skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Predictive Neural Network Based Adaptive Controller for Grid-Connected PV Systems Supplying Pulse-Load

Abstract

This paper presents an adaptive controller for grid-tie DC-AC inverter in grid-connected Photovoltaic (PV) power system supplying a pulse AC load. The proposed controller oversees regulating the dc-bus voltage, managing the injected power to the grid, and minimizing the injected harmonics. The controller parameters are optimized and adaptively tuned using predictive neural network controller (PNNC). The PNNC predicts the control parameters by tracking the mean square errors of grid currents and dc-bus voltage and eliminating these errors in a very short finite time. The proposed controller was implemented in MATLAB environment and tested under different dynamic conditions, including step variation of irradiance level and the application of pulse loads. The controller performance was investigated in comparison with a base-case, in which the controller parameters are arbitrarily tuned. The results showed that the proposed adaptive controller offers faster dynamic response with less settling time and maximum overshoot for both current and voltage variables. Furthermore, the injected harmonics to the grid were significantly reduced showing a 1.97% total harmonic distortion (THD) in comparison with the conventional controller with 5.06% THD, which makes the PV system compatible with the requirements in the IEEE 519 international standard for harmonic control in electrical power systems.

Authors:
ORCiD logo [1];  [2];  [2];  [2]
  1. National Renewable Energy Laboratory (NREL), Golden, CO (United States)
  2. Zagazig University
Publication Date:
Research Org.:
National Renewable Energy Lab. (NREL), Golden, CO (United States)
Sponsoring Org.:
USDOE Office of Energy Efficiency and Renewable Energy (EERE)
OSTI Identifier:
1569448
Report Number(s):
NREL/JA-5400-75047
DOE Contract Number:  
AC36-08GO28308
Resource Type:
Journal Article
Journal Name:
Solar Energy
Additional Journal Information:
Journal Volume: 193
Country of Publication:
United States
Language:
English
Subject:
14 SOLAR ENERGY; 24 POWER TRANSMISSION AND DISTRIBUTION; adaptive control; DC-AC inverter; grid-connected; photovoltaic (PV) system; predictive neural network controller (PNNC)

Citation Formats

Mohamed, Ahmed A, Metwally, Hamid, El-Sayed, Ahmed, and Selema, S. I. Predictive Neural Network Based Adaptive Controller for Grid-Connected PV Systems Supplying Pulse-Load. United States: N. p., 2019. Web. doi:10.1016/j.solener.2019.09.018.
Mohamed, Ahmed A, Metwally, Hamid, El-Sayed, Ahmed, & Selema, S. I. Predictive Neural Network Based Adaptive Controller for Grid-Connected PV Systems Supplying Pulse-Load. United States. doi:10.1016/j.solener.2019.09.018.
Mohamed, Ahmed A, Metwally, Hamid, El-Sayed, Ahmed, and Selema, S. I. Fri . "Predictive Neural Network Based Adaptive Controller for Grid-Connected PV Systems Supplying Pulse-Load". United States. doi:10.1016/j.solener.2019.09.018.
@article{osti_1569448,
title = {Predictive Neural Network Based Adaptive Controller for Grid-Connected PV Systems Supplying Pulse-Load},
author = {Mohamed, Ahmed A and Metwally, Hamid and El-Sayed, Ahmed and Selema, S. I.},
abstractNote = {This paper presents an adaptive controller for grid-tie DC-AC inverter in grid-connected Photovoltaic (PV) power system supplying a pulse AC load. The proposed controller oversees regulating the dc-bus voltage, managing the injected power to the grid, and minimizing the injected harmonics. The controller parameters are optimized and adaptively tuned using predictive neural network controller (PNNC). The PNNC predicts the control parameters by tracking the mean square errors of grid currents and dc-bus voltage and eliminating these errors in a very short finite time. The proposed controller was implemented in MATLAB environment and tested under different dynamic conditions, including step variation of irradiance level and the application of pulse loads. The controller performance was investigated in comparison with a base-case, in which the controller parameters are arbitrarily tuned. The results showed that the proposed adaptive controller offers faster dynamic response with less settling time and maximum overshoot for both current and voltage variables. Furthermore, the injected harmonics to the grid were significantly reduced showing a 1.97% total harmonic distortion (THD) in comparison with the conventional controller with 5.06% THD, which makes the PV system compatible with the requirements in the IEEE 519 international standard for harmonic control in electrical power systems.},
doi = {10.1016/j.solener.2019.09.018},
journal = {Solar Energy},
number = ,
volume = 193,
place = {United States},
year = {2019},
month = {9}
}