A Data-Driven Fault-Tolerant Control Approach for Ship Power Condenser System

With the advancement of intelligence and automation in ship power systems, the integration and complexity of its power system are becoming increasingly high, which poses higher demands on safety and reliability. Effective fault diagnosis and fault-tolerant control technologies act as vital means of preventing faults in the ship power system and ensuring the safety during navigation, and they are also essential components for achieving intelligent ships. Existing fault-tolerant control methods for power systems require precise models, which are difficult to implement. This study investigates the fault diagnosis and fault-tolerant control issues of condensers within ship power systems and proposes a data-driven real-time optimization fault-tolerant control method. The method employs Youla parameterization theory and updates parameters in real-time using gradient descent to mitigate the impact of faults on system performance. System parameters at the condenser working point are obtained through an adaptive observer, and a real-time optimized fault-tolerant controller is designed for actuator faults in the condenser. Simulation results confirm the effectiveness of the proposed method, significantly enhancing system reliability, which is of great importance for improving the safety of ship power.