基于数据驱动的流域集控中心语音报警优化

Data-driven Optimization of Voice Alarm for Watershed Centralized Control Center

生产异常信息的及时发现与控制是持续性电力生产过程中的重点与难点。针对集控中心不能及时发现异常导致系统运行风险增大的问题,提出了基于报警频率、信息特征等优化语音设置的方法。该方法通过分析语音播报的频次、时长、空间分布等特点,优化语音配置,改善文本表述,采用自动化生成技术,使生产报警的及时率、有效性均得到提升。经运行证明了该方法的有效性。

Timely detection and control of abnormal production information is among the key and difficult points in the continuous power production.A method based on alarm frequency,information features,and other optimization of voice settings is proposed to address the issue that the central control center is not able to detect anomalies in a timely manner,which increases the risks of system operation.This method enables the center to analyze the frequency,duration,spatial distribution and other characteristics of voice broadcasting,optimize voice configuration,improve text expression,and adopt automated generation technology so as to improve the timeliness and effectiveness of production alarms,of which the effectiveness has been proven upon operation.