Estimating the amounts of the solar energy: A multiple linear regression approach

The solar energy has some kind of effects in many fields connected to the human life as health, electricity generation and industrial, in a lot of countries around the world. Iraq is one of the countries, which has a large amount of the solar radiation throughout a year that can be invested to compe...

Full description

Saved in:
Bibliographic Details
Main Author: Saleh, Hussein H.
Format: Conference Proceeding
Language:English
Subjects:
Electricity distribution
Estimation
Independent variables
Indicators
Mean square errors
Prediction models
Radiation
Regression analysis
Relative humidity
Root-mean-square errors
Solar energy
Solar radiation
Statistical analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The solar energy has some kind of effects in many fields connected to the human life as health, electricity generation and industrial, in a lot of countries around the world. Iraq is one of the countries, which has a large amount of the solar radiation throughout a year that can be invested to compensate the shortfall in the electric power particularly. In this paper, the analysis models in the four years from 2017 to 2020 are applied to estimate the amounts of the daily solar radiation using Multiple Linear Regression (Mtp_LR) method in Anbar, Iraq. Furthermore, two climate variables relative humidity (%) and air pressure Pascal unit (Pa) have been implemented as an input dataset to the predictive models. The results are evaluated via two statistical indicators Root Mean Square Error (RMSE) and Relative Root Mean Square Error (RRMSE). The models of the four years acquire remarkable accuracy in solar energy forecasting with RMSE values of the statistical indicators RMSE (0.8915, 1.0827, 0.7757 and 0.5055) kilowatt hours per square meter (kWh/m2) in 2017, 2018, 2019, and 2020 respectively, that determines a strong estimation outcome with the best relationship between the independent variables.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0125118