The thrust control malfunction (TCM) in aeroengines poses significant threats to aircraft safety, and traditional detection methods rely on precise mathematical models that are limited by model errors and complex operating conditions. This paper proposes a digital twin-based approach for automatic detection and fault-tolerant control of TCM in fuel systems. By constructing a fuel control system model for a certain type of engine, a digital twin algorithm is employed to monitor the fuel control simulation system in real time, automatically detecting TCM. A neural network fitting method is used to improve the operation speed of the monitoring model. Additionally, a worry-free switching control method based on initial integral value reset optimizes the fault-tolerant control of TCM. Experiments validate the effectiveness of the proposed method through four fault modes including fuel flow increasing fault, fuel flow instability, fuel flow unadjustability, and fuel flow random delay, thereby significantly improving aeroengine operational safety.