Analyzing Oxygen Production from Photovoltaic Electrolysis Using Machine Learning During the COVID-19 Period: A Case Study

In the context of energy conservation and emissions reduction, using photovoltaic electrolysis offers a promising approach for oxygen ( \(O_{2}\) ) generation using renewable energy sources. This study aims to evaluate the potential for oxygen production using photovoltaic electrolysis at the Mohammed VI University Hospital Center, located in Oujda City in the Oriental region of Morocco, identified by latitude and longitude coordinates 34.656455837015315, -1.9105162863259066. Using an installation equipped with 5 MW of solar panels, the study compares the results with the oxygen consumption from November 2020 to August 2021, a period marked by COVID-19. The study employs two algorithms: a time series model (Prophet) and a non-time series model (Support Vector Regression) to predict hourly oxygen production. Results indicate that SVR outperforms Prophet, showing higher R-squared values and lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values on the test sets. The study estimates that up to 429,629.88 normal cubic meters of oxygen could be produced in the period between November 2020 and August 2021, underscoring the advantages of adopting photovoltaic electrolysis for oxygen production in hospitals. This study and its findings can provide a foundation for upcoming research initiatives.