Application Research Of Data Mining In Power Distribution GIS System

Data Mining is a newly technology to found useful information and knowledge from digging up and analyzing a large amount of data, in various industries it has a wide range of applications. Here combine with the power system, the result of forecasting the short-term power load will have great practical significance on current planning and scheduling.The load forecast is the electric power plan foundation, accurate load forecasting will greatly increase the stability and security of power supply system, reduce the resource waste generation costs, and provides supplementary opinions for power planning and scheduling.This paper research the data mining theory and the main impact factor about load forecasting, considering, there are lots of disadvantages in the traditional neural network prediction ways, including be sensitive to the initial network weights, easy to run into the local minimum point, etc. this paper brings forward genetic algorithm to the BP neural network, optimizing the initial network weights. in order to improve the accuracy ,then use the Bagging method integrated the results, aim to improve its accuracy. This is a new experiment of combine the advantages in different algorithms.This paper describes the structure and function of the power distribution GIS system at last, and forecasting the load with the model. Using the spatial analysis function of the GIS platform, deal with the relevant factors impact the load. The result show that not only the Bagging method and Genetic neural network can avoid the disadvantages in the traditional BP network and inherit its good learning and training abilities, but also have stronger generalization ability ,and improve the prediction precision, and achieved satisfactory predictions.