Condition based monitoring for fault detection in windmill gear box using artificial neural network

Abstract

Gears are condemning element in a diverse industrial application such as machine tool and gearboxes. Many contributing losses occur due to an unexpected breakdown of the gears. Recently, in many researches fault diagonsis has been the most contributing content in the gears. Vibration analysis has been used as a deciding tool for several machinery maintenance decisions. The most efficient form of indication is by the increment in the vibration level. By measuring and analysing the machine's vibration, it is possible to determine both the nature and severity of the defect, and hence predict the machine's failure. The vibration signal of a gearbox carries the signature of the fault in the gears and early fault detection of the gearbox is possible by analysing the vibration signal using different signal processing techniques. The present work depicts about the predictive maintenance program implementation based on the experimental data set of the gear box assembly in wind turbines by using the Artificial neural network and comprises of the prediction of variation in the accuracy of the experimental data set. The code values are generated using python 3.7 version software. The accuracy of the Model was found to be 61.86%. This accuracy indicates the functioning of the gears and can also be used for predictions in real time scenario.