摘要
针对基于确定权重的评估方法难以适应风电机组特征量多且相互关系不明确的特点,提出基于多种类证据体的风电机组健康状态评估方法。首先考虑风速大小随机变化对机组振动和温度特征量波动范围的影响,提出基于动态阈值的评估指标劣化度量化方法;其次针对证据问信息可能存在的高冲突问题,按照证据间的优先度分布情况,依据证据源修正思想建立基于证据推理的风电机组健康状态评估模型;最后,以某风电机组健康状态评估为例,与传统的证据评估方法对比分析,验证所提出评估方法可均衡兼顾各评估指标的状态信息获得有效的风电机组健康状态。
In view of the fact that the evaluation method based on deterministic weight is difficult to adapt to the large amount of wind turbine features and their mutual relations are not clear, a method to assess the health status of wind turbines based on multiple evidences was proposed. Firstly, considering the influence of the uncertainty of wind speed and wind direction on the fluctuation range change of vibration characteristic and temperature characteristic parameters of generator, the quantification method of degradation index based on dynamic threshold was presented. Secondly, according to the possible conflicts between the evidences, according to the distribution of the priority among the evidences and the revision of the evidence source, the assessment model of health status of wind turbines based on evidence reasoning was established. Finally, taking the assessment of the health status of certain wind turbine as an example, comparing with the traditional evidence evaluation methods, the validation of proposed evaluation method can obtain the effective health status of the wind turbines by balancing the status information of each evaluation index.
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2018年第2期331-341,共11页
Acta Energiae Solaris Sinica
基金
“国际科技合作专项(2013DFG61520)
国家自然科学基金项目(51377184)
重庆市重点产业共性关键技术创新专项(cstc2015zdcy-ztzx0212)
重庆市科技新星培育工程项目(KJXX2017009)
关键词
风电机组
状态评估
指标量化
证据冲突
证据修正
wind turbine generator system
conditions assessment
indices quantification
evidences conflict
evidential reasoning