Prediction of wind turbine generator bearing failure through analysis of high-frequency vibration data and the application of support vector machine algorithms

Abstract

Innovations which help facilitate predictive maintenance strategies have the potential to greatly reduce costs associated with wind turbine O&M by driving efficiency and increasing wind turbine availability. This study uses multiple examples of the same generator bearing failure to provide insight into how condition monitoring systems can be used in to train machine learning algorithms with the ultimate goal of predicting failure and remaining useful life. Results show that by analysing high-frequency vibration data and extracting key features to train support vector machine algorithms, an accuracy of 67% can be achieved in successfully predicting failure 1–2 months before occurrence. This study reflects on the limitations surrounding a generalised training approach, taking advantage of all available data, showing that if too many different examples are considered of different wind turbines and operating conditions, the overall accuracy can be diminished.

Keywords

• vibrations
• remaining life assessment
• power generation faults
• wind turbines
• mechanical engineering computing
• turbogenerators
• condition monitoring
• support vector machines
• learning (artificial intelligence)
• fault diagnosis
• machine bearings
• wind turbine generator
• high-frequency vibration data
• support vector machine algorithms
• wind turbine downtime
• asset management
• onshore developments
• offshore developments
• multiple examples
• generator bearing failure
• condition monitoring systems
• remaining useful life
• generalised training approach
• different wind turbines
• time 1.0 month to 2.0 month