Improving the Accuracy of Multi-Axis Machines Through On-Line Error Compensation Using Neural Networks

Author(s):  
Abderrazak El Ouafi ◽  
Michel Guillot ◽  
Abdellah Bedrouni

Abstract This research is devoted to one of the most fundamental problems in precision engineering: multi-axis machines accuracy. The paper presents a new approach designed to support the implementation of software error compensation of geometric, thermal and dynamic errors for enhancing the accuracy of multi-axis machines. The accuracy of multi-axis machines can be significantly improved using an intelligent integration of sensor information to perform the compensation function. The compensation process consists of the following major steps carried out on-line: continuous monitoring of the machine conditions using position, force, speed and temperature sensors mounted on the machine structure. Error forecasting through sensor fusion. Volumetric error synthesis and software compensation. To improve the effectiveness of error modeling, an artificial neural network is extensively applied. Implemented on a turning center, the compensation approach has enabled improvement of the machine accuracy by reducing the maximum dimensional error from 70 μm initially to less than 4 μm.

2007 ◽  
Vol 359-360 ◽  
pp. 219-223
Author(s):  
Li Ming Xu ◽  
Lun Shi ◽  
Xiao Ming Zhao ◽  
De Jin Hu

Spindle thermal deformation is the main error source of many precision profile grinders. In this paper, the relationship between spindle temperature and either radial or axial thermal deformation is studied based on experiments. The placement and amount of temperature sensors are optimized. Then a kind of thermal error modeling method based on support vector machine is presented and applied in the modeling of thermal error of profile grinding. The result shows the model is robust and the on-line accurate prediction of grinding thermal error is realized based on monitoring of temperature rise of spindle. Finally, the error compensation strategy is discussed for further application of thermal error modeling.


2011 ◽  
Vol 189-193 ◽  
pp. 4145-4148
Author(s):  
Qian Jian Guo ◽  
Lei He ◽  
Guang Ming Zhu

Thermal errors are the major contributor to the dimensional errors of a workpiece in precision machining. Error compensation technique is a cost-effective way to reduce thermal errors. Accurate modeling of errors is a prerequisite of error compensation. In this paper, a thermal error model was proposed by using projection pursuit regression (PPR). The PPR method improves the prediction accuracy of thermal deformation in the CNC turning center.


2013 ◽  
Vol 718-720 ◽  
pp. 1388-1393
Author(s):  
Abderrazak El Ouafi ◽  
Noureddine Barka

In order to improve multi-axis machine accuracy, error compensation techniques have been widely applied. However, the lack of reliable methods for direct, global and comprehensive estimation implies that all compensation techniques are based on off-line sequential error components measurement. These measurements provide static results, and cannot reflect the actual machine conditions. Thus, these results are not representative of the real working conditions because of disturbances from thermal distortions and dynamic perturbations. This paper presents an on-line error identification and compensation approach for CNC multi-axis machine tools. Based on the simultaneous measurement of error components, the proposed identification scheme is built to ensure volumetric error prediction for an adaptive error compensation system. Implemented on a moving bridge type CMM, the approach led to a significant improvement of the three-dimensional measurement accuracy.Compared to the conventional off-line error compensation techniques, the proposed identification and compensation approach can further improve the compensation adaptability and efficiency.


Author(s):  
Henry Krumb ◽  
Dhritimaan Das ◽  
Romol Chadda ◽  
Anirban Mukhopadhyay

Abstract Purpose Electromagnetic tracking (EMT) can partially replace X-ray guidance in minimally invasive procedures, reducing radiation in the OR. However, in this hybrid setting, EMT is disturbed by metallic distortion caused by the X-ray device. We plan to make hybrid navigation clinical reality to reduce radiation exposure for patients and surgeons, by compensating EMT error. Methods Our online compensation strategy exploits cycle-consistent generative adversarial neural networks (CycleGAN). Positions are translated from various bedside environments to their bench equivalents, by adjusting their z-component. Domain-translated points are fine-tuned on the x–y plane to reduce error in the bench domain. We evaluate our compensation approach in a phantom experiment. Results Since the domain-translation approach maps distorted points to their laboratory equivalents, predictions are consistent among different C-arm environments. Error is successfully reduced in all evaluation environments. Our qualitative phantom experiment demonstrates that our approach generalizes well to an unseen C-arm environment. Conclusion Adversarial, cycle-consistent training is an explicable, consistent and thus interpretable approach for online error compensation. Qualitative assessment of EMT error compensation gives a glimpse to the potential of our method for rotational error compensation.


