Analysis And Research On Line Loss Of County Distribution Power Network

The county distribution power network has the characteristics of low voltage level,imperfect topology,many power users,infrastructure construction and power detection equipment.It makes the SDADA system unable to completely collect the information of all power users,which brings great difficulties to reduce line losses.The prediction model can provide a theoretical basis for reducing line loss.Therefore,the practical and effective line loss rate prediction model is of great significance for the development of loss reduction.Firstly,the abnormal district in the county distribution power network is analyzed and preprocessed.Based on the analysis of high-loss and negative-loss,the typical abnormal district screening model and load abnormal district identification model are established.The model can effectively identify the types of abnormal district,and determine the set of abnormal districts.According to the results,a comprehensive model is established.The model can eliminate the abnormal district and improve the quality of the research data.Secondly,this paper proposes a model based on NCA algorithm,it can build a feature core index system.The features are used to clustering analysis,and all low-voltage users are divided into several clusters with different properties.The cluster quality evaluation model is used to determine the best number of clusters,which can provide the basic conditions for further research.Thirdly,the BP-ANN and Elman-ANN line loss rate prediction model are established.Through the error analysis to predict the results,a prediction model that satisfies the demand for line loss rate prediction is obtained.The BP-ANN prediction model is applied to each cluster district,comparing the prediction results when the original data set is used as the input.The error between the prediction value and the real value is smaller by using k-means clustering algorithm first.Finally,on the basis of country distribution power network in Jiangsu Province,the stations can be divided into four types by using the k-means algorithm.Applying BP-ANN line loss rate prediction model to each cluster class can get high accuracy prediction results.This provides a new method for the prediction and calculation of the line loss rate,which is extremely beneficial to the economic and efficient operation of the power grid.