Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm
Thermal error of a CNC machine tool has a serious effect on its machining accuracy. In order to reduce the thermal error, establishing an accurate and robust thermal error model is necessary. A new data-driven thermal error model is proposed based on gray wolf optimizer (GWO) and least squares suppo...
| Glavni autori: | , , , |
|---|---|
| Format: | Članak |
| Jezik: | English |
| Izdano: |
Elsevier
2024-01-01
|
| Serija: | Case Studies in Thermal Engineering |
| Teme: | |
| Online pristup: | http://www.sciencedirect.com/science/article/pii/S2214157X23011644 |
| _version_ | 1827383101198696448 |
|---|---|
| author | Yang Li Yue Yang Jiaqi Wang Fusheng Liang |
| author_facet | Yang Li Yue Yang Jiaqi Wang Fusheng Liang |
| author_sort | Yang Li |
| collection | DOAJ |
| description | Thermal error of a CNC machine tool has a serious effect on its machining accuracy. In order to reduce the thermal error, establishing an accurate and robust thermal error model is necessary. A new data-driven thermal error model is proposed based on gray wolf optimizer (GWO) and least squares support vector machine (LSSVM). While, three temperature-sensitivity points (TSPs) were picked by grouping search method. With optimizing the hyperparameters γ and σ2 by GWO algorithm, LSSVM is applied to thermal error modeling, which has advantages in dealing with small samples and nonlinear data. The experiments were conducted, and the results showed that the error prediction model constructed by GWO-LSSVM achieved an accuracy of 94.39 % meeting the requirements for error compensation. Then, the stability of the proposed error model is verified. Finally, the LSSVM model is compared with MLR and BP models commonly used in error modeling, and the advantages and applicable occasions are analyzed through experiments. |
| first_indexed | 2024-03-08T14:36:17Z |
| format | Article |
| id | doaj.art-7cccf21e21a746bd8b600436f117d021 |
| institution | Directory Open Access Journal |
| issn | 2214-157X |
| language | English |
| last_indexed | 2024-03-08T14:36:17Z |
| publishDate | 2024-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Case Studies in Thermal Engineering |
| spelling | doaj.art-7cccf21e21a746bd8b600436f117d0212024-01-12T04:56:42ZengElsevierCase Studies in Thermal Engineering2214-157X2024-01-0153103858Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithmYang Li0Yue Yang1Jiaqi Wang2Fusheng Liang3School of Mechanic Engineering, Northeast Electric Power University, Jilin, Jilin Province, 132012, China; Corresponding author.School of Mechanic Engineering, Northeast Electric Power University, Jilin, Jilin Province, 132012, ChinaSchool of Mechanic Engineering, Northeast Electric Power University, Jilin, Jilin Province, 132012, ChinaJiangsu Provincial Key Laboratory of Advanced Robotics and Collaborative Innovation Center of Suzhou Nano Science and Technology, College of Mechanical and Electrical Engineering, Soochow University, Suzhou, Jiangsu Province, 215021, ChinaThermal error of a CNC machine tool has a serious effect on its machining accuracy. In order to reduce the thermal error, establishing an accurate and robust thermal error model is necessary. A new data-driven thermal error model is proposed based on gray wolf optimizer (GWO) and least squares support vector machine (LSSVM). While, three temperature-sensitivity points (TSPs) were picked by grouping search method. With optimizing the hyperparameters γ and σ2 by GWO algorithm, LSSVM is applied to thermal error modeling, which has advantages in dealing with small samples and nonlinear data. The experiments were conducted, and the results showed that the error prediction model constructed by GWO-LSSVM achieved an accuracy of 94.39 % meeting the requirements for error compensation. Then, the stability of the proposed error model is verified. Finally, the LSSVM model is compared with MLR and BP models commonly used in error modeling, and the advantages and applicable occasions are analyzed through experiments.http://www.sciencedirect.com/science/article/pii/S2214157X23011644Thermal error modelingTemperature-sensitivity pointsGray wolf optimizerLeast squares support vector machineCNC machine tools |
| spellingShingle | Yang Li Yue Yang Jiaqi Wang Fusheng Liang Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm Case Studies in Thermal Engineering Thermal error modeling Temperature-sensitivity points Gray wolf optimizer Least squares support vector machine CNC machine tools |
| title | Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm |
| title_full | Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm |
| title_fullStr | Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm |
| title_full_unstemmed | Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm |
| title_short | Thermal error modeling of servo axis based on optimized LSSVM with gray wolf optimizer algorithm |
| title_sort | thermal error modeling of servo axis based on optimized lssvm with gray wolf optimizer algorithm |
| topic | Thermal error modeling Temperature-sensitivity points Gray wolf optimizer Least squares support vector machine CNC machine tools |
| url | http://www.sciencedirect.com/science/article/pii/S2214157X23011644 |
| work_keys_str_mv | AT yangli thermalerrormodelingofservoaxisbasedonoptimizedlssvmwithgraywolfoptimizeralgorithm AT yueyang thermalerrormodelingofservoaxisbasedonoptimizedlssvmwithgraywolfoptimizeralgorithm AT jiaqiwang thermalerrormodelingofservoaxisbasedonoptimizedlssvmwithgraywolfoptimizeralgorithm AT fushengliang thermalerrormodelingofservoaxisbasedonoptimizedlssvmwithgraywolfoptimizeralgorithm |