Digital Twinning and LSTM-based Forecasting Model of Solar PV Power Output

Omar Adil Al-Isawi, Lutfi Hatem Amirah, Omar Ahmed Al-Mufti, Chaouki Ghenai

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    5 Scopus citations

    Abstract

    The efficiency, return on investment, and output of Solar power plants are all significantly impacted by the development of Digital Twin technology. Digital Twining of the Solar PV system will help to improve the predictability of the power output, gather the necessary information for energy forecasting (predict ahead the power output to balance supply and demand) and power system default detection. The main objective of this study is to develop a Digital Twin Solar PV system. The technical approach used to develop a Solar PV Digital Twin includes: (1) real time data collection of power output from Solar PV system using smart sensors, (2) development, calibration, and validation of a Solar PV system model using MATLAB (3) development and testing the accuracy of LSTM-based forecasting model to predict ahead (15 mins) the Solar PV electrical power output using historical data. The results show that the Solar PV Digital Twin compares well with the targeted data from the real system (data from sensors of the physical Solar PV system). The accuracy of the models was assessed using the coefficient of correlation (R). The results show that the Solar PV Digital Twin simulation and forecasting models have R-values of 0.99893 and 0.99427, respectively. To help an asset owner or operator to visualize the performance of Digital Twin, a dashboard application was developed using MATLAB App Designer. The dashboard shows current, historical, and continuously forecasts future energy consumption and production to balance supply and demand. The development of Digital Twin technology, artificial intelligence and machine learning predictive tools for monitoring, forecasting and maintenance of Solar PV plants are keys to reach the Net Zero Energy Building goals. Additionally, to ensure the reproducibility of the research, the irradiance measurements data was published in a GitHub repository.

    Original languageBritish English
    Title of host publication2023 Advances in Science and Engineering Technology International Conferences, ASET 2023
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    ISBN (Electronic)9781665454742
    DOIs
    StatePublished - 2023
    Event2023 Advances in Science and Engineering Technology International Conferences, ASET 2023 - Dubai, United Arab Emirates
    Duration: 20 Feb 202323 Feb 2023

    Publication series

    Name2023 Advances in Science and Engineering Technology International Conferences, ASET 2023

    Conference

    Conference2023 Advances in Science and Engineering Technology International Conferences, ASET 2023
    Country/TerritoryUnited Arab Emirates
    CityDubai
    Period20/02/2323/02/23

    Keywords

    • Digital Twin
    • Forecasting
    • LSTM
    • Modeling & Simulation
    • Net-Zero Energy Building
    • Solar PV

    Fingerprint

    Dive into the research topics of 'Digital Twinning and LSTM-based Forecasting Model of Solar PV Power Output'. Together they form a unique fingerprint.

    Cite this