Flow field distribution and structural strength performance evaluation of fixed offshore wind turbine based on digital twin technology

Abstract

Due to the adverse impact of the offshore environment, the cost of on-site monitoring, operation and maintenance of offshore wind turbines is greatly increased. Here, a digital twin (DT) method based on digital model and computational fluid dynamics (CFD) simulation database is proposed, which is used to rapidly predict and synchronously display the distribution of the wake field, structural deformation and stress of fixed OWTs. First, a large number of data have been calculated by using CFD method. Second, the three-dimensional finite element model is reduced to a digital model by the proper orthogonal decomposition method, all data is stored in a multi-source heterogeneous database. Furthermore, the anisotropic inverse distance weighted interpolation and particle swarm optimization algorithm methods are used to obtain uncalculated data results and supplement database. Then the Bayesian regularization-back propagation neural network method is developed to correct the data with large errors. The results show that the learning error of the flow field and structural strength is less than 10% and 4% compared with the CFD method, respectively. This study could provide a rapid monitoring engineering reference for the safety assessment of turbine flow field distribution and structural strength.