key: cord-1042051-dunmqn6l authors: Luo, Shihua; Hu, Weihao; Liu, Wen; Liu, Zhou; Huang, Qi; Chen, Zhe title: Flexibility enhancement measures under the COVID-19 pandemic – a preliminary comparative analysis in Denmark, the Netherlands, and Sichuan of China date: 2021-09-30 journal: Energy (Oxf) DOI: 10.1016/j.energy.2021.122166 sha: 315e44aa1bcbc1d89bab322fffc87652b1c0ac5e doc_id: 1042051 cord_uid: dunmqn6l The COVID-19 pandemic affects all the aspects of modern society worldwide, especially in the power sector. Measures of flexibility enhancement are regarded as solutions to guarantee reliable and flexible electricity supply in such an emergency. This study aims at investigating the impact of flexibility enhancement measures (electricity storage and flexible demand) in different situations of the preliminary COVID-19 pandemic. Case studies in different regions (Denmark, the Netherlands, and the Sichuan province of China) are conducted and assessed using the hourly simulation tool EnergyPLAN. These regions own different electricity supply mix and level of renewable electricity. It is found that the flexible demand measure within one day or one week can hardly eliminate the electricity imbalance caused by either the pandemic or the increasing renewable electricity. The monthly flexible demand is effective for balancing, but its potential in these regions is not enough. However, electricity storage measure enhances the electricity balance even during the most extreme situation of the pandemic. From the economic perspective, electricity storage measure leads to an increase of up to 15% in total system costs, while flexible demand measure has a negligible effect on costs. This study serves as the first step to understand the performance of flexibility enhancement measures in the power sector under the shock of a pandemic. the share of wind power and PV reached 45% in Denmark and 20% in Spain by the end of 2016, and it is expected to double to higher than 10% in large power systems (such as China and the US) by 2022 [15] . However, strong dependence on the weather results in the intermittence of renewable electricity. A higher penetration share of renewable electricity in the power sector can aggravate the electricity imbalance. A reliable and resilient electricity supply plays an essential role during this kind of emergency. Maintaining a sustainable electricity supply and its contribution to people's regular lives should be an urgent priority. Measures of ensuring flexibility in the power sector are considered a good choice. They include supply-side and demand-side management, grid-side and system-wide storage measures [16, 17] . For the demand side, flexible electricity demand measure provides opportunities for reducing or temporally defer some load to extend the flexibility of electricity demand. The importance of providing demand-side flexibility is highlighted in the power system regarding electricity balance and system costs. It is estimated by International Energy Agency (IEA) that nearly 185GW of flexible demand can realize cost-effectively by 2040 [18] . Electricity storage measure serves as one of the most mature and widely used flexibility enhancement measures. It includes different types of technologies such as pumped hydro storage, utility-scale battery, compressed air energy, and so on [19] . The existing analyses on the impact of these flexibility enhancement measures on electricity supply and demand focused mainly on the individual measure and less on the comparison between flexible demand and electricity storage measures on electricity balance, especially in the context of COVID-19. Flexibility enhancement measures could reduce the temporal electricity imbalance caused by both the shock of pandemic and the expansion of renewable electricity. Researches concerning renewable energy integration and flexibility enhancement measure over the world widely exist [20, 21] . However, it is not clear the extent to which the COVID-19 pandemic affects the power sector in terms of renewable electricity supply, especially the impacts on an hourly basis. Our hypothesis is that the renewable power supply share will increase because of COVID-19. The level of changes varies in different types of supply and different power systems. It is interesting to investigate and compare the change of renewable power supply in power systems with different J o u r n a l P r e -p r o o f renewable power share. Second, the role of electricity storage and flexible electricity demand measures in securing the electricity balance has not been evaluated in the context of the COVID-19 pandemic. The effectiveness of these measures in promoting largescale renewable energy integration in a pandemic situation is not apparent. Based on the discussion above, the purpose of this paper is twofold: 1) to investigate the impacts of the COVID-19 pandemic on the existing power sectors in regions with different electricity supply mix and level of renewable electricity and 2) to evaluate the capability of two measures in ensuring the electricity balance under the influence of two factors, i.e., renewable electricity integration and the COVID-19 pandemic. Three regions are selected as case studies, namely Denmark, the Netherlands and the Sichuan province of China. Hourly simulations for the electricity supply and demand are conducted for these regions by the EnergyPLAN model. First, the impact of the preliminary COVID-19 pandemic in the power sector in the above regions is simulated and analyzed, emphasizing the influence on renewable electricity. Second, we investigate how flexibility enhancement measures (electricity storage and flexible demand) could affect the electricity balance and determine the capacity requirement of these measures under the impact of higher renewable electricity penetration shares and more difficult pandemic situations. The main novelty and contribution of this work are listed below:  It quantitively presents and compares the impacts of COVID-19 on the renewable power supply in different power systems.  