Nonlinear Autoregressive Exogenous Neural Network Research Articles

A Comparative Study of Activation Functions of NAR and NARX Neural Network for Long-Term Wind Speed Forecasting in Malaysia

Since wind power is directly influenced by wind speed, long-term wind speed forecasting (WSF) plays an important role for wind farm installation. WSF is essential for controlling, energy management and scheduled wind power generation in wind farm. The proposed investigation in this paper provides 30-days-ahead WSF. Nonlinear Autoregressive (NAR) and Nonlinear Autoregressive Exogenous (NARX) Neural Network (NN) with different network settings have been used to facilitate the wind power generation. The essence of this study is that it compares the effect of activation functions (namely, tansig and logsig) in the performance of time series forecasting since activation function is the core element of any artificial neural network model. A set of wind speed data was collected from different meteorological stations in Malaysia, situated in Kuala Lumpur, Kuantan, and Melaka. The proposed activation functions tansig of NARNN and NARXNN resulted in promising outcomes in terms of very small error between actual and predicted wind speed as well as the comparison for the logsig transfer function results.

Decoding knee angle trajectory from electroencephalogram signal using NARX neural network and a new channel selection algorithm

Objectives: The aim of this research was to reveal the electroencephalogram (EEG) signal changes to obtain the knee angle change trajectory during a movement. Approach: Initially, a number of recorded EEG channels were selected using a new proposed EEG channel selection method. The signals were recorded from 10 healthy subjects in two states of movement imagination and implementation. Then, a NARX (Nonlinear Autoregressive Exogenous) neural network estimated the motion pattern of knee angle using the selected channels of EEG data. Main results: The results indicated that movement information extracted from the selected channels in mu rhythm was more accurate. Significance: This research suggests an approach to design the desired motion trajectory of the knee joint using the information emerging from the motor control process.

Data Driven Heating Energy Load Forecast Modeling Enhanced by Nonlinear Autoregressive Exogenous Neural Networks

Data Driven Heating Energy Load Forecast Modeling Enhanced by Nonlinear Autoregressive Exogenous Neural Networks

Energy Modeling with Nonlinear-Autoregressive Exogenous Neural Network

The model-based predictive control (MPC) is considered to be an effective tool for optimal control of building heating, ventilation, and air-conditioning (HVAC) systems. MPC need to update the operating set points of the local control loops that have a significant influence on the energy performance of the system. Performance of MPC relies on the accuracy of the system performance model. There are two commonly used modeling approach – conventional or analytical approach that is the way of process modeling for some time, but it tends to increase the online computational load as it requires a full mathematical description of the real system. Furthermore, such techniques rely on different simplifying assumptions that limit the accuracy of the performance model. A second commonly used technique is the data-driven approach. The neural network (NN) is the most potent data-driven approach. NN can accurately model complex nonlinear systems without even knowing the structure of the system and it also addresses the problem of the online computational load since the computational load moves to the offline training step. In order to set up neural network model-based predictive control (NNMPC), it is important to build a reliable energy model of HVAC system that can be used to perform multi-step-ahead prediction of system energy performance. In this paper, the energy modeling of the chiller plant is conducted. Data for the training of chiller plant energy model is generated from HVAC testbed build in TRNSYS simulation environment. The nonlinear-autoregressive neural network with exogenous input (NARX) is used to model the energy performance of the chiller plant. The NARX is a powerful method for forecasting of time series data and dynamic control problems. NARX model is first trained in the open-loop form with the actual output instead of feedback, using back-propagation with the Levenberg-Marquardt method; this model can be used to perform only one-step-ahead prediction. Open-loop NARX model is transformed into a closed-loop form, by connecting the internal feedback, i.e. actual output is replaced by predicted output, to perform multi-step-ahead prediction (for predictive control). Comparative analysis of developed NARX-based chiller model is carried out with respect to process data from testbed, which demonstrated the good accuracy of the NARX-based chiller model.

The use of radial basis function and non-linear autoregressive exogenous neural networks to forecast multi-step ahead of time flood water level

Many different Artificial Neural Networks (ANN) models of flood have been developed for forecast updating. However, the model performance, and error prediction in which forecast outputs are adjusted directly based on models calibrated to the time series of differences between observed and forecast values, are very interesting and challenging task. This paper presents an improved lead time flood forecasting using Non-linear Auto Regressive Exogenous Neural Network (NARXNN), which shows better performance in term of forecast precision and produces minimum error compared to neural network method using Radial Basis Function (RBF) in examined 12-hour ahead of time. First, RBF forecasting model was employed to predict the flood water level of Kelantan River at Kuala Krai, Kelantan, Malaysia. The model is tested for 1-hour and 7-hour ahead of time water level at flood location. The same analysis has also been taken by NARXNN method. Then, a non-linear neural network model with exogenous input promoted with enhancing a forecast lead time to 12-hour. Both about the performance comparison has briefly been analyzed. The result verified the precision of error prediction of the presented flood forecasting model.

