Due to the special operating environment for offshore wind turbines

it is often to replace parts as a whole when a failure occurs. Implementing condition-based maintenance and selecting suitable replacement cycles is an effective strategy for reducing operational costs. But the condition monitoring and health diagnosis technology overseas are not reliable. Aiming at the problem

based on operation and maintenance logs

and the time interval between potential fault and functional fault

the time-delay theory is introduced into the offshore wind turbine maintenance

and the law of development from potential failure to functional failure is modeled. Based on this

an irregular preventive maintenance strategy is proposed

and the parts replacement cycle is optimized. Case analysis shows that the proposed method can determine a set of optimal operational strategies

maintain the parts unit cost expectation at a low level. In the meantime

the results of sensitivity analysis can be used to make operation risk assessment for the optimized strategy. This work is supported by Science and Technology Innovation Project of Shanghai Science and Technology Commission(No. 14DZ1200905).