基于PSC-9000系统平台的换流站阀冷数据智能预警模型  

Intelligent Alarm Model of Valve Cooling Data in Converter Station Based on PCS-9000 System Platform

在线阅读下载全文

作  者:范富江 田维文 黄仕鹏 Fan Fujiang;Tian Weiwen;Huang Shipeng(Guiyang Bureau of CSG EHV Power Transmission Company,Guiyang 550001,China)

机构地区:[1]中国南方电网有限责任公司超高压输电公司贵阳局,贵阳550001

出  处:《机电工程技术》2023年第9期211-214,共4页Mechanical & Electrical Engineering Technology

摘  要:为提高换流站阀冷数据预警的准确性,缩短阀冷数据的计算时间,提出了一种基于PCS-9000系统的阀冷分析模型。首先利用粗糙集方法收集PCS-9000系统中的阀冷数据,剔除冗余数据,计算不同数据之间的相关性,并将数据进行标准化处理,实现不同数据结构之间的转换。然后用能量守恒理论对液位、温度、流速进行监测,通过设定阈值识别异常值,确定异常值出现的原因和位置。最后比较异常值,结合PCS-9000系统中的日志、报错信息,确定异常值的危险等级并进行告警。结果显示,所提分析模型能充分利用PCS-9000系统数据,缩短计算时间,提高告警准确性,结果显著优于单独的PCS-9000系统分析,符合换流站的告警要求。To improve the accuracy of valve cooling data early warning in converter stations and shorten the calculation time of valve cooling data,a valve cooling analysis model based on the PCS-9000 system is proposed.Firstly,this paper uses the rough set method to collect valve cooling data in the PCS-9000 system,eliminates redundant data,calculates the correlation between different data,and standardizes the data to achieve conversion between different data structures.Then,the energy conservation theory is used to monitor the liquid level,temperature,and flow rate.By setting thresholds,abnormal values are identified and the cause and location of the abnormal values are determined.Finally,by comparing the outliers,and combining with the logs and error messages in the PCS-9000 system,the hazard level of the outliers can be determined and issue relevant alarm.The results show that the proposed analysis model can fully utilize the PCS-9000 system data,shorten the calculation time,improve alarm accuracy,and significantly outperform the individual PCS-9000 system analysis,thus it meets the alarm requirements of the converter station.

关 键 词:阀冷数据 智能告警 PCS-9000 系统平台 

分 类 号:TM863[电气工程—高电压与绝缘技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象