Knowledge and Data Fusion-driven for Offshore Wind Turbine Gearbox Fault Diagnosis

At present, the cumulative installed capacity of offshore wind power in China has reached 27.26 million kilowatts, which promotes the clean and low-carbon transformation of energy and helps China achieve the goal of "carbon peak and carbon neutrality". However, the gearbox of offshore wind turbine is affected by its structure, working condition and environment, which leads to high gearbox failure. Traditional fault diagnosis methods are difficult to meet the requirements of offshore wind turbine equipment diagnosis with variable working conditions and multiple fault types. This paper proposes a knowledge and data fusion-driven fault diagnosis method for offshore wind turbine gearbox. It not only classifies the operating data of offshore wind turbine gearbox through convolutional neural networks (CNN), but also uses knowledge graph to display detailed information of faults and perform intelligent question answering. The innovation of this method is that the fault diagnosis method based on data monitoring and the fault type reasoning using knowledge graph are combined to form a comprehensive fault diagnosis model of offshore wind turbine gearbox. This method is applied to the fault diagnosis of gearboxes in Jiangsu offshore wind farms. For the types of cracks, wear and missing teeth of offshore wind turbine gearbox, the accuracy of fault diagnosis based on convolutional neural network model reaches 95%. At the same time, the visual display and intelligent question answering of offshore wind turbine gearbox faults are realized by using the constructed knowledge graph. The results show that the knowledge and data fusion-driven fault diagnosis method for offshore wind turbine gearbox has a good application effect on the intelligent operation and maintenance of offshore wind turbines.

Keywords: Offshore wind turbine gearbox, Knowledge and data fusion-driven, Fault diagnosis, Convolutional neural network, Knowledge graph.

Download PDF