Catálogo de publicaciones - libros

We propose a search region prediction method using a Frequency Sensitive Competitive Learning(FSCL) algorithm for the adaptive vector quantization of the motion vector. We train the motion vector codebook using the first two successive images of a sequence of images and utilize it for search region prediction. The proposed method can reduce computation time by using a smaller number of search points compared to other methods, and also decreases the bits required to represent motion vectors. The experimental results show that it provides competitive PSNR values compared to other block matching algorithms.

Forecast of the flow of data packets between client and server for a website traffic analysis is viewed as a part of web analytics. Thousands of web-smart businesses depend on web analytics to improve website conversions, reduce marketing costs, website optimization, website monitoring and provide a higher level of service to their customers and partners. This paper particularly intends to develop a high-accuracy prediction approach as the need for a website traffic analysis. The proposed composite model (ASVR-ANFIS/NGARCH) is schemed to build a systematic structure such that it is not only to improve the predictive accuracy because of resolving the problems of the overshoot and volatility clustering simultaneously, but also to boost website tracking capacity helping each webmaster to optimize their website, maximize online marketing conversions and lead campaign tracking.

This paper presents two approaches to symbolic interval time series forecasting. The first approach is based on the autoregressive moving average (ARMA) model and the second is based on a hybrid methodology that combines both ARMA and artificial neural network (ANN) models. In the proposed approaches, two models are respectively fitted to the mid-point and range of the interval values assumed by the symbolic interval time series in the learning set. The forecast of the lower and upper bounds of the interval value of the time series is accomplished through the combination of forecasts from the mid-point and range of the interval values. The evaluation of the proposed models is based on the estimation of the average behaviour of the and in the framework of a Monte Carlo experiment.

With the Peak Ground Velocity 283 records in three dimensions, the velocity attenuation relationship with distance was discussed by neural network in this paper. The earthquake magnitude, epicenter distance, site intensity and site condition were considered as basic input element for the network. By using Bayesian Regularization Back Propagation Neural Networks (BRBPNN), the over-fitting phenomenon was reduced to some extent. The horizontal velocity was discussed. The PGV predicted by neural networks can simulate the detail difference with distance, while the PGV given by other traditional attenuation relationship only give a reduction relation with distance. The importance of each input factor was compared by the square weight of the input layer of the network. The order may be earthquake magnitude, epicenter distance and soil condition.

This paper proposes a hybrid machine learning model for electricity demand forecasting, based on Bayesian Clustering by Dynamics (BCD) and Support Vector Machine (SVM). In the proposed model, a BCD classifier is firstly applied to cluster the input data set into several subsets by the dynamics of load series in an unsupervised manner, and then, groups of 24 SVMs for the next day’s electricity demand curve are used to fit the training data of each subset. In the numerical experiment, the proposed model has been trained and tested on the data of the historical load from New York City.

In this paper, a short-term load forecasting model using a Multiscale BiLinear Recurrent Neural Network with an adaptive learning algorithm (M-BLRNN(AL)) is proposed. The proposed M-BLRNN(AL) model is based on a wavelet-based neural network architecture formulated by a combination of several individual BLRNN models. The wavelet transform adopted in the M-BLRNN(AL) is employed to decompose the load curve into a mutiresolution representation. Each individual BLRNN model is used to forecast the load signal at each resolution level obtained by the wavelet transform. The learning process is further improved by applying an adaptive learning algorithm at each resolution level. Experiments and results on load data from the North-American Electric Utility (NAEU) show that the proposed M-BLRNN(AL) model converges faster and archives better forecasting performance in comparison with other conventional models.

A new approach to electrical load forecasting is investigated. The method is based on the semi-parametric spectral estimation method that is used to decompose a signal into a harmonic linear signal model and a non-linear part. A neural network is then used to predict the non-linear part. The final predicted signal is then found by adding the neural network predicted non-linear part and the linear part. The performance of the proposed method seems to be more robust than using only the raw load data.

To predict short-term power load in an effective and fast way, the forecasting model of least square support vector machine (LSSVM) based on chaotic time series is established. According to A. Wolf method, Lyapunov exponents are worked out, and then the embedding dimension and time delay are also determined. And then the continuous power load data are transformed into data matrix by using the theory of phase-space reconstruction. Finally, LSSVM is used to train and predict the power load data. In order to prove the rationality of chosen dimension, another two random dimensions are selected to compare with the calculated dimension. And to prove the effectiveness and fast operating speed of the model, standard SVM algorithm and BP are used to compare with the model of LSSVM. The results show that the model is highly accurate and faster operating speed in short-term power load forecasting.

Extended linear programming (ELP) is an extension of classic linear programming in which the decision vector varies within a set. In previous studies in the neural network community, such a set is often assumed to be a box set. In the paper, the ELP problem with a general polyhedral set is investigated, and three recurrent neural networks are proposed for solving the problem with different types of constraints classified by the presence of bound constraints and equality constraints. The neural networks are proved stable in the Lyapunov sense and globally convergent to the solution sets of corresponding ELP problems. Numerical simulations are provided to demonstrate the results.

In this paper, a recurrent neural network model is proposed for solving non-smooth convex programming problems, which is a natural extension of the previous neural networks. By using the non-smooth analysis and the theory of differential inclusions, the global convergence of the equilibrium is analyzed and proved. One simulation example shows the convergence of the presented neural network.