Solar-powered adaptive battery charger for IoT solutions using ESP32

Abstract

The growth of Internet of Things (IoT) applications has intensified the demand for energy-autonomous devices, highlighting the limitations of conventional chargers. The paper addresses the design of a charging system for rechargeable batteries -including Li-ion, LiPo, NiCd and lead-acid- oriented to energy autonomous contexts using solar energy. An analysis of existing charging methods (Constant Current - Constant Voltage, step charging and pulsed charging), and their limitations when applied to multiple battery types in IoT environments is performed. A system based on an adaptive charging Current algorithm similar to step charging, managed by the ESP32 development board and powered by an 80 W solar panel is proposed. The system allows setting the maximum charging voltage and dynamically regulates the current depending on the battery status, with two thresholds (450 mA and 2000 mA), integrating temperature and humidity sensors, data storage in microSD and Bluetooth communication for monitoring and control. The test included the analysis of the thermal and electrical behavior of a lead-acid battery ($12 \mathrm{~V}, 7 \mathrm{Ah}$) under stable and variable solar irradiation conditions, demonstrating operational stability, current regulation and absence of overheating. Additionally, tests were carried out with Li-Ion, LiPo and NiCd batteries, showing stable behavior and capacity under stable and variable solar irradiation conditions.