Opportunistic maintenance optimisation for offshore wind farm with considering random wind speed
提出一个联合维护决策框架，通过k-means聚类筛选历史风速数据并建立马尔可夫链，模拟维修船等待时间，将风机部件按有效年龄分为四状态，以最小化维护成本为目标，比较三种机会维护策略，发现基于年龄的机会维护（ABOM）更有效且适合长期计划。
A joint maintenance decision-making framework is proposed to optimise the long-term maintenance plan and lower the maintenance cost for offshore wind farms. The historical wind speed data are screened by using the method of k-means clustering, and Markov chains are established for the wind speed in different seasons. On this basis, the approach of Markov chain Monte Carlo is applied to simulate the distribution of repair vessel's waiting time for maintenance, where the impact of wind speed on maintenance availability is considered. Moreover, the components in wind turbines are divided into four states according to their effective ages, i.e. young, mature, old and failed, respectively. A maintenance decision model is established, with the objective to minimise maintenance cost. Besides, three types of opportunistic maintenance are considered, i.e. failure-based opportunistic maintenance (FBOM), event-based opportunistic maintenance (EBOM) and age-based opportunistic maintenance (ABOM), respectively. The enhanced elitist genetic algorithm (SEGA) is adopted to solve the optimisation problem. The results indicate that among the three types of opportunistic maintenance, ABOM can reduce maintenance cost more effectively, and it is more suitable for long-term maintenance plans of offshore wind farm.