It is the first attempt to evaluate the effectiveness of two measures in securing electricity balance in the context of COVID-19.  It facilitates the decision-making process for energy suppliers and policy-makers in a similar urgent crisis. The rest of this paper is organized as follows. Section 2 displays the case areas and their lockdown measures during the preliminary COVID-19 pandemic. Section 3 presents the methodology including the modelling tool and relevant methods. Section 4 shows the data and impact analysis of preliminary COVID-19 pandemic on the power sector. Section 5 displays the results of the two measures. The discussion and conclusion are shown in Section 6 and Section 7, respectively. The reasons for taking the regions can be summarized as follows: 1) The three regions are all affected by the COVID-19 pandemic in 2020; 2) The three regions have different renewable electricity shares, which represents that these regions are in different stages of renewable energy development; 3) The three regions have different electricity supply mix. This is helpful to the flexibility enhancement analysis under different situations. A brief introduction of these reasons is provided as follows. Denmark is one of the northern European countries, owns an area of 43,000 km 2 . In 2018, the total population in Denmark was around 5.7 million and the population density reached 135.6 people/ km 2 (Fig. 1a) . In the past two decades, the electricity supply in Denmark had a significant change: the considerable decline of coal electricity generation and the rapid expansion of wind power. Moreover, Denmark is excellent at integrating variable renewable energy due to the high interconnection and flexible power system [22] (see Fig. 1d ). The proportion of electricity generation of each technology in Denmark in 2019 is shown in Fig. 1g . The first medical case of the COVID-19 pandemic in Denmark was confirmed on February 27th, and then the pandemic rapidly spread in early March. Lockdown measures such as temporary border control, traffic restriction and closure of public places were claimed by the government to curb the rise of infected people. Daycare, schools, and educational institutions have been closed since March 13th, and the closure period was originally two weeks, which has been extended to April 13th [23] . The number of J o u r n a l P r e -p r o o f infected people started to fall at the end of April and the daily number only reached 22 on May 31st. However, due to the relaxation of restrictions, the situation of the pandemic was likely to worsen again and the total infected patients in Denmark were around 16600 by the end of August [24] . The period from March to May in 2020, which was most affected by the lockdown measures, is selected as the sample. Located in the western part of the European continent, the Netherlands has an area of 41,000 km 2 and the total population was 17.4 million in 2019, with a population density of nearly 515 people/ km 2 (Fig. 1b) . The electricity supply in the Netherlands relies heavily on fossil fuels. Natural gas serves as the most crucial fuel in the energy supply mix and accounts for a large share of electricity generation. Renewable energy in the Netherlands developed rapidly, and the share of renewable energy to total energy supply doubled between 2008 and 2019 [25] . The share of renewable electricity to the electricity generation from 2005 to 2019 is presented in Fig. 1e , along with the proportion of electricity generation of each technology in the Netherlands in Fig. 1h . Similar to Denmark, the first case of the pandemic in the Netherlands occurred on February 27th. Lockdown measures were implemented by the government in Netherland. Different from Denmark as the first European country to ease the restriction, the Netherlands extended the effective period of stringent lockdown measures to April 28th [26] . In short, the number of infected people in the Netherlands shares a similar increase tendency with that of Denmark. The daily number decreased in April and was only 185 on May 31st [27] . Thus, the period from March to May in the Netherlands is chosen as the sample for the same reason. Sichuan is one of the largest provinces in China with an area of 486,000 km 2 and its population has reached 83.8 million by the end of 2019 (Fig. 1c) . The electricity supply in the power sector of Sichuan province is hydro-dominated, because of the rich hydropower resources. Other renewable energy resources such as wind power and solar power are also abundant, but they are not fully exploited in the present [28] . It is worth mentioning that renewable electricity penetration in