考虑用户数据隐私保护的分布式加密台区线损计算方法研究  

Research on a distributed encrypted method for non-technical loss calculation considering user data privacy protection

在线阅读下载全文

作  者:卜龙敏 解玉满 陈红 陈建华 徐慧婷 陈浩 颜钟宗 BU Longmin;XIE Yuman;CHEN Hong;CHEN Jianhua;XU Huiting;CHEN Hao;YAN Zhongzong(State Grid Hunan Electric Power Company,Changsha 410000,China;Hunan Provincial Key Laboratory of Intelligent Electric Measurement and Application Technology,Changsha 410000,China;College of Electrical and Information Engineering,Hunan University,Changsha 410012,China)

机构地区:[1]国网湖南省电力有限公司,湖南长沙410000 [2]智能电气量测与应用技术湖南省重点实验室,湖南长沙410000 [3]湖南大学电气与信息工程学院,湖南长沙410012

出  处:《中国测试》2024年第S2期230-234,240,共6页China Measurement & Test

基  金:国家电网公司科技项目资助(5400-202323233A-1-1-ZN)

摘  要:随着电力系统的数字化和智能化发展,配电网中的台区线损计算成为电力公司重点关注的问题。然而,伴随着数据隐私问题的日益凸显,传统的线损计算方法面临用户数据泄露的风险。为此,该文研究了一种兼顾用户数据隐私保护的配电台区线损计算方法。该方法通过在XGBoost算法中引入多方秘密共享(Secret Sharing)技术,重构了切分点查找算法,并将模型叶子节点权重的计算过程进行分布式处理,从而实现了在多方参与计算时不泄露用户信息的目标。该文所研究的方法已在实际台区线损数据中进行了测试,结果表明,该算法不仅能够准确估计台区线损,还能有效保护用户隐私。As the digitization and intelligence of power systems continue to develop,the calculation of nontechnical losses in distribution networks has become a critical concern for power companies.However,with the increasing prominence of data privacy issues,traditional line loss calculation methods face the risk of user data leakage.Therefore,this paper explores a method for feeder line loss calculation in distribution networks that considers user data privacy protection.By introducing the multi-party secret sharing technique into the XGBoost algorithm,the splitting point search process within the XGBoost algorithm is reconstructed,and the calculation of the model’s leaf node weights is performed in a distributed manner.This ensures that user information is not leaked during multi-party computations.The proposed method was tested on actual feeder line loss data,and the results demonstrate that the algorithm can accurately estimate line losses while protecting privacy.

关 键 词:线损计算 隐私保护 秘密分享 XGBoost 

分 类 号:TP309[自动化与计算机技术—计算机系统结构] TM744[自动化与计算机技术—计算机科学与技术]

 

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