Self-constructing fuzzy neural network-based fault diagnosis and monitoring for asynchronous motors

Induction motors play an essential role in industrial applications due to their reliability, efficiency, and durability. Ensuring their continuous, trouble-free operation is critical to maintaining productivity and reducing maintenance costs. This work presents an intelligent framework for diagnosing and monitoring faults in three-phase induction motors using a Self-Constructing Fuzzy Neural Network (SCFNN). It dynamically adjusts its structure and parameters based on input data, enabling it to recognize deviations from the motor’s normal behaviour effectively. The proposed system combines real-time monitoring with adaptive learning to detect and classify faults. The MATLAB/Simulink software was used to model the induction motor and implement the SCFNN-based controller. Simulation results demonstrated the robustness of the SCFNN under various operating conditions, highlighting its ability to accurately track reference signals and detect faults early. This approach offers a promising solution for predictive maintenance in industrial environments, reducing system downtime and enhancing motor reliability.