【作者】 于春雨； 郭建英； 苑会娟； 苏子美； 李红梅；

【Author】 YU Chunyu;GUO Jianying;YUAN Huijuan;SU Zimei;LI Hongmei;Institute of Sensor and Reliability Engineering,Harbin University of Science and Technology;AVIC Harbin Bearing Corporation LTD;

【机构】 哈尔滨理工大学传感器与可靠性工程研究所； 中航工业哈尔滨轴承有限公司；

【摘要】 风力发电机组是典型的复杂机电系统,针对其运行工况复杂、故障频发、维修困难等问题,研究了基于指数故障/指数维修的串联系统可用度评价方法,分别给出了Bayes方法和经典方法的串联系统可用度评估模型,基于现场数据计算出了无信息先验下风力发电机组系统可用度Bayes近似置信限及其经典近似置信限。根据信息等价转换原则,首先确定各单元的指数故障与指数维修数据,并求解出各单元的可用度一阶矩、二阶矩,再根据矩相等原则,利用风力发电机组系统可用性模型求解出系统可用度的一阶矩、二阶矩。分别引入Bayes方法和经典方法,将串联系统的可用性信息折算成风力发电机组正常运行次数和停机维修次数,利用不完全Beta函数及分布与F分布分位数间的关系,分别解析出二种方法对应的风力发电机组系统可用度近似置信限。以某风场风力发电机组为例,验证了该方法的可行性与适用性。更多还原

【Abstract】 Wind turbine(WT) is a typical complex electromechanical system with high frequency faults and difficult maintenance work,running in a hard and complicated working condition.Availability assessment method of series system is studied based on exponential lifetime and repair time,availability models are given based on Bayes and classical method respectively,and Bayes approximate confidence limits under non-information prior and classical approximate confidence limits for WT availability are computed according to the field data.First exponential fault and repair data of each unit are identified using information equivalent transformation principle,meanwhile,the first and second moments of every unit availability are computed.Further more,the first and second moment of WT availability are obtained using system availability model according to moment equivalent principle.The availability information of series system is translated into WT’s successful operation times and maintenance times integrating Bayes and classical methods.Bayes and classical approximate limits of WT availability are derived by incomplete beta function and the relationship between and F distribution quantile.A numerical example is applied to prove the effectiveness of this method using WT operation and maintenance data from a domestic wind farm.更多还原