With the increasing demand for urban rail transit among Chinese residents, in order to solve the problem of traditional vehicle braking control models being unable to brake normally under abnormal conditions, intelligent methods are studied to adjust the parameters of the braking model under abnormal conditions and construct an intelligent braking control model. The performance verification of the improved brake control model was conducted, and the experimental results showed that under abnormal signal collection conditions, the error of the brake cylinder pressure change curve of the improved brake control model was 8kPa, and the error of the pressure relief curve was 5kPa. Its brake performance and pressure relief performance were better than traditional brake models. In summary, the improved brake control model can meet the pressure control needs of traditional models in different situations, and has practical value