Version 1
: Received: 30 September 2024 / Approved: 1 October 2024 / Online: 3 October 2024 (07:56:39 CEST)
How to cite:
Son, N.; Kang, E. Enhancing Short- and Medium-term Solar Power Generation Forecasting Based on Prophet and Gate Recurrent Unit. Preprints2024, 2024100030. https://doi.org/10.20944/preprints202410.0030.v1
Son, N.; Kang, E. Enhancing Short- and Medium-term Solar Power Generation Forecasting Based on Prophet and Gate Recurrent Unit. Preprints 2024, 2024100030. https://doi.org/10.20944/preprints202410.0030.v1
Son, N.; Kang, E. Enhancing Short- and Medium-term Solar Power Generation Forecasting Based on Prophet and Gate Recurrent Unit. Preprints2024, 2024100030. https://doi.org/10.20944/preprints202410.0030.v1
APA Style
Son, N., & Kang, E. (2024). Enhancing Short- and Medium-term Solar Power Generation Forecasting Based on Prophet and Gate Recurrent Unit. Preprints. https://doi.org/10.20944/preprints202410.0030.v1
Chicago/Turabian Style
Son, N. and Eunjoo Kang. 2024 "Enhancing Short- and Medium-term Solar Power Generation Forecasting Based on Prophet and Gate Recurrent Unit" Preprints. https://doi.org/10.20944/preprints202410.0030.v1
Abstract
Accurate solar power forecasting is pivotal for the effective planning, management, and operation of power systems, ensuring a sustainable energy supply for consumers while optimizing the integration of renewable energy sources and the functioning of electricity markets. Recent advancements have focused on developing predictive models that offer precise daily forecasts for solar power generation, which are critical for power usage planning and production efficiency. These models typically leverage a diverse range of data inputs, including solar power metrics and meteorological variables such as temperature, humidity, precipitation, solar radiation, and wind speed—factors that significantly influence solar power generation due to their weather-dependent nature. This study introduces a novel hybrid forecasting model that synergizes the capabilities of the Prophet model and Gated Recurrent Units (GRU), leveraging their respective strengths to enhance predictive performance. The proposed model is rigorously evaluated across short-term (2 days, 7 days) and medium-term (15 days, 30 days) forecasting horizons. Experimental results demonstrate that the hybrid model significantly outperforms the existing models, delivering superior accuracy in solar power generation forecasts.
Keywords
Photovoltaic solar power forecasting; GRU; LSTM; Prophet; Meteorological data
Subject
Environmental and Earth Sciences, Environmental Science
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.