Carlos A. M. Pinheiro

Forecasting models for prediction in time series

This paper presents the study of three forecasting models—a multilayer perceptron, a support vector machine, and a hierarchical model. The hierarchical model is made up of a self-organizing map and a support vector machine—the latter on top of the former. The models are trained and assessed on a time series of a Brazilian stock market fund. The results from the experiments show that the performance of the hierarchical model is better than that of the support vector machine, and much better than that of the multilayer perceptron.

A neural model in anti-spam systems

The paper proposes the use of the multilayer perceptron model to the problem of detecting ham and spam e-mail patterns. It also proposes an intensive use of data pre-processing and feature selection methods to simplify the task of the multilayer perceptron in classifying ham and spam e-mails. The multilayer perceptron is trained and assessed on patterns extracted from the SpamAssassin Public Corpus. It is required to classify novel types of ham and spam patterns. The results are presented and evaluated in the paper.

Improving the performance of fuzzy rules-based forecasters through application of FCM algorithm

Prediction models based on artificial intelligence techniques have been widely used in Time Series Forecasting in several areas. They are often fuzzy models or neural networks. This paper describes the development of neural and fuzzy models for forecasting time series of practical examples, and shows the comparisons of results between models, including the results of statistical modeling. The use of data clustering algorithms like Fuzzy C-Means is considered in fuzzy models.

A hierarchical hybrid neural model with time integrators in long-term load forecasting

A novel hierarchical hybrid neural model to the problem of long-term load forecasting is proposed in this paper. The neural model is made up of two self-organizing map nets—one on top of the other—and a single-layer perceptron. It has application into domains which require time series analysis. The model is compared to a mutilated architecture of it, and to a multilayer perceptron. The hierarchical, the mutilated, and the multilayer perceptron models are trained and assessed on load data extracted from a North-American electric utility. They are required to predict either once every week or once every month the electric peak-load during the next two years. The results from the experiments show that the performance of HHNM on long-term load forecasts is better than that of the mutilated model, and much better than that of the MLP model.

Time series forecasting through rule-based models obtained via rough sets

Prediction models based on artificial intelligence techniques have been widely used in Time Series Forecasting in several areas. They are often fuzzy models or neural networks. However, the use of rough sets based models have not yet been explored. The aim of this work is to introduce a new approach which uses rough set concepts to obtain rule-based models capable to perform time series forecasting.

A neural model in intrusion detection systems

The paper proposes the use of the multilayer perceptron model to the problem of detecting attack patterns in computer networks. The multilayer perceptron is trained and assessed on patterns extracted from the files of the Third International Knowledge Discovery and Data Mining Tools Competition. It is required to classify novel normal patterns and novel categories of attack patterns. The results are presented and evaluated in the paper.

A hybrid neural model in long-term electrical load forecasting

A novel hierarchical hybrid neural model to the problem of long-term electrical load forecasting is proposed in this paper. The neural model is made up of two self-organizing map nets — one on top of the other —, and a single-layer perceptron. It has application into domains which require time series analysis. The model is compared to a multilayer perceptron. Both the hierarchical and the multilayer perceptron models are endowed with time windows in their input layers. They are trained and assessed on load data extracted from a North-American electric utility. The models are required to predict once every week the electric peak-load and mean-load during the next two years. The results are presented and evaluated in the paper.

Forecasting financial series using clustering methods and support vector regression

This paper proposes a two-stage model for forecasting financial time series. The first stage uses clustering methods in order to segment the time series into its various contexts. The second stage makes use of support vector regressions (SVRs), one for each context, to forecast future values of the series. The series used in the experiments is composed of values of an equity fund of a Brazilian bank. The proposed model is compared to a hierarchical model (HM) presented in the literature. In this series, the HM presented prediction results superior to both a support vector machine (SVM) and a multilayer perceptron (MLP) models. The experiments show that the proposed model is superior to HM, reducing the forecasting error of the HM by 32%. This means that the proposed model is also superior to the SVM and MLP models. An analysis of the construction and use of clusters associated with a series volatility study shows that data obtained from only one type of volatility (low or high) are enough to provide sufficient knowledge to the model so that it is able to forecast future values with good accuracy. Another analysis on the quality of the clusters formed by the model shows that each cluster carries different information about the series. Furthermore, there is always a group of SVRs capable of making adequate forecasts and, for the most part, the SVR used in forecasting is a SVR belonging to this group.

A secondary control based on fuzzy logic to frequency and voltage adjustments in islanded microgrids scenarios

This paper aims to propose a secondary control algorithm based on Fuzzy Logic to adjust frequency and voltage in islanded microgrids. It is well-known, that, when a microgrid changes its operation from grid-connected to islanded mode, the time domain variability of the loads will change the frequency and voltage profiles of the system. In this sense, the operation of the system might become unsafe and out of the acceptable ranges of quality. It is essential that the MGCC be able to restore the frequency and voltage through the secondary control. In this context, a Fuzzy approach is proposed to change the frequency and the voltage reference values in the droop equation of the Voltage Sources Inverters to correct them. The main idea of this paper is to consider these variables uncoupled and then correct each one separately, taking into account that the frequency is global and the voltage is local variables of the system. As results of this work, this paper brings the frequency and voltage profiles of a modified IEEE Feeder Test during a random day, showing the behavior of theses variables with and without the proposed algorithm.

Controle Neural de Sistemas Não Lineares por Resposta em Freqüência

A switching circuit controls at least one semiconductor switch TR1 using drive signals generated from logic circuitry 4 that generates complementary drive signals on an upper and lower drive path. Validation circuitry is provided between the logic circuitry and the semiconductor switch that inhibits any uncontrolled switching of the semiconductor switch TR1 due to noise, such as lightning strike, appearing at the load.