You have no items in your shopping cart.
ABSTRACT
The global adoption and use of solar power as the main source of energy is the key to realising the UN Millennium Development Goals on Green Energy. The technology is projected to contribute about 20% of the world’s energy supply by 2050, over 60% by 2100 and leading to a 50% reduction in global CO2 emissions, but it is threatened by its poor performance in tropical climates. This work aims to model solar charging system performance under tropical weather conditions.
Data was taken on various weather parameters such as Humidity, Atmospheric-Temperature, Luminous-Intensity and Cloud-Cover over two seasons (rainy and dry culminations in one tropical year). Data consisted of various charging parameters such as Charge-Rate, Charge-Current, Battery-Voltage, Hourly Added Electric Charge (Ampere-Hours) and PV-Voltage (Voltage of the Panel). The data was then analyzed using machine learning algorithms to develop a model to define the relationships between the solar charging system’s performance and the various tropical weather parameters.
At the end of the project, a machine-learning model that can predict the charging pattern of deep-cycle batteries in tropical regions was developed.
