Nonlinear Autoregressive Exogenous Research Articles

FORECASTING OIL PALM PRODUCTION BASED ON A NONLINEAR AUTOREGRESSIVE EXOGENOUS (NARX) NEURAL NETWORK MODEL

time series data analysis and prediction tool that is able to predict the yield of oil palm is needed to ensure an acceptable forecasting quality. An attempt was made in this study to develop a Nonlinear Autoregressive Exogenous (NARX) neural network model of oil palm production using MATLAB. This NARX model was used to predict the yield of oil palm in the states of Kelantan, Johor, Sabah and Sarawak in Malaysia. The performance of the NARX model was tested and validated using the Levenberg-Marquardt (LM) training algorithm and was compared with the Autoregressive Integrated Moving Average (ARIMA) model. The best performance of the NARX model was achieved at 70 per cent: 15 per cent: 15 per cent, with 10 neurons in the hidden layers and a delay value of four for Sarawak. For Kelantan and Johor, the NARX model produced the best result using the parameters of 70 per cent:10 per cent:20 per cent, with 13 neurons in the hidden layers and a delay value of four. The best result for Sabah was achieved using the parameters of 70 per cent: 15 per cent: 15 per cent, with 13 neurons in the hidden layers and a delay value of four. The results demonstrated that the proposed NARX model was more effective in modeling and forecasting time series data than the ARIMA model. The NARX model registered a minimum mean square error and mean absolute percentage error with a maximum average accuracy percentage and correlation coefficient. Keywords: Oil palm cultivation, yield predictions, nonlinear autoregressive exogenous neural network, autoregressive integrated moving average.

Cooling Load Forecasting via Predictive Optimization of a Nonlinear Autoregressive Exogenous (NARX) Neural Network Model

Accurate calculations and predictions of heating and cooling loads in buildings play an important role in the development and implementation of building energy management plans. This study aims to improve the forecasting accuracy of cooling load predictions using an optimized nonlinear autoregressive exogenous (NARX) neural network model. The preprocessing of training data and optimization of parameters were investigated for model optimization. In predictive models of cooling loads, the removal of missing values and the adjustment of structural parameters have been shown to help improve the predictive performance of a neural network model. In this study, preprocessing the training data eliminated missing values for times when the heating, ventilation, and air-conditioning system is not running. Also, the structural and learning parameters were adjusted to optimize the model parameters.

Weighed query-specific distance and hybrid NARX neural network for video object retrieval

Abstract The technical revolution in the field of video recording using the surveillance videos has increased the amount of the video databases that caused the need for an efficient video management system. This paper proposes a hybrid model using the nearest search algorithm (NSA) and the Levenberg–Marquardt (LM)-based non-linear autoregressive exogenous (NARX) neural network for performing the video object retrieval using the trajectories. Initially, the position of the objects in the video are retrieved using NSA and NARX individually, and they are averaged to determine the position of the object. The positions determined using the hybrid model is compared with the original database, and the trajectories of the objects are retrieved based on the minimum distance, which depends on the weighed query-specific distance. Experiments have been carried out using seven videos taken from the CAVIAR dataset, and the performance of the proposed method is compared with the existing methods. This proposed method found to be better than the existing method with respect to multiple object tracking precision (MOTP), multiple object tracking accuracy (MOTA), average tracking accuracy (ATA), precision, recall and F-measure that results a greater MOTP rate of 0.8796, precision rate of 0.8154, recall rate of 0.8408, the F-measure at a rate of 0.8371, MOTA of 0.8459 and ATA of 0.8324.

