Library Subscription: Guest

ISSN Online: 2379-1748

7th Thermal and Fluids Engineering Conference (TFEC)
SJR: 0.152 SNIP: 0.14 CiteScore™:: 0.5

Indexed in

Clarivate CPCI (Proceedings) Scopus
May, 15-18, 2022 , Las Vegas, NV, USA

SUPPORT VECTOR REGRESSION BASED THERMAL PERFORMANCE ASSESSMENT OF PHOTOVOLTAIC PANELS

Get access (open in a dialog) pages 615-626
DOI: 10.1615/TFEC2022.mpm.041171

Abstract

In India, the solar installed capacity is 40 GW as of March 2021, and it has set a target to reach 100 GW by the end of 2022, which will cause solar PV to emerge as the largest renewable power source in the country. The photovoltaic panel converts only ~18% of incident solar radiation energy into electrical energy, while remaining lost as heat energy. For every 1°C rise in the PV panel temperature, its lifetime significantly decreases, and it decreases the output power by ~0.4% depending on the type of PV cell technology used. Different cooling strategies for PV panel cooling can be broadly classified into active and passive cooling methods. Passive cooling technologies are more economical than active cooling. In the present study, passive cooling methods with fin and without fin-based systems have been investigated, considering a 40 Wp solar PV panel to analyze and compare the thermal performances. A support vector regression algorithm is trained by utilizing the weather data, and temperature distribution values obtained through numerical investigations conducted using computational fluid dynamics. The trained model is then used to predict the temperature distribution at the panel's surface and observe its effects on the thermal performance of the PV panel on any day of the year. Localized cooling techniques could be employed at the hotspot locations identified through numerical simulations.