Sichuan has passed the stage of rapid development and reached a very high share of 88.4% in 2018 (see Fig. 1f ). The proportion of electricity generation of each technology to the whole electricity supply in Sichuan in 2018 is provided in Fig. 1i . The first medical case in Sichuan appeared on January 11th. The government advocated behaviors (such as wearing masks) and implemented stringent lockdown measures as soon as possible to prevent the further spread of the epidemic [29] . After around three months of epidemic control and management, the medical cases related to the COVID-19 pandemic in Sichuan province reduced to zero. This indicates that people's lives affected by the pandemic gradually returned to normal. The number of infected people reached its peak in early February and the total number of COVID-19 pandemic patients in Sichuan was 660 as of the end of August [30] . Thus, the most serious period (from January to March) of the pandemic in Sichuan is selected to compare with the same period in 2019 to show the impact of the preliminary COVID-19 pandemic. J o u r n a l P r e -p r o o f , and Sichuan (f); the proportion of each electricity generation technology to total electricity generation: Denmark (g), the Netherlands (h) and Sichuan (i). The simulation steps intend to explore the performance of the flexibility enhancement measures under the background of the COVID-19 context and increased renewable electricity. The first step is the scenario design in the three regions. It aims at simulating the impact of the preliminary COVID-19 pandemic on the power sector, and it also considers the electricity demand uncertainty. The second step is the renewable electricity penetration in the three regions. The critical factor in this step is the determination of the maximum renewable electricity penetration share. The lockdown measures implemented by governments led to a significant impact on the actual power system. The most significant impact of the preliminary COVID-19 pandemic on the power sector is the change in electricity demand and hourly distribution. Due to the lack of extensive storage measures in the power system, it is required that the electricity supply and demand be the same in real-time. As a result, the electricity supply is required to change during the period of this pandemic, which may result in an imbalance of electricity. Actually, how the COVID-19 pandemic affects the power sector is complicated. Many studies utilized the changes in electricity demand and hourly distribution to reflect the impact of the pandemic by comparing the electricity demand in 2019 and 2020 [31, 32] . Thus, the data of electricity demand and hourly distribution in the three regions are collected and summarized to explore the effectiveness of flexibility enhancement measures in the context of COVID-19. However, it is not certain that whether the electricity demand will fall or rise in the power sector facing the next pandemic outbreak as the measures will vary. Considering this uncertainty, four alternative scenarios are constructed with the electricity demand increased or decreased by 5% and 10%, respectively. In addition, a reference scenario is established for the purpose of comparison. Details of the above scenarios in regions are presented in Table 1 J o u r n a l P r e -p r o o f As indicated in the introduction, the current renewable electricity penetration shares in regions are different: Denmark (60%), the Netherlands (25%) and Sichuan (88%). Here, the renewable electricity penetration share is identified as the proportion of renewable electricity to the total electricity supply. In this study, it is essential to determine the maximum renewable electricity penetration share based on the electricity supply structure in each region. A certain percentage of thermal electricity production in the total electricity production has been assigned to ensure grid stabilization. In general, thermal electricity production will decrease if renewable electricity increases in the electricity supply. However, owing to the intermittency of renewable electricity and the requirement of grid stabilization, a larger renewable electricity share may inversely increase thermal power production. In this case, renewable electricity should not be added to the system because the increase of renewable electricity will increase the need for thermal power and cannot realize the decrease of CO2 emissions. Thus, marginal thermal electricity production is chosen as the factor in this study, influencing the feasible penetration share of renewable energy. The maximum renewable electricity penetration share is determined by the formula (1) as below: where MTE(α) i refers to the marginal thermal electricity production of the power sector under the renewable electricity share of α (%) in scenario i; TE(α) i and TE(α-5%) i represent the thermal electricity production of the power sector in scenario i when the renewable electricity share reaches α and α-5%, respectively; α indicates the renewable electricity penetration share, its initial value serves as the penetration share when a mismatch of electricity supply and demand first occurs. The maximum share is identified as α before the MTE(α) becomes positive during the increase of renewable energy penetration. The MTE(α) i of scenarios in the three regions is calculated when the renewable