Portal de Periódicos da CAPES

Improving the traditional levelized cost of electricity approach by including the integration costs in the techno‐economic evaluation of future photovoltaic plants

2021; Wiley; Volume: 45; Issue: 6 Linguagem: Inglês

10.1002/er.6456

1099-114X

Elisa Veronese, Giampaolo Manzolini, David Moser,

Hybrid Renewable Energy Systems

International Journal of Energy ResearchVolume 45, Issue 6 p. 9252-9269 RESEARCH ARTICLEOpen Access Improving the traditional levelized cost of electricity approach by including the integration costs in the techno-economic evaluation of future photovoltaic plants Elisa Veronese, Corresponding Author Elisa Veronese elisa.veronese@eurac.edu orcid.org/0000-0003-2797-5716 Institute for Renewable Energy, EURAC Research, Bolzano, Italy Correspondence Elisa Veronese, Institute for Renewable Energy, EURAC Research, Bolzano 39100, Italy. Email: elisa.veronese@eurac.eduSearch for more papers by this authorGiampaolo Manzolini, Giampaolo Manzolini Energy Department, Politecnico di Milano, Milan, ItalySearch for more papers by this authorDavid Moser, David Moser Institute for Renewable Energy, EURAC Research, Bolzano, ItalySearch for more papers by this author Elisa Veronese, Corresponding Author Elisa Veronese elisa.veronese@eurac.edu orcid.org/0000-0003-2797-5716 Institute for Renewable Energy, EURAC Research, Bolzano, Italy Correspondence Elisa Veronese, Institute for Renewable Energy, EURAC Research, Bolzano 39100, Italy. Email: elisa.veronese@eurac.eduSearch for more papers by this authorGiampaolo Manzolini, Giampaolo Manzolini Energy Department, Politecnico di Milano, Milan, ItalySearch for more papers by this authorDavid Moser, David Moser Institute for Renewable Energy, EURAC Research, Bolzano, ItalySearch for more papers by this author First published: 04 February 2021 https://doi.org/10.1002/er.6456Citations: 2 Funding information: Provincia autonoma di Bolzano-Alto Adige 2014-2020, Grant/Award Number: EFRE/FESR AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Summary The levelized cost of electricity (LCOE) is a techno-economic parameter used to evaluate the cost of a kilowatt-hour of energy produced from a selected power plant. The initial investment, annual operation and maintenance costs together with the annual energy production are some of the input data needed to determine the LCOE. The most typical approach to calculate the LCOE does not account for the interaction of the new power plant with the existing energy system, assuming indirectly the power plant as stand-alone. This can be misleading in scenarios with high variable renewable energy sources (VRES) penetration as costs related to overproduction, reinforcement of the grid and additional efforts of existing fossil fuels power plants to satisfy the electricity demand that is not instantly covered by VRES production are not accounted for. The aim of this work is to define a general methodology of easy application for the estimation of these additional costs, called integration costs, of the photovoltaic (PV) technology with the corresponding parameter called system LCOE. In order to demonstrate the importance of the new definition, the methodology is applied to the future Italian energy system and PV sector foreseen for the year 2030. The Italian PV LCOE in 2030 calculated with the usual methodology ranges from 12.55 to 15.93 €/MWh, while the system LCOE can be as high as 22 €/MWh with a relevant increase by on average 50%. In case of addition of storage to PV systems, the system LCOE after the addition of the integration costs ranges from 45 to 51 €/MWh. However, even when batteries and integration costs are included, PV remains competitive compared to the market price. List of Abbreviations BESS battery energy storage system BIPV building-integrated photovoltaics CCGT combined cycle gas turbine DSO distribution system operator GSE Gestore dei Servizi Energetici LCOE levelized cost of electricity O&M operation and maintenance PNIEC Piano Nazionale Integrato Energia e Clima PV photovoltaic RES renewable energy sources TSO transmission system operator VRES variable renewable energy sources 1 INTRODUCTION Following the Paris Agreement of the year 2015, it has become clear that countries must face a substantial and fast energy transition to reduce environmental impacts and the unsustainable consumption of nonrenewable natural resources. In particular, the electricity production shall change from a centralized configuration based on fossil fuels to a more distributed system based on RES, among which solar and wind should play the major role. At the same time, high levels of safety and reliability of the electricity grid shall be guaranteed even with this strong penetration of intermittent and not directly controllable generation sources.1 Hence, new services will be asked from VRES sector in terms of better production forecast and dispatchability, for example, through the installation of storage systems, and active participation to the electricity market.2 The expected energy transition will bring undoubted benefits for both the environment and the society, thanks to the improvement of air and environment quality and the potential creation of new business sectors and jobs.3 However, it requires additional costs for the current energy system since investments in new generation capacity, grid infrastructures, and digitalization are necessary to adapt to this significant and rapid change.1 This study focuses on how these additional costs should be included in the future techno-economic evaluation of power plants to avoid that they will be completely socialized and paid by the community. The techno-economic evaluation of power plants is usually based on the Levelized Cost