The surge in electric vehicle usage has expanded the number of charging stations, intensifying demands on their operation and maintenance. Public charging stations, often exposed to harsh weather and unpredictable human factors, frequently encounter malfunctions requiring prompt attention. Current methods primarily employ data-driven approaches or rely on empirical expertise to establish warning thresholds for fault prediction. While these approaches are generally effective, the artificially fixed thresholds they employ for fault prediction limit adaptability and fall short in sensitivity to special scenarios, timings, locations, and types of faults, as well as in overall intelligence. This paper presents a novel fault prediction model for charging equipment that utilizes adaptive dynamic thresholds to enhance diagnostic accuracy and reliability. By integrating and quantifying Environmental Influence Factors (EF), Scenario Influence Factors (SF), Fault Severity Factors (FF), and Charging Equipment Status Factors (CF) into a cohesive predictive framework, our model dynamically adjusts thresholds based on a comprehensive analysis of these factors. Using a dataset of 560,000 charging records from Hangzhou, the model employs a batch offline reinforcement learning approach based on a Markov Decision Process (MDP). Threshold adjustments are optimized via a Deep Q-learning Network (DQN) to maximize long-term rewards. The proposed system is evaluated through metrics such as advance warning time, alert precision, and recall rates. Results demonstrate the model’s ability to provide timely, accurate fault detection and enhance alert effectiveness, thereby improving the reliability and efficiency of electric vehicle charging networks.