Enhancing Visual Evoked Potentials Detection with Use of Computational Intelligence Tools
International Journal of Biomedical Data Mining (Print), 2011
The analysis of evoked potentials (EPs) in the electroencephalogram (EEG) is usually inspected vi... more The analysis of evoked potentials (EPs) in the electroencephalogram (EEG) is usually inspected visually and demands subjective interpretation of the results. This paper aims at combining an statistical criterion based on the magnitude square multiple coherence (MSMC) estimate with computational intelligence methods in order to estimate the EPs detection rate (DR) using only portions of the frequency spectrum. Thus, networks were used to predict the DR in EEG signals of 15 normal subjects during stroboscopic stimulation. The algorithms were designed to receive the spectral information of two, four or six EEG derivations as the input and DR as the output. Our best result shows that the artificial neural networks can estimate DR with correlation coefficient of 0.97 compared with MSMC, even when a reduced amount of spectral information from the data is available.
Cyberbullying Classification Using Extreme Learning Machine Applied to Portuguese Language
Communications in computer and information science, 2017
Increasing accessibility to virtual environments has resulted in higher incidences of Cyberbullyi... more Increasing accessibility to virtual environments has resulted in higher incidences of Cyberbullying attacks and the consequences of these attacks can affect the victim’s life for a long time or even permanently. For this reason, it is extremely important to develop tools to inhibit such practices. In this paper, a solution to the Cyberbullying classification problem using machine learning (Extreme Learning Machine) is proposed. The application of the proposed method to a database of sentences in Portuguese language shows promising results where compared to a standard method available in the literature.
State-of-the-art technologies in neural speech decoding utilize data collected from microwires or... more State-of-the-art technologies in neural speech decoding utilize data collected from microwires or microarrays implanted directly into the cerebral cortex. Yet as a tool accessible only to individuals with implanted electrodes, speech decoding from devices of this nature is severely limited in its implementation, and cannot be considered a viable solution for widespread application. Speech decoding from non-invasive EEG signals can achieve relatively high accuracy (70-80%), but only from very small classification tasks, with more complex tasks typically yielding a limited (20-50%) classification accuracy. We propose a novel combination of technologies in which transcranial magnetic stimulation (TMS) is first applied to augment the neural signals of interest, producing a greater signal-to-noise ratio in the EEG data. Next, delay differential analysis (DDA) – a cutting-edge computational method based on nonlinear dynamics – is implemented to capture the widest range of information avai...
The extraction of electrophysiological features that reliably forecast the occurrence of seizures... more The extraction of electrophysiological features that reliably forecast the occurrence of seizures is one of the most challenging goals in epilepsy research. Among possible approaches to tackle this problem is the use of active probing paradigms in which responses to stimuli are used to detect underlying system changes leading up to seizures. This work evaluates the theoretical and mechanistic underpinnings of this strategy using two coupled populations of the well-studied Wendling neural mass model. Different model settings are evaluated, shifting parameters (excitability, slow inhibition, or inter-population coupling gains) from normal towards ictal states while probing stimuli are applied every 2 seconds to the input of either one or both populations. The correlation between the extracted features and the ictogenic parameter shifting indicates if the impending transition to the ictal state may be identified in advance. Results show that not only can the response to the probing sti...
Anais do 14º Simpósio Brasileiro de Automação Inteligente, 2019
This paper presents a strategy to calculate a suboptimal solution to the finite-time linear-quadr... more This paper presents a strategy to calculate a suboptimal solution to the finite-time linear-quadratic control problem of Markov jump linear systems. When the mode is fixed, the system has parameter-varying matrices that take values in a polutopic set. An upper bound for the cost function is obtained and, as a consequence, the feedback gains are calculated by solving an optimization problem written in terms of linear matrix inequalities. A numerical example is given to illustrate the effectiveness of the proposed method. Resumo: Este trabalho apresenta uma estratégia para calcular uma solução subótima para o problema de controle linear quadrático de tempo finito de sistemas lineares com saltos Markovianos. Quando os modos são fixados, o sistema tem matrizes com parâmetros variantes, os quais tomam valores em um conjunto politópico. Um limitante superior para a função custoé obtido, como consequência, os ganhos de realimentação são calculados resolvendo um problema de otimização escrito em termos de desigualdades matriciais lineares. Um exemplo numéricoé dado para ilustrar a eficácia do método proposto.