A randomized two-stage iterative method for switched nonlinear systems identification

This paper addresses the identification of discrete time switched nonlinear systems, which are collections of discrete time nonlinear continuous systems (modes) indexed by a finite-valued variable defining the current mode. In particular, we consider the class of Switched Nonlinear AutoRegressive eXogenous (Switched NARX, or SNARX) models, where the continuous dynamics are represented by NARX models. Given a set of input–output data, the identification of a SNARX model for the underlying system involves the simultaneous identification of the mode sequence and of the NARX model associated to each mode, configuring a mixed integer non-convex optimization problem, hardly solvable in practice due to the large combinatorial complexity. In this paper, we propose a black-box iterative identification method, where each iteration is characterized by two stages. In the first stage the identification problem is addressed assuming that mode switchings can occur only at predefined time instants, while in the second one the candidate mode switching locations are refined. This strategy allows to significantly reduce the combinatorial complexity of the problem, thus allowing an efficient solution of the optimization problem. The combinatorial optimization is addressed using a randomized method, whereby the sample-mode map and the SNARX model structure are characterized by a probability distribution, which is progressively tuned via a sample-and-evaluate strategy, until convergence to a limit distribution concentrated on the best SNARX model of the system generating the observed data.

Multivariate NARX neural network in prediction gaseous emissions within the influent chamber of wastewater treatment plants

Several potential problems associated with emissions such as odour nuisance may occur in sewer systems and wastewater treatment plants (WWTP). Improved understanding of prediction of emitted gases (e.g. hydrogen sulphide, H2S) is of great importance for better evaluation of such problems in WWTPs. The present paper provides a survey of the feasibility of utilising soft computing models in predicting emission factors (gaseous H2S) of based on five input parameters including the total dissolved sulphides, biochemical oxygen demand (BOD5), temperature, flow rate and pH. Multivariate nonlinear autoregressive exogenous (NARX) neural networks were developed and applied to predict weekly H2S in four WWTPS. Due to the nonlinearity nature of the H2S emission, the principle component analysis, and the average mutual information (AMI) method was employed to determine the effective independent parameters and the optimal time lags in state space for the H2S data. In addition to the NARX models, three standard regressive models (multiple linear regression model, stepwise regression and two-variate logarithmic regression model) were also applied to the same dataset for evaluating the capability and reliability of the developed NARX models. Based on several statistical measures (RMSE, MAPE, PCC, NSE & GRI), it was found that the NARX models were superior to the other applied models in predicting emitted H2S. Although the temperature and wastewater flow rate were found to be the most effective parameters for modelling H2S, but the results indeed showed that taking into account of all wastewater parameters would lead to more accurate H2S prediction.

Extended linear and non-linear auto-regressive models for forecasting the urban water consumption of a fast-growing city in an arid region

Accurate water consumption forecasting models are required for fast-growing cities in arid regions. In the present study, two forecasting models are proposed for urban-water consumption using extended Auto-Regressive Integrated Moving Average (ARIMA) and Nonlinear Auto-Regressive Exogenous (NARX) methods. Various extension of these models by adding new forecaster factors were tested to find superior linear and nonlinear models. The ARIMA and NARX model accuracy with and without additional forecaster factors (minimum, maximum and mean temperature, precipitation, mean relative humidity, sunshine hours, mean dew point temperature, mean wind speed, and population) were compared using accuracy indices. The ARIMA model which includes sunny hour in addition to the base model predictors was selected as the superior linear model and also the model NARX which includes sunny hours and population in addition to the base model’s predictors was selected as the superior nonlinear model. This infers that sunny hour has linear performance and population predictor has non-linear performance in forecasting water consumption. This shows adding sunny hours and population as forecaster factors can robust forecasting performance considerably. By extending the models with these additional forecasting factors, the precision of daily water consumption forecasts can be improved for providing a better urban water supply.

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.

Predicting the spread of influenza epidemics by analyzing twitter messages

Seasonal influenza epidemics affect millions of people with respiratory illnesses and cause 250,000 to 500,000 deaths worldwide each year. Rapidly predicting the outbreak of epidemics leads to an earlier detection and control. In this study, we predicted an influenza-like illness (ILI) based on social media data derived from Twitter. Tweets and patients do not always have a linear correlation; therefore, we employed nonlinear methods including autoregressive with exogenous inputs (ARX), autoregressive-moving-average with exogenous inputs (ARMAX), nonlinear autoregressive exogenous (NARX), deep multilayer perceptron (DeepMLP), and a convolutional neural network (CNN). Two new features employed to significantly reduce the prediction errors are products of the tweets and Centers for Disease Control and Prevention (CDC) data and of the tweets and Google data. Furthermore, we introduced a new method based on entropy that decreased the errors as well as time complexity. Among the available methods and features, the best results were obtained with the newly developed features in the deep neural network methods and the entropy-based method that decreased the mean average error by up to 25%. The entropy method also reduced the time complexity. Applying the above-mentioned methods to the Twitter datasets from 2009 to 2010 and 2011–2014 revealed that the ILI outbreak can be predicted 2–4 weeks earlier than by the CDC.

