Prediction of vibration in milling of thin-walled aluminum alloy parts using neural network model

Machining chatter is likely to occur during milling of thin-walled parts. The structural differences in thin-walled parts and the magnitude of the milling force can lead to varying degrees of chatter in different areas of the machining process. Predicting machining stability using dynamic modeling methods can be time-consuming. In this study, a method for establishing a particle swarm optimization-back propagation (PSO-BP) neural network model is proposed to predict the modal parameters of thin-walled parts and the surface vibration of machined parts. Based on measurements of the length, height, wall thickness, and position of the thin-walled parts, the modal parameters of the workpiece were predicted using the PSO-BP neural network model. Additionally, the average milling force was included as an input parameter to predict the displacement of surface vibrations on thin-walled parts using the PSO-BP model. The predictive results of the modal parameters and surface vibration displacement are evaluated using the evaluation function, which indicates that the PSO-BP neural network model can reliably predict the modal parameters and surface vibration depth of thin-walled parts.