The present study proposes a multi-objective framework for structure selection of nonlinear syste... more The present study proposes a multi-objective framework for structure selection of nonlinear systems which are represented by polynomial NARX models. This framework integrates the key components of Multi-Criteria Decision Making (MCDM) which include preference handling, Multi-Objective Evolutionary Algorithms (MOEAs) and a posteriori selection. To this end, three well-known MOEAs such as NSGA-II, SPEA-II and MOEA/D are thoroughly investigated to determine if there exists any significant difference in their search performance. The sensitivity of all these MOEAs to various qualitative and quantitative parameters, such as the choice of recombination mechanism, crossover and mutation probabilities, is also studied. These issues are critically analyzed considering seven discretetime and a continuous-time benchmark nonlinear system as well as a practical case study of non-linear wave-force modeling. The results of this investigation demonstrate that MOEAs can be tailored to determine the correct structure of nonlinear systems. Further, it has been established through frequency domain analysis that it is possible to identify multiple valid discrete-time models for continuous-time systems. A rigorous statistical analysis of MOEAs via performance sweet spots in the parameter space convincingly demonstrates that these algorithms are robust over a wide range of control parameters.
Downhole pressure is an important process variable in the operation of gas-lifted oil wells. The ... more Downhole pressure is an important process variable in the operation of gas-lifted oil wells. The device installed in order to measure this variable is often called a Permanent Downhole Gauge (PDG). Replacing a faulty PDG is often not economically viable and to have an alternative estimate of the downhole pressure is an important goal. Using data from operating PDGs, this paper describes a number of issues dealt with in the development of soft sensors for several deepwater gas-lifted oil wells. Some of the tested models include nonlinear polynomials, neural networks, committee machines, unscented Kalman filters and filter banks. The variety of model classes used in addition to the diversity of oil wells considered brings to light some of the key-problems that have to be faced and reveal the strengths and weaknesses of each alternative solution. A major constraint throughout the work was the use of historical data, hence no specific tests were performed at any time. The aim of this work is to discuss the procedures, pros and cons of the tested solutions and to point to possible future directions of research.
Proceedings of the 4th Wseas International Conference on Applied Mathematics and Computer Science, Apr 25, 2005
In this paper a new approach to model climatic variations in the Plio-Pleistocene is presented. I... more In this paper a new approach to model climatic variations in the Plio-Pleistocene is presented. In a recent reference, Rial in [1] introduced the working hypothesis that frequency modulation (FM) of the orbital eccentricity forcing may be an important source of the nonlinearities observed in the δ 18 O time series from deep-sea sediment cores. Two models are proposed based on the ANFIS (Adaptive Neuro Fuzzy Inference System) structure. The first model uses only past values of the time series under investigation. The second model uses information on the orbital eccentricity forcing and an artificially generated FM which is an extension of the FM signal proposed by Rial. The two models are compared in the light of long term predictions.
Caracterização De Agrupamentos De Termos Na Seleção De Estrutura De Modelos Polinomiais Narx: Uma Abordagem Por Meio Das Características Estáticas
Resumo. A modelagem de sistemas está presente em praticamente todas as áreas da ciência. Todavia,... more Resumo. A modelagem de sistemas está presente em praticamente todas as áreas da ciência. Todavia, nem sempre é uma tarefa fácil modelar sistemas por meio de leis físicas que descrevem a dinâmica do processo. Nesse contexto, a identificação de sistemas surge como uma alternativa bastante viável. Pois, por meio de medições de dados de entrada e saída encontram-se modelos que descrevem as características dinâmicas desejadas do processo sob investigação. Apesar da fácil parametrização desses modelos, o estudo da determinação de uma estrutura de modelo que melhor se ajuste aos dados gerados pelo sistema ainda está em aberto. Esse problema se agrava em aplicações que necessitam de modelos não-lineares, tais como modelos NARX. Por esta razão, a escolha da estrutura do modelo é fundamental no sentido de evitar problemas de so-breparametrização. Como solução, este trabalho propõe uma nova metodologia para seleção de estruturas de modelos NARX polinomiais. Utilizando o conceito de agrupamento...
Diffeomorphical equivalence vs topological equivalence among Sprott systems
Chaos, Aug 1, 2021
In 1994, Sprott [Phys. Rev. E 50, 647-650 (1994)] proposed a set of 19 different simple dynamical... more In 1994, Sprott [Phys. Rev. E 50, 647-650 (1994)] proposed a set of 19 different simple dynamical systems producing chaotic attractors. Among them, 14 systems have a single nonlinear term. To the best of our knowledge, their diffeomorphical equivalence and the topological equivalence of their chaotic attractors were never systematically investigated. This is the aim of this paper. We here propose to check their diffeomorphical equivalence through the jerk functions, which are obtained when the system is rewritten in terms of one of the variables and its first two derivatives (two systems are thus diffeomorphically equivalent when they have exactly the same jerk function, that is, the same functional form and the same coefficients). The chaotic attractors produced by these systems-for parameter values close to the ones initially proposed by Sprott-are characterized by a branched manifold. Systems B and C produce chaotic attractors, which are observed in the Lorenz system and are also briefly discussed. Those systems are classified according to their diffeomorphical and topological equivalence.