Prediction of short and medium term PM10 concentration using artificial neural networks

Purpose There has been a growing concern about air quality because in recent years, industrial and vehicle emissions have resulted in unsatisfactory human health conditions. There is an urgent need for the measurements and estimations of particulate pollutants levels, especially in urban areas. As a contribution to this issue, the purpose of this paper is to use data from measured concentrations of particulate matter and meteorological conditions for the predictions of PM10. Design/methodology/approach The procedure included daily data collection of current PM10 concentrations for the city of São Carlos-SP, Brazil. These data series enabled to use an estimator based on artificial neural networks. Data sets were collected using the high-volume sampler equipment (VFA-MP10) in the period ranging from 1997 to 2006 and from 2014 to 2015. The predictive models were created using statistics from meteorological data. The models were developed using two neural network architectures, namely, perceptron multilayer (MLP) and non-linear autoregressive exogenous (NARX) inputs network. Findings It was observed that, over time, there was a decrease in the PM10 concentration rates. This is due to the implementation of more strict environmental laws and the development of less polluting technologies. The model NARX that used as input layer the climatic variables and the PM10 of the previous day presented the highest average absolute error. However, the NARX model presented the fastest convergence compared with the MLP network. Originality/value The presentation of a given PM10 concentration of the previous day improved the performance of the predictive models. This paper brings contributions with the NARX model applications.

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.

Mapping urban land surface temperature using remote sensing techniques and artificial neural network modelling

ABSTRACTThe objective of the present study is to monitor and predict the changes in land surface temperature (LST) in the North of Jordan during the Period 2000 to 2016. Due to political instability in the nearby countries Syria and Iraq, Jordan has witnessed increased influxes of refugees, starting from the year 2003, which has been led to the urban expansion in the area that reflected on the climatic conditions and affected the LST values. Satellite images were used for providing LST, the acquired images represented two seasons of each year, namely summer and winter. Simulation and prediction of LST values for the next 10 years were carried out using nonlinear autoregressive exogenous (NARX) artificial neural network (ANN) model. The inputs to the model consist of meteorological data collected from eight stations in the study area, population, and land use and land cover (LULC). In fact, LULC was expressed in terms of normalized difference building index (NDBI) and normalized difference vegetation index (NDVI) that were obtained from satellite images. The model showed a high correlation between these parameters and the values of simulated LST, where the correlation coefficient for the training set, validation set, testing set and for the entire data ranged from 0.91 to 0.92. Based on the predicted LST values, LST maps for the next 10 years were developed and compared with the present actual LST maps for the year 2016. The comparison has shown an average increase of 1.1 °C in the average LST values, which is considered a significant increase compared with previous studies.

Short-Term Prediction of Carbon Monoxide Concentration Using Artificial Neural Network (NARX) Without Traffic Data: Case Study: Shiraz City

