Skip to main content

ORIGINAL RESEARCH article

Front. Energy Res.
Sec. Solar Energy
Volume 12 - 2024 | doi: 10.3389/fenrg.2024.1446422
This article is part of the Research Topic Ensuring the Reliability of Solar Photovoltaics View all 7 articles

A Hybrid Model Based on Photovoltaic Conversion Model and Artificial Neural Network Model for Short-term Photovoltaic Power Forecasting

Provisionally accepted
Chen ran Chen ran 1*Gao Shaowei Gao Shaowei 2zhao yao zhao yao 2*Li dongdong Li dongdong 2*lin shunfu lin shunfu 2*
  • 1 State Grid Shanghai Electric Power Research Institute, Shanghai, China
  • 2 Shanghai University of Electric Power, Shanghai, China

The final, formatted version of the article will be published soon.

    Photovoltaic (PV) power is greatly uncertain due to the random meteorological parameters. Therefore, the accurate PV power forecasting results are significant for the dispatching of power and improving of system stability. In this paper, a hybrid forecasting model is proposed for one-day ahead PV power forecasting under different cloud amount conditions. The proposed model consists of an improved artificial neural network (ANN) algorithm and a PV power conversion model. First, the ANN model is designed to forecast plane of array (POA) irradiance and ambient temperature. Backpropagation methods, Gradient descent methods, and L2 regularization methods are applied in the structure of the ANN model to achieve the best weights, improve the prediction accuracy, and alleviate the effect of the over-fitting. Second, the PV power conversion model employs the forecasted results of POA irradiance and ambient temperature to determine the PV power produced by a PV module. In addition to the basic temperature factor, an environmental efficiency and a reflection efficiency are incorporated into the conversion model to account for real PV module losses. The performance of the proposed model is validated with real weather and PV power data from Alice Spring and Climate Data Store. Results show that the model improves forecast accuracy compared to four benchmark models. Specifically, it reduces Root Mean Square Error (RMSE) and nRMSE by up to 25% under cloudy conditions and offers a 3% shorter training time compared to Extreme Gradient Boosting.

    Keywords: artificial neural network, Hybrid model, Photovoltaic power forecasting, Photovoltaic conversion model, Short-term forecasting

    Received: 09 Jun 2024; Accepted: 25 Nov 2024.

    Copyright: © 2024 ran, Shaowei, yao, dongdong and shunfu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence:
    Chen ran, State Grid Shanghai Electric Power Research Institute, Shanghai, China
    zhao yao, Shanghai University of Electric Power, Shanghai, China
    Li dongdong, Shanghai University of Electric Power, Shanghai, China
    lin shunfu, Shanghai University of Electric Power, Shanghai, China

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.