Railway trains represent a vital component of transportation infrastructure in numerous countries. Railway travel offers individuals convenient and secure transportation options. The paramount concern for the general public is the safety of railway trains, particularly the dependability and security of the door control system. The motor encoder is essential, and in the event of a malfunction, it has the potential to result in abnormal door operation, which could lead to significant safety risks. To prevent potential malfunctions, the door control system must be designed to detect any anomalous noise and promptly halt operation. In the event of encoder detachment or poor contact with the encoder wire, which is a common occurrence, the system must be capable of detecting such anomalies with immediate effect. In this study, a multiscale wavelet transform is proposed to transform the velocity data of the motor encoder. The resulting curve identifies and accentuates the distinctive attributes of anomalous signals, thereby facilitating the expeditious detection of encoder malfunctions in the door control system. The wavelet transform converts the standard encoder velocity signals for learning and analysis. Following this, any noise caused by encoder detachment or poor contact with the encoder wire is identified. Subsequently, the controller utilizes the distinctive characteristics of anomalous encoder signals to guarantee the prompt cessation of door operation in case of a motor encoder failure. Verification has demonstrated that this method not only rapidly converts and highlights abnormal encoder signals but also effectively ensures the safe operation of the railway train door system.