In order to study the variation of brake squeal incidence with braking conditions under steady magnetic field,a prediction model of friction brake squeal incidence based on long-short term memory(LSTM) network was proposed.The measurement scheme of brake squeal signal was designed,and a large number of simulated brake tests were carried out.Based on the deep learning network of LSTM gated loop unit,an intelligent prediction model for the incidence of brake squeal under different braking conditions was constructed.The accuracy of the prediction model was verified by sample data and experiments.The results show that the prediction model can accurately predict the incidence of brake squeal.The steady magnetic field can effectively reduce the incidence of brake squeal,however,when the magnetic induction intensity increases to a certain threshold,the influence degree of magnetic field on friction behavior tends to saturation,and the incidence of brake squeal gradually tends to be stable.The results provide a theoretical basis and data model for the development of active brake squeal suppression technology based on magnetic field.