Reliability analysis of floating offshore wind turbine generator based on failure prediction and preventive maintenance

Aiming at the failure problems of a floating offshore wind turbine (FOWT) generator, a preventive maintenance model based on failure prediction is established. Firstly, the fault tree method is adopted to calculate the generator reliability value under run or failure states. Secondly, quantitative analysis of the generator in a maintainable state is carried out by the Markov method. Finally, a system optimization model is established based on the failure prediction and preventive maintenance (FPPM) strategy. The single-objective teaching-learning-based optimization (TLBO) algorithm and multi-objective genetic algorithm (MOGA) are respectively used to solve the optimization problem. The results show that FPPM strategy can greatly improve the availability and mean time between failure (MTBF) value, which increase the system working time and efficiency. Multi-objective analysis can get the best combination of the system maintenance cost, availability, and risk than single-objective. The time interval of failure prediction is calculated as 19788 h and the system is in a safe state if the remaining life is predicted to be more than 101 h. Sensitivity analysis of the model is carried out to determine that the optimization objectives are less sensitive to variables, which verifies the rationality and accuracy of the proposed model.

Automated summary

This study establishes a preventive maintenance model based on failure prediction for a floating offshore wind turbine (FOWT) generator. The reliability value of the generator under run or failure states is calculated using the fault tree method, and quantitative analysis of the generator in a maintainable state is carried out using the Markov method. The results show that the FPPM strategy can improve the system's availability and mean time between failure (MTBF) value.