Thermal error modeling of motorized spindle based on Elman neural network optimized by sparrow search algorithm

The thermal error of the motorized spindle is an essential factor affecting machining accuracy high-speed numerically controlled machines. establishment a model for compensation can effectively improve impact errors on machine tool. This paper proposes sparrow search algorithm to optimize Elman neural network predict in spindles. First simulation analysis characteristics A02 spindle. Based results, position temperature measuring points arranged and experiment displacement data at different rotational speeds were collected; secondly, method combining pedigree clustering k-means used perform cluster each measurement point, grey correlation degree determine between error. Three temperature-sensitive screened from ten points, which reduced collinearity number independent variables model. Finally, weights thresholds are optimized by algorithm, prediction based SSA-Elman established compared with Particle Swarm Optimized Neural Network results show that has highest exhibits good stability generalization ability.