Dynamic Feedforward Neural Network Research Articles

A seasonal feedforward neural network to forecast electricity prices

In power industry and management, given the peculiarity and high complexity of the time series, it is highly requested to make models more flexible and well adapted to the data, in order to give them the ability to capture and recognize the most complex patterns, and consequently gain more accuracy in forecasting. To fulfill this aim, we define the seasonal autoregressive neural network (SAR-NN) as a dynamic feedforward artificial neural network (ANN), essentially conceived to forecast electricity prices. The SAR-NN is an ANN-based autoregressive model that considers that autoregressors are only those that are lagged by a multiple of the period p of the dominating seasonality. Moving forward this neural network, step-by-step, allows to generate multiple-steps-ahead reliable forecasts. This model is specifically designed to robustly overcome the strong seasonal effects and the other nonlinear patterns that often harm ANNs forecasting performance. The strategy is tested and compared to a number of homologous models throughout empirical experiments. As a case study, we focus on the Nord Pool Scandinavian power market, which is one of the most mature energy markets worldwide.

A dynamic semisupervised feedforward neural network clustering

AbstractAn efficient single-layer dynamic semisupervised feedforward neural network clustering method with one epoch training, data dimensionality reduction, and controlling noise data abilities is discussed to overcome the problems of high training time, low accuracy, and high memory complexity of clustering. Dynamically after the entrance of each new online input datum, the code book of nonrandom weights and other important information about online data as essentially important information are updated and stored in the memory. Consequently, the exclusive threshold of the data is calculated based on the essentially important information, and the data is clustered. Then, the network of clusters is updated. After learning, the model assigns a class label to the unlabeled data by considering a linear activation function and the exclusive threshold. Finally, the number of clusters and density of each cluster are updated. The accuracy of the proposed model is measured through the number of clusters, the quantity of correctly classified nodes, and F-measure. Briefly, in order to predict the survival time, the F-measure is 100% of the Iris, Musk2, Arcene, and Yeast data sets and 99.96% of the Spambase data set from the University of California at Irvine Machine Learning Repository; and the superior F-measure results in between 98.14% and 100% accuracies for the breast cancer data set from the University of Malaya Medical Center. We show that the proposed method is applicable in different areas, such as the prediction of the hydrate formation temperature with high accuracy.

A Dynamic Feedforward Neural Network Based on Gaussian Particle Swarm Optimization and its Application for Predictive Control

A dynamic feedforward neural network (DFNN) is proposed for predictive control, whose adaptive parameters are adjusted by using Gaussian particle swarm optimization (GPSO) in the training process. Adaptive time-delay operators are added in the DFNN to improve its generalization for poorly known nonlinear dynamic systems with long time delays. Furthermore, GPSO adopts a chaotic map with Gaussian function to balance the exploration and exploitation capabilities of particles, which improves the computational efficiency without compromising the performance of the DFNN. The stability of the particle dynamics is analyzed, based on the robust stability theory, without any restrictive assumption. A stability condition for the GPSO+DFNN model is derived, which ensures a satisfactory global search and quick convergence, without the need for gradients. The particle velocity ranges could change adaptively during the optimization process. The results of a comparative study show that the performance of the proposed algorithm can compete with selected algorithms on benchmark problems. Additional simulation results demonstrate the effectiveness and accuracy of the proposed combination algorithm in identifying and controlling nonlinear systems with long time delays.

Artificial neural network based system identification and model predictive control of a flotation column

The paper describes the design of a neural network based model predictive controller for controlling the interface level in a flotation column. For the system identification, the tailings valve opening is subjected to a pseudo-random ternary signal and response of the interface level is recorded over a period of time. The data so generated is used to develop a dynamic feed forward neural network model. The model uses two past values and one present value of the tailings valve opening as well as interface level as inputs and predicts the future interface level. This model is used for the design of a model predictive controller to control the interface level. The controller was tested both for liquid–gas system as well as liquid–gas–solid system and was found to perform very satisfactorily. The performance of the controller was compared with that of a conventional PI controller for a two-phase system and was found to be better.

Functional nodes in dynamic neural networks for bioprocess modelling.

This contribution presents a novel method for the direct integration of a-priori knowledge in a neural network and its application for the online determination of a secondary metabolite during industrial yeast fermentation. Hereby, existing system knowledge is integrated in an artificial neural network (ANN) by means of 'functional nodes'. A generalized backpropagation algorithm is presented. For illustration, a set of ordinary differential equations describing the diacetyl formation and degradation during the cultivation is incorporated in a functional node and integrated in a dynamic feedforward neural network in a hybrid manner. The results show that a hybrid modelling approach exploiting available a-priori knowledge and experimental data can considerably outperform a pure data-based modelling approach with respect to robustness, generalization and necessary amount of training data. The number of training sets were decreased by 50%, obtaining the same accuracy as in a conventional approach. All incorrect decisions, according to defined cost criteria obtained with the conventional ANN, were avoided.

An Adaptive State Filtering Algorithm for Systems With Partially Known Dynamics

On-line filtering of stochastic variables that are difficult or expensive to directly measure has been widely studied. In this paper a practical algorithm is presented for adaptive state filtering when the underlying nonlinear state equations are partially known. The unknown dynamics are constructively approximated using neural networks. The proposed algorithm is based on the two-step prediction-update approach of the Kalman Filter. The algorithm accounts for the unmodeled nonlinear dynamics and makes no assumptions regarding the system noise statistics. The proposed filter is implemented using static and dynamic feedforward neural networks. Both off-line and on-line learning algorithms are presented for training the filter networks. Two case studies are considered and comparisons with Extended Kalman Filters (EKFs) performed. For one of the case studies, the EKF converges but it results in higher state estimation errors than the equivalent neural filter with on-line learning. For another, more complex case study, the developed EKF does not converge. For both case studies, the off-line trained neural state filters converge quite rapidly and exhibit acceptable performance. On-line training further enhances filter performance, decoupling the eventual filter accuracy from the accuracy of the assumed system model.

A neural network approach to the construction of Delaunay tessellation of points in R/sup d/

Since a neural network may be designed directly from either the Delaunay tessellation (DT) or its abstract dual, the Voronoi diagram, the procedure advanced here for training a dynamic feedforward neural network to generate the DT of specified points representing exemplars in multidimensional feature space, contributes toward the goal of an all-neural approach to the synthesis of neural networks. As the expected number of simplexes in the DT over n points is linear in n, the procedure is convenient for real-time implementation of pattern classifiers.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>