A Data and Model Driven Method for Low-Voltage Distribution Area Topology Identification

Topology identification of low-voltage distribution areas is the important basis for intelligent low-voltage distribution network management. At present, topology relationship identification for distribution areas faces challenges, such as confusion, loss and inaccuracy of electricity user information. Traditional topology identification methods for low-voltage distribution areas has a series of problems, such as low accuracy and low efficiency. Therefore, combining the operating mechanism and measured data of distribution areas, this paper presented a data and model driven topology identification method for low-voltage distribution areas. Firstly, hierarchical clustering is utilized to cluster the voltage sequences of observer meters and electricity users respectively. Then, Pearson correlation coefficient is utilized to measure the correlation among voltage series of distribution area transformers, observer meters and electricity users, in order to realize the phase identification of distribution areas. Finally, based on the phase identification results of distribution areas, the mathematical programming model of the line identification of distribution areas is established through the electricity conservation mechanism, and the optimal solution of the model is obtained by the CPLEX optimization solver to realize the line identification of distribution areas. The real-life experiment results show that the proposed method can accurately, quickly and automatically realize the topology identification of distribution areas, which has high application value and potential in practical applications.