Zastosowanie radialnych sieci neuronowych w detekcji uszkodzeń wirnika silnika indukcyjnego

Diagnostics of electrical machines is complicated process based on such elements as: measurements of chosen signals and parameters of the motor, transformation of the obtained results in order to separate fault symptoms and the fault detector and classifier design. In this paper fault detectors and classifiers based on neural networks with radial activation function are implemented for diagnostics of rotor damages in induction motors. The main stages of the design methodology of the radial basis neural detectors are described. Furthermore, influence of neural networks complexity and parameters of neuronal activation function on quality of data classification is shown. Presented neural detectors are tested with measurement data obtained in laboratory setup contained of converter-fed induction motor and changeable rotors with different degree of damages.