Font Size: a A A

The Research On Grid Space Load Forecasting Method Based On Big Data Technology

Posted on:2020-09-17Degree:MasterType:Thesis
Country:ChinaCandidate:X J ZhangFull Text:PDF
GTID:2392330590483023Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
Since the “Twelfth Five-Year Plan”,China has gradually implemented grid planning for distribution networks.Grid load density prediction is an important prerequisite for refined grid planning.Accurate grid load forecasting can improve the efficiency of investment,energy utilization,the environment and the optimal allocation of grid power capacity.The development of grid load is related to many factors,and many data cannot be provided by the planning department,which brings difficulties to grid load forecasting.Therefore,the grid space load forecasting of big data technology is of great significance to the distribution network.This paper first introduces the source of power big data,uses data mining technology and web crawler technology to obtain external data related to power load,and then combines internal data of power,and builds power through technologies such as data cleaning,data processing and data fusion.The data collection is the basis for the work to be carried out later.Secondly,the correlation factors of load release characteristics and release characteristics are analyzed.The correlation coefficients of load release characteristics are analyzed.A load sample classification method based on hierarchical analysis and clustering algorithm is proposed.According to the clustered sample and power load history data,the correlation coefficient of the classification load release characteristics is obtained by least squares fitting,which lays a foundation for the grid load density correction.Finally,a correlation analysis method of load release characteristics based on rough set information entropy theory is proposed,and the correlation factors of load release characteristics are analyzed.Finally,considering the release characteristics of grid load,a grid load forecasting method based on open source information is proposed.The total load forecasting load of the city is obtained by the integrated load forecasting method.The relative load density of each grid is obtained according to the grid load quantity.The grid load density is obtained by optimizing the relative load density,and the grid load density is corrected according to the load release characteristics.It provides a new idea for grid load forecasting of distribution network,and provides certain theoretical reference value and practical significance for distribution network construction.
Keywords/Search Tags:Big Data, Power Grid, Load Forecasting, Distribution Network Planning
PDF Full Text Request
Related items