For the increasingly prominent problems of wind turbine maintenance, using edge cloud collaboration technology to construct wind farm equipment operation and maintenance framework is proposed, digital twin is used for fault prediction and diagnosis. Framework consists of data source layer, edge computing node layer, public or private cloud. Data source layer solves acquisition and transmission of wind turbine operation and maintenance data, edge computing node layer is responsible for on-site data cloud computing, storage and data transmission to cloud computing layer, receiving cloud computing results, device driving and control. The cloud computing layer completes the big data calculation and storage from wind farm, except that, based on real-time data records, continuous simulation and optimization, correct failure prediction mode, expert database and its prediction software, and edge node interaction and shared intelligence. The research explains that wind turbine uses digital twin to do fault prediction and diagnosis model, condition assessment, feature analysis and diagnosis, life prediction, combining with the probabilistic digital twin model to make the maintenance plan and decision-making method.
A Digital Twin Based Self-Calibration Tool for Fault Prediction of FDM Additive Manufacturing Systems
