Abstract:Aiming at the problem of uncertainty quantification and maintenance decisionmaking of the failure correlation effects of offshore wind turbine components，an optimisation method of preventive maintenance strategy for offshore wind turbines considering the failure correlation uncertainty is proposed. Firstly，the interval analysis method is used to quantify the uncertainty of failure correlation，and the failure correlation interval model is constructed to obtain the reliability interval of components under the influence of uncertainty. Secondly，according to the running state information of the components，the neural network is used to capture the spatiotemporal characteristics of the fault correlation graph，and a fault correlation prediction model based on GAT-GRU is constructed to form an interval fault correlation coefficient matrix.Finally， taking the minimum maintenance cost as the goal and the component reliability index as the constraint，a preventive maintenance model considering the uncertainty of fault correlation is constructed to optimize the preventive maintenance time interval of each component. The simulation results verify the effectiveness and economy of the proposed method.