of Electricity (LCOE) that is equivalent to the cost of producing a kilowatt-hour with a selected type of power plant. In general, the LCOE is calculated as the total costs incurred by the power plant divided by the total energy produced during the lifetime. The costs typically include the initial investment, the operation and maintenance (O&M) expenditures, the fuel and consumable costs (when applicable), while the total amount of energy produced can be adjusted considering the degradation rate of the power plant and its components. This definition is described in Reference 4. For PV systems, other LCOE formulations have been proposed by References 5-7. The basic formulation can be extended as in Reference 5 with more details regarding the calculation of the annual electricity production and substituting the discount rate with the WACC (Weighted Average Cost of Capital). The formulation is here given just as an example of classical LCOE: (1)where CAPEX is the total investment expenditure of the system in the year t = 0, OPEX(t) the operation and maintenance expenditure in year t, WACCnom the nominal weighted average cost of capital per year, WACCreal the real weighted average cost of capital per year, Utilization0 is the initial utilization in the year t = 0 (without considering degradation), Degradation is the annual degradation of the nominal power of the system, n is economic lifetime of the system, and t the year of lifetime (1,2,…,n). In Reference 6, the study is focused on ground-mounted PV systems and the authors added the land costs (ie, costs related to the acquisition of the land required for the installation of the PV plant), the insurance costs, the tracking factor, and performance factor. The tracking factor adjusts the solar resource to the real incident solar energy as a function of the PV plant orientation and it is equal to one for optimally inclined and south-oriented modules. The performance factor, instead, converts the total available solar resource into the real amount of electricity produced by the system per Watt installed. In Reference 7, the residual value concept has been introduced to include the possible earnings coming from the disposal or resale of the power plant at the end of its useful life. The authors defined also a more complex LCOE involving other financial parameters that can be applied for the commercial and industrial PV sector such as: the project costs minus any investment tax credit or grant, depreciation, interest paid, loan payment, and the tax rate. It is possible to notice from these approaches that no parameters accounting for the interaction between the new power plant and the existing energy system are included. These simplified approaches can be appropriate for engineers to select the power generation technologies during the planning phase but they can be misleading when the LCOE is used by public institutions to identify the optimal energy strategies among different technologies because (a) it is strongly affected by the cost of capital and to a lesser extent by the technological parameters and (b) it considers the power plant as stand-alone not accounting for the impacts of connecting the power plant to the existing grid and energy system. Therefore, the classical formulation of the LCOE is not able to reflect the technical and economic challenges that must be faced with the significant increase of VRES production in terms of grid stability and change of the usual operating conditions of thermal power plants, which eventually affect the electricity market and prices. Some research studies8-12 have been proposing methodologies to overcome these limits by estimating the impacts of high VRES penetration on the LCOE. They have determined the so-called integration costs that can be combined with the LCOE to include the impacts of adding new intermittent generation in the existing energy systems. In this way, a new parameter can be defined which in this work is called "system LCOE" consistently with a previous work done by other authors in Reference 8. The system LCOE is defined as follows (2)where Δ represents the integration costs as the sum of balancing costs, grid costs, adequacy costs (or backup costs), full-load hours reduction, and overproduction costs. All these cost components are determined by the authors in Reference 8, with a top-down approach and a formalization of system LCOE with a mathematical definition of integration costs that directly relates to economic theory. For Reference 8, the integration costs are all the additional costs for the nonrenewable part of the power system when VRES are added. However, the formulation as given in Reference 8, might not be of easy implementation and the integration costs may also be split into different costs components applying a bottom-up approach instead of a top-down approach. For example, examining the geographical distribution of the added installed VRES capacity, there could be the need to extend as well as to reinforce the existing grid infrastructure to prevent problems like overvoltage or reverse power flows. This necessity is reflected into investments in transmission and distribution grids that are referred as grid costs.8-11 Additionally, since the VRES production is not directly controllable and programmable, its fluctuations are currently managed by fossil fuel power plants that change their role from baseload to peak plants when VRES production is not available or not enough. This means that fossil fuel power plants will be forced to operate at partial load conditions and the consequent costs arising are called balancing costs8, 9, 12 since additional balancing services are necessary to overcome the unpredictable VRES electricity output. The nonprogrammability of