Use of artificial neural networks as explicit finite difference operators

Results of a numerical exercise, substituting a numerical operator by an artificial neural network (ANN) are presented in this paper. The numerical operator used is the explicit form of the finite difference (FD) scheme. The FD scheme was used to discretize the one-dimensional transport equation, which included both the advection and dispersion terms. Inputs to the ANN are the FD representation of the transport equation, and the concentration was designated as the output. Concentration values used for training the ANN were obtained from analytical solutions. The numerical operator was reconstructed from a back calculation of the weights of the ANN. Linear transfer functions were used for this purpose. The ANN was able to accurately recover the velocity used in the training data, but not the dispersion coefficient. This capability was improved when numerical dispersion was taken into account; however, it is limited to the condition: C/P<0.5 , where C is the Courant number and P , the Peclet number (i.e., the restriction imposed by the Neumann stability condition).

Hybrid neural network—finite element river flow model

Results obtained from a hybrid neural network—finite element model are reported in this paper. The hybrid model incorporates artificial neural network (ANN) nodes into a numerical scheme, which solves the two-dimensional shallow water equations using finite elements (FE). First, numerical computations are carried out on the entire numerical model, using a larger mesh. The results from this computation are then used to train several preselected ANN nodes. The ANN nodes model the response for a part of the entire numerical model by transferring the system reaction to the location where both models are connected in real time. This allows a smaller mesh to be used in the hybrid ANN-FE model, resulting in savings in computation time. The hybrid model was developed for a river application, using the computational nodes located at the open boundaries to be the ANN nodes for the ANN-FE hybrid model. Real-time coupling between the ANN and FE models was achieved, and a reduction is CPU time of more than 25% was obtained.

Comparison between response surface models and artificial neural networks in hydrologic forecasting

Developing an efficient and accurate hydrologic forecasting model is crucial to managing water resources and flooding issues. In this study, response surface (RS) models including multiple linear regression (MLR), quadratic response surface (QRS), and nonlinear response surface (NRS) were applied to daily runoff (e.g., discharge and water level) prediction. Two catchments, one in southeast China and the other in western Canada, were used to demonstrate the applicability of the proposed models. Their performances were compared with artificial neural network (ANN) models, trained with the learning algorithms of the gradient descent with adaptive learning rate (ANN-GDA) and Levenberg-Marquardt (ANN-LM). The performances of both RS and ANN in relation to the lags used in the input data, the length of the training samples, long-term (monthly and yearly) predictions, and peak value predictions were also analyzed. The results indicate that the QRS and NRS were able to obtain equally good performance in runoff prediction, as compared with ANN-GDA and ANN-LM, but require lower computational efforts. The RS models bring practical benefits in their application to hydrologic forecasting, particularly in the cases of short-term flood forecasting (e.g., hourly) due to fast training capability, and could be considered as an alternative to ANN

Optimal design of traffic signal controller using neural networks and fuzzy logic systems

This paper aims at optimally adjusting a set of green times for traffic lights in a single intersection with the purpose of minimizing travel delay time and traffic congestion. Neural network (NN) and fuzzy logic system (FLS) are two methods applied to develop intelligent traffic timing controller. For this purpose, an intersection is considered and simulated as an intelligent agent that learns how to set green times in each cycle based on the traffic information. The training approach and data for both these learning methods are similar. Both methods use genetic algorithm to tune their parameters during learning. Finally, The performance of the two intelligent learning methods is compared with the performance of simple fixed-time method. Simulation results indicate that both intelligent methods significantly reduce the total delay in the network compared to the fixed-time method.

Intelligent cuckoo search optimized traffic signal controllers for multi-intersection network

Traffic congestion in urban roads is one of the biggest challenges of 21 century. Despite a myriad of research work in the last two decades, optimization of traffic signals in network level is still an open research problem. This paper for the first time employs advanced cuckoo search optimization algorithm for optimally tuning parameters of intelligent controllers. Neural Network (NN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) are two intelligent controllers implemented in this study. For the sake of comparison, we also implement Q-learning and fixed-time controllers as benchmarks. Comprehensive simulation scenarios are designed and executed for a traffic network composed of nine four-way intersections. Obtained results for a few scenarios demonstrate the optimality of trained intelligent controllers using the cuckoo search method. The average performance of NN, ANFIS, and Q-learning controllers against the fixed-time controller are 44%, 39%, and 35%, respectively.

