13 th International Conference on Computational Intelligence and Communication Networks 978-1-7281-7696-3/21/$31.00©2021 IEEE 66 DOI: 10.1109/CICN51697.2021.13 Duck Curve with Renewable Energies and Storage Technologies Giovani Manuel Pitra, K.S. Sastry, Musti Department of Electrical and Computer Engineering Namibia University of Science and Technology Windhoek, Namibia e-mail: giovanipitra24@gmail.com, mks.sastry@gmail.com Abstract—Duck curve phenomena occurs when solar energy in higher quantities is integrated into the power grid. This results in excess generation that cannot be delivered during peak hours and a part of the load that cannot be supplied during off-peak hours. This paper proposes a novel, 2-step methodology to determine the effects of duck curve and also to flatten the same. This methodology uses two well-known opensource platforms - SAM (System Advisory Model) and IRENA FlexTool. Data for the energy capacity addition is obtained from SAM and optimization is done with FlexTool. A simple system is considered with a typical load profile and different energy sources. A few case scenarios are considered to demonstrate the effectiveness of the proposed approach and results are summarized. Keywords-Duck curve; Renewable energy technologies; System Advisory Model; IRENA FlexTool; Unit commitment; Energy mix. I. INTRODUCTION Over the last two decades, significant efforts have been invested in renewable energy technologies (RETs). Among the different RETs, solar photovoltaic (PV) is usually an attractive option due to its unique relationship to system integration, its decreasing cost, performance improvements, and its huge unexploited resource potential [1]. Traditionally, the increase and decrease of the load is met by optimally dispatching the units of a coal fired, natural gas, or oil-fired power plant, as well as from hydroelectric power plants. However, with the advent of RETs, it is common to use solar PV segment of energy to meet peak loading conditions. In many cases, excess amounts of solar energy during peak hours can result in negative loading conditions since supply can be far greater than the load itself. This leads to the formation of duck curve phenomena, which is generally illustrated using a typical plot of load and generation over a day and a pattern that resembles a duck can be seen. Fig. 1 shows a typical formation of duck curve condition as result of varying solar PV penetrations. Naturally, PV power starts ramping up as the sun rises in the morning, reaches its peak around mid-afternoon, and starts dropping as the sun sets in the late afternoon. This declining power in the late afternoon is unfortunately accompanied by a rising load, demanding an immediate supply of peaking power from fossil fuel-based power plants or any other possible sources. Thus, the addition of high penetration of PV to the power grid increases operational complexity and in fact results in contrasting conditions of lower load demand during mid noon and higher load demand in the evening hours. The net load, that is, the difference between electricity demand (load) and the portion supplied by the generated PV power, is what is referred to as the duck curve. Figure 1. Load Curve wih different levels of PV penetration. Fig. 2 demonstrates the duck curve at different levels of solar PV penetration. As it can be seen in the figure, the net load decreases significantly with the increase in PV penetration during peak hours. This may result in the curtailment of solar PV because of excess generation (when compared to load), as base load power plants are chosen to run 24/7 because of their ramping up and down constraints [11]. In most cases solar PV plants are privately owned by Independent Power Producers (IPPs) and the energy is sold through well-structured power purchase agreements that may not provide flexibility to utility engineers in dealing with operating conditions. Furthermore, industrial and commercial consumers are also prosumers, and this causes the duck curve to become even stepper [12], resulting in greater ramping problems [13]. 2021 13th International Conference on Computational Intelligence and Communication Networks (CICN) | 978-1-7281-7696-3/21/$31.00 ©2021 IEEE | DOI: 10.1109/CICN51697.2021.9574671