Neural network thermal error compensation of a machining center

Journal of Manufacturing Systems

This paper presents a new modelling methodology for compensation of the thermal errors on a gantrytype 5-axis CNC machine tool. The method uses a "Grey Neural Network Model with Convolution Integral" (GNNMCI(1, N)), which makes full use of the similarities and complementarity between Grey system models and artificial neural networks (ANNs) to overcome the disadvantage of applying either model in isolation. A Particle Swarm Optimisation (PSO) algorithm is also employed to optimise the proposed Grey neural network. The size of the data pairs is crucial when the generation of data is a costly affair, since the machine downtime necessary to acquire the data is often considered prohibitive. Under such circumstances, optimisation of the number of data pairs used for training is of prime concern for calibrating a physical model or training a black-box model. A Grey Accumulated Generating Operation (AGO), which is a basis of the Grey system theory, is used to transform the original data to a monotonic series of data, which has less randomness than the original series of data. The choice of inputs to the thermal model is a non-trivial decision which is ultimately a compromise between the ability to obtain data that sufficiently correlates with the thermal distortion and the cost of implementation of the necessary feedback sensors. In this study, temperature measurement at key locations was supplemented by direct distortion measurement at accessible locations. This form of data fusion simplifies the modelling process, enhances the accuracy of the system and reduces the overall number of inputs to the model, since otherwise a much larger number of thermal sensors would be required to cover the entire structure. The Z-axis heating test, C-axis heating test, and the combined (helical) movement are considered in this work. The compensation values, calculated by the GNNMCI(1, N) model were sent to the controller for live error compensation. Test results show that a 85% reduction in thermal errors was achieved after compensation.

Applied Soft Computing, 2014

Thermal errors can have significant effects on CNC machine tool accuracy. The errors come from thermal deformations of the machine elements caused by heat sources within the machine structure or from ambient temperature change. The effect of temperature can be reduced by error avoidance or numerical compensation. The performance of a thermal error compensation system essentially depends upon the accuracy and robustness of the thermal error model and its input measurements. This paper first reviews different methods of designing thermal error models, before concentrating on employing an adaptive neuro fuzzy inference system (ANFIS) to design two thermal prediction models: ANFIS by dividing the data space into rectangular sub-spaces (ANFIS-Grid model) and ANFIS by using the fuzzy c-means clustering method (ANFIS-FCM model). Grey system theory is used to obtain the influence ranking of all possible temperature sensors on the thermal response of the machine structure. All the influence weightings of the thermal sensors are clustered into groups using the fuzzy c-means (FCM) clustering method, the groups then being further reduced by correlation analysis.

International Journal of Machine Tools and Manufacture, 2003

Thermal error in machine tools has been observed to be closely linked to the temperature of critical elements of the machine. It was found, in the experiments conducted herein, that there was a significant increase in the axis positioning error on account of an increase in the temperature of the machine elements due to continuous operation. However, it was also observed that the specific operating parameters of the test cycles carried out also significantly affected the positioning error. Different sets of operating parameters generated significantly different error values even though the temperature of the machine elements generated by those operating conditions was similar. As such, this observation forms a very important consideration in the development of a generic thermal error compensation system. It appears to indicate that such a system needs to be capable of evaluating the compensation depending upon the temperature values of the machine elements as well as account for the effect that different operating parameters induce upon the positioning error under the same thermal condition of the machine. This paper attempts to analyse the thermal behaviour of a three-axis vertical machining centre under the influence of various operating parameters and, through the experimental results obtained, point out and explain the effect of these parameters on the axis positioning error. This behaviour forms the basis of an improved modelling methodology that is presented separately in Part II of this paper.

MM Science Journal, 2021

Machine tool thermal errors are an important element in machined workpiece inaccuracies. In the past few decades, thermal errors have been successfully reduced by software compensation techniques such as multiple linear regression analysis, finite element method, neural network and transfer function (TF). A phenomenological approach based on TFs is used for thermal error compensation in this research. This approach respects basic heat transfer mechanisms in the MT structure and requires a minimum of additional gauges. Every MT is unique due to manufacturing inaccuracies in components, assembly processes (different preloads in bolts, deviations in the assembly and seating of components, etc.) or working environment (air-conditioned room or ordinary production hall). The aim of this paper is to investigate thermal error compensation model transfer between machines of the same product line to improve thermal error model prediction. Several milling centers of the same product line were ...

Thermal error modeling using the temperature data obtained from different thermal key points is one of the most basic requirements for thermal error compensation in CNC machining centers. Effective thermal error compensation relies on the accurate prediction of the time-variant thermal error during machining. The increase in number of thermal key points not only increases prediction noise in the thermal error model but is also very cost effective. Hence, optimization of thermal key points is essential for accurate prediction of thermal behaviour of machine tool. In this study, the thermal key points of horizontal feed drives of a vertical CNC machine are optimized based on sensitivity analysis. A non-linear statistical regression thermal error prediction model is developed by using the temperature data obtained from the optimized key points. Thermal errors predicted before and after sensitivity analysis of thermal key points using statistical model are presented. The developed model...

