基于多源监控数据的设备状态感知方法

Asset State Perception Method Based on Multi-source Monitoring Data

近年来,随着智能电网的发展,电网资产及相应的量测数据呈级数增长。但复杂且非结构性电网数据对电网公司安全、稳定运行提出挑战,亟需在电网企业调度控制中心系统中构建多源异构数据平台。本文提出了一种基于电网多源监控数据的设备状态感知方法。首先建立监控数据模型及数据平台;接着采用基于聚类的深度学习方法开展设备监控调度系统状态感知和辅助决策研究;最后对所提出的状态感知方法进行实际验证。所提出方法可有效对设备运行状态进行判断与感知,实现检修业务的智能辅助决策。

In recent years,with the development of smart grid,the assets of power grid and the corresponding measurement data grow exponentially. Consequently,the complex and unstructured grid data pose a challenge to the safe and reliable operation of a power grid company,and a multi-source heterogenous data platform is urgently needed in the company’s dispatching and control center system. In this paper,an asset state perception method is proposed based on the multi-source monitoring data of power grid. First,a monitoring data model and a data platform are established. Then,state perception and assisted decision-making of the assets monitoring and dispatching system are developed using the deep learning technique based on clustering. Finally,the proposed method is verified on a real system,indicating that it can effectively judge and sense the operating states of assets and realize the intelligent assisted decision-making for assets maintenance.