electricity penetration share (i.e. α) increases from its initial value. The results are shown in Appendix A. The indicator of marginal thermal power production is created to illustrate the comprehensive effects of increasing renewable electricity, including the benefits of decreasing thermal power production and CO2 emissions. The result of maximum renewable share cannot reflect the future renewable electricity penetration in the actual power system but aims at displaying the possibility of renewable electricity expansion based on the current power system structure. This indicator is discussed further in the Discussion section. The maximum penetration share of scenarios for three regions is provided in Table 2 . In the three regions, electricity demand is divided into individual sectors, namely the residential sector, the commercial and the industrial sector. Firstly, electricity demand is uniformly classified by sector to assess the flexible demand potential in each sector. Then, a technical method pioneered and proposed in [33, 34] is applied to assess the flexible demand potential. This method divides the electricity demand into several processes firstly and estimates the flexible electricity demand potential by determining whether the electricity demand of individual processes can be transferred temporally. Storability and controllability are identified J o u r n a l P r e -p r o o f as two main factors that determine the value of flexible electricity demand in the residential and commercial sectors. Particularly, the processes in the industrial sector account for an additional criterion: independence. An industrial process that is closely related to other processes rather than independent may be a hurdle for the flexibility of electricity demand. For the residential sector, the electricity demand of refrigerator and washing equipment is responsible for a large share of electricity demand in this sector and have the potential of being converted to flexible demand. Electrical processes linked to the behavior of consumers meet neither of the requirements of storability and controllability. Their electricity demand is regarded as inflexible. For the commercial sector, the demand for refrigerators, ventilation and space heating can be transferred to flexible electricity demand. As for the industrial sector, ventilation, fans and refrigerator processes have the possibility of being flexible demand. However, ventilation and refrigerator processes in the industrial sector have other uses, which are related mainly to other industrial processes. Thus, only half the electricity demand of this kind of process can be classified as flexible demand. A summary of flexible electricity demand types by sector is shown in Fig. 3 . EnergyPLAN tool can simulate the flexible electricity demand in the power sector with three frames: one day, one week and one month (4 weeks). The flexible demand is distributed evenly by the time frame and is not allowed to delay from one to another. Actual flexible electricity demand is mainly determined within a time frame of some hours or one day, and rarely one week and one month. The share of the electricity demand of individual sectors to total electricity demand for each region is collected to quantify the flexible electricity demand potential. For the share of electricity demand that can be transferred into flexible demand in each sector, industrial investigations and correlation researches are employed to obtain and summarize the transferred share in Denmark, the Netherlands and Sichuan, respectively. Key assumptions of the above shares are provided in Appendix B. The estimated flexible electricity demand potentials of the three regions during the three months period of the COVID-19 pandemic are presented in Table 3 . The energy system simulation and analysis tool EnergyPLAN are widely used to formulate energy system strategies, determine the appropriate penetration share of renewable energy and evaluate the feasibility of advanced technologies [35, 36] . The reasons for taking the EnergyPLAN tool are listed as following:1) It includes various conventional and advanced electricity generation technologies to meet the need of the power sector simulation; 2) It supports the implementation of flexibility enhancement measures such as electricity storage and flexible electricity demand in the power sector; 3) It simulates the power sector on an hourly basis, and calculate the hourly, weekly, monthly and yearly electricity balance between supply and demand. As for the endogenous logic, EnergyPLAN is a deterministic input/output tool that relies on manual heuristics rather than iteration, and it can obtain the simulation results in a few seconds with a time resolution of one hour [37] [38] [39] . The schematic diagram of the EnergyPLAN tool is shown in Fig. 4 . The hourly profile of electricity demand and supply in Denmark and the Netherlands is gathered from the Transparency Platform of the European Network of Transmission System Operators of Electricity (ENTSO) [40] , as well as the installed capacity of electricity generation technologies. The installed capacity and hourly distribution data of electricity supply in Sichuan are obtained from [41] . The cost data in the EnergyPLAN tool includes fuel cost, investment cost, fixed and flexible operation and maintenance (O&M) cost. The