Line loss calculation of low-voltage districts based on improved K-Means

The low-voltage districts is complex, large in number and have various types of users, which makes the calculation of line loss more complicated. Therefore, combining the research status of line loss calculation at home and abroad with the related algorithms of data mining, based on the power consumption data of low-voltage districts, we proposed a prediction model of low-voltage districts line loss based on improved K-Means algorithm. First of all, cluster the data pre-processed. Secondly, establish multivariate linear regression prediction model for each clustered data. What's more, input the sample data and determine which type of the districts they belongs by Euclidean distance so as to predict the line loss rate. Finally, analyze the prediction error of the model by comparing the actual value with the predicted value. The analysis of the actual data of low-voltage districts shows that the improved K-Means has the characteristic of high prediction accuracy and it's simple, fast and practical. It has superior performance in analyzing and processing the data in low-voltage districts, which can effectively improve the management, standardization and refinement level of low-voltage districts.