VRES production introduces also issues related to the system reliability and security of supply that can be represented by the capacity costs,9, 12 also called adequacy costs.8 Other aspects that can be included in the integration costs are related to storage addition,10, 11 VRES production curtailment,11 and profile costs,8 which put together the effects of VRES production on the fossil fuel power plants in terms of full-load hours decrease, overproduction costs, and backup costs. Other more recent attempts to go beyond the classical LCOE formulation are represented by Levelized Avoided Cost of Electricity developed by the US Energy Information Administration 13 and the value-adjusted LCOE (or VALCOE) developed by the IEA.14 LACE is an alternative assessment of economic competitiveness between generation technologies which is gained by considering the avoided cost, a measure of what it would cost the grid to generate the electricity that would be displaced by a new generation project. Avoided cost, which provides a proxy measure for the annual economic value of a candidate project, may be summed over its financial life and converted to a level annualized value that is divided by average annual output of the project to develop its levelized avoided cost of electricity. The value-adjusted LCOE (or "VALCOE") incorporates information about both costs and the value provided to the system. Based on the LCOE, estimates of energy, capacity, and flexibility value are incorporated to provide a metric of competitiveness for power generation technologies. This metric provides a more robust approach to compare dispatchable technologies and variable renewables. Both methodologies focus on the added value of power plants (eg, market electricity price, flexibility in providing services in terms of regulation, reserve power, and capacity) and not on the costs related to the integration of renewables in the energy system. Starting from the new emerging approach of system LCOE to the profitability analysis of VRES power plants, the aim of this study is to define a methodology to estimate the overall integration costs of VRES and include them to the classical LCOE calculation, understanding how the future VRES generation costs will be affected. The developed methodology is based on the bottom-up approach: starting from the effects of VRES on the power system, the integration costs are split into different cost components that are then mathematically defined. An energy modeling tool is used in this study to simulate future energy scenarios and evaluate the integration costs. Since these are strictly related to the power system and grid infrastructure, the methodology, which is general and applicable to any energy system, needs to be tested on a real case with actual numbers of investments and technologies. In this work, it is applied to the Italian case study, focusing the attention on the utility-scale PV sector and its future generation costs. 2 METHODOLOGY 2.1 System LCOE general definition Starting from the LCOE methodological improvements shown in References 8-12 and discussed in the previous section, the system LCOE is defined in this paper as the sum of power plant costs, that is, PV, or PV plus storage, and the integration costs. Integration costs are divided into two main subcategories, grid and balancing costs. The grid costs consider the investments required for the transmission and distribution grids to foster the transport of additional VRES electricity while guaranteeing at the same time the grid reliability and security of supply. Therefore, the grid costs are determined as the sum of reinforcing of transmission/distribution network, adequacy, and curtailment costs. The balancing costs, instead, accounts for impacts of additional VRES production on the existent fossil fuel power plants in terms of the efficiency decay and the start-up costs. Grid costs and balancing costs are strictly dependent on the energy system configuration, that is, grid infrastructure, number, and geographical distribution of the existing power plants. The system LCOE structure is schematically represented in Figure 1 and the corresponding formulation is reported in Equation (7). (3)where Cpp is the power plant costs, Cdistr the reinforcing distribution network costs, Ctrans the reinforcing transmission network costs, Cadequacy the adequacy costs, Ccurt the curtailment costs, Cdecay the decay of efficiency costs, and Cstart-up the start-up costs. All these cost components are expressed in €/MWh and are mathematically expressed in the following subsections. FIGURE 1Open in figure viewerPowerPoint Schematic representation of the system LCOE [Colour figure can be viewed at wileyonlinelibrary.com] The following section details the different costs components of the system LCOE. It must be stressed out that the definitions provided hereafter coincide theoretically for the PV and PV plus storage cases. However, the calculated values will be different between the two cases as the adoption of BESS significantly reduces the integration issues. This point will be better discussed in the following sections. 