摘要
为了能够最大限度地提高海上风电机组的利用率,降低维护成本,在考虑海上风机运行情况的基础上,对风机状态进行评定和趋势预测,并按照实际需求优化维护计划。首先根据风机状态监测统计数据或典型的n状态左?右无跨越模型,采用马尔科夫状态转移方程预测风机状态。然后考虑维护动作费用、天气窗口的等待时间及备货期的电价收益损失等因素,以运行寿命内折扣费用最低为准则,建立基于半马尔科夫决策过程的维护策略优化模型,确定任意时刻的维护方式和检测时间间隔。最后,以某海上风机为例,验证分析了该维护策略的有效性和模型的适应性。
In order to improve availability of offshore wind power generations and reduce maintenance cost, a model for optimizing maintenance strategy of offshore wind turbine based on condition monitoring and prediction is presented. Firstly, according to condition monitoring data or result of typical n state model without left and right crossing, condition of offshore wind turbine is predicted using Markov state transition equation. Then several factors, such as maintenance cost, weather window waiting time, production loss in lead time for spare parts and so forth are considered in the model. Aiming at minimizing maintenance discount costs during operational cycle, the maintenance strategy model based on semi-Markov decision process(SMDP) is developed, in which maintenance action and time interval of detection are determined at any time. Finally, a case of offshore wind turbine is demonstrated to prove the effectiveness of the proposed strategy.
出处
《电网技术》
EI
CSCD
北大核心
2015年第11期3292-3297,共6页
Power System Technology
基金
上海市重点支撑公关计划项目(13160500800)~~
关键词
海上风机
维护策略
状态预测
马尔科夫决策过程
offshore wind turbine
maintenance strategy
state prediction
Markov decision process