No abstract

The main purpose of this paper is to achieve a comparative analysis among Autoregressive Integrated Moving Average model, Artificial Neural Networks and Adaptive NeuroFuzzy System techniques for load demand forecasting in distribution substations. The system inputs are three load demand time series, which are composed by data measured at intervals of five minutes each, during seven days, from substations located at Andradina, Ubatuba and Votuporanga. Autoregressive Integrated Moving Average models with suitable results have been analyzed, whereas several input configurations and different architectures have been investigated for Artificial Neural Networks and Adaptive Neuro-Fuzzy System techniques aiming the forecasting of twelve further steps. The results showed the Artificial Neural Network based technique superiority for such forecasting, followed by Autoregressive Integrated MovingAverage model and Adaptive Neuro-Fuzzy approach. The load demand forecasting can minimize costs of energy generation as well as improve the electric power system safety.Index Terms--Autoregressive integrated moving average processes, feedforward neural networks, fuzzy systems, intelligent systems, load forecasting, time series.

The induction motor is considered one of the most important elements in manufacturing processes. The use of strategies based on intelligent systems capable to classify the presence or absence of failures and also to determine its origin for the diagnosis and faults prediction is widely investigated in three phase induction motors. Thus, the aim of this paper is to present a methodology of bearing failures classification based on artificial neural networks, by using voltage and electric currents values in the time domain. Experimental results collected at real industrial process are presented to validate this proposal.

No abstract

No abstract

This paper evaluates the behavior related to the main variables that influence in the quantification of potential points for installation of fault indicators along electric power distribution feeders. Moreover, based on these behavioral characteristics, fuzzy inference systems are also used to estimate the best positions to allocate fault indicators, which take into account the distance in that a particular bus is in relation to more adjacent protection devices, load profile and short-circuit current levels of the downstream system to the respective points. Results with real data highlight the efficiency of the proposed methodology.Index Terms--Artificial intelligence, fault indicator, fuzzy inference system, power system protection.

No abstract

This research presents the development of an alternative pre-processing technique of signals based on the fractal dimension calculation, entropy and signal energy that will be applied to disturbances classification occurring in an electrical power system (EPS). With respect to the power quality disturbances, the voltage sags, voltage swells, oscillatory transients and interruptions were considered for this application. In order to test and validate the proposed technique, a representative database has been obtained through computer simulations of a real EPS using the ATP software. Through windows data and the pre-processing technique proposed, the data were directed to Artificial Neural Network (ANN) architecture to classify the power quality events. The results shown that, combining both the data pre-processing techniques and ANNs, a satisfactory performance of all the proposed methodology can be obtained.