Prediction interval-based control of nonlinear systems using neural networks

Prediction interval (PI) is a promising tool for quantifying uncertainties associated with point predictions. Despite its informativeness, the design and deployment of PI-based controller for complex systems is very rare. As a pioneering work, this paper proposes a framework for design and implementation of PI-based controller (PIC) for nonlinear systems. Neural network (NN)-based inverse model within internal model control structure is used to develop the PIC. Firstly, a PI-based model is developed to construct PIs for the system output. This model is then used as an online estimator for PIs. The PIs from this model are fed to the NN inverse model along with other traditional inputs to generate the control signal. The performance of the proposed PIC is examined for two case studies. This includes a nonlinear batch polymerization reactor and a numerical nonlinear plant. Simulation results demonstrated that the proposed PIC tracking performance is better than the traditional NN-based controller.

Design of distributed adaptive neural traffic signal timing controller by cuckoo search optimization

This paper focuses on designing an adaptive controller for controlling traffic signal timing. Urban traffic is an inevitable part in modern cities and traffic signal controllers are effective tools to control it. In this regard, this paper proposes a distributed neural network (NN) controller for traffic signal timing. This controller applies cuckoo search (CS) optimization methods to find the optimal parameters in design of an adaptive traffic signal timing control system. The evaluation of the performance of the designed controller is done in a multi-intersection traffic network. The developed controller shows a promising improvement in reducing travel delay time compared to traditional fixed-time control systems.

Aprendizado relacional por um modelo neural

As técnicas que formam o campo da Descoberta de Conhecimento em Bases de Dados (DCBD) surgiram devido à necessidade de se tratar grandes volumes de dados. O processo completo de DCBD envolve um elevado grau de subjetividade e de trabalho não totalmente automatizado. Podemos dizer que a fase mais automatizada é a de Mineração de Dados (MD). Uma importante técnica para extração de conhecimentosa partir de dados é a Programação Lógica Indutiva (PLI), que se aplica a tarefas de classificação, induzindo conhecimento na forma da lógica de primeira ordem. A PLI tem demonstrado as vantagens de seu aparato de aprendizado em relação a outras abordagens, como por exemplo, aquelas baseadas em aprendizado proposicional Os seus algorítmos de aprendizado apresentam alta expressividade, porém sofrem com a grande complexidade de seus processos, principalmente o teste de corbertura das variáveis. Por outro lado, as Redes Neurais Artificiais (RNs) introduzem um ótimo desempenho devido à sua natureza paralela. às RNs é que geralmente são "caixas pretas", o que torna difícil a obtenção de um interpretação razoável da estrutura geral da rede na forma de construções lógicas de fácil compreensão Várias abordagens híbridas simbólico-conexionistas (por exemplo, o MNC MAC 890 , KBANN SHA 94 , TOW 94 e o sistema INSS OSO 98 têm sido apresentadas para lidar com este problema, permitindo o aprendizado de conhecimento simbólico através d euma RN. Entretanto, estas abordagens ainda lidam com representações atributo-valor. Neste trabalho é apresentado um modelo que combina a expressividade obtida pela PLI com o desempenho de uma rede neural: A FOLONET (First Order Neural Network).

ONNIS-GI: uma rede neural oscilatória para segmentação de imagens implementada em arquitetura maciçamente paralela

