Evolutionary algorithms have been used to try to solve distribution network reconfiguration for loss reduction problem with a certain degree of success. But some problems, specially related to a codification that is able to represent and work with a complex multiconstraint and combinatorial problem such as this one, have prevented the use of the full potential of these algorithms to find quality solutions for large systems with minor computational effort. This paper proposes a solution to this problem, with a new codification and using an efficient way for implementing the operator of recombination to guaranty, at all times, the production of new radial topologies. The algorithm is presented and tested in a real distribution system, showing excellent results and computational efficiency.

A method for spatial electric load forecasting using a reduced set of data is presented. The method uses a cellular automata model for the spatiotemporal allocation of new loads in the service zone. The density of electrical load for each of the major consumer classes in each cell is used as the current state, and a series of update rules are established to simulate S-growth behavior and the complementarity among classes. The most important features of this method are good performance, few data and the simplicity of the algorithm, allowing for future scalability. The approach is tested in a real system from a mid-size city showing good performance. Results are presented in future preference maps.

A multi-agent system for spatial electric load forecasting, especially suited to simulating the different social dynamics involved in distribution systems, is presented. This approach improves the spatial forecasting techniques that usually consider the service zone as a static entity to model or simulate the spatial electric load forecasting in a city. This paper aims to determine how the electric load will be distributed among the sub-zones in the city. For this, the service zone is divided into several subzones, each subzone considered as an independent agent identified with a corresponding load level, and their relationships with the neighbor zones are represented through development probabilities. These probabilities are considered as input data for the simulation. Given this setting, different kinds of agents can be developed to simulate the growth pattern of the loads in distribution systems in parallel. The approach is tested with data from a real distribution system in a mid-size city; the results show a low spatial error when compared to real data. Less than 6% of the load growth was identified 0.71 km outside of its correct location on the test system.

This paper presents an assessment of the potential behind the BiGRU-CNN artificial neural network to be used as an electric power theft detection tool. The network is based on different architecture layers of the bidirectional gated recurrent unit and convolutional neural network. The use of such a tool with this classification model can help energy sector companies to make decisions regarding theft detection. The BiGRU-CNN artificial neural network singles out consumer units suspected of fraud for later manual inspections. The proposed artificial neural network was programmed in python, using the keras package. The best detection model was that of the BiGRU-CNN artificial neural network when compared to multilayer perceptron, recurrent neural network, gated recurrent unit, and long short-term memory networks. Several tests were carried out using data of an actual electricity supplier, showing the effectiveness of the proposed approach. The metric values assigned to their classifications were 0.929 for accuracy, 0.885 for precision, 0.801 for recall, 0.841 for F1-Score, and 0.966 for area under the receiver operating characteristic curve.