In recent years, neural network methods with different architectures and training strategies are widely used in machine tool thermal error compensation field, but there are still many problems such as low model accuracy, long training time and bad generalized ability. An integrated neural network classifier is proposed for compensation of thermal error in the paper. The investigation shows that the proposed method has higher classification precision and reliability, and is an ideal pattern classifier. Real cutting experiments are conducted on a CNC turning machine to validate the effectiveness of the method. Both simulation and experiment indicate that the proposed method is quite effective and ubiquitous.

Volume 3: Joint MSEC-NAMRC Symposia, 2016

Additive manufacturing (AM) processes involve the fabrication of parts in a layer-wise manner. The layers of material are deposited using a variety of established methodologies, the most popular of which involve either the use of a powerful laser to sinter/melt successive layers of metal/alloy/polymer powders or, the deposition of layers of polymers through a heated extrusion head at a controlled rate. The thermal nature of these processes causes the development of temperature gradients throughout the part and as a result, the part undergoes irregular deformations which ultimately leads to dimensional inaccuracies in the manufactured part. An Artificial Neural Network (ANN) based methodology is presented in this paper to directly compensate the part geometric design which will help to counter the thermal deformations in the manufactured part. A feed-forward ANN model is trained using backpropagation algorithm to study part deformations resulting from the AM process. The trained netw...

International Journal of Mechanical Engineering and Mechatronics, 2012

The thermal behaviour of machine tools can profoundly affect the dimensional accuracy of manufactured parts. Consequently, reducing its influence has become increasingly important to modern manufacturing. One of the most convenient and effective ways to reduce thermal error and enhance working accuracy is volumetric error compensation, especially for CNC machine tools. However, currently available compensation methods are highly complicated and difficult to apply in industry. They require a high-precision laser interferometer to measure 21 geometric error components and numerous thermocouples to monitor temperature variations. These requirements make such techniques costly and time consuming. This paper describes the development of a simplified and economical method of compensating for thermally induced errors. Compensation is implemented on the basis of only three axial positioning errors, which are assumed to be functions of ball screw nut temperature and travel distance. Instead of an expensive laser interferometer and many thermocouples, a simple laser Doppler displacement meter and only three thermocouples are required for the proposed method. Results show good agreement between measured and predicted thermally induced errors.

Expert Systems with Applications, 2011

In this paper, the regression analysis (RA) and artificial neural network (ANN) are presented for the prediction of tool-chip interface temperature depends on cutting parameters in machining. The RA and ANN model for prediction tool-chip interface temperature are developed and mathematical equations derived for tool-chip interface temperature prediction are obtained. The tool-chip interface temperature results obtained from mathematical equations with RA and ANN model and the experimental results available in the literature obtained by using AISI 1117 steel work piece with embedded K type thermocouple into the uncoated cutting tool (Korkut, Boy, Karacan, & S ßeker, 2007) are compared. The coefficient of determination (R 2) both training and testing data for temperature prediction in the ANN model are determined as 0.999791289 and 0.997889303 whereas; R 2 for both training and testing data in the RA model are computed as 0.999063 and 0.999427, respectively. The correlation obtained by the training ANN model are better than the one obtained by training RA model. The training ANN model with the Levenberg-Marquardt (LM) algorithm provides more accurate prediction and is quite useful in the calculation of tool-chip interface temperature when compared with the trained RA method in machining. On the other hand, prediction values obtained the testing RA model is slightly better performance than the testing ANN model. The results show that the tool-chip interface temperature equation derived from RA and ANN model can be used for prediction.

Precision Engineering, 2020

Achieving high workpiece accuracy is the long-term goal of machine tool designers. There are many causes for workpiece inaccuracy, with thermal errors being the most common. Indirect compensation (using prediction models for thermal errors) is a promising strategy to reduce thermal errors without increasing machine tool costs. The modelling approach uses transfer functions to deal with this issue; it is an established dynamic method with a physical basis, and its modelling and calculation speed are suitable for real-time applications. This research presents compensation for the main internal and external heat sources affecting the 5-axis machine tool structure including spindle rotation, three linear axes movements, rotary C axis and time-varying environmental temperature influence, save for the cutting process. A mathematical model using transfer functions is implemented directly into the control system of a milling centre to compensate for thermal errors in real time using Python programming language. The inputs of the compensation algorithm are indigenous temperature sensors used primarily for diagnostic purposes in the machine. Therefore, no additional temperature sensors are necessary. This achieved a significant reduction in thermal errors in three machine directions X, Y and Z during verification testing lasting over 60 h. Moreover, a thermal test piece was machined to verify the industrial applicability of the introduced approach. The results of the transfer function model compared with the machine tool's multiple linear regression compensation model are discussed.