A presente tese apresenta a concepção de uma rede neural oscilatória e sua realização em arquitetura maciçamente paralela, a qual é adequada à implementação de chips de visão digitais para segmentação de imagens. A rede proposta, em sua versão final, foi denominada ONNIS-GI (Oscillatory Neural Network for Image Segmentation with Global Inhibition) e foi inspirada em uma rede denominada LEGION (Locally Excitatory Globally Inhibitory Oscillator Network), também de concepção recente. Inicialmente, é apresentada uma introdução aos procedimentos de segmentação de imagens, cujo objetivo é o de situar e enfatizar a importância do tema abordado dentro de um contexto abrangente, o qual inclui aplicações de visão artificial em geral. Outro aspecto abordado diz respeito à utilização de redes neurais artificiais em segmentação de imagens, enfatizando as denominadas redes neurais oscilatórias, as quais têm apresentado resultados estimulantes nesta área. A implementação de chips de visão, integrando sensores de imagens e redes maciçamente paralelas de processadores, é também abordada no texto, ressaltando o objetivo prático da nova rede neural proposta. No estudo da rede LEGION, são apresentados resultados de aplicações originais desenvolvidas em segmentação de imagens, nos quais é verificada sua propriedade de separação temporal dos segmentos. A versão contínua da rede, um arranjo paralelo de neurônios baseados em equações diferenciais, apresenta elevada complexidade computacional para implementação em hardware digital e muitos parâmetros, com procedimento de ajuste pouco prático. Por outro lado, sua arquitetura maciçamente paralela apresenta-se particularmente adequada à implementação de chips de visão analógicos com capacidade de segmentação de imagens. Com base nos bons resultados obtidos nas aplicações desenvolvidas, é proposta uma nova rede neural, em duas versões, ONNIS e ONNIS-GI, as quais suplantam a rede LEGION em diversos aspectos relativos à implementação prática. A estrutura dos elementos de processamento das duas versões da rede, sua implementação em arquitetura maciçamente paralela e resultados de simulações e implementações em FPGA são apresentados, demonstrando a viabilidade da proposta. Como resultado final, conclui-se que a rede ONNIS-GI apresenta maior apelo de ordem prática, sendo uma abordagem inovadora e promissora na solução de problemas de segmentação de imagens, possuindo capacidade para separar temporalmente os segmentos encontrados e facilitando a posterior identificação dos mesmos. Sob o ponto de vista prático, a nova rede pode ser utilizada para implementar chips de visão digitais com arquitetura maciçamente paralela, explorando a velocidade de tais topologias e apresentando também flexibilidade para implementação de procedimentos de segmentação de imagens mais sofisticados.

Application of multivariable control using artificial neural networks in a debutanizer distillation column

LOPES, Jose Soares Batista et al. Application of multivariable control using artificial neural networks in a debutanizer distillation column.In: INTERNATIONAL CONGRESS OF MECHANICAL ENGINEERING - COBEM, 19, 5-9 nov. 2007, Brasilia. Anais... Brasilia, 2007

Controle preditivo neural aplicado à processos petroquímicos

A pesquisa tem como objetivo desenvolver uma estrutura de controle preditivo neural, com o intuito de controlar um processo de pH, caracterizado por ser um sistema SISO (Single Input - Single Output). O controle de pH é um processo de grande importância na indústria petroquímica, onde se deseja manter constante o nível de acidez de um produto ou neutralizar o afluente de uma planta de tratamento de fluidos. O processo de controle de pH exige robustez do sistema de controle, pois este processo pode ter ganho estático e dinâmica nãolineares. O controlador preditivo neural envolve duas outras teorias para o seu desenvolvimento, a primeira referente ao controle preditivo e a outra a redes neurais artificiais (RNA s). Este controlador pode ser dividido em dois blocos, um responsável pela identificação e outro pelo o cálculo do sinal de controle. Para realizar a identificação neural é utilizada uma RNA com arquitetura feedforward multicamadas com aprendizagem baseada na metodologia da Propagação Retroativa do Erro (Error Back Propagation). A partir de dados de entrada e saída da planta é iniciado o treinamento offline da rede. Dessa forma, os pesos sinápticos são ajustados e a rede está apta para representar o sistema com a máxima precisão possível. O modelo neural gerado é usado para predizer as saídas futuras do sistema, com isso o otimizador calcula uma série de ações de controle, através da minimização de uma função objetivo quadrática, fazendo com que a saída do processo siga um sinal de referência desejado. Foram desenvolvidos dois aplicativos, ambos na plataforma Builder C++, o primeiro realiza a identificação, via redes neurais e o segundo é responsável pelo controle do processo. As ferramentas aqui implementadas e aplicadas são genéricas, ambas permitem a aplicação da estrutura de controle a qualquer novo processo

A Learning Function for Parameter Reduction in Spiking Neural Networks with Radial Basis Function

Spiking neural networks - networks that encode information in the timing of spikes - are arising as a new approach in the artificial neural networks paradigm, emergent from cognitive science. One of these new models is the pulsed neural network with radial basis function, a network able to store information in the axonal propagation delay of neurons. Learning algorithms have been proposed to this model looking for mapping input pulses into output pulses. Recently, a new method was proposed to encode constant data into a temporal sequence of spikes, stimulating deeper studies in order to establish abilities and frontiers of this new approach. However, a well known problem of this kind of network is the high number of free parameters - more that 15 - to be properly configured or tuned in order to allow network convergence. This work presents for the first time a new learning function for this network training that allow the automatic configuration of one of the key network parameters: the synaptic weight decreasing factor.

