| Popis: |
An inefficient operation of the utility grid can be inferred from low power quality indexes. Under such conditions, reductions in the useful life of equipments and loads, grid instabilities, service interruptions and breakdowns can be expected, which would produce meaningful economic losses for the end-users as well as for the power supply company. In this context, it is useful to have a diagnosis tool able to identify the underlying cause of a given disturbance. This article proposes a one-step extreme learning machine-based method for the diagnosis of voltage dips detected on an electrical distribution grid. The proposed method was evaluated on the basis of synthetic data generated by the simulation of a real power grid. The performance of 10 hyperparameter combinations on the diagnosis of 20 different classes of voltage dips was assessed through two different strategies, i.e.: 1) a one-for-all approach, with a common classifier for all disturbance classes; and 2) a one-against-all approach, with a classifier per disturbance class. It was proven that strategy 2) has a better generalization ability and lower training and testing times. The obtained results suggest the potential of the proposed approach for power quality disturbance diagnosis, and open the challenge of formulating further hypotheses to be assessed. |
