Abstract: To make the maintenance decision based on reliability consistent with the actual operational and maintenance process,and to improve the comprehensive economic benefit of wind farm,the failure rate function update model of wind turbine was provided based on the existing imperfect maintenance models considering the principle of generalized renewal process.Furthermore,the multi-objective periodic dynamic imperfect preventive maintenance model of wind turbine was developed on the premise of considering set-up cost and ensuring the basic availability of wind farm.The optimal solution was achieved with genetic algorithm,which compared with two traditional maintenance strategies that were perfect preventive maintenance strategy and maintenance strategy without the influence of number of minimal maintenance on set-up cost.The sensitivity analysis was made,and the economic benefit,effectiveness and feasibility of the proposed model were validated.

Key words: imperfect preventive maintenance, wind turbines, maintenance decision, virtual age factor, failure rate accelerated factor, genetic algorithms

摘要： 为了使基于可靠度的维修决策与风电场实际运维一致,达到提高风电场整体经济效益的目的,在非完美维修模型的基础上,基于广义更新过程思想构建了风电机组的故障率函数更新模型,进而在考虑维修准备成本和保证风电场基本可用度的前提下提出风电机组的多目标定周期动态非完美预防性维修决策。通过遗传算法求得模型最优解,并将结果与传统完美维修策略和不考虑小修次数对维修准备成本影响的维修策略进行对比,同时对其进行灵敏性分析,验证了模型的有效性、经济性和可行性。

关键词: 非完美预防性维修, 风电机组, 维修决策, 虚龄因子, 故障率加速因子, 遗传算法