With the threat of climate change looming, it is time to embrace renewables on a larger scale. Photovoltaic systems, which generate electricity from a nearly unlimited supply of sunlight energy, are one of the most promising ways to generate clean energy. However, integrating PV systems into existing power grids is not a straightforward process. Since energy production for PV systems is highly dependent on environmental conditions, power plant and network managers need estimates of how much energy will be injected by PV systems in order to plan optimal generation and maintenance schedules, among other important operational aspects.
In keeping with recent trends, if something needs to be predicted, you can safely bet that AI will come up. To date, there are several algorithms that can estimate the power produced by PV systems several hours in advance by learning from past data and analyzing current variables. One of them, called Adaptive Neural Fuzzy Inference System (ANFIS), has been widely applied to predict the performance of complex renewable energy systems. Since its inception, many researchers have combined ANFIS with a variety of machine learning algorithms to further improve its performance.
In a recent study published in Renewable and Sustainable Energy CommentsA research team led by Jung Wan-ho of Incheon National University, Korea, has developed two new models based on ANFIS to better estimate the energy generated by PV systems as early as a full day. These two models are “hybrid algorithms” because they combine the traditional ANFIS approach with two different particle swarm optimization methods, which are powerful and computationally efficient strategies for finding optimal solutions to optimization problems.
To assess the performance of their models, the team compared them to other hybrid algorithms based on ANFIS. They tested the predictive capabilities of each model using real data from an actual photovoltaic system deployed in Italy in a previous study. The results, said Dr. Hu, were very promising: “One of the two models we developed outperformed all the hybrid models tested, thus showing great potential for short- and long-term prediction of PV power for solar systems.”
The results of this study could have immediate implications in the field of photovoltaic systems from a software and production perspective. “In terms of software, our models can be turned into applications that accurately estimate PV system values, resulting in improved performance and grid operation. In terms of production, our methods can translate into a direct increase in PV by helping to identify variables that can be used in the design of The photovoltaic system,” explains Dr. Hu. Let’s hope this work helps us transition to sustainable energy sources!
Research: Photovoltaic cells could make the world free of fossils faster than expected
Mosbeh R.Kaloop et al, The new application of adaptive swarm intelligence techniques coupled with a grid-based fuzzy inference system in PV forecasting, Renewable and Sustainable Energy Comments (2021). DOI: 10.1016 / j.rser.2021.111315
Presented by Incheon National University
the quote: Machine learning models to help PV systems find their place (2021, July 20) Retrieved on July 20, 2021 from https://techxplore.com/news/2021-07-machine-photovoltaic.html
This document is subject to copyright. Notwithstanding any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.