A process to apply the method of images for a charge located in a three-layer medium is presented. The images are found according to the boundary conditions between the layers for the electric field. The characteristics of the electric potential are also considered, thus the number of unknown variables becomesa guide to set the image charges needed to solve the problem. The results are compared with finite element simulations through the use of the software FEMM 4.2, showing good agreement. The found limitations of the process are also noted, mainly in regards to the dependence of the images on the coordinates where the field is to be calculated. The model obtained was applied to different cases, where it was seen that it was not limited to three material media only. Finally, the null potential boundary condition was applied, showing how the method of images could be applied to this type of problems.

This research compares four machine learning techniques: linear regression, support vector regression, random forests, and artificial neural networks, with regard to the determination of mechanical stress in power transformer winding conductors due to three-phase electrical faults. The accuracy compared with finite element results was evaluated for each model. The input data were the transient electrical fault currents of power system equivalents with impedances from low to high values. The output data were the mechanical stress in the conductors located in the middle of the winding. To simplify the design, only one hyperparameter was varied on each machine learning technique. The random forests technique had the most accurate results. The highest errors were found for low-stress values, mainly due to the high difference between maximum and minimum stresses, which made the training of the machine learning models difficult. In the end, an accurate model that could be used in the continuous monitoring of mechanical stress was obtained.

The purpose of this research is the evaluation of artificial neural network models in the prediction of stresses in a 400 MVA power transformer winding conductor caused by the circulation of fault currents. The models were compared considering the training, validation, and test data errors’ behavior. Different combinations of hyperparameters were analyzed based on the variation of architectures, optimizers, and activation functions. The data for the process was created from finite element simulations performed in the FEMM software. The design of the Artificial Neural Network was performed using the Keras framework. As a result, a model with one hidden layer was the best suited architecture for the problem at hand, with the optimizer Adam and the activation function ReLU. The final Artificial Neural Network model predictions were compared with the Finite Element Method results, showing good agreement but with a much shorter solution time.

No abstract

In Ecuador, a country with several active volcanoes and with four eruptions in the last decade in the continental arc, it is very likely that high-voltage transmission lines cross volcanic hazard zones on their routes. Here, we quantify the impact of fresh volcanic ash from the hydromagmatic Cotopaxi-2015 and the magmatic Tungurahua-2016 eruptions on the dielectric characteristics of ANSI 52–3 suspension insulators made of porcelain and glass, under moist conditions. The experiments include two methodologies to measure the performance of the insulators in real-time: the minimum insulator flashover voltage (FOVmin) and the dielectric loss factor angle. Both allow quantifying i) the critical voltage that the insulators can withstand prior to flashover occurrence and, ii) the strong fluctuating behavior that the insulators undergo in an ashy environment. Based on six contamination scenarios, we found that there is a higher chance of flashover if the insulators are completely blanketed (top and bottom) even with a fine ash layer (1 mm), than if they are covered just at the top. Our results further show that the ash of Cotopaxi-2015 eruption has a higher chance of leading to insulator failure because of its higher conductivity (i.e. higher leachate content) than that of Tungurahua-2016. Additionally, we identify two critical voltages prior to electrical flashover on the insulators of 28.25 kV and 17.01 kV for the 230 kV and 138 kV Ecuadorian transmission lines, respectively. Finally, we present a simple impact evaluation for the main Ecuadorian transmission lines based on the outcomes of this research and the official volcanic hazard maps for Cotopaxi and Tungurahua volcanoes.