceedings - 2010 6th International Conference on Natural Computation, ICNC 2010, 8(Icnc), 4305-4309. https://doi.org/10.1109/ICNC.2010.5583620

  • Liu, H., Miao, E. M., Wei, X. Y., & Zhuang, X. D. (2017). Robust modeling method for thermal error of CNC machine tools based on ridge regression algorithm. International Journal of Machine Tools and Manufacture, 113, (November 2016), 35-48. https://doi.org/10.1016/j. ijmachtools.2016.11.001

  • Liu, H., Miao, E., Zhang, L., Li, L., Hou, Y., & Tang, D. (2020). Thermal error modeling for machine tools: Mechanistic analysis and solution for the pseudocorrelation of temperature-sensitive points. IEEE Access, 8, 63497-63513. https://doi. org/10.1109/ACCESS.2020.2983471

  • Zhang, Y., Yang, J., & Jiang, H. (2012). Machine tool thermal error modeling and prediction by grey neural network. International Journal of Advanced Manufacturing Technology, 59(9-12), 1065-1072.https://doi.org/10.1007/s00170-011 -3564-3


Abstract Views: 115

PDF Views: 0




  • Thermal Error Modeling of Machine Tool Spindle Through an Ensemble Approach

Abstract Views: 115  |  PDF Views: 0

Authors

Anirban Tudu
Indian Institute of Technology Madras, Chennai, India
Rupavath Manikanta
Indian Institute of Technology Madras, Chennai, India
D. S. Srinivasu
Indian Institute of Technology Madras, Chennai, India

Abstract


Thermal error compensation of machine tool is cost-effective than other methods. Towards this, data-driven machine learning (ML) algorithms have been used to produce accurate prediction models. However, ML models have limitations, such as overfitting, requiring a large data etc. In present work, a hybrid model is proposed by exploiting the linear regression (LR), support vector machine (SVM), neural network (NN), and decision tree (DT) models. For this purpose, the optimum weights to each constituent model is identified by cosine similarity maximization. The developed models are validated against the experimental data. The prediction results with optimized weight are compared with equal weights and the root means square error (RMSE) for both methods are 1.8879 and 2.8978, respectively. The RMSE shows that the hybrid model produces good accuracy for both small and large data sets compared to individual models.


Keywords


Hybrid Model, Cosine Maximization, Thermal Error, Support Vector Machine, Linear Regression, Neural Network

References