Abstract: In order to predict the deformation of thin-walled parts during side milling, a new method based on SAPSO-BP neural network is proposed to predict the deformation of thin-walled parts in side milling. Through the establishment of T-shaped thin-walled workpiece side milling simulation model, and experimental verification of its feasibility, provide training samples for the follow-up neural network; then introduce simulated annealing particle swarm optimization(SAPSO) to optimize the initial weights and thresholds of BP neural network, establish the deformation prediction model of thin-walled workpiece side milling based on SAPSO-BP, and verify and analyze its reliability.