Predicción inmediata de la actividad económica con datos de pagos electrónicos. Un enfoque de modelado predictivo

La predicción inmediata de la actividad económica (es decir, hacer estimaciones del período actual), es conveniente porque la mayoría de las medidas tradicionales de actividad económica se realizan con rezagos sustanciales. Nuestro objetivo es predecir el ISE, un indicador de actividad económica de corto plazo en Colombia. Las entradas son los rezagos del ISE y un conjunto de datos de pagos realizados entre individuos, empresas y el gobierno central, mediante transferencias electrónicas y cheques. Bajo un enfoque de modelado predictivo, empleamos un modelo de red neuronal exógena autoregresiva no lineal. Los resultadossugieren que nuestra elección de insumos y el método predictivo, nos permiten difundir la actividad económica con una precisión razonable. Además, validamos que los datos de pagos electrónicos reducen significativamente el error de predicción en un modelo de red neuronal autoregresiva de referencia. La predicción inmediata de la actividad económica con datos de los instrumentos de pago electrónicos, no solo contribuye a la toma de decisiones de los agentes y al modelado económico, sino que también soporta nuevas vías de investigación sobre cómo utilizar dichos datos en modelos actuales.

Power Quality Events Classification using ANN with Hilbert Transform

With the evolution of Smart Grid, Power Quality issues have become prominent. The urban development involves usage of computers, microprocessor controlled electronic loads and power electronic devices. These devices are the source of power quality disturbances. PQ problems are characterized by the variations in the magnitude and frequency in the system voltages and currents from their nominal values. To decide a control action, a proper classification mechanism is required to classify different PQ events. In this paper we propose a hybrid approach to perform this task. Different Neural topologies namely Cascade Forward Backprop Neural Network (CFBNN), Elman Backprop Neural Network (EBPNN), Feed Forward Backprop Neural Network (FFBPNN), Feed Forward Distributed Time Delay Neural Network (FFDTDNN) , Layer Recurrent Neural Network (LRNN), Nonlinear Autoregressive Exogenous Neural Network (NARX), Radial Basis Function Neural Network (RBFNN) along with the application of Hilbert Transform are employed to classify the PQ events. A meaningful comparison of these neural topologies is presented and it is found that Radial Basis Function Neural Network (RBFNN) is the most efficient topology to perform the classification task. Different levels of Additive White Gaussian Noise (AWGN) are added in the input features to present the comparison of classifiers.

A Nonlinear Autoregressive Exogenous (NARX) Neural Network Model for the Prediction of the Daily Direct Solar Radiation

The solar photovoltaic (PV) energy has an important place among the renewable energy sources. Therefore, several researchers have been interested by its modelling and its prediction, in order to improve the management of the electrical systems which include PV arrays. Among the existing techniques, artificial neural networks have proved their performance in the prediction of the solar radiation. However, the existing neural network models don’t satisfy the requirements of certain specific situations such as the one analyzed in this paper. The aim of this research work is to supply, with electricity, a race sailboat using exclusively renewable sources. The developed solution predicts the direct solar radiation on a horizontal surface. For that, a Nonlinear Autoregressive Exogenous (NARX) neural network is used. All the specific conditions of the sailboat operation are taken into account. The results show that the best prediction performance is obtained when the training phase of the neural network is performed periodically.

Performance analysis of NARX neural network backpropagation algorithm by various training functions for time series data

This study seeks to investigate the various training functions with non-linear auto regressive eXogenous neural network (NARXNN) to forecasting the closing index of the stock market. An iterative approach strives to adjust the number of hidden neurons of a NARXNN model. This approach systematically constructs different NARXNN models from simple architecture to complex architecture with different training functions and finds the optimum NARXNN model. The effectiveness of the proposed approach was seen to be a step ahead of Bombay Stock Exchange (BSE100) closing stock index of the Indian stock market. This approach has identified optimum neuron counts in the hidden layer for every training function with NARXNN, which reduces neural network (NN) structure and training time and increases the convergence speed. The experimental result reveals that neuron counts in the hidden layer cannot be identified by some rule of thumb.