Author(s):  
Siyu Zhang ◽  
R. Ganesan ◽  
T. S. Sankar

Abstract The problem of estimating an unknown multivariate function from on-line vibration measurements, for determining the conditions of a machine system and for estimating its service life is considered. This problem is formulated into a multiple-index based trend analysis problem and the corresponding indices for trend analysis are extracted from the on-line vibration data. Selection of these indices is based on the simultaneous consideration of commonly-observed faults or malfunctions in the machine system being monitored. A neural network algorithm that has been developed by the present authors for multiple-index based regression is adapted to perform the trend analysis of a machine system. Applications of this neural network algorithm to the condition monitoring and life estimation of both a bearing system as well as a gearbox are fully demonstrated. The efficiency and computational supremacy of the new algorithm are established through comparing with the performance of Self-Organizing Mapping (SOM) and Constrained Topological Mapping (CTM) algorithms. Further, the usefulness of multiple-index based trend analysis in precisely predicting the condition and service life of a machine system is clearly demonstrated. Using on-line vibration signal to constitute the set of variables for trend analysis, and employing the newly-developed self-organizing neural algorithm for performing the trend analysis, a new approach is developed for machinery monitoring and diagnostics.


Author(s):  
Yi Zhang ◽  
Jianguo Yang ◽  
Sitong Xiang ◽  
Huixiao Xiao

This article intends to provide an error compensation system for five-axis machine tools. A volumetric error model is established with homogeneous transformation matrix method, from which compensation values of both orientation and position errors can be obtained. Thirty-seven errors on a five-axis machine tool are classified into three categories – functional, random, and negligible errors, among which the effect of the first one on volumetric accuracy is considered as great enough to be included in this model. Some typical modeling methods are discussed on positioning and straightness errors, considering both geometric and thermal effects. Then, we propose a compensation implementation technique based on the function of external machine zero point shift and Ethernet data communication protocol for machine tools. Finally, laser diagonal measurements have been conducted to validate the effectiveness of the proposed volumetric error compensation system.


2020 ◽  
Vol 239 ◽  
pp. 11001
Author(s):  
M. Herman ◽  
D.A. Brown ◽  
M.B. Chadwick ◽  
W. Haeck ◽  
T. Kawano ◽  
...  

A new paradigm for nuclear reaction data evaluations is proposed to produce adjusted libraries that take into account integral experiments on the same footing as the differential ones. These evaluations will provide comprehensive covariance matrices including cross-correlations among different materials/reactions that are critical for realistic propagation of data uncertainties to integral quantities. The new approach should also reduce error compensation issues and facilitate updating of the library to account for new or corrected experiments and advances in reaction modeling.


Author(s):  
Xiong Zhao ◽  
Lianyu Zheng ◽  
Yuehong Zhang

Abstract Mirror error compensation is usually employed to improve the machining precision of thin-walled parts. However, this zero-order method may result in inadequate error compensation, due to the time-varying cutting condition of thin-walled parts. To cope with this problem, an on-line first-order error compensation method is proposed for thin-walled parts. With this context, firstly, the time-varying cutting condition of thin-walled parts is defined with its in-process geometric and physical characteristics. Based on it, a first-order machining error compensation model is constructed. Then, during the process planning, the theory geometric and physical characteristic of thin-walled parts are respectively obtained with CAM software and structure dynamic modification method. After process performing, the real geometric characteristic of thin-walled parts is measured, and it is used to calculate the dimension error of thin-walled parts. Next, the error compensated value is evaluated based on the compensation model, from which, an error compensation plane is constructed to modify the tool center points for next process step. Finally, the machining error is compensated by performing the next process step. A milling test of thin-walled part is employed to verify the proposed method, and the experiment results shown that the proposed method can significantly improve the error compensation effect for low-stiffness structure, and thickness precision of thin-walled parts is improved by 71.4 % compared with the mirror error compensation method after machining.


Sign in / Sign up

Export Citation Format

Share Document