Approximation of machine tool experimental thermal characteristics by neural network

Abstract

The article evaluates the effectiveness of an artificial neural network use for mathematical processing of the machine tool experimental thermal characteristics. To improve the quality of approximation and the accuracy of forecasting, two types of neural networks were used, namely a network of radially basis functions and a generalized regression neural network. The results of a full-scale thermal experiment of the idling 400V machine tool are presented. Computational experiments of the thermal testing results of a metal cutting machine were carried out to build and test the models under study. The results showed that neural networks perform better than the classical power polynomial model in terms of accuracy and approximation quality. Thus, neural networks can be used to approximate the experimental thermal characteristics of the machine tool in real time.

Spindle Thermal Error Prediction Based on LSTM Deep Learning for a CNC Machine Tool

“…Multiple linear regression (MLR) is a common algorithm for creating mathematical prediction models of thermal displacement [17][18][19][20][21][22][23]. Some neural network modeling techniques have also been proposed to obtain more robust and accurate predictions [24][25][26][27][28][29][30][31][32]. In addition, how to select and decide the representative temperature-sensitive points among numerous Initial Temperature Points (ITPs) is crucial.…”