Comparison of logistic and neural network models to fit to the egg production curve of White Leghorn hens

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Análise de desempenho da rede neural artificial do tipo multilayer perceptron na era multicore

Artificial neural networks are usually applied to solve complex problems. In problems with more complexity, by increasing the number of layers and neurons, it is possible to achieve greater functional efficiency. Nevertheless, this leads to a greater computational effort. The response time is an important factor in the decision to use neural networks in some systems. Many argue that the computational cost is higher in the training period. However, this phase is held only once. Once the network trained, it is necessary to use the existing computational resources efficiently. In the multicore era, the problem boils down to efficient use of all available processing cores. However, it is necessary to consider the overhead of parallel computing. In this sense, this paper proposes a modular structure that proved to be more suitable for parallel implementations. It is proposed to parallelize the feedforward process of an RNA-type MLP, implemented with OpenMP on a shared memory computer architecture. The research consistes on testing and analizing execution times. Speedup, efficiency and parallel scalability are analyzed. In the proposed approach, by reducing the number of connections between remote neurons, the response time of the network decreases and, consequently, so does the total execution time. The time required for communication and synchronization is directly linked to the number of remote neurons in the network, and so it is necessary to investigate which one is the best distribution of remote connections

A neural network approach for robust nonlinear parameter estimation in presence of unknown-but-bounded errors

Systems based on artificial neural networks have high computational rates due to the use of a massive number of simple processing elements and the high degree of connectivity between these elements. This paper presents a novel approach to solve robust parameter estimation problem for nonlinear model with unknown-but-bounded errors and uncertainties. More specifically, a modified Hopfield network is developed and its internal parameters are computed using the valid-subspace technique. These parameters guarantee the network convergence to the equilibrium points. A solution for the robust estimation problem with unknown-but-bounded error corresponds to an equilibrium point of the network. Simulation results are presented as an illustration of the proposed approach. Copyright (C) 2000 IFAC.

Neural network based estimation of torque in induction motors for real-time applications

Induction motors are largely used in several industry sectors. The selection of an induction motor has still been inaccurate because in most of the cases the load behavior in its shaft is completely unknown. The proposal of this article is to use artificial neural networks for torque estimation with the purpose of best selecting the induction motors rather than conventional methods, which use classical identification techniques and mechanical load modeling. Since proposed approach estimates the torque behavior from the transient to the steady state, one of its main contributions is the potential to also be implemented in control schemes for real-time applications. Simulation results are also presented to validate the proposed approach.

Neural Network-Based Approach for Identification of the Harmonic Content of a Nonlinear Load in a Single-Phase System

In this paper an alternative method based on artificial neural networks is presented to determine harmonic components in the load current of a single-phase electric power system with nonlinear loads, whose parameters can vary so much in reason of the loads characteristic behaviors as because of the human intervention. The first six components in the load current are determined using the information contained in the time-varying waveforms. The effectiveness of this method is verified by using it in a single-phase active power filter with selective compensation of the current drained by an AC controller. The proposed method is compared with the fast Fourier transform.

Uma análise da aplicação do modelo de Rede Neural RePART em Comitês de classificadores

RePART (Reward/Punishment ART) is a neural model that constitutes a variation of the Fuzzy Artmap model. This network was proposed in order to minimize the inherent problems in the Artmap-based model, such as the proliferation of categories and misclassification. RePART makes use of additional mechanisms, such as an instance counting parameter, a reward/punishment process and a variable vigilance parameter. The instance counting parameter, for instance, aims to minimize the misclassification problem, which is a consequence of the sensitivity to the noises, frequently presents in Artmap-based models. On the other hand, the use of the variable vigilance parameter tries to smoouth out the category proliferation problem, which is inherent of Artmap-based models, decreasing the complexity of the net. RePART was originally proposed in order to minimize the aforementioned problems and it was shown to have better performance (higer accuracy and lower complexity) than Artmap-based models. This work proposes an investigation of the performance of the RePART model in classifier ensembles. Different sizes, learning strategies and structures will be used in this investigation. As a result of this investigation, it is aimed to define the main advantages and drawbacks of this model, when used as a component in classifier ensembles. This can provide a broader foundation for the use of RePART in other pattern recognition applications

Classification and characterization of places for mapping of environment using hierarchical neural network and omnivision

Mobile robots need autonomy to fulfill their tasks. Such autonomy is related whith their capacity to explorer and to recognize their navigation environments. In this context, the present work considers techniques for the classification and extraction of features from images, using artificial neural networks. This images are used in the mapping and localization system of LACE (Automation and Evolutive Computing Laboratory) mobile robot. In this direction, the robot uses a sensorial system composed by ultrasound sensors and a catadioptric vision system equipped with a camera and a conical mirror. The mapping system is composed of three modules; two of them will be presented in this paper: the classifier and the characterizer modules. Results of these modules simulations are presented in this paper.

