Design of the Wind Turbine Gearbox Fault Diagnosis System Based on LabVIEW

Abstract

In the research history of the wind turbine gearbox fault diagnosis, artificial neural network plays an important role. However, due to the shortcomings of single neural network, the accuracy of fault diagnosis is not high. The fault diagnosis system of wind turbine gearbox based on LabVIEW was designed to monitor the speed, the power, the oil temperature of the bearing, the pressure, the vibration signal etc. Mathematical morphology was used to preprocess the collected data, such as filtering, and then four different neural networks used for primary diagnosis. The diagnosis results of each network were smoothed by mathematical morphology operation for the final diagnosis result. The experimental results show that the system has an excellent human-computer interaction interface to analyze and diagnose its operational status. The application of mathematical morphology combined with the multi-neural network in the fault diagnosis of wind turbine gearbox is conducive to reducing the uncertainty. It can correctly diagnose the fault type and improve the accuracy in the case of fault characteristics which are similar.