基于模糊贝叶斯的风机部件失效风险评估方法

A Fuzzy Bayesian Based Risk Assessment Method for Wind Turbine Component Failure

针对失效模式与影响分析(FMEA)方法难以精确描述失效风险的问题,提出了一种基于模糊贝叶斯的风机部件失效风险评估方法。根据风机的工作原理、相关文献确定了风机的9个关键部件及其主要失效模式;利用隶属函数对发生度、严重度、检测度的评价值进行模糊化处理,建立含置信度的IF-THEN推理规则表示部件失效属性和风险程度之间的关系,并利用贝叶斯网络推理技术实现置信推理规则的转换;通过贝叶斯网络得到风机部件失效的风险程度,对风机部件失效风险程度进行排名。结果表明:叶片、齿轮箱、发电机的失效风险程度排前三,且提出的方法较传统方法更为精确和合理。

To address the problem of cannot precisely describe the failure risk using failure modes and effects analysis(FMEA),a fuzzy Bayesian based risk assessment method for wind turbine components failure is proposed.According to the working mechanism of wind turbine and related literatures,nine key components of the wind turbine and their main failure modes are determined.The membership function is used to describe the occurrence,severity,and detection.IF-THEN reasoning rules with belief structures are established to represent the relationship between component failure properties and degree of risk.And use Bayesian network reasoning technology to realize the conversion of belief reasoning rules;obtain the risk degree of fan component failure through Bayesian network,and rank the risk degree of fan component failure.The results show that the failure of blades,gearboxes and generators ranks rank top three,and the proposed method is more accurate and reasonable than traditional methods.