Vibration failure is one of the key factors affecting the safe operation of rail vehicles. Due to factors such as wear and tear of wheels and tracks, aging of equipment, and defects in vehicle manufacturing, it is inevitable that vehicle vibration problems will occur. The frequency of vibration failures is even higher, posing a threat to the safe operation of rail vehicles. By detecting vibration faults in real-time, potential safety hazards can be identified and resolved in a timely manner, ensuring the safe operation of trains. Therefore, design an intelligent detection system for vibration faults in urban rail transit vehicles. Scientifically select and configure the vibration sensor and main control hardware, and design the connection circuit diagram between the two. Develop a program for collecting vibration signals of railway vehicles, effectively collect vibration signals, use low-pass filters to remove noise signals from vibration signals, extract time-domain and frequency-domain features of vibration signals, introduce feedforward neural networks to construct a railway vehicle vibration fault detection model, extract important features of vibration signals from complex vibration signals, input the vibration signals of the tested railway vehicles into the trained model, and the output result is the railway vehicle vibration fault detection result. The experimental results show that the vibration signal features extracted by the application design system are consistent with the actual features, and the vibration fault detection results are consistent with the actual results, indicating that the design system has good detection capability.