无监督学习下移动电费收缴数据异常波动辨识  

Identification of abnormal fluctuation of mobile electricity charge data collection based on unsupervised learning

在线阅读下载全文

作  者:何小宇 汤闰 HE Xiao-yu;TANG Run(Guangzhou Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Guangzhou 510000,China)

机构地区:[1]广东电网有限公司广州供电局,广州510000

出  处:《信息技术》2021年第7期150-154,共5页Information Technology

摘  要:移动电费收缴系统数据存在差异特征量,易出现异常波动,提出无监督学习下移动电费收缴数据异常波动辨识研究。无监督学习环境下,构建数据采集模型,对采集的系统数据进行统计分析和自相关特征匹配处理,通过分析采集数据的差异性特征量,结合线性拟合和多元参数融合方法,融合调度差异性特征量,调度关联规则特征集,提取系统数据异常波动谱特征量,建立波动振荡分析模型,结合状态参数监测,实现移动电费收缴系统数据异常波动辨识。实验结果表明,对移动电费收缴系统数据异常波动辨识的准确性较高,波动特征量提取精度较好,提高了移动电费收缴的异常监测能力。There are different characteristic quantities in the data of mobile electricity collection system,which are prone to show abnormal fluctuations.Therefore,this paper proposes a study on the identification of abnormal fluctuations of mobile electricity collection data under unsupervised learning.Under unsupervised learning environment,the data acquisition model is constructed,and the collected system data are statistically analyzed and the autocorrelation feature matching is processed.The abnormal data fluctuation identification of mobile electricity collection system is realized by a series of processes,including analyzing the different characteristic quantity of the collected data,and combining with the linear fitting and multi parameter fusion method.The scheduling difference characteristic quantity and the scheduling association rule feature set are fused as well,and the abnormal fluctuation spectrum characteristic quantity of the system data is extracted,and the fluctuation vibration is established based on the analysis model,and combined with state parameter monitoring.The experiment results show that the accuracy of abnormal fluctuation identification of mobile electricity collection system is high,and the extraction accuracy of fluctuation characteristic quantity is better,which improves the abnormal monitoring ability of mobile electricity charge collection.

关 键 词:无监督学习 移动电费 收缴系统 数据异常波动 辨识 

分 类 号:TP391[自动化与计算机技术—计算机应用技术]

 

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