Hybrid Clustering-GWO-NARX neural network technique in predicting stock price

Prediction of stock price is one of the most challenging tasks due to nonlinear nature of the stock data. Though numerous attempts have been made to predict the stock price by applying various techniques, yet the predicted price is not always accurate and even the error rate is high to some extent. Consequently, this paper endeavours to determine an efficient stock prediction strategy by implementing a combinatorial method of Grey Wolf Optimizer (GWO), Clustering and Non Linear Autoregressive Exogenous (NARX) Technique. The study uses stock data from prominent stock market i.e. New York Stock Exchange (NYSE), NASDAQ and emerging stock market i.e. Malaysian Stock Market (Bursa Malaysia), Dhaka Stock Exchange (DSE). It applies K-means clustering algorithm to determine the most promising cluster, then MGWO is used to determine the classification rate and finally the stock price is predicted by applying NARX neural network algorithm. The prediction performance gained through experimentation is compared and assessed to guide the investors in making investment decision. The result through this technique is indeed promising as it has shown almost precise prediction and improved error rate. We have applied the hybrid Clustering-GWO-NARX neural network technique in predicting stock price. We intend to work with the effect of various factors in stock price movement and selection of parameters. We will further investigate the influence of company news either positive or negative in stock price movement. We would be also interested to predict the Stock indices.

Modeling nonlinear behavior of Buckling-Restrained Braces via different artificial intelligence methods

Five AI models are presented to model the dynamic nonlinear behavior of Buckling-Restrained Braces (BRBs). The AI techniques utilized in the models are: Time-Delayed Neural Networks (TDNN), Nonlinear Auto-Regressive eXogenous (NARX) neural networks, Gaussian-Mixture Models Regression (GMMR), Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Polynomial Classifier Regression (PCR). The models are developed using time-delayed brace displacements inputs and brace force outputs to predict updated brace forces during load reversals. The training and testing of the AI models are performed using experimental data from BRB specimens tested at the Pacific Earthquake Engineering Research (PEER) Center. The training stage for every method makes use of the experimental data from one specimen. In order to assess the models’ learning and generalization capabilities, three sets of experimental data for different specimens are used. To arrive at an optimized architecture that best models the phenomenon, the model performance with different parameters is evaluated. The brace force predicted by the proposed model shows excellent resemblance to the experimental results for the training sample, for all techniques. The predicted behavior of the testing samples shows noticeable accuracy and further demonstrates the generalization and prediction capability of the proposed modeling techniques. The various techniques are compared on the basis of selected performance criteria. It is found that the performance of two AI techniques standout among the others: the NARX and the PCR. Although the NARX demonstrates a slight advantage in the prediction accuracy over the PCR, the latter is far more superior in terms of computational efficiency. Thus, the PCR would be recommended for scenarios where online training is needed. The BRB design and performance investigation processes can be facilitated by the developed modeling techniques thus minimizing the need for, and extent of, experimental testing.

MODIS NBAR Time Series Modeling With Two Statistical Methods and Application to Leaf Area Index Recursive Estimation

The inconsistent data quality of remote sensing observation, which is largely a result of atmospheric conditions, presents problems in the application of these data. Pixel reflectance in remote sensing observation varies with the type of land cover and the observation time. For land cover types that cycle yearly, such as vegetation, the variations in surface reflectance usually have dynamic periodic characteristics. In this study, we modeled the temporal feature of Moderate-Resolution Imaging Spectroradiometer (MODIS) Nadir BRDF-adjusted reflectance (NBAR) time series data of typical forest and cropland areas using two statistical methods: season-trend and seasonal autoregressive (AR) integrated moving average (SARIMA). The fitting values of these models were used in the recursive estimation of leaf area index (LAI) time series based on a nonlinear AR exogenous (NARX) neural network. This suppressed interferences from observational data noise and missing values. The results from 6 years (2008–2013) of MODIS NBAR modeling indicate that the two statistical methods are effective to model the NBAR time series of the vegetation surface; the season-trend model can extract both seasonal and trend components of long time series, and the SARIMA model has a good fitting capacity for general time series data. The NARX neural network performs well with the improved NBAR time series input, and the estimated LAI time series is more continuous than the MODIS LAI. Comparison with field data reveals the reliability of the estimated LAI values.

Ensemble Nonlinear Autoregressive Exogenous Artificial Neural Networks for Short-Term Wind Speed and Power Forecasting.

Short-term wind speed and wind power forecasts (for a 72 h period) are obtained using a nonlinear autoregressive exogenous artificial neural network (ANN) methodology which incorporates either numerical weather prediction or high-resolution computational fluid dynamics wind field information as an exogenous input. An ensemble approach is used to combine the predictions from many candidate ANNs in order to provide improved forecasts for wind speed and power, along with the associated uncertainties in these forecasts. More specifically, the ensemble ANN is used to quantify the uncertainties arising from the network weight initialization and from the unknown structure of the ANN. All members forming the ensemble of neural networks were trained using an efficient particle swarm optimization algorithm. The results of the proposed methodology are validated using wind speed and wind power data obtained from an operational wind farm located in Northern China. The assessment demonstrates that this methodology for wind speed and power forecasting generally provides an improvement in predictive skills when compared to the practice of using an “optimal” weight vector from a single ANN while providing additional information in the form of prediction uncertainty bounds.

