1 May 2023 in IEEE Access
Line loss management and evaluation is a core issue in the management of power grid projects. However, with the deepening of Chinas power system reform, there are too many human factors in the traditional line loss management and evaluation system. There is a lack of analysis and evaluation of the correlation between indicator data and line loss, and between indicator data and data. It is necessary to establish a new indicator relationship. This article applies knowledge graph technology to the selection of evaluation indicators of line loss management. This strategy uses recursive feature elimination and random forest fusion algorithm to filter evaluation indicators and builds intelligent evaluation indicators for line loss lean management system. Finally, the combination of the indicator system and the evaluation method is applied to the evaluation of line loss management in the municipalities of the China Southern Power Grid Municipal enterprises. The analysis of comparative examples shows that the constructed lean and intelligent evaluation index system has a better prediction accuracy than other methods, and can objectively and reasonably reflect the actual status of line loss management. Also, it can be used as a basis for future line loss management evaluation. It can be used as a decision basis for future line loss management evaluation.