Vibration analysis system applied to fault detection in wind turbines

Abstract

Wind power is one of the most important renewable energy sources due to its vast availability. Wind turbines are the equipment required to take advantage of the wind energy potential; therefore, a low reliability of these turbines limits the maximum power obtained from the wind. Different techniques and methodologies have been developed to monitor and detect failures in wind turbines in order to prevent undesirable conditions due to different operating conditions. This work presents a system designed to detect failures in wind turbines caused by mechanical vibrations, this system allows to diagnose, online, different structural failures in the wind turbine through a statistical frequency analysis based on LabView and Matlab. The designed system is validated by online measurements, obtained by 3-axis vibrations sensors in a domestic wind turbine. A graphical interface is developed in LabView in order to facilitate the online location and isolation of the detected failures.

A Wind Turbine Vibration Monitoring System for Predictive Maintenance Based on Machine Learning Methods Developed under Safely Controlled Laboratory Conditions

“…Several studies have presented different application methods that have been developed for wind turbine failure analysis based on accelerometer measurements. The use of accelerometers has several advantages since, besides being economical, they can capture a wide range of frequencies, thus providing reliable results for identifying wear in mechanisms and predicting possible errors occupying various databases (which are available for research) in order to prevent possible failure [8,20,[26][27][28][29][30].…”