Research On On-line Loss Analysis System Of Active Distribution Network

The power industry is the main industry of energy consumption in China.Analysis of distribution network loss has profound implications for energy conservation and loss reduction in the power system.At present,there are three problems in the analysis of distribution network loss analysis: newnergy’s intermittent and unstable characteristics adversely affect power quality in the distribution network,so that the disturbance of power quality to power consumption cannot be ignored.The loss analysis of the distribution network is mostly based on the loss rate calculation and simple statistics,which cannot be obtained the power quality factors that affect the loss of the distribution network through the existing loss analysis,thus the loss reduction work has a certain degree of blindness.With the rapid development of active distribution network,the scale of the power network is showing an exponential increase,and the large data of the distribution network is formed.The existing analysis system can not meet the real-time requirements of the active distribution network loss analysis system.In order to solve the above problems,this paper studies the online loss analysis system of active distribution network.The main contents are as follows:By studying the architecture design of the distribution network loss calculation system,the Neo4j-CIM model is proposed to store the active distribution network big data and adopt the parallel architecture programming to solve the problems such as the difficulty of translation of the relational database model and low read-write concurrency,which improve the time scale of power network storage and topology analysis to the millisecond level.Aiming at the lack of guidance of current loss analysis methods for power quality control,this paper proposes a parallel analysis method of active power network loss based on the balanced KNN algorithm.First,the K-means algorithm is used to divide loss scenes of massive power loss historical data.Then,the balanced KNN algorithm is used to balance the loss scenarios and make decisions for the loss scenes.Finally,the power quality factors are weighted using the anti-entropy method.Using the anti-entropy weight method to weight the power quality factors,And using this value to determine the impact of power quality factors on the distribution network loss rate.In view of the shortage of computing resources in traditional computers,this paper builds a distributed storage and computing platform based on virtual private cloud,which provides a highly available and highly concurrent architecture for data migration and storage in power networks.On this basis,the cloud computing technology is used to parallelize the loss analysis algorithms of the grid,and the on-line analysis of the loss of the active distribution network is realized.