| 引用本文: | 黄玲玲,马永杰,应飞祥,王全德,刘璐洁.基于剩余寿命预测信息的风电场动态成组维护策略研究[J].电力系统保护与控制,2024,52(16):178-187.[点击复制] |
| HUANG Lingling,MA Yongjie,YING Feixiang,et al.Dynamic group maintenance strategy for a wind farm based on residual life prediction information[J].Power System Protection and Control,2024,52(16):178-187[点击复制] |
|
| 摘要: |
| 现有的风电场成组维护优化研究中，较少考虑维护时间窗内的部件实时状态信息动态变化影响，针对此问题，提出了一种考虑剩余寿命预测信息动态更新的风电场成组维护策略。首先，利用实时状态信息获得各部件剩余寿命预测结果，基于实时剩余寿命预测结果优化最小平均维修成本，构建单部件最优维修时间窗。其次，考虑风电机组部件结构相关性及部件备件库存约束，以节省维修成本最大为目标，建立风电场成组维护模型，并采用遗传算法进行成组维护策略优化。最后，采用滚动时间窗模型实时更新机组部件的剩余寿命预测信息，动态调整原有维修方案。一个实际风电场案例的分析结果表明，所提策略能够实时更新风电场维修计划，实现维修计划的动态优化，有助于降低维修成本。 |
| 关键词: 风电场 剩余寿命预测 相关性 动态成组维护 遗传算法 |
| DOI：10.19783/j.cnki.pspc.231397 |
| 投稿时间：2023-10-31修订日期：2024-02-28 |
| 基金项目:国家自然科学基金项目资助(52177097) |
|
| Dynamic group maintenance strategy for a wind farm based on residual life prediction information |
| HUANG Lingling1,MA Yongjie2,YING Feixiang2,WANG Quande2,LIU Lujie1 |
| (1. Engineering Research Center of Offshore Wind Technology Ministry of Education (Shanghai University of Electric Power),
Shanghai 200090, China; 2. College of Electrical Engineering, Shanghai University of Electric Power, Shanghai 200090, China) |
| Abstract: |
| There is a problem that the dynamic change of real-time condition information of the components in a maintenance time window is rarely considered in existing research on wind farm group maintenance optimization. Thus this paper proposes a wind farm group maintenance strategy that considers the dynamic update of the remaining life prediction information. First, the residual life prediction results of each component are obtained using real-time condition information. Based on these results, the minimum average maintenance cost is optimized, and the optimal maintenance time window of single component is constructed. Secondly, considering the structural correlation of wind turbine components and the inventory constraints of spare parts, a group maintenance model of the wind farm is established to save the maximum maintenance cost, and a genetic algorithm is used to optimize the group maintenance strategy. Finally, a rolling time window model is used to update the remaining life prediction information of wind turbine components in real time and dynamically adjust the original maintenance plan. The analysis of an actual wind farm case shows that the proposed strategy can update the wind farm maintenance plan in real time, realize the dynamic optimization of the maintenance plan, and help to reduce the maintenance cost. |
| Key words: wind farm residual life prediction correlation dynamic group maintenance genetic algorithm |
