In the global energy landscape, renewable energies such as solar energy have become increasingly important as a sustainable and environmentally beneficial replacement for fossil fuels. The utilization of energy derived from solar radiation and meteorological conditions is essential for power grids across the globe. For effective planning and optimization of solar energy generation in the future, energy output projections must be improved. This study explored three AI approaches for predicting power generation in PV solar systems: the hybrid LR-RF model, KNN, and LSTM. Using REMCL data from Indian Railways, the hybrid LR-RF model significantly outperformed both LSTM and KNN. Its strength lies in effectively handling time-series data, leading to superior solar power forecasting accuracy. The forecasted load was given to the grid using a hardware prototype, and results were shown. This study also introduces a novel integration of predicted solar power generation with a hardware prototype, allowing the predictive models to be validated in real-world scenarios. The hardware prototype provides insights into the usefulness of solar forecasting and makes a significant contribution to the growth of renewable energy integration.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Boosting Solar Load Forecasting Using Artificial Intelligence and Hardware Prototype




    Publication date :

    2024-07-31


    Size :

    996901 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Marine pollution diagnosis and forecasting system using artificial intelligence

    BAE JAE HWAN | European Patent Office | 2021

    Free access

    Artificial Intelligence and Forecasting of Traffic Accidents Using GIS

    Qarrot, Amro / Alramlawi, Khaled | Springer Verlag | 2023



    A Prototype Artificial Intelligence Driven Marine Propulsor Design Tool

    Hambric, S. / Dai, C. / Mulvihill, L. et al. | British Library Conference Proceedings | 1994