Abstract:With the large number of new energy access and the wide participation of users, the data of power grid enterprises is growing exponentially. Power grid planning departments urgently need to use big data analysis methods to improve the accuracy of planning decisions. In this paper, the physical business of the municipal power supply company is taken as the entry point.Based on geographic location information and data fusion, this paper uses data visualization technology to dynamically display regional grid load distribution and grid supply capacity in multi-dimension, and assists planners to quickly discover the law of load distribution and grid vulnerability. A load forecasting model based on load characteristic analysis is constructed by using big data analysis and mining method to support grid project and operation mode arrangement. This paper breaks through the single evaluation method of power grid load problem, establishes a measurable power grid comprehensive evaluation system and model, and provides quantitative basis for power grid project co-ordination. The lean analysis platform of power grid planning based on big data can effectively improve the efficiency of power grid diagnosis and analysis through visualization of power grid load. The prediction accuracy of regional maximum load is improved by using big data analysis method, and the digital evaluation of power grid problems is realized, and the precision of investment decision in power grid planning is improved.