向家坝电站运行设备异常状态识别功能的设计与实现

Design and Implementation of the Function of Identifying Abnormal State of the Operation Equipment of Xiangjiaba Power Station

针对水电站计算机监控系统常规限值报警发生时,设备已经出现严重故障影响电站安全稳定运行的问题,以水电站历史运行数据为基础,运用人工智能与数据挖掘技术自动学习运行设备的健康特征,实现水电站运行设备异常状态识别功能,在设备运行参数尚未到达故障报警限值前发现异常,达到设备异常早发现早排除的目的,并对辅助系统渗漏异常、开停机流程异常、温度异常、电压与频率异常等关键技术进行了讨论。该功能通过水电站生产设备异常运行状态的感知与报警,实现了安全生产隐患的早期发现,对于保障水电站安全稳定运行具有重要的意义。

In respect to the problem of the equipment failures when limit alarm occurs in SCADA system of hydropower station,artificial intelligence and data mining technology are used to automatically obtain the health features of the operating equipment based on the historical operating data of the hydropower station,and to realize the function of identifying the abnormal state of the operating equipment of the hydropower station,find the abnormality before the equipment operating parameters reach the fault alarm limit,achieve the purpose of early detection and early elimination of the equipment abnormality,and key technologies such as abnormal leakage of the auxiliary system,abnormal startup and shutdown procedures,abnormal temperature,and voltage frequency anomalies are discussed.The function realizes the early detection of potential safety hazards through the perception and alarm of the abnormal operation status of the production equipment of the hydropower station,which is of great significance for ensuring the safe and stable operation of the power station.