摘要
故障预警研究作为资产管理的新方向,吸引着全球学者的目光.非线性状态估计(non-linear state estimation technique,NSET)以设备历史状态为基础,使用非线性组合的方式表征当前设备的预警值;并将3σ准则进一步推广,得到故障预警阈值;最终通过预警值与实际值的残差,得出设备状态是否处于故障的结论.依托现场的真实风电机组数据,经数据预处理,建立NSET预警模型,并通过设定预警阈值验证了NSET预警故障的可行性与预警策略的可实施性.
As a new direction of asset management,fault early warning research has attracted the attention of scholars from all over the world.The non-linear state estimation technique(NSET)is based on the historical state of the equipment and uses a non-linear combination to characterize the early warning value of the current equipment;the 3σ criteria are further promoted to obtain the fault early warning threshold.Finally,the early warning value is obtained.The residual error with the actual value and warning value is the observation object,and concluded whether the equipment state is in failure.Relying on the real wind turbine data on site,the NSET early warning model was established after data preprocessing,and the feasibility of NSET for warning failure and the implementability of the early warning strategy were verified by setting the early warning threshold.
作者
张冬琪
Zhang Dongqi(School of Big Data Science,Hebei Finance University,Baoding,Hebei 071051,China)
出处
《保定学院学报》
2020年第6期107-112,共6页
Journal of Baoding University
关键词
非线性状态估计
故障预警
风电机组
non-linear state estimation technique
fault warning
wind turbines