电网多源异构缺失数据最优投影整合算法研究  

Research on Optimal Projection Integration Algorithm for Multi-Sources Heterogeneous Missing Data in Power Grid

作　　者：杨晶 妥建军 李昊 廖翯 马雅蓉 YANG Jing;TUO Jianjun;LI Hao;LIAO He;MA Yarong(Development Division of State Grid Gansu Electric Power Company(Economic and Technological Research Institute),Lanzhou 730046,China)

机构地区：[1]国网甘肃省电力公司发展事业部(经济技术研究院),甘肃兰州730046

出　　处：《自动化仪表》2024年第4期76-81,共6页Process Automation Instrumentation

基　　金：甘肃省自然科学基金资助项目(2019GS100237)。

摘　　要：电网多源异构数据分散性较强,导致数据随机缺失问题严重,影响电网整体运行的管理质量。研究了一种电网多源异构缺失数据最优投影整合算法。首先,引入拉格朗日差值方法填充电网多源异构数据中的缺失值。其次,采用经验模态分解方法消除电网多源异构数据中的噪声。最后,建立投影指标函数,将高维电网多源异构数据经过投影处理后转变为一维的投影值。计算模糊聚类迭代适应度,利用混沌文化差分进化算法寻找电网多源异构数据的最优投影方向,完成缺失数据的整合。试验结果表明,所提算法的数据去噪效果较好,去噪后电网多源异构数据并未出现失真现象,并且能够在提高整合精度的同时确保算法运行的稳定性。所提算法数据整合精度受数据缺失的影响较小。该研究能够提高电网多源异构数据的质量,有助于提高电网运行管理水平。The strong dispersion of multi-source heterogeneous data in power grids leads to a serious problem of random missing data,which affects the management quality of the overall operation of power grids.A power grid multi-source heterogeneous missing data optimal projection integration algorithm is studied.Firstly,the Lagrange difference method is introduced to fill the missing values in the multi-source heterogeneous data of power grid.Secondly,an empirical mode decomposition method is used to eliminate the noise in the grid multi-source heterogeneous data.Finally,the projection index function is established to transform the high-dimensional power grid multi-source heterogeneous data into one-dimensional projection values after projection processing.The iterative adaptation of fuzzy clustering is calculated,and utilize the chaotic cultural difference evolution algorithm to find the optimal projection direction of the power grid multi-source heterogeneous data to complete the integration of the missing data.The experimental results show that the proposed algorithm has a better data denoising effect,and there is no distortion of the power grid multi-source heterogeneous data after denoising,and it can ensure the stability of the algorithm operation while improving the integration accuracy.The data integration accuracy of the proposed algorithm is less affected by missing data.The research can improve the quality of the multi-source heterogeneous data of the power grid and help to improve the operation and management level of the power grid.

关 键 词：电网多源异构数据 数据填充 投影指标函数 拉格朗日差值方法 数据整合 最优投影方向 

分 类 号：TH123-1[机械工程—机械设计及理论]