TY - JOUR ID - 141135 TI - Power demand estimation during pandemic times: The case of the COVID-19 in Tehran, Iran JO - Journal of Energy Management and Technology JA - JEMT LA - en SN - AU - Najibi, Sasan AU - Najafi, Mojtaba AU - Mallaki, Mehrdad AU - Cheragh Shirazi, Najmeh AD - Department of Electrical Engineering, Bushehr Branch, Islamic Azad University, Bushehr, Iran Y1 - 2022 PY - 2022 VL - 6 IS - 3 SP - 176 EP - 187 KW - COVID-19 pandemic KW - Power demand forecasting KW - principal component regression KW - twin support vector machine KW - quantile regression DO - 10.22109/jemt.2021.278673.1289 N2 - The impact of the COVID-19 pandemic on power demand has been studied in some countries. In this study, we investigate the effects of the COVID-19 pandemic on power demand in Tehran, Iran. Hence, power demand variations between 2016 and 2020 are investigated in this research. Results indicate that the effects of the COVID-19 pandemic on power demand vary from month to month and day to day, depending on various factors such as government limitations and the COVID-19 mortality. It is observed that power demand annual growth is changed during both the COVID-19 pandemic and financial crisis. For instance, the average power demand growth in 2020 is 1.03%, while was 4.96% in 2019. Also, most power demand forecasting algorithms have been developed for the normal situation; therefore, we introduce two forecasting algorithms to forecast power demand. The first algorithm is developed based on the principal component regression (PCR), and the second is developed based on the twin support vector machine and quantile regression (TWSVQR). The PCR is selected due to its simplicity and high performance. The proposed PCR model considers daily, annual, and biennial power demand variation rates. The advantage of the TWSVQR method is that it is so accurate, requires a small training dataset, considers various factors for forecasting power demand, and is robust against outlier data. Finally, we investigate our proposed algorithms to forecast power demand in Tehran. Results illustrate proposed algorithms can predict power demand UR - https://www.jemat.org/article_141135.html L1 - https://www.jemat.org/article_141135_f19d7bf0b92bdf94206ae2be291e8003.pdf ER -