Carlos R. Hall Barbosa

Neural Network Simulation and Evolutionary Synthesis of QCA Circuits

CMOS technology miniaturization limits have promoted research on new alternatives which can keep the technologically advanced level of the last decades. Quantum-dot cellular automata (QCA) is a new technology in the nanometer scale that has been considered as one of these alternatives. QCA have a large potential in the development of circuits with high space density and low heat dissipation and allow the development of faster computers with lower power consumption. Differently from conventional technologies, QCA do not codify information by means of electric current flow, but rather by the configuration of electrical charges in the interior of the cells. The Coulomb interaction between cells is responsible for the flow of information. This paper proposes the use of computational intelligence techniques in the simulation and in the automatic synthesis of QCA circuits. The first results show that these techniques may play an important role in this research area since they are capable of simulating efficiently and fast, synthesizing optimized circuits with a reduced number of cells. Such optimization reduces the possibility of failures and guarantees higher speed

Quantum-Inspired Evolutionary Algorithms and Its Application to Numerical Optimization Problems

This work proposes a new kind of evolutionary algorithm inspired in the principles of quantum computing. This algorithm is an extension of a proposed model for combinatorial optimization problems which uses a binary representation for the chromosome. This extension uses probability distributions for each free variable of the problem, in order to simulate the superposition of solutions, which is intrinsic in the quantum computing methodology. A set of mathematical operations is used as implicit genetic operators over those probability distributions. The efficiency and the applicability of the algorithm are demonstrated through experimental results using the F6 function.

Cultural operators for a quantum-inspired evolutionary algorithm applied to numerical optimization problems

This work presents the application of cultural algorithms operators to a new quantum-inspired evolutionary algorithm with numerical representation. These operators (fission, fusion, generalization and specialization) are used in order to provide better control over the quantum-inspired evolutionary algorithm. We also show that the quantum-inspired evolutionary algorithm with numerical representation behaves in a very similar manner to a pure cultural algorithm and we propose further investigations concerning this aspect.

Neural networks for inflow forecasting using precipitation information

This work presents forecast models for the natural inflow in the Basin of Iguacu River, incorporating rainfall information, based on artificial neural networks. Two types of rainfall data are available: measurements taken from stations distributed along the basin and ten-day rainfall forecasts using the ETA model developed by CPTEC (Brazilian Weather Forecating Center). The neural nework model also employs observed inflows measured by stations along the Iguacu River, as well as historical data of the natural inflows to be predicted. Initially, we applied preprocessing methods on the various series, filling missing data and correcting outliers. This was followed by methods for selecting the most relevant variables for the forecast model. The results obtained demonstrate the potential of using artificial neural networks in this problem, which is highly non-linear and very complex, providing forecasts with good accuracy that can be used in planning the hydroelectrical operation of the Basin.

Improvement of a technique for localization of steel needles in humans using a SQUID magnetometer

A technique was previously developed, based on magnetic field measurements, to localize hypodermic and sewing needles lost in the human body, with the purpose of surgical extraction. The measurements are performed using a SQUID magnetometer, which detects the magnetic field associated with the remanent magnetization of the needle. The technique allowed easy surgical localization of the needles with good precision in all six clinical cases studied so far. The procedure greatly decreases the surgery time for foreign body extraction, and also reduces the generally high odds of failure. This paper presents an improvement of the original algorithm, which is now independent of any constant magnetic field component, thus overcoming the main experimental difficulty usually found, namely that a SQUID system does not measure absolute fields.

Image processing techniques for NDE SQUID systems

Image processing techniques were used to enhance magnetic data generated by SQUID nondestructive evaluation of conducting plates. The process consists of low-pass filtering the acquired data with a FIR filter with order depending on the standoff distance and background noise. In addition a high-pass FIR filter is used to enhance the magnetic field disturbance due to the flaw. The technique was tested on magnetic field data generated through a dc current applied to flawed aluminum test samples, and detected by using an RF-SQUID system. It was observed that when the standoff distance exceeds 5 times the size of the flaw, image inspection is difficult. Applying the enhancement algorithm, magnetic field disturbances could be visualized at distances 10 times the size of the flaw.<<ETX>>

Molecular circuit design

With a more realistic vision of a possible limit for miniaturization of components with conventional CMOS technology, a new technology has surfaced -molecular electronics which, from the bottom-up approach, aims at the construction of electrical devices to implement computation using individual or small collections of molecules, offering an alternative way to build nanoscale circuits. These circuits have the potential to reduce device size and fabrication costs by several orders of magnitude relative to conventional CMOS. Recently, some mechanisms have been considered as a basis to molecular electronic systems design. Two terminal molecular devices that work as diodes have been synthesized, with similar behavior to rectifying and tunnel diodes. In this article, a study on the synthesis of molecular circuits is presented, integrating simulated molecular devices as the molecular Tour-Reed diode based on the evolvable hardware (EHW) paradigm.

Tail-Chopped Lightning Impulses Time Parameters Estimated According to Standard IEC 60060-1:2010

The International Electrotechnical Commission (IEC) issued in 2010 a revision of the family of standards IEC 60060, describing, among other requirements, how to calculate the various amplitude and time parameters used to evaluate high-voltage impulse waveforms. However, the standards are not clear in some aspects, especially when describing tail-chopped lightning impulses. This paper aims at describing in detail how to properly estimate the time parameters of such impulses.

Modelagem da sensibilidade de amostras GMI por redes neurais

Over the past few years, several studies have been developed in order to quantitatively model the GMI effect (Giant Magnetoimpedance). However, these models adopt simplifications that significantly affect its theoretical-experimental performance and its generalization capability, and models that incorporate parameters that generate asymmetry - AGMI (asymmetric GMI) - such as the DC level of the excitation current of the GMI samples are still rare. This work aims to develop a new model, sufficiently general, which also incorporates the asymmetry induced by the DC level of the excitation current, capable of guiding the experimental procedures of characterization of the GMI samples. Thus, this paper proposes, presents and discusses the use of a computational model based on feedforward Multilayer Perceptron Neural Networks to model the impedance magnitude sensitivity and impedance phase sensitivity, of the GMI effect, as functions of the magnetic field, for Co70Fe5Si15B10 ferromagnetic amorphous alloys. The proposed model allows obtaining these sensitivities based on some of the main parameters that affect it: length of the samples, DC level and frequency of the excitation current and the external magnetic field.

Decision Support Methods

This section presents a summary of the main concepts on which evolutionary algorithms are based. First, the operating principle of Genetic Algorithms (GAs) is explained and their main parts and their evolution parameters described. Next, a description of Cultural Algorithms (CAs) is presented and its main components are pointed out.