investment cost of different technologies in Denmark and the Netherlands is summarized on the basis of studies of the Danish Energy Agency [42, 43] . The investment cost of different technologies in Sichuan is gathered from industrial reports and researches [44] . The investment and O&M cost of technologies in the three regions are also presented in Tables 4. The fuel cost is based on the projection of IEA [45] , and the fuel cost of each region is shown in Table 5 . In addition, the interest rate in this study is set at 6%. The validation of the reference model of the three regions plays a vital role in the flexibility enhancement measure analysis. The simulation results are compared with the actual data to verify the accuracy of the EnergyPLAN tool. The comparison of electricity supply and demand between the actual value and EnergyPLAN simulation in Denmark, the Netherlands and Sichuan are given in Appendix C. It can be seen that the modeled electricity generation from thermal plants, wind and PV are all within the expected margins (lower than 1% difference). Therefore, the reference model can correctly simulate the power system in the above regions. Changes in the power sector in the Netherlands in terms of the electricity supply and demand, and concrete proportions during the COVID-19 pandemic are illustrated in Fig. 6 . The electricity demand in the Netherlands has an apparent reduction each month during the pandemic: March contributes 0.53 TWh (decrease by 5.1%), April 0.7 TWh (7.7%) and May 0.9 TWh (9.3%). The reduction in electricity demand is increasing with the spread of the pandemic. This tendency causes the share of electricity demand in May to reduce to 32.36%, with the largest decline among these three months (Fig. 6b ). In addition, the total electricity demand decreases by 2.13 TWh, which is 7.3% lower than the electricity demand in the same period in 2019. Thermal electricity production in the Netherlands during the pandemic gains a massive decline of 3. Critical excess electricity production (CEEP), caused by the temporal mismatches of electricity demand and supply, performs the key indicator in evaluating the electricity balance. CEEP of these scenarios under different renewable electricity penetration share is illustrated in Fig. 8 . Note that the transmission lines connected to neighboring countries or provinces are allowed in this study. Electricity imbalance will occur when excess electricity production has no choice of exporting or storage. From the perspective of renewable electricity, the amount of CEEP increases with the improvement of renewable electricity penetration share in each scenario. It indicates that higher renewable electricity integration will lead to a huge electricity imbalance in all three regions. On the other hand, CEEP in the scenario with greater electricity demand is higher than the scenario with less demand in the same renewable penetration share. This is because renewable electricity production is required to increase to reach the same penetration share in a scenario with greater electricity demand. The CEEP is not proportional to the renewable electricity share and the electricity demand but increases growth rate. Meanwhile, the growth rate of CEEP differs by region. As indicated in Fig. 8a , CEEP increases from 0.01 TWh to 0.03TWh when the penetration share of renewable electricity increases from 55% to 60% in scenario EDD-c. However, the growth of CEEP is 0.19 TWh with the share increasing from 60% to 65% and the growth comes to 0.7 TWh in the next increase in the penetration share. When looking at different regions, the Netherlands has the highest CEEP of 12.08 TWh in scenario EDN-e after realizing the significant growth in renewable energy integration. Despite the most minor power system among these regions, Denmark's CEEP growth is higher than Sichuan, and the highest CEEP occurs in the EDD-b scenario, reaching 8.16 TWh at 80% renewable electricity penetration. Due to the minor increase in penetration share (increase only by around 5%), the CEEP growth in Sichuan is the lowest. The capacity requirements of electricity storage for achieving the electricity balance in three regions under different shares of renewable electricity penetration are presented in Fig. 10 . Specific values of the capacity requirement of electricity storage measure are provided in Appendix D. Given that the CEEP in the Netherlands is relatively small in the early stage of renewable penetration, a feasible storage capacity of below 15 GWh can ensure the electricity balance with a CEEP of 0.1 TWh. When the penetration share exceeds 40% in each scenario in the Netherlands, the electricity storage capacity increases significantly with an increasing installed capacity to cope with the CEEP over 1 TWh. These additional storage capacities are obliged to store excess electricity and allocate it during off-peak hours. It means that the storage measure can provide an opportunity to temporally storing excessive renewable electricity, contributing to further CEEP curtailment. Moreover, the highest storage capacity in the Netherlands occurs in the EDN-e scenario at 55% renewable electricity penetration, reaching 291 GWh and the highest capacity in Denmark occurs in the EDD-b scenario at 80% penetration, 410 GWh. Compared with the Denmark scenarios (Fig. 10a) , the storage capacity