Harmonic Analysis in Distribution Systems Using a Multi-Step Prediction with NARX

This paper proposes a multi-step prediction method for the total harmonic distortion (THD) in three-phase networks with nonlinear loads using a nonlinear autoregressive network with exogenous inputs (NARX).The prediction is performed in three stages: in an open-loop neural network, closed-loop mode, and open-loop mode with the removal of the delay time. To verify the accuracy of the proposed method for identifying the harmonic source, comparisons are performed with two types of neural networks, a feed-forward back-propagation (FFBP) neural network and a cascaded feed-forward back-propagation (CFFBP) neural network. Moreover, three different cases of nonlinear loads are studied to validate the effectiveness of the proposed technique for harmonic prediction. A MATLAB/Simulink program is used for the modelling and simulation of the distribution system and all neural networks.

[1]  T.A. George,et al.  Harmonic power flow determination using the fast fourier transform , 1991, IEEE Power Engineering Review.

[2]  Yi-Chun Du,et al.  Levenberg-Marquardt Neural Network Algorithm for Degree of Arteriovenous Fistula Stenosis Classification Using a Dual Optical Photoplethysmography Sensor , 2018, Sensors.

[3]  V. Suresh Kumar,et al.  Determining true harmonic contributions of sources using neural network , 2016, Neurocomputing.

[4]  B. Blazic,et al.  Harmonic Contributions Evaluation With the Harmonic Current Vector Method , 2008, IEEE Transactions on Power Delivery.

[5]  Luis Sainz,et al.  Characterization of non-linear load behavior , 2008 .

[6]  Varun Arora,et al.  Hourly electric load forecasting using Nonlinear AutoRegressive with eXogenous (NARX) based neural network for the state of Goa, India , 2015, 2015 International Conference on Industrial Instrumentation and Control (ICIC).

[7]  Zoltan Corba,et al.  Control of grid connected converter with improved power quality characteristic , 2015 .

[8]  Marcelo Godoy Simões,et al.  Evaluating the Long-Term Impact of a Continuously Increasing Harmonic Demand on Feeder-Level Voltage Distortion , 2012, IEEE Transactions on Industry Applications.

[9]  Christopher Heard,et al.  Wind Speed Prediction Using a Univariate ARIMA Model and a Multivariate NARX Model , 2016 .

[10]  Daozhi Xia,et al.  Harmonic source identification and current separation in distribution systems , 2004 .

[11]  Dongpu Cao,et al.  Levenberg–Marquardt Backpropagation Training of Multilayer Neural Networks for State Estimation of a Safety-Critical Cyber-Physical System , 2018, IEEE Transactions on Industrial Informatics.

[12]  Rinat R. Nasyrov,et al.  Study and analysis of power quality situation in electrical power network. Case study: Lattakia-Syria , 2018, 2018 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus).

[13]  Lin Tao,et al.  2m+1 point estimate method for probabilistic harmonic power flow , 2016, 2016 IEEE Power and Energy Society General Meeting (PESGM).

[14]  Wilsun Xu,et al.  A "critical impedance"-based method for identifying harmonic sources , 2004, IEEE Transactions on Power Delivery.

[15]  M. Braun,et al.  Dimensioning and design of a Modular Multilevel Converter for drive applications , 2012, 2012 15th International Power Electronics and Motion Control Conference (EPE/PEMC).

[16]  Sónia Ferreira Pinto,et al.  Cascade multilevel STATCOM as a solution to improve the voltage profile of a power grid , 2018, 2018 International Young Engineers Forum (YEF-ECE).

[17]  Mark Beale,et al.  Neural Network Toolbox™ User's Guide , 2015 .

[18]  M. Eladawy,et al.  Prediction of the true harmonic current contribution of nonlinear loads using NARX neural network , 2017, Alexandria Engineering Journal.

[19]  Marcelo Godoy Simões,et al.  Evaluating the Long-Term Impact of a Continuously Increasing Harmonic Demand on Feeder-Level Voltage Distortion , 2014 .

[20]  Y. Liu,et al.  An Investigation on the Validity of Power Direction Method for Harmonic Source Determination , 2002, IEEE Power Engineering Review.

[21]  R.G. Harley,et al.  Neural Network Based Method for Predicting Nonlinear Load Harmonics , 2007, IEEE Transactions on Power Electronics.

[22]  Angelo Baggini,et al.  Handbook of Power Quality , 2008 .

[23]  H. Lin.  Intelligent Neural Network-Based Fast Power System Harmonic Detection , 2007, IEEE Transactions on Industrial Electronics.

[24]  Juriy Plotkin,et al.  Harmonic current pollution source determination in a grid connected wind farm , 2015, 2015 17th European Conference on Power Electronics and Applications (EPE'15 ECCE-Europe).

[25]  Haritza Camblong,et al.  A Nonlinear Autoregressive Exogenous (NARX) Neural Network Model for the Prediction of the Daily Direct Solar Radiation , 2018 .

[26]  Zvonimir Klaić,et al.  Harmonic Distortion Prediction Model of a Grid-Tie Photovoltaic Inverter Using an Artificial Neural Network , 2019, Energies.