Thermal error prediction of numerical control machine based on improved particle swarm optimized back propagation neural network

Thermal error is a primary factor affecting the working accuracy of numerical control machine, whereas it can´t be measured on-line. So an accurate predictive error is critical to final product quality. In order to improve the predictive accuracy of thermal error, a predictive method based on improved particle swarm optimized back propagation neural network is proposed in this paper. Because of back propagation neural network has the disadvantages of low convergence rate, easy to fall into local optimization and so on. An improved updating strategy of particle position based on the stand particle swarm method is built. Then with the improved particle swarm algorithm, the threshold and weight of the neural network is optimized to form the thermal error predictive method. The simulation results conducted on MATLAB shows that the proposed thermal error predictive method has a higher predictive accuracy, better generalization ability compared with stand Back Propagation(BP) Neural Network and Support Vector Machine.