A method to estimate the (positive) largest Lyapunov exponent (LLE) from data using interval exte... more A method to estimate the (positive) largest Lyapunov exponent (LLE) from data using interval extensions is proposed. The method differs from the ones available in the literature in its simplicity since it is only based on three rather simple steps. Firstly, a polynomial NARMAX is used to identify a model from the data under investigation. Secondly, interval extensions, which can be easily extracted from the identified model, are used to calculate the lower bound error. Finally, a simple linear fit to the logarithm of lower bound error is obtained and then the LLE is retrieved from it as the third step. To illustrate the proposed method, the LLE is calculated for the following well-known benchmarks: sine map, Rössler Equations, and Mackey-Glass Equations from identified models given in the literature and also from two identified NARMAX models: a chaotic jerk circuit and the tent map. In the latter, a Gaussian noise has been added to show the robustness of the proposed method.
Displacement in the parameter space versus spurious solution of discretization with large time step
Journal of Physics A: Mathematical and General, 2004
In order to investigate a possible correspondence between differential and difference equations, ... more In order to investigate a possible correspondence between differential and difference equations, it is important to possess discretization of ordinary differential equations. It is well known that when differential equations are discretized, the solution thus obtained ...
IEEE Transactions on Circuits and Systems I: Regular Papers
The present study proposes a simple grey-box identification approach to model a real DC-DC buck c... more The present study proposes a simple grey-box identification approach to model a real DC-DC buck converter operating in continuous conduction mode. The problem associated with the information void in the observed dynamical data, which is often obtained over a relatively narrow input range, is alleviated by exploiting the known static behavior of buck converter as a priori knowledge. A simple method is developed based on the concept of term clusters to determine the static response of the candidate models. The error in the static behavior is then directly embedded into the multi-objective framework for structure selection. In essence, the proposed approach casts greybox identification problem into a multi-objective framework to balance bias-variance dilemma of model building while explicitly integrating a priori knowledge into the structure selection process. The results of the investigation, considering the case of practical buck converter, demonstrate that it is possible to identify parsimonious models which can capture both the dynamic and static behavior of the system over a wide input range.
Abstract Here we propose a solution for artifact removal from electroencephalogram (EEG) signals... more Abstract Here we propose a solution for artifact removal from electroencephalogram (EEG) signals. To this end two techniques were employed: the Local Singular Spectrum Analysis1 (Local SSA), and a procedure for automatic artifact removal that uses as a criterion the standard deviation at each signal derivation. The combined uses of the Local SSA, which removes the influence of high amplitude electrooculogram (EOG) signals, together with the method based on standard deviation, which identifies and eliminates instrumental artifacts, generated during signal acquisition, such as baseline and muscular movement, provides an interesting tool for the practical problem of artifact removal in EEG signals. The proposed solution enables treating EEG signals in such a way as to avoid the presence of too many artifacts without the need to exclude excessive samples and within a reasonable computational cost. The procedure was here applied to real data.
Consensus of multi-agent systems with nonuniform non-differentiable time-varying delays
In this paper the consensus problem for continuous time multi-agent systems in the presence of ti... more In this paper the consensus problem for continuous time multi-agent systems in the presence of time-delay is addressed. A novel sufficient condition for the case of nonuniform non-differentiable time-varying delays with minimum value greater than zero and a method to compute an estimate of the convergence rate are given. Simulation examples are given to show the performance of the proposed method.
29th IEEE Conference on Decision and Control, 1990
... As the power plant operates in a boiler following mode, all changes in the BID signal have to... more ... As the power plant operates in a boiler following mode, all changes in the BID signal have to be iniciated in the turbo generator steam ... Levy's identification method 111 was then used to determine the fuel control system transfer function coefficients for each operating point. ...
Dynamic reconfiguration of control and estimation algorithms for induction motor drives
Proceedings of the IEEE International Symposium on Industrial Electronics ISIE-02, 2002
In this paper, the authors present the concept of dynamic reconfiguration (DR) of algorithms appl... more In this paper, the authors present the concept of dynamic reconfiguration (DR) of algorithms applied to induction motor drives. A classification according to several criteria is proposed as well as two practical examples. The first example concerns a partial dynamic reconfiguration by combination of rotor flux estimation algorithms for a field oriented control (FOC) strategy. The second example presents a
Uploads
Papers by Eduardo Mendes