Towards an Effective Artificial Intelligence Systems for Condition Monitoring of Off-Shore Wind Turbines: The Application of Sensor Fusion

Abstract

AbstractIn the face of increasing energy demand and the upsurge in the recent energy prices post Covid-19 pandemic, scientists and technologists around the world are working to develop more efficient renewable energy alternatives. Among such technologies, wind turbines play an important role as a very mature clean energy technology. But minimising maintenance costs and downtime is critical for off-shore wind turbines; and researchers around the world are trying to develop comprehensive online and real time monitoring systems to monitor the health of wind turbines to advance condition-based maintenance (CBM) strategies in order to reduce cost and enhance availability. There is a need to use sensor fusion since a single type of sensor is not expected to capture the needed information regarding the health of the wind turbine due to the complexity of the operational conditions such as wind speed, wind direction, power output, environmental temperatures; in addition to many other factors. Industrial case study will be presented in this paper to explore the sensor fusion option and discus how to select the most suitable sensors to detect a specific fault, or group of faults, among hundreds of sensors. This is considered a critical step for the development of an artificial intelligence CBM system. The paper presents the use of the ASPS approach (Automated Sensor and Signal Processing Selection). The results show that the suggested methodology could easily identify the sensors and signal processing methods that are sensitive to fault conditions for future diagnostics and prognostics.