requirement in Sichuan is more minor, even with a higher J o u r n a l P r e -p r o o f CEEP in the power sector (Fig. 10c) . When faced with a similar CEEP of about 0.2 TWh (0.21TWh in scenario EDD-c at 65% renewable penetration, 0.22 TWh in scenario EDS-d at 85% renewable penetration), the storage capacity in Sichuan serves as only 10 GWh, lower than that in Denmark by 8 GWh. This is because the transmission line capacity in Sichuan is far beyond the line capacity in Denmark and the Netherlands. Consequently, the reduction of CEEP can be realized in Sichuan by few storage capacities, as shown in Fig. 10c . Appendix Tables D2 and D3 . Flexible demand within one week works better than daily flexible demand, but it is also powerless in decreasing large CEEP in the power sector. As for the flexible demand within one month, it has the best performance in ensuring electricity balance than daily and weekly flexible demand. However, the monthly flexible demand requirement is responsible for nearly 28.6% in Denmark, 17.9% in the Netherlands and 9.3% in Sichuan of total electricity demand. Actual monthly flexible demand cannot account for such a high share, and it is considered unsuitable for eliminating electricity imbalance. J o u r n a l P r e -p r o o f The system cost by cost types of each scenario after introducing electricity storage is provided in Fig. 12 . The cost growth rate due to the introduction of electricity storage is shown in Fig. 13 . in negligible costs in the existing power system. Therefore, the cost of the power system after applying flexible demand measures is considered the same as the original power system. From an economic perspective, the flexible demand measure performs best than electricity storage. Interconnection capacity can strongly influence the storage requirement of flexibility enhancement measures. For each region, the impact of different interconnection capacities is evaluated. In the sensitivity analysis, the interconnection capacity is changed from 0.9 times of actual capacity to 1.1 times. An overview of values in the sensitivity analysis is presented in Table 6 . The scenario of decreasing electricity demand by 5% in the three regions (i.e. EDD-b, EDN-b and EDS-b) is selected. The different interconnection capacity is applied in each scenario for conducting this sensitivity analysis. The results of sensitivity analysis for interconnection capacity are provided in Fig. 14 and Fig. 15 . J o u r n a l P r e -p r o o f It can be seen from the figures that a lower interconnection capacity leads to the growth of capacity requirements of flexibility enhancement measures. In comparison, a higher interconnection capacity leads to the reduction of capacity requirements. Taking Denmark as an example, increasing the interconnection capacity from 3500 MW to 3850 MW results in a reduction of storage capacity of electricity storage measure from 35 GWh to 24 GWh at the renewable share of 70%, which is responsible for nearly 31.43%. However, the reduction of the interconnection capacity from 3500 GWh to 3150 GWh results in an expansion of storage capacity by 23 GWh, accounting for 65.71% of the initial storage capacity requirement. The difference of storage capacity fluctuation between the interconnection capacity expansion and reduction is because of the hourly distribution of excess electricity production. The expansion of excess electricity production due to lower interconnection capacity is larger than the decrease due to higher interconnection capacity. Therefore, decreasing the interconnection capacity in each scenario brings a more significant change in storage capacity requirements. Meanwhile, a similar tendency occurs in the performance of flexible demand measure under such situations. Increasing the interconnection capacity from 6130 MW to 5520 MW leads to decreased flexible demand by 2.3 TWh in scenario EDN-b at the renewable share of 35% while decreasing the interconnection capacity from 6130 MW to 5520 MW leads to an expansion by 3.2 TWh. This should also be attributed to the distribution of excess electricity distribution. Considering that the daily flexible demand requirement is relatively small, changing the interconnection capacity can cause a more significant influence on flexible demand measure than electricity storage measure in ensuring the electricity balance. In summary, lower interconnection capacity rather than higher interconnection capacity has a more apparent effect on the capacity requirement of flexibility enhancement measures. Compared with electricity storage measure, flexible demand measure reacts more significantly to a change in interconnection capacity. The hydropower in Sichuan has seasonal features based on its generation. Seasons can correspond to three periods, namely low- Table 7 . It is worthwhile to mention that the data used in previous scenarios in Sichuan is in the low-water period, since the research time range in this paper is from January to March. The results of sensitivity analysis for interconnection capacity are provided in Tables 8-9 . Obviously, varying from the low-water period to the other two periods significantly increases the capacity requirements of Sichuan's two flexibility enhancement measures. In the low-water period, the storage capacity requirement of electricity storage measure in scenario EDS-a for eliminating