Performance of different structures of artificial neural network systems for harmonic estimation of grid-tied power conversion systems

Alghamdi, Thamer A. H. and Anayi, Fatih ORCID: https://orcid.org/0000-0001-8408-7673 2022. Performance of different structures of artificial neural network systems for harmonic estimation of grid-tied power conversion systems. Presented at: 6th International Conference on Power and Energy Engineering (ICPEE), Shanghai, China, 25-27 November 2022. Proceedings of 6th International Conference on Power and Energy Engineering. IEEE, pp. 187-190. 10.1109/ICPEE56418.2022.10050331

Full text not available from this repository.

Abstract

Practically, it is difficult to estimate the real harmonic emissions of a grid-connected solar Photovoltaic (PV) power inverter using conventional metering systems. As a result, the authors have recently suggested a strategy that uses an Artificial Neural Network (ANN) system and incorporates location-specific data. This paper compares the performance and computational requirements of various neural networks, including Multilayer Perceptron (MLP), Recurrent Neural Networks (RNN), Echo State Networks (ESN), Linear Auto-Regressive eXogenous (NARX), and Adaptive Wavelet Neural Network (AWNN) systems, to assess the effectiveness of various neural network structures for this application. It has been shown that compared to the NARX and AWNN, the MLP, RNN, and ESN have superior prediction accuracy and require less training and prediction time with a comparatively fewer number of neurons.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Publication
Status:	Published
Schools:	Engineering
Publisher:	IEEE
ISBN:	9781665464765
Last Modified:	06 Apr 2023 11:15
URI:	https://orca.cardiff.ac.uk/id/eprint/158388

Actions (repository staff only)

Edit Item