Modeling Method for Thermal Drift of CNC Machine Tool Spindle Based on fa‑lssvm

Abstract

The invention discloses a FA‑LSSVM-based thermal drift modeling method of a CNC machine tool spindle, which comprises the following steps: S1, collecting sample data and performing normalization processing on it; S2, adopting a fuzzy mean value clustering grouping method and based The linear regression search algorithm selects the input variables for thermal drift modeling from the sample data; S3, using the fireworks algorithm to obtain the optimal FA‑LSSVM combined model parameters and the optimal thermal drift prediction model; S4, evaluating the correctness of the method . The present invention optimizes the two core parameters of the LSSVM model based on the fireworks algorithm, and the search range of the feasible solution space is large, which is conducive to finding a more reasonable parameter combination. At the same time, the algorithm has a fast optimization speed, and the optimized FA can be obtained without too many iterations ‑LSSVM combination model accurately predicts the thermal error of the CNC machine tool spindle, which can more effectively reduce the thermal error of the CNC machine tool and significantly improve the machining accuracy of the CNC machine tool.

Description

technical field [0001] The invention belongs to the technical field of mechanical manufacturing, and in particular relates to the design of a FA-LSSVM-based thermal drift modeling method for a CNC machine tool spindle. Background technique [0002] CNC milling and boring machine is a very important machine tool in the processing and manufacturing industry. As the requirements for the machining accuracy of parts are getting higher and higher, the maintenance and improvement of the machining accuracy of CNC machine tools has also been paid more and more attention. Studies have shown that thermal error accounts for about 40% to 70% of the total error of the machine tool, and has become one of the most important factors affecting the machining accuracy of the machine tool. With the development of machine tools towards high spindle speed and high cutting feed rate, the impact of thermal errors on machine tool machining accuracy is more significant. Therefore, in order to improve...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that the kernel function parameters and penalty parameters of the LSSVM model in the prior art have a significant impact on the prediction accuracy, but the prediction accuracy and robustness of the current optimization method are still not ideal, and a method based on FA-LSSVM Modeling Method for Thermal Drift of CNC Machine Tool Spindle

Images