Performance prediction of the Micro Solar Dome in different climatic regions of India from pilot-scale by Random Forest algorithm. (August 2022)

• Record Type: Journal Article
• Authors: GhoshThakur, Richik; Basu, Aman; Haque, Zinia; Bhattacharya, Biswarup; GonChaudhuri, Santipada; Balachandran, Srinivasan
• Is Part Of: Sustainable energy technologies and assessments. Volume 52:Part B(2022)
• Publication Date: 2022-08
• Subjects: Solar PV -- Standalone lighting system -- Machine learning -- Hybrid solar lighting -- Model comparisons -- Köppen-Geiger classification; Renewable energy sources -- Periodicals; Energy development -- Technological innovations -- Periodicals; Electric power production -- Periodicals; Energy storage -- Periodicals; 333.79
• DOI: 10.1016/j.seta.2022.102163
• Languages: English
• ISSNs: 2213-1388

Abstract

Highlights: Affordable Standalone hybrid Solar indoor Lighting solution for rural houses. Pilot-scale implementation and performance testing in 8 states of India. Global Horizontal Solar irradiance primarily affects the performance of the device. Random forest was the best performing model for Köppen Classification. Device operates 4.6hrs in Bwa region, 2.4 hrs in Cfa region averaging around 3.5hrs. Abstract: As people living in remote areas still lack access to indoor lighting, the intervention of appropriate need-based innovations to improve basic indoor lighting facilities are essential. The present work discusses the development of a hybrid lighting device, Micro Solar Dome, that utilizes both the active and passive forms of solar energy to ensure sustainable lighting facilities for rural houses. The work elaborates the design, simulation, and pilot-scale installation of the device in 8 states in India. Based on the performance, a Principal Component Analysis was done, which indicated that more than 93% variation of the performance of the device was due to Global Horizontal Irradiance and Relative Humidity. A novel predictive model utilizing the Random Forest algorithm was developed for major Köppen climatic zones in India to determine the night-time performance of the device and validate the Principal Component Analysis results. As per the prediction, the average operation time of the device is about 3.5hrs at night, with the highest of 4.6 hrs in the Hot Desert Climate and the lowest of 2.4 hrs in the Humid Subtropical climate when 9 Köppen Zones are considered covering more than 95% of the geographic area of the country.