Otimização de superfícies seletivas de frequência com elementos pré-fractais utilizando rede neural MLP e algoritmos de busca populacional

This thesis describes design methodologies for frequency selective surfaces (FSSs) composed of periodic arrays of pre-fractals metallic patches on single-layer dielectrics (FR4, RT/duroid). Shapes presented by Sierpinski island and T fractal geometries are exploited to the simple design of efficient band-stop spatial filters with applications in the range of microwaves. Initial results are discussed in terms of the electromagnetic effect resulting from the variation of parameters such as, fractal iteration number (or fractal level), fractal iteration factor, and periodicity of FSS, depending on the used pre-fractal element (Sierpinski island or T fractal). The transmission properties of these proposed periodic arrays are investigated through simulations performed by Ansoft DesignerTM and Ansoft HFSSTM commercial softwares that run full-wave methods. To validate the employed methodology, FSS prototypes are selected for fabrication and measurement. The obtained results point to interesting features for FSS spatial filters: compactness, with high values of frequency compression factor; as well as stable frequency responses at oblique incidence of plane waves. This thesis also approaches, as it main focus, the application of an alternative electromagnetic (EM) optimization technique for analysis and synthesis of FSSs with fractal motifs. In application examples of this technique, Vicsek and Sierpinski pre-fractal elements are used in the optimal design of FSS structures. Based on computational intelligence tools, the proposed technique overcomes the high computational cost associated to the full-wave parametric analyzes. To this end, fast and accurate multilayer perceptron (MLP) neural network models are developed using different parameters as design input variables. These neural network models aim to calculate the cost function in the iterations of population-based search algorithms. Continuous genetic algorithm (GA), particle swarm optimization (PSO), and bees algorithm (BA) are used for FSSs optimization with specific resonant frequency and bandwidth. The performance of these algorithms is compared in terms of computational cost and numerical convergence. Consistent results can be verified by the excellent agreement obtained between simulations and measurements related to FSS prototypes built with a given fractal iteration

Especificação de um bloco funcional implementando rede neural para redes industriais foundations fieldbus

This work proposes the specification of a new function block according to Foundation Fieldbus standards. The new block implements an artificial neural network, which may be useful in process control applications. The specification includes the definition of a main algorithm, that implements a neural network, as well as the description of some accessory functions, which provide safety characteristics to the block operation. Besides, it also describes the block attributes emphasizing its parameters, which constitute the block interfaces. Some experimental results, obtained from an artificial neural network implementation using actual standard functional blocks on a laboratorial FF network, are also shown, in order to demonstrate the possibility and also the convenience of integrating a neural network to Fieldbus devices

Contribuição para o estudo do embarque de uma rede neural artificial em field programmable gate array (FPGA)

This study shows the implementation and the embedding of an Artificial Neural Network (ANN) in hardware, or in a programmable device, as a field programmable gate array (FPGA). This work allowed the exploration of different implementations, described in VHDL, of multilayer perceptrons ANN. Due to the parallelism inherent to ANNs, there are disadvantages in software implementations due to the sequential nature of the Von Neumann architectures. As an alternative to this problem, there is a hardware implementation that allows to exploit all the parallelism implicit in this model. Currently, there is an increase in use of FPGAs as a platform to implement neural networks in hardware, exploiting the high processing power, low cost, ease of programming and ability to reconfigure the circuit, allowing the network to adapt to different applications. Given this context, the aim is to develop arrays of neural networks in hardware, a flexible architecture, in which it is possible to add or remove neurons, and mainly, modify the network topology, in order to enable a modular network of fixed-point arithmetic in a FPGA. Five synthesis of VHDL descriptions were produced: two for the neuron with one or two entrances, and three different architectures of ANN. The descriptions of the used architectures became very modular, easily allowing the increase or decrease of the number of neurons. As a result, some complete neural networks were implemented in FPGA, in fixed-point arithmetic, with a high-capacity parallel processing