Dynamic modelling of an automotive variable speed air conditioning system using nonlinear autoregressive exogenous neural networks

In this paper, a nonlinear autoregressive neural network with exogenous inputs (NARX) is proposed to model the dynamic behaviour of an automotive air conditioning (AAC) system equipped with a variable speed compressor. Two popular network architectures of artificial neural network namely multilayer perceptron (MLP) and radial basis network (RBN) are adopted to simulate the transient cabin temperature under random modulation of compressor speed. The input and output data required for the system identification are collected from an experimental bench made up of the original components of an AAC system. Optimization of network structure is conducted for the respective network architecture to study the effect of model structures on the model predictive performance. One-Step-Ahead (OSA) and Model Predicted Output (MPO) prediction tests are used to validate and verify the model. Investigation results show that NARX model with the investigated network architectures can be a powerful tool in identifying the nonlinear AAC system. However, a comparison between them signifies the advantage of MLP model over RBN model in term of network complexity, generalization capability and robustness against the presence of noise in the training data. The identified model of the AAC system can be used for further implementation in a model based intelligent control of an AAC system equipped with a variable speed compressor.

Comparison of NARX Neural Network and Classical Modelling Approaches

Classical optimization tools are effective when precise mechanistic models exist to support their design and implementation. However, most of the real-world processes are complex due to either nonlinearities or uncertainties (or both) and environmental variations, thus making realizing accurate mathematical models for such processes quite difficult or often impossible. Black box approach tends to present a better alternative in such situations. This paper presents a comparison of nonlinear autoregressive with eXogenous (NARX) neural network and traditional modelling techniques [autoregressive with exogenous input (ARX) and autoregressive moving average with exogenous input (ARMAX)]. The models were validated using experimental data from full-scale plants. Simulation results revealed that the performance of the NARX neural network is better compared to the ARMAX and ARX. The NARX neural network may serve as a valuable forecasting tool for the plants.

A nonparametric approach using artificial intelligence in vibration and noise reduction of flexible systems

The main aim of this paper is to broaden the application’s area of artificial intelligence including fuzzy logic and multiobjective evolutionary algorithm into real-time control area. Wiper system is a high order, nonlinear model with single-input and multi-outputs so that rise time, maximum overshoot, and end-point vibration of wiper blade are observed in conflict as the faster response leads to the larger level of undesired noise and vibration. The first part of this paper centers acquiring experimental data from a passenger automobile wiper system during its operation and using a reliable nonlinear system identification, namely, nonlinear autoregressive exogenous Elman neural network. Knowing that in a practical environment, where the loading conditions of the flexible wiper blade may be varied due to rain, snow, or wind lift in high-speed driving, causing changes in the characteristics of the system, the system performance with a fixed conventional controller scheme will not be satisfactory. The main contribution of this work is presented in second part where a novel multiobjective, bilevel adaptive-fuzzy controller is proposed for an automobile wiper system. The system’s parameters are tuned simultaneously by a multiobjective genetic algorithm based on fitness sharing whereby an automobile wiper blade is moved within its sweep workspace in the least amount of time with minimum noise and vibration.

Simulation Model of Magnetic Levitation Based on NARX Neural Networks

In this paper, we present analysis of different training types for nonlinear autoregressive neural network, used for simulation of magnetic levitation system. First, the model of this highly nonlinear system is described and after that the Nonlinear Auto Regressive eXogenous (NARX) of neural network model is given. Also, numerical optimization techniques for improved network training are described. It is verified that NARX neural network can be successfully used to simulate real magnetic levitation system if suitable training procedure is chosen, and the best two training types, obtained from experimental results, are described in details. Index Terms — Neural Network, Magnetic Levitation System, Nonlinear Model, Neural Network Training

NARX neural networks for nonlinear analysis of structures in frequency domain

The Volterra series is a powerful tool in the analysis of nonlinear systems. The nonlinear behavior of a system can be interpreted from investigation of Fourier transforms of Volterra kernels (so called higher‐order frequency response function, HFRFs). The Volterra series are highly promising for nonlinear analysis of civil structures where the nonlinear nature of structures must often be considered. However, a major limitation of Volterra series is the difficulty involved in the calculation of Volterra kernels. Chance et al. (1998) developed a method to obtain HFRFs of a single‐input system directly from the weights of the NARX (Nonlinear AutoRegressive with eXogenous) neural network which identified the system. In order to analyze nonlinear seismic behavior of multi‐input large civil structures such as bridges, this study extends the work of Chance et al. (1998) to a multi‐input Volterra series. Moreover, a numerical example and a real application example (nonlinear seismic behavior analysis of Bai‐Ho bridge, the first seismic isolation bridge in Taiwan) demonstrate the feasibility of applying the proposed method for nonlinear analysis of a multi‐input system.