Francisco García-Lagos

FPGA implementation of a systems identification module based upon Hopfield networks

The aim of this contribution is to implement a hardware module that performs parametric identification of dynamical systems. The design is based upon the methodology of optimization with Hopfield neural networks, leading to an adapted version of these networks. An outstanding feature of this modified Hopfield network is the existence of weights that vary with time. Since weights can no longer be stored in read-only memories, these dynamic weights constitute a significant challenge for digital circuits, in addition to the usual issues of area occupation, fixed-point arithmetic and nonlinear functions computations. The implementation, which is accomplished on FPGA circuits, achieves modularity and flexibility, due to the usage of parametric VHDL to describe the network. In contrast to software simulations, the natural parallelism of neural networks is preserved, at a limited cost in terms of circuitry cost and processing time. The functional simulation and the synthesis show the viability of the design. In particular, the FPGA implementation exhibits a reasonably fast convergence, which is required to produce accurate parameter estimations. Current research is oriented towards integrating the estimator within an embedded adaptive controller for autonomous systems.

Hopfield networks for identification of delay differential equations with an application to dengue fever epidemics in Cuba

Abstract This work is aimed at proposing an algorithm, based upon Hopfield networks, for estimating the parameters of delay differential equations. This neural estimator has been successfully applied to models described by Ordinary Differential Equations, whereas its application to systems with delays is a novel contribution. As a case in point, we present a model of dengue fever for the Cuban case, which is defined by a delay differential system. This epidemiological model is built upon the scheme of an SIR (susceptible, infected, recovered) population system, where both delays and time-varying parameters have been included. The latter are thus estimated by the proposed neural algorithm. Additionally, we obtain an expression of the Basic Reproduction Number for our model. Experimental results show the ability of the estimator to deal with systems with delays, providing plausible parameter estimations, which lead to predictions that are coherent with actual epidemiological data. Besides, when the Basic Reproduction Number is computed from the estimated parameter values, results suggest an evolution of the epidemic that is consistent with the observed infection. Hence the estimation could help health authorities to both predict the future trend of the epidemic and assess the efficiency of control measures.

Optimal phasor measurement unit placement using genetic algorithms

Many methods to codify Artificial Neural Networks have been developed to avoid the defects of direct encoding schema, improving the search into the solutions space. A method to estimate how the search space is covered and how are the movements along search process applying genetic operators is needed in order to evaluate the different encoding strategies for Feedforward Neural Networks. A first step of this method is considered with two encoding strategies, a direct encoding method and an indirect encoding scheme based on graph grammars: generative capacity, how many different architectures the method is able to generate.

Sine-fitting multiharmonic algorithms implemented by artificial neural networks

A new method designed to perform high-accuracy spectral analysis, based on ADALINE artificial neural networks (ANNs), is proposed. The proposed network is able to accurately calculate the fundamental frequency and the harmonic content of an input signal. The method is especially useful in high-precision digital measurement systems in which periodical signals are involved, i.e. digital watt meters. Most of these systems use spectral analysis algorithms as an intermediate step for the computation of the magnitudes of interest. The traditional spectral analysis methods require synchronous sampling, which introduce limitations to the sampling circuitry. Sine-fitting multiharmonics algorithms resolve the hardware limitations concerning the synchronous sampling but have some limitations with regard to the phase of the array of samples. The new implementation of sine-fitting multiharmonics algorithms based on ANN eliminates these limitations.

Contingency evaluation and monitorization using artificial neural networks

In this paper, different neural network-based solutions to the contingency analysis problem are presented. Contingency analysis is examined from two perspectives: as a functional approximation problem obtaining a numerical evaluation and ranking contingencies; and as a graphical monitoring problem, obtaining an easy visualization system of the relative severity of the contingencies. For the functional evaluation problem, we analyze the use of different supervised feed-forward artificial neural networks (multilayer perceptron and radial basis function networks). The proposed systems produce a very accurate evaluation and ranking, and so present a high applicability. For the graphical monitoring problem, unsupervised artificial neural networks such as self-organizing maps by Kohonen have been used. This solution allows both a rapid, easy and simultaneous visualization of the severity level of the complete contingency set. The proposed solutions avoid the main drawbacks of previous neural network approaches to this problem, which are explicitly analyzed here.

Non Spontaneous Saccadic Movements Identification in Clinical Electrooculography Using Machine Learning

In this paper we evaluate the use of the machine learning algorithms Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Classification and Regression Trees (CART) and Naive Bayes (NB) to identify non spontaneous saccades in clinical electrooculography tests. Our approach tries to solve problems like the use of manually established thresholds present in classical methods like identification by velocity threshold (I-VT) or identification by dispersion threshold (I-DT). We propose a modification to an adaptive threshold estimation algorithm for detecting signal impulses without the need of any user input. Also, a set of features were selected to take advantage of intrinsic characteristics of clinical electrooculography tests. The models were evaluated with signals recorded to subjects affected by Spinocerebellar Ataxia type 2 (SCA2). Results obtained by the algorithm show accuracies over 97%, recalls over 97% and precisions over 91% for the four models evaluated.

Self-organizing maps for contingency analysis: visual classification and temporal evolution

In this paper an analysis of the applicability of Kohonens self-organizing maps (SOMs) to contingency analysis in power systems is presented. We show the applicability of this artificial neural paradigm for both visualization and graphic monitoring of contingency severity, and the prediction of the system evolution to a future possible dangerous state. Both bidimensional and linear SOMs have been studied using as reference standard IEEE-14 and IEEE-118 electrical networks. Among the advantages of linear SOMs with respect to bidimensional SOMs and other classical methods we highlight the following ones: (1) a greater number of contingencies may be represented in one only screen and they may be more easily analyzed by a human operator; (2) the architecture and training process complexity of the SOM does not significantly increase with the power system size; and (3) the operation model is carried out in real time.

Neural Networks for Contingency Evaluation and Monitoring in Power Systems

In this paper an analysis of the applicability of different neural paradigms to contingency analysis in power systems is presented. On one hand, unsupervised Self-Organizing Maps by Kohonen have been implemented for visualization and graphic monitoring of contingency severity. On the other hand, supervised feed-forward neural paradigms such as Multilayer Perceptron and Radial Basis Function, are implemented for severity numerical evaluation and contingency ranking. Experiments have been performed with successfully result in the case of Kohonen and Multilayer Perceptron paradigms.

Spectrum analysis of asynchronously sampled signals by means of an ANN method

A method, based on ADALINE Artificial Neural Networks (ANNs), for spectrum analysis and fundamental frequency estimation of asynchronously sampled signals is compared with standard multiharmonics four parameter sine fit algorithm (4PSF). The performance of the method is demonstrated on real sinusoidal and real harmonically distorted sampled signals. The method resolves convergence problems of standard 4PSF and improves the repeatability of the measurements. Furthermore, the method shows high immunity to errors in the initial value of the fundamental frequency and negligible dependence on the calculated number of harmonics.

Harmonics and interharmonics spectral analysis by ANN

In this paper we present a new method based on ADALINE artificial neural networks (ANNs) for the spectra analysis and frequencies estimation of signals with harmonic and interharmonic content. First tests indicate that the method is able to obtain high accurate results without any information about the signal under analysis.