论文部分内容阅读
在电力信息系统中,中间件的运维工作需要从传统的事后排查提升为故障预判和智能预警,面向电力中间件的故障预测与健康管理(PHM)技术成为当前迫切需要研究的课题;分析和利用PHM技术的数据处理流程,突破性将该技术应用于智能化运维管理平台的中间件集群管理;以灰色状态下的马尔科夫预测模型为核心预测算法,结合时间切片管理和动态置信阈值技术,设计并实现了面向中间件集群管理的故障预警模型;以湖北省电力公司构建的智能化运维管理平台上的实施应用为实例,该方法实现对中间件故障准确预警,并大大降低了虚警率;实验结果证明了该方法的可行性和有效性。
In the power information system, the operation and maintenance of middleware need to be upgraded from traditional after-investigation to fault prediction and intelligent early warning. Fault prediction and health management (PHM) technology for power middleware has become an urgent issue to be studied; And using the PHM technology data processing flow, this technology is applied to the middleware cluster management of the intelligent operation and maintenance management platform in a breakthrough manner. Taking the Markov forecasting model with gray status as the core prediction algorithm, combined with time slice management and dynamic confidence Threshold technology to design and implement a fault early warning model for middleware cluster management. Taking the implementation and application of the intelligent operation and maintenance management platform built by Hubei Electric Power Company as an example, this method can accurately detect the middleware failures and greatly reduce The false alarm rate was tested. The experimental results proved the feasibility and effectiveness of this method.