Phase topology identification in low-voltage distribution networks: A Bayesian approach

Knowledge of customer phase connection in low-voltage distribution networks is important for Distribution Operators (DSOs). This paper presents a novel data-driven phase identification method based on Bayesian, which uses load consumption profiles as inputs. This method uses a non-linear function to establish the of a customer being connected to a given phase, based on variations in the customer’s consumption those in the phase feeders. Owing to the Bayesian inference, the proposed method can provide up-to-date about the phase connection of each customer. To improve the detection of those customers that are difficult to identify, after obtaining the up-to-date certainty for all users, the consumption of those who an up-to-date certainty above a certain percentile compared with the rest of the substation (those that are likely to be correctly classified) is subtracted from the phase in which they are classified. The performance the proposed method was evaluated using a real (non-synthetic) low-voltage distribution network. Favourable (with accuracies higher than 97 %) were obtained in almost all cases, regardless of the percentage of Meter penetration and the size of the substation. A comparison with other state-of-the-art methods showed the proposed method outperforms (or equals) them. The proposed method does not necessarily require labelled data; however, it can handle them even if they contain errors. Having previous information (partial or complete) increases the performance of phase identification, making it possible to correct erroneous labelling.