Research On Evaluation Techniques Of Reactive Voltage Support Capability Based On Big Data

With the expansion of the scale of the power grid,the possibility of voltage collapse accidents is increasing.Power grid reactive power and voltage current has many problems,which requires the voltage supporting ability of the quantitative indicators,the complex power grid data in data mining,the formation of visual data,so that the operators will take care of the entire network reactive voltage operation has a global control,so the research based on large data grid voltage support capability assessment and visualization technology research has important practical significance.Based on the actual project,this paper deeply mines the potential key information of the global reactive voltage data,carries out the technical research on the evaluation of the reactive voltage support capacity,and studies the implementation method of the visualization of power grid data.Firstly,this paper studies the data value mining technology of smart grid,expounds several types of data mining technology under the background of smart grid,and analyzes the process of data preprocessing technology and its connotation including data standardization,abnormal data identification and reconstruction.An isolated forest algorithm based on sliding window technology is proposed as a fast resolution algorithm for abnormal data of power grid.Secondly,this paper analyzes the defects of power grid data acquisition and monitoring control system,and emphasizes the necessity of data preprocessing by means of clustering and interpolation.This paper studies two kinds of clustering indexes: contour coefficient and in-cluster error variance,analyzes the advantages and disadvantages of K-means clustering algorithm and density based robust to noise clustering algorithm,proposes a comprehensive K-means clustering algorithm and density based robust to noise clustering algorithm,and studies cubic spline interpolation method.The integrated clustering and interpolation method is used to process and visualize the grid data,and the statistical data is formed,which paves the way for the final evaluation and visualization.Finally,the paper analyzes the connotation of assessment questions,studied the method of subjective experience and objective method,two methods of establishing the index system of data,reactive voltage support ability evaluation index system is put forward,dynamic resolution coefficient method was proposed based on improved grey correlation degree analysis method,based on improved grey correlation degree method of reactive voltage supporting ability assessment.Based on the example,the section evaluation of the reactive voltage support capability is carried out,and the source of the existing problems is realized by analyzing the index weight.Combined with the grid location information,the 3d visualization of the grid data under two working conditions is presented,and the visualization of the grid data under fault state is presented in layers and partitions.