Grid Performance Analysis Based On Data Mining

Performance analysis is one of the key links of power grid lean operation. By choosing appropriate method of data mining, the performance can be assessed in the grid planning, investment, transformation, operation business, etc. from quantitative analysis and scientific evaluation of coordination between grid and load. It would be found about the weak links of power grid development and the best way of controlling the cost, effective use of funds and improving the grid performance, also it can provide strong theoretical support on the grid planning and construction, and realize the benign interaction between the grid and economic development.Base on a survey of the performance assessment and regulation for power enterprise, a grid performance assessment method based on DEA (Data Envelopment Analysis) is developed, considering the power grid operation characteristics, the performance assessment requirements, and the data availability. In the method, R-clustering technology is used for input and output selection. Based on the input and output index, Q-clustering technology is used to classify the power grids. And then the validity of the classification is tested by statistical criterions. Furthermore, the power grids belonging to the same cluster are analyzed. The slack variables relating with an ineffective DMU are projected to illustrate the optimal route for the performance improvement by a delicacy DEA performance method. In the end, real data of the11th-Five-Year Plan period and2011provided by China Electricity Council are used to illustrate the method effectiveness and practicability.A power grid with the increasing demand should not be expanded with a uniform speed. The quantitative coordination analysis of network expansion considering load growth rate to obtain maximum benefits has become a significant issue for power system planning. In this paper, a system Granger Causality based method on coordination analysis between grid capacity and electric power demand is developed. To meet the times series stationary requirement of Granger Causality, the differential method was used to process the raw data. Based on the State Grid11th-Five-Year statistical project, the data from1993to2011are used to illustrate the method effectiveness and practicability.The effect of grid performance analysis is affected by the quality of statistical data, the quality of power statistical personnel also affects the quality of statistical data directly. In this paper, relying on the power of statistical personnel quality investigation and analysis, a kind of analytic hierarchy process based on Delphi to improve is proposed to make a comprehensive evaluation for power statistical personnel quality. According to the evaluation index system and comprehensive evaluation model, a final evaluation result will be promoted layer by layer.. At last, the analysis method of quality of statistical personnel is verified to be scientific and reasonable, using the training test scores of power statistical personnel from Zhejiang province in2013.