Pedro Pereira Barbeiro

Abstract

Abstract
The increasing connection of new assets in LV networks will surely require a better monitoring of these systems in a real-time manner. In order to meet this purpose, a Distribution State Estimator (DSE) module clearly appears as the most cost-effective and possibly the only reliable option available. In this sense, in the scope of the evolvDSO project, a DSE tool exploiting the concept of ELM-AE was developed and tested in two distinct real LV distribution networks. In this paper the main results achieved with the proposed DSE tool are presented and discussed.

Abstract
The LV SCADA project aimed at the development of advanced technical, commercial and regulatory solutions to contribute for an effective smart grid implementation. One of the biggest challenges of the project was related with the lack of characterization that usually exists in LV networks, together with the almost non-existing observability. In order to overcome these issues, a LV management system integrating a state estimation tool based on artificial intelligence techniques was developed. The tool is currently installed in one pilot demonstration site that aggregates 2 MV/LV substations. In this paper the performance of tool in real environment is evaluated and the results gathered from the pilot site are analyzed.

Abstract

Abstract
In this paper a Distribution State Estimator (DSE) tool suitable for real-time monitoring in poorly characterized low voltage networks is presented. An Autoencoder (AE) properly trained with Extreme Learning Machine (ELM) technique is the "brain" of the DSE. The estimation of system state variables, i.e., voltage magnitudes and phase angles is performed with an Evolutionary Particle Swarm Optimization (EPSO) algorithm that makes use of the already trained AE. By taking advantage of historical data and a very limited number of quasi real-time measurements, the presented approach turns possible monitoring networks where information of topology and parameters is not available. Results show improvements in terms of estimation accuracy and time performance when compared to other similar DSE tools that make use of the traditional back-propagation based algorithms for training execution.