electricity imbalance is 3 GWh. The capacity requirement in the normal-water period serves as 25 GWh, increased by 22 GWh and the capacity requirement in the high-water period is 310 GWh, increased by 307 GWh. The capacity requirement of different periods doesn't increase proportionally but increases with a dramatic growth rate. Improving the hydropower generation only with electricity demand unchanged will lead to more significant excess electricity production as the time of hydropower generation is concentrated and the power generation per unit time is huge, which easily exceeds the electricity demand. The seasonal hydro can affect the capacity requirement of electricity storage measure and strongly influence that of flexible demand measure. For example, changing the low-water period to the normal-water period raises the capacity requirement of flexible demand from 6.4 TWh to 21 TWh in scenario EDS-a. In the high-water period, abundant hydropower makes the renewable electricity share in Sichuan reaching nearly 100%. Electricity storage measure can ensure electricity balance in the high-water period with a higher storage capacity, while flexible demand measure is powerless. From the perspective of technical feasibility, the electricity storage measure performs better than the flexible demand measure facing the condition of seasonal hydro in Sichuan. The higher storage capacity requirement of electricity storage measure translates into more reliability and flexibility in the Sichuan J o u r n a l P r e -p r o o f power system. The influence of the preliminary COVID-19 pandemic on the power sector varies greatly in different regions. The electricity demand in the Netherlands and Sichuan reduce significantly while electricity demand in Denmark is minor. For countries like the Netherlands, the pandemic increases renewable electricity production. Especially the PV technology, considering its falling price in recent years, the rapid expansion of PV power is considered to be sustainable in the Netherlands. As for the regions owning a high share of renewable penetration, such as the Sichuan province of China, the decline in electricity demand and the implementation of confinement measures may reemphasize the importance of thermal electricity production. The requirement of gird stabilization in such emergencies and the seasonal fluctuation of renewable electricity can result in the idleness of electricity generation equipment and the increase of renewable investment burden. As for the impact of flexibility enhancement measures on the power sector during the preliminary COVID-19 pandemic, it is found that the flexible electricity demand measure has a limited effect on eliminating the electricity mismatch under high renewable electricity penetration and severe pandemic situations. Both the flexible electricity demand within one day and one week contribute little to maintain the electricity balance. The monthly flexible electricity demand is helpful but unrealistic due to its poor potential in all regions. Nevertheless, more electricity demand flexibility can release the strong dependence of electricity supply on neighboring regions by decreasing electricity importation and exportation. More importantly, this measure can be realized without infrastructure establishment and huge investment costs, which performs a more cost-effective option. Electricity storage measure is a more effective measure to guarantee the electricity balance in different regions. Even in a scenario suffering the most severe pandemic influence, electricity storage with a storage capacity of 420 GWh can reduce the CEEP to zero. Moreover, increasing the transmission line capacity, though it can diminish storage capacity requirement, is not regarded as the solution to ensure the electricity balance. The harm of electricity imbalance has not been eliminated or solved by exporting electricity but is shifted to neighboring regions. The assumption of maximum renewable electricity penetration share in scenarios plays an important role in the flexibility enhancement analysis. The indicator for determining the maximum penetration share is the change in thermal electricity production. The technical assumption of maximum renewable electricity has been conducted in many studies [48, 49] . The difference between this study and previous studies is the indicator. Previous studies use the COMP (namely, the ratio between the marginal primary energy supply (PES) to the marginal critical electricity excess production (CEEP)) as the indicator to determine the feasible penetration level of renewable electricity. These two indicators are similar, and their difference occurs in: The indicator of marginal thermal electricity production focuses on the change in thermal electricity and CO2 emission. The indicator of COMP focuses on the change in the primary energy supply and system fuel efficiency. Thus, the marginal thermal electricity production indicator in evaluating the maximum share of renewable electricity is regarded as a reasonable assumption in this study. It needs to be noted that this paper is exploratory simulation research aiming at advancing more knowledge of flexibility enhancement measures if the pandemic becomes more severe and the renewable electricity increases rapidly. This study focuses more on the flexibility enhancement analysis based on