Air pollution is one of the most widespread and important issues in mechanical civilization, and it has made supervision and control of air quality an ineluctable issue that has been introduced as a principal national problem. This study investigates the ability of dynamic neural networks, particularly the nonlinear autoregressive exogenous (NARX) network, in predicting air carbon monoxide concentration in Shiraz in the absence of traffic data since there are no accurate statistical data on traffic volume (as one of the primary sources for air pollution modeling). Dynamic networks have been utilized to model time-variable patterns as they have time memory through the history of concentration volume implicitly containing traffic characteristics. To begin this study, meteorological data including temperature, moisture content, rainfall amount, and wind velocity and direction at a 3-h mean basis were obtained from the Bureau of Meteorology at the Shiraz Airport. Moreover, air pollutant concentration data due to Setad Square’s measurement station between 2005 and 2008 were prepared from the Fars Department of Environmental Protection. According to the results obtained from the static neural network, the correlation coefficients (R) for the training, validation, and test datasets are estimated as 0.49, 0.37, and 0.41, respectively. Moreover, the R2 correlation coefficient, the root mean square error (RMSE), and mean absolute percentage error (MAPE) are 0.31, 0.43, and 51%, respectively. However, the correlation coefficients achieved from NARX model for the training, validation, and test datasets are estimated as 0.77, 0.76, and 0.80, respectively, while the R2 correlation coefficient, RMSE, and MAPE are 0.72, 0.05, and 7%, respectively. The results demonstrate the dynamic neural network’s high performance in modeling carbon monoxide concentration in the absence of traffic data. Moreover, sensitivity analysis indicates the stability of the model to the noisy data.

Comparison of NARX and Dual Polarization Models for Estimation of the VRLA Battery Charging/Discharging Dynamics in Pulse Cycle

The following work presents the model-assisted research on Valve-Regulated Lead-Acid (VRLA) Absorbent Glass Mat (AGM) battery in pulse operation cycle. The experimental research was conducted for a constant value of State of Charge (SOC) of the battery, for values ranging from 0.2 to 0.8. Based on the conducted test stand research, the parameters of the battery were identified, which were later used to model the battery using the equivalent circuit based on dual polarization (DP) model with double Resistive-Capacitive (RC) loop. Simulations were performed for the identified parameters of the battery which are described by the general form of the polynomial. The second part contains the research on utilization of Nonlinear AutoRegressive eXogenous (NARX) recurrent neural network to predict SOC and a terminal voltage of the battery. Obtained validation results with the use of the identified parameters of the double RC loop and NARX model were discussed in the following work. The article also features the advantages and disadvantages of NARX model and DP model utilization for the use of in Battery Managements Systems (BMS) and micro-installations based on renewable energy sources. Furthermore, the advantages of the addition of more RC loops to describe the dynamic states of batteries in pulse states were discussed in the article.

Economic dispatch of energy storage systems in dc microgrids employing a semidefinite programming model

Abstract A mathematical optimization approach for the optimal operation focused on the economic dispatch for dc microgrid with high penetration of distributed generators and energy storage systems (ESS) via semidefinite programming (SDP) is proposed in this paper. The SDP allows transforming the nonlinear and non-convex characteristics of the economic dispatch problem into a convex approximation which is easy for implementation in specialized software, i.e., CVX. The proposed mathematical approach contemplates the efficient operation of a dc microgrid over a period of time with variable energy purchase prices, which makes it a practical methodology to apply in real-time operating conditions. A nonlinear autoregressive exogenous (NARX) model is employed for training an artificial neural network (ANN) for forecasting solar radiation and wind speed for renewable generation integration and dispatch considering periods of prediction of 0.5 h. Four scenarios are proposed to analyze the inclusion of ESS in a dc microgrid for economic dispatch studies. Additionally, the results are compared with GAMS commercial optimization package, which allows validating the accuracy and quality of the proposed optimizing methodology.

Visual-Manual Distraction Detection Using Driving Performance Indicators With Naturalistic Driving Data

This paper investigates the problem of driver distraction detection using driving performance indicators from onboard kinematic measurements. First, naturalistic driving data from the integrated vehicle-based safety system program are processed, and cabin camera data are manually inspected to determine the driver’s state (i.e., distracted or attentive). Second, existing driving performance metrics, such as steering entropy, steering wheel reversal rate, and lane offset variance, are reviewed against the processed naturalistic driving data. Furthermore, a nonlinear autoregressive exogenous (NARX) driving model is developed to predict vehicle speed based on the range (distance headway), range rate, and speed history. For each driver, the NARX model is then trained on the attentive driving data. We show that the prediction error is correlated with driver distraction. Finally, two features, steering entropy and mean absolute speed prediction error from the NARX model are selected, and a support vector machine is trained to detect driving distraction. Prediction performances are reported.

