<p>This paper investigates the adaptive fault-tolerant stability control problem for uncertain nonlinear systems subject to novel actuator faults and external disturbances. Such faults arise from unknown changes in input power within the actuator module, causing the system’s input power to deviate from its rated operating point and exhibit unknown values, which generally does not be equal to one. In order to construct the effective adaptive fault-tolerant controller and improve the fault-tolerant stability control performance, it is necessary to estimate the unknown input power fault and obtain its precise information. Furthermore, using the neural networks, a novel adaptive fault-tolerant compensation control strategy is designed to guarantee that the whole system is asymptotical stable and all the closed-loop signals converge to the origin. What’s more, the unknown system input power fault parameters can be estimated online through the constructed estimation signals, and some assumptions proposed in the existing results are relaxed in the presented control strategy. Finally, the simulation results demonstrate the effectiveness of the proposed control scheme.</p>