Self-optimizing thermal error compensation models with adaptive inputs using Group-LASSO for ARX-models

Zimmermann, Nico; Büchi, Tobias; Mayr, Josef; Wegener, Konrad

doi:10.1016/j.jmsy.2022.04.015

Cited by 29 publications

(23 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thanks to such detection the machine tool productivity loss was reduced by 78% with no significant decrease in self-compensation system precision. It appears from the latest research on the autonomous compensation system that a system based on the adaptive selection of input variables and the updating of the ARMA model by means of the Group-LASSO method combined with the PSO algorithm is in a long time horizon (a few months) more precise than the previously investigated system by at least 20% [60]. Fig.…”

Section: Case Studymentioning

confidence: 95%

“…K-means clustering combined with the timeseries cluster kernel (TCK) increased model resistance, whereby machine tool productivity loss was reduced by 45% [45]. In comparison to TCK, Group-Lasso combined with particle swarm optimization (PSO) reduced the mean error from ±14μm to ±11μm (a reduction of about 21%) along the linear axes and from ±32 μm/m to ±22 μm/m (a reduction of about 31%) along the rotational axes [60]. Thus it was demonstrated that the Group-Lasso method combined with PSO is significantly better than k-means clustering combined with TCK.…”

Section: Data Processingmentioning

confidence: 99%

“…Several attempts at realizing long-term thermal error compensation have been made [27,45,47,48,60,70]. One of the ways to realize long-term compensation is to train and test the model on data from different seasons of the year [27].…”

Section: Case Studymentioning

confidence: 99%

“…There is always a risk that the model trained in this way will become inaccurate in the future due to a change in the conditions. In order to prevent this, self-learning has been employed [45,47,48,60,70]. The R-Test procedure was used to measure the thermal error by means of a measuring probe.…”

Section: Case Studymentioning

confidence: 99%

See 3 more Smart Citations

Application of Machine Learning in the Precise and Cost-Effective Self-Compensation of the Thermal Errors of CNC Machine Tools – A Review

Czwartosz¹,

Jędrzejewski²

2022

Journal of Machine Engineering

View full text Add to dashboard Cite

The current development of production engineering takes place through the innovative improvement of machine tools and machining processes at the constantly growing application of intelligent self-improvement functions. Machine learning opens up possibilities for machine tool self-improvement in real time. This paper discusses the state of knowledge relating to the application of machine learning for precise and cost-effective thermal error self-compensation. Data acquisition and processing, models and model learning and self-learning methods are also considered. Three highly effective error compensation systems (supported with machine learning) are analysed and conclusions and recommendations for future research are formulated.

show abstract

Section: Case Studymentioning

confidence: 95%

Section: Data Processingmentioning

confidence: 99%

Section: Case Studymentioning

confidence: 99%

Section: Case Studymentioning

confidence: 99%

See 2 more Smart Citations

Application of Machine Learning in the Precise and Cost-Effective Self-Compensation of the Thermal Errors of CNC Machine Tools – A Review

Czwartosz¹,

Jędrzejewski²

2022

Journal of Machine Engineering

View full text Add to dashboard Cite

show abstract

“…Studies show that the thermal deformation may bring about thermal errors of the spindle along the axial and radial directions, thereby affecting the accuracy of the machine tool. [1][2][3] In order to resolve this problem, numerous methods, including replacing the steel material of some parts in the spindle with low-thermal expansion coefficient materials (e.g., ceramic, carbon-fiber-reinforced plastics), [4][5][6] and thermal error compensation 7,8 have been proposed. Heat removal strategy is also an important and effective 1 way to reduce the temperature rise in the rotating parts and control the thermal error.…”

Section: Introductionmentioning

confidence: 99%

A heat dissipation enhancing method for the high-speed spindle based on heat conductive paths

Bai

et al. 2023

Advances in Mechanical Engineering

View full text Add to dashboard Cite

High-speed spindle systems could generate sufficient heat when they are operating and would cause thermal deformation that influences spindle accuracy. Heat dissipation is a common and effective way to remove the generated heat. In this paper, heat conductive paths were created on the bearing seat and the spindle housing along the radial direction. Along these paths, the heat was transferred directly from heat sources inside the spindle system to outside and the heat dissipation was enhanced. To limit the heat transferred along the axial direction, the inner wall of the path was coated with a thermal insulation material. Based on an annular plate model, the influences of the affecting parameters of paths on the heat transfer performance were studied. The temperature distributions of the models with and without heat conductive paths were numerically and experimentally investigated. It was found that after using heat conductive paths, the heat dissipation enhanced significantly. Moreover, it is found that as the number and diameter of paths increases, the temperature of internal parts decreases and the gap between the highest and the lowest temperature is narrowed.

show abstract

Thermal Error Compensation Models Utilizing the Power Consumption of Machine Tools

Lang

Zimmermann

Mayr

et al. 2023

Lecture Notes in Production Engineering

View full text Add to dashboard Cite

Thermal errors are among the most significant contributors to deviations of products manufactured on modern machine tools (MTs). Reducing them is typically achieved through either design adaptation, active cooling of the MT and its environment, or compensation using measurements or model-based predictions. Model-based compensation strategies promise to have the lowest environmental footprint by far. In general, a compensation model needs to be accurate, robust to changing boundary conditions and must require only minimal experimental efforts as this reduces the productivity of the MT. Model inputs such as temperature measurements or the power consumption of various components, can be used to predict the thermal errors. The temperature inputs require additional sensors, effort and cost for the MT manufacturer to install and ensure up-time while the power consumption could be logged and are typically provided from the control system anyway. Adaptive compensation models are created using four different sets of inputs consisting of 13 temperature sensors and 7 power measurements. While the best results were obtained with all 20 inputs, the 7 energy recordings give similar results as the 13 temperature sensors if the environmental temperature is considered. The volumetric RMSE was reduced by 72% and the maximal error from 32.75 µm to 9.5 µm. ARX models proved to be suitable and even outperform more complex model structures such as LSTM and especially those without time dependency such as feed forward neural networks.

show abstract

Self-optimizing thermal error compensation models with adaptive inputs using Group-LASSO for ARX-models

Cited by 29 publications

References 39 publications

Application of Machine Learning in the Precise and Cost-Effective Self-Compensation of the Thermal Errors of CNC Machine Tools – A Review

Application of Machine Learning in the Precise and Cost-Effective Self-Compensation of the Thermal Errors of CNC Machine Tools – A Review

A heat dissipation enhancing method for the high-speed spindle based on heat conductive paths

Thermal Error Compensation Models Utilizing the Power Consumption of Machine Tools

Contact Info

Product

Resources

About