the actual data of electricity demand and hourly profile during the pandemic in the three regions. The initial findings at the regional level can help policy designers and electricity suppliers to understand the performance of flexibility enhancement measures under severe crisis and higher renewable energy integration. The study results are helpful to have a benchmark for the setup of a secure electricity supply system and infrastructure to withstand the extreme crisis in the future. From an economic perspective, the introduction of electricity storage measure brings about cost expansion in the three regions. The highest cost growth due to the implementation can reach nearly 15%, which occurs in Denmark. Meanwhile, flexible demand measure results in a negligible cost expansion in the existing power system as it makes effects without adding extra devices. Considering the sensitivity analysis of interconnection capacity and seasonal hydro, flexible demand measure reacts more significantly to a change in interconnection capacity in the three regions compared with electricity storage measure. Flexible demand measure cannot deal with the extensive excess electricity production due to seasonal hydro, while electricity storage measure can ensure electricity balance in the high-water period with a higher storage capacity. To sum up, flexible demand measure is more economically effective and is helpful to reduce the exported and imported electricity. However, electricity storage measure performs better than flexible demand measure in terms of ensuring electricity balance when facing similar crises like the preliminary COVID-19 pandemic and facing changes in interconnection capacity and seasonal hydro. Therefore, the long-term plan for deploying more electricity storage measures, especially in regions with low development levels of renewable energy, should be prioritized if more renewable electricity integration is to be accomplished in the power sector. Industrial sector 59. 17 5 Appendix C. The validation of EnergyPLAN model in the simulation of the power sector. WHO. Coronavirus Disease (COVID-19) Outbreak Situation Coronavirus Travel Restrictions and Bans Globally: Updating List Isolation, quarantine, social distancing and community containment: Pivotal role for old-style public health measures in the novel coronavirus (2019-nCoV) outbreak Impact of COVID-19 outbreak measures of lockdown on the Italian Carbon Footprint A hybrid multi-objective optimizer-based model for daily electricity demand prediction considering COVID-19 IEA. Covid-19 impact on electricity Dynamic dependence and risk connectedness among oil and stock markets: New evidence from time-frequency domain perspectives When pandemics impact economies and climate change: Exploring the impacts of COVID-19 on oil and electricity demand in China Bangladesh power supply scenarios on renewables and electricity import Flexibility categorization, sources, capabilities and technologies for energy-flexible and grid-responsive buildings: State-of-the-art and future perspective Low carbon transition pathway of power sector with high penetration of renewable energy Resilience of standalone hybrid renewable energy systems: The role of storage capacity Power system flexibility quantitative evaluation based on improved universal generating function method: A case study of Zhangjiakou The value of flexible load in power systems with high renewable energy penetration The electricity market in a renewable energy system The Danish government has taken measures to implement temporary border control operations The measures taken by the Dutch government to curb the spread of the COVID-19 pandemic may be extended COVID-19 pandemic in the Netherlands Research on some major issues of medium and long-term power development in Sichuan province How is the prevention and control of the COVID-19 pandemic in Sichuan Real-time dynamics of the COVID-19 pandemic in China Impact analysis of COVID-19 responses on energy grid dynamics in Electricity demand during pandemic times: The case of the COVID-19 in Spain Demand side management potential a case study for Germany Assessment and evaluation of flexible demand in a Danish future energy scenario Positive interactions between electric vehicles and renewable energy sources in CO2-reduced energy scenarios: The Italian case Decarbonizing a large City's heating system using heat pumps: A case study of Beijing EnergyPLAN: advanced energy systems analysis computer model The economic crisis and sustainable development: The design of job creation strategies by use of concrete institutional economics Energy system analysis of 100% renewable energy systems-The case of Denmark in years 2030 and 2050 Sichuan electric power production and market operation information report for the first quarter Exploring the pathways towards a sustainable heating system -A case study of Utrecht in the Netherlands Heat roadmap china: New heat strategy to reduce energy consumption towards 2030 Large-scale integration of wind power into the existing Chinese energy system Large scale integration of renewable energy sources (RES) in the future Colombian energy system This work was supported by the National Key Research and Development Program of China (2018YFE0127600). Appendix A. The marginal thermal electricity production of each renewable penetration share.