A novel dynamic modeling of an industrial gas turbine using condition monitoring data

For some of the existing gas turbines, there is no standard documentation, so non-OEM (Original Equipment Manufacturer) companies will encounter problems for upgrading the systems. The problem even is getting worse for companies focusing on old gas turbines control system retrofit. The main purpose of this paper is to present a gas turbine dynamic modeling procedure in start-up and loading modes, considering some performance information with minimum access to the design and related technical datasheets of the turbine. To specify the characteristics of the shaft speed during system start-up, there are numerous complexities due to uncertainties and unspecified parameters. In this paper, for the start-up mode, the neural network (NN) method and the derived functions of linear regression (LR), shaft dynamic (SD)-based function, and nonlinear auto-regressive exogenous (NARX) and Hammerstein-Wiener (HW) fitted structures are used in addition to the time-delayed transfer functions in order to generate the dynamic model. Also, the operational information of an operating power plant is utilized to generate the dynamic model of the system in loading and unloading modes.

A Hybrid Feedforward-Feedback Hysteresis Compensator in Piezoelectric Actuators Based on Least-Squares Support Vector Machine

Hysteresis nonlinearity of piezoelectric actuators degrades the positioning accuracy of micro/nanopositioning systems. To overcome this problem, an innovative hysteresis compensator based on least-squares support vector machine (LSSVM) is proposed in this paper. First, the LSSVM hysteresis modeling is presented using nonlinear auto regressive eXogenous (NARX) structure. To compensate for the hysteresis behavior, two feedforward control schemes according to different inputs of NARX model are proposed and analyzed separately. Then, a hybrid feedforward controller combining both the control schemes is put forward to revise the model input. To further improve the tracking performance, the hybrid feedforward control combined with the feedback control is realized. The comparative study reveals the superior tracking performance of feedforward-feedback control scheme over hybrid feedforward control or feedback control. Moreover, the hybrid feedforward-feedback control scheme is capable of tracking different testing waveforms with negligible errors, which confirms the effectiveness and generalization ability of the proposed approach.

A Novel Modeling Method for Aircraft Engine Using Nonlinear Autoregressive Exogenous (NARX) Models Based on Wavelet Neural Networks

Abstract A novel modeling method for aircraft engine using nonlinear autoregressive exogenous (NARX) models based on wavelet neural networks is proposed. The identification principle and process based on wavelet neural networks are studied, and the modeling scheme based on NARX is proposed. Then, the time series data sets from three types of aircraft engines are utilized to build the corresponding NARX models, and these NARX models are validated by the simulation. The results show that all the best NARX models can capture the original aircraft engine’s dynamic characteristic well with the high accuracy. For every type of engine, the relative identification errors of its best NARX model and the component level model are no more than 3.5 % and most of them are within 1 %.

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.

The cortical focus in childhood absence epilepsy; evidence from nonlinear analysis of scalp EEG recordings

ObjectiveTo determine the origin and dynamic characteristics of the generalised hyper-synchronous spike and wave (SW) discharges in childhood absence epilepsy (CAE). MethodsWe applied nonlinear methods, the error reduction ratio (ERR) causality test and cross-frequency analysis, with a nonlinear autoregressive exogenous (NARX) model, to electroencephalograms (EEGs) from CAE, selected with stringent electro-clinical criteria (17 cases, 42 absences). We analysed the pre-ictal and ictal strength of association between homologous and heterologous EEG derivations and estimated the direction of synchronisation and corresponding time lags. ResultsA frontal/fronto-central onset of the absences is detected in 13 of the 17 cases with the highest ictal strength of association between homologous frontal followed by centro-temporal and fronto-central areas. Delays consistently in excess of 4 ms occur at the very onset between these regions, swiftly followed by the emergence of “isochronous” (0–2 ms) synchronisation but dynamic time lag changes occur during SW discharges. ConclusionsIn absences an initial cortico-cortical spread leads to dynamic lag changes to include periods of isochronous interhemispheric synchronisation, which we hypothesize is mediated by the thalamus. SignificanceAbsences from CAE show ictal epileptic network dynamics remarkably similar to those observed in WAG/Rij rats which guided the formulation of the cortical focus theory.