Aging, degradation, or damage to internal insulation materials often contribute to transformer failures. Furthermore, combustible gases can be produced when these insulation materials experience thermal or electrical stresses. This paper presents an artificial neural network for pattern recognition (PRN) to classify the operating conditions of power transformers (normal, thermal faults, and electrical faults) depending on the combustible gases present in them. Two network configurations were presented, one with five and the other with ten neurons in the hidden layer. The main advantage of applying this model through artificial neural networks is its ability to capture the nonlinear characteristics of the samples under study, thus avoiding the need for iterative procedures. The effectiveness and applicability of the proposed methodology were evaluated on 815 real data samples. Based on the results, the PRN performed well in both training and validation (for samples that were not part of the training), with a mean squared error (MSE) close to expected (0.001). The network was able to classify the samples with a 98% accuracy rate of the 815 samples presented and with 100% accuracy in validation, showing that the methodology developed is capable of acting as a tool for diagnosing the operability of power transformers.

This study aimed to evaluate the Artificial Neural Network (ANN) to establish a classification and analysis of degraded soils and its recovery in response to lime and gypsum application. The analyzed degraded soil was classified as Oxisol, and the physical attributes considered were: soil density, soil porosity (macroporosity and microporosity) and soil penetration resistance. The ANN used in this study is the backpropagation composed of two layers, the middle layer and the output layer, with supervised training. The network has four inputs, that are the physical attributes of the soil, in the middle layer the network contains ten neurons and the output layer only one neuron, which has the function of informing if the soil was recovered (R), partially recovered (PR) or not recovered (NR). The analyzed data come from the year 2012, concerning the depths 0.0 - 0.1 m, 0.1 - 0.2 m and 0.2 - 0.4 m. Considering the performance of ANN, it was verified that the network obtained an adequate training to classify the degraded soils, showing low mean square error of analyzed data. Therefore, ANN is considered an interesting alternative and a powerful automatic tool to classify degraded soils during recovery process.

This paper presents an alternative methodology to obtain the precise amount of shunt reactive power compensation in order to simultaneously reduce the total reactive power losses, improve the voltage profile and increase the loading margin of power systems. The amount of shunt reactive power compensation to be allocated is determined based on the curve of the total reactive power losses versus the voltage magnitude of a chosen bus. The best places for shunt reactive compensation are defined by the load bus participation factors of the critical mode provided by the static modal analysis. Simulation results employing shunt capacitors obtained with the new approach for the IEEE test systems (14, 57 and 300 buses) show that the procedure leads to a reduction in total reactive and active power losses and simultaneously improves the voltage profile and loading margin.