Analysis of Times Series Clear Sky Global Horizontal Irradiance dataset for Application in Photovoltaic System Design

Abstract

In this study, analysis of the times series clear sky global horizontal irradiance dataset for application in photovoltaic (PV) system design is presented. Specifically, the Clear Sky Global Horizontal Irradiance (CKGHI) dataset spanning from 2003 to 2023 for Akwa Ibom State University, Obio Akpa campus, located at latitude and longitude 4.964487 and 7.759773, is considered. The data is retrieved from the National Aeronautics and Space Administration (NASA) Data Access Viewer (DAV) online platform for meteorological data. The case study of the 21-year daily CKGHI dataset has 7670 daily data records with a daily mean value of 6.2549 kW-hr/m²/day and a 0.4961 standard deviation, and all the 40 potential outliers identified occurred in the values below the lower fence of 4.8, and the outliers ranged from 3.996 to 4.76. The results showed that as the time span for the moving average increases from 3, 5, 7 and up to 365 days, the lower boundary value for outliers increases, whereas the upper boundary values decrease. The two values tend towards the mean daily value of the entire dataset, which is about 6.2549. In all, the results showed that with the use of moving average and outlier determination, it is possible to develop heuristic mechanisms that can be used to select a suitable solar radiation value for sizing a solar power system so that the desired loss of load specification can be achieved for the given days of power autonomy.