Performance analysis of photovoltaic panel using machine learning method

Demand for energy is increasing as the world’s population grows, fossil fuels deplete on a daily basis, and climate conditions change. Renewable energy is more important than ever. Solar energy is the most accessible and cost-effective renewable energy source available today. Photovoltaic (PV) cells are the most promising way to convert solar energy into electricity. Wind speed, ambient temperature, incident radiation rate, and dust deposition are some of the internal and external variables that affect photovoltaic panel performance. Unwanted heat from the sun’s rays raises panel temperatures, reduces the amount of energy that solar cells can produce, and lowers conversion efficiency. Solar panels must be adequately cooled. The current research is focused on improving photovoltaic panel performance. The experimental system includes a fully automated photovoltaic panel, a microcontroller (NodeMCU8266), a DC pump, voltage and temperature sensors. The experiment was carried out with and without cooling of the PV panel. The findings suggest that keeping PV panel temperatures close to ambient temperatures improves performance. The Wi-Fi module collects real-time data on PV panel temperature, irradiation, ambient temperature, water temperature, and PV panel power output. The collected data was analyzed using machine learning. The PV panel’s performance was analyzed using the linear regression method.