<![CDATA[In the dynamic realm of electrical system traction, when Electric Vehicles (EVs) operate at various speeds or require high levels of accuracy and reliability in propulsion, malfunctions or faults might occur. Therefore, the drive system must be capable of detecting, estimating, and accommodating these faults using the designed controllers. This paper proposes an efficient Fault-Tolerant Control (FTC) based on the Adaptive Neuro-Fuzzy Inference System (ANFIS) and an integrated Luenberger Observer (LO) for speed tracking control of an EV driven by a Double-Fed Asynchronous Motor (DFAM). The ANFIS controller and LO are employed to play two functions: One for sensorless control and the other for estimating the fault that affect the machine. The performance metrics and accuracy of the ANFIS process are tested using statistical parameters, sush as Root Mean Square Error (RMSE), and convergence analysis. We use a High-Order Sliding Mode Controller (HOSMC), as a nominal control for DFAM. Moreover, the efficacy of the suggested control is demonstrated by comparing its performance with conventional FTC. We have found that ANFIS improves both the precision and responsiveness of the FTC, demonstrating no peak overshoot as well. The obtained results prove that the FTC-based on ANFIS was more enhanced fault estimation accuracy, reduced error, and faster convergence than the conventional FTC methods. Finally, these significant improvments underscore the effectiveness of the suggested algorithm.]]>
Copyrights © 2025