Modeling thermal conductivity, specific heat, and density of milk: A neural network approach

The accurate determination of thermophysical properties of milk is very important for design, simulation, optimization, and control of food processing such as evaporation, heat exchanging, spray drying, and so forth. Generally, polynomial methods are used for prediction of these properties based on empirical correlation to experimental data. Artificial neural networks are better Suited for processing noisy and extensive knowledge indexing. This article proposed the application of neural networks for prediction of specific heat, thermal conductivity, and density of milk with temperature ranged from 2.0 to 71.0degreesC, 72.0 to 92.0% of water content (w/w), and 1.350 to 7.822% of fat content (w/w). Artificial neural networks presented a better prediction capability of specific heat, thermal conductivity, and density of milk than polynomial modeling. It showed a reasonable alternative to empirical modeling for thermophysical properties of foods.

Weather-based prediction of anthracnose severity using artificial neural network models

Data were collected and analysed from seven field sites in Australia, Brazil and Colombia on weather conditions and the severity of anthracnose disease of the tropical pasture legume Stylosanthes scabra caused by Colletotrichum gloeosporioides. Disease severity and weather data were analysed using artificial neural network (ANN) models developed using data from some or all field sites in Australia and/or South America to predict severity at other sites. Three series of models were developed using different weather summaries. of these, ANN models with weather for the day of disease assessment and the previous 24 h period had the highest prediction success, and models trained on data from all sites within one continent correctly predicted disease severity in the other continent on more than 75% of days; the overall prediction error was 21.9% for the Australian and 22.1% for the South American model. of the six cross-continent ANN models trained on pooled data for five sites from two continents to predict severity for the remaining sixth site, the model developed without data from Planaltina in Brazil was the most accurate, with >85% prediction success, and the model without Carimagua in Colombia was the least accurate, with only 54% success. In common with multiple regression models, moisture-related variables such as rain, leaf surface wetness and variables that influence moisture availability such as radiation and wind on the day of disease severity assessment or the day before assessment were the most important weather variables in all ANN models. A set of weights from the ANN models was used to calculate the overall risk of anthracnose for the various sites. Sites with high and low anthracnose risk are present in both continents, and weather conditions at centres of diversity in Brazil and Colombia do not appear to be more conducive than conditions in Australia to serious anthracnose development.

Prediction via neural networks of the residual hydrogen peroxide used in photo-fenton processes for effluent treatment

This communication proposes the use of neural networks in the prediction of residual concentrations of hydrogen peroxide from the treatment of effluents through Advanced Oxidative Processes (AOP's), in particular, the photo-Fenton process. To verify the efficiency of the oxidative process, the Chemical Oxygen Demand (COD) parameter, the values of which may be modified by the presence of oxidizing agents such as residual hydrogen peroxide, is frequently taken in account. The analysis of the H2O2 interference was performed by spectrophotometry at 450 nm wavelength, via the monitoring of the reaction of ammonia with metavanadate. The results of the hydrogen peroxide residual concentration were modeled via a feedforward neural network, with the correlation coefficients between actual and predicted values above 0.96, indicating good prediction capacity.

Optimization of Radial Basis Function neural network employed for prediction of surface roughness in hard turning process using Taguchi's orthogonal arrays

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Aircraft interior failure pattern recognition utilizing text mining and neural networks

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

An analog implementation of radial basis neural networks (RBNN) using BiCMOS technology

This paper describes a analog implementation of radial basis neural networks (RBNN) in BiCMOS technology. The RBNN uses a gaussian function obtained through the characteristic of the bipolar differential pair. The gaussian parameters (gain, center and width) is changed with programmable current source. Results obtained with PSPICE software is showed.

Using artificial intelligence for the control of medium voltage substations

This paper introduces a method for the supervision and control of devices in electric substations using fuzzy logic and artificial neural networks. An automatic knowledge acquisition process is included which allows the on-line processing of operator actions and the extraction of control rules to replace gradually the human operator. Some experimental results obtained by the application of the implemented software in a simulated environment with random signal generators are presented.