2.1.1 Power plant costs The power plant costs (Cpp) represent the electricity generation costs of a certain power production technology and are usually evaluated with the LCOE, the calculation method of which depends on the technology considered. In the case of PV power plant without storage system, these costs are calculated with Equation (1) as shown before; whereas, if the combined PV plus storage system is considered, the methodology proposed in Reference 13 is applied as summarized in Equation (4): (4)where Csystem,t is the total cost of the PV plus storage system at time t in €, Esystem,t is the sum of the electricity delivered by the storage at time t (Estorage), and the energy produced by the PV plant and directly consumed by the load at time t (Epvdirect) in MWh, Cpvsurplus the cost at time t for generating the PV surplus energy in €, Cstorage the storage cost at time t in €, Cpvdirect the cost at time t for generating the energy directly consumed by the load in €. The power plant costs are calculated subdividing the reference geographical area into smaller regions, called macro regions, to consider the change of the solar irradiation at different latitudes and obtain a more precise geographical estimation of Esystem,t and the corresponding LCOE. 2.1.2 Reinforcing distribution network costs PV technology is largely used on residential and commercial buildings to increase the self-consumption and reduce the energy bill costs. Therefore, most of them are connected to the distribution grid, that is, the low and medium voltage grid. However, the electricity network was originally meant to transport electricity from the power production units, typically large-scale fossil fuel power plants, to the distributed final users, allowing the electricity to flow in one direction through the distribution grid. For this reason, the widespread diffusion of PV plants at this lower voltage level may bring about problems related to the injection of electricity in a part of the grid infrastructure not designed for accepting it. To solve this issue, the distribution network might need to be reinforced and renovated with certain investment costs especially outside of urban areas. The reinforcing distribution network costs may need some network computations (eg, a power flow model of the distribution grid, but also short-circuit computations, protection settings, voltage drop computations) to be correctly estimated, but the creation of it is beyond the scope of this analysis. Another approach can consist of using investment costs determined in other works/projects. In this work, the figures determined in the PV Parity Project,10 which provides some numerical examples of reinforcing distribution network costs for different countries are adopted. 2.1.3 Reinforcing transmission network costs The transmission grid, like the distribution network, has faced new challenges in the recent years due to the spread of VRES production. On one hand, the increase of self-consumption at distribution level modifies the national demand profile1 and the residual VRES production injected might cause problems of overvoltage, reverse power flows at the connection points between distribution and transmission grids and dynamic issues related to the decrease of the system inertia and short-circuit levels. On the other hand, the diffusion of large-scale VRES power plants connected to the transmission grid stresses the grid infrastructure that shall accept an additional share of variable and intermittent electricity. Similarly to the distribution grid, sometimes these issues might be fixed by reinforcing grid infrastructures and expanding transport capacity of powerlines; however environmental concerns make more and more difficult to get permissions to build new lines, especially overhead lines. Other possible solutions are control systems and devices able to reroute power flows in such a way that, whenever possible, power system security is kept acceptable in the presence of large VRES injections. Further to new infrastructures (lines and substations), many flexibility tools can be used for this goal, ranging from new control devices, that is, phase-shifters and flexible AC transmission systems (FACTS), to tools provided by the electricity markets (Ancillary Service markets, for example, demand response, etc). The former solutions are characterized by investment costs that can be embedded into reinforcing transmission network costs.; the latter solutions have a cost currently very difficult to estimate, given that there is no experience about them all over the world. Like for the reinforcing distribution network costs, in this case, it is also necessary to compute the reinforcement actions needed to keep the power system security in any operating conditions. This computation requires many detailed network studies, basically power flow computations as well as dynamic stability studies, for the most frequent operating conditions. This makes it necessary, for each envisaged operating conditions, to carry out steady-state power flow computations, and to run dynamic simulations for the most probable and the most significant perturbations, in order to identify any reasonable bottleneck to be removed of mitigated. This should be done as a function of the VRES and storage systems penetration. For each binding technical constraint (either a congestion or an instability issue), it is necessary to identify the yearly number of hours it occurs, and the energy not supplied due to it (ie, its cost); at the same time, it is necessary to identify possible solutions and their cost. The TSO, then, can run a cost-benefit analysis (CBA) to identify the most effective solutions to mitigate or remove the most expensive technical issues at the transmission level. This task is of course possible only for a TSO, which has full knowledge of the massive data needed. A simpler approach which is adopted in this work being the development of a power flow model beyond the scope, consists of estimating the transmission costs from the investment planned by the national transmission system operator (TSO) for RES integration as given in its Annual Development Plan. The resulting formulation is the following Equation (5). (5)where InvTSO,RES int (m) is the total investment made by the national TSO for RES integration in the macro region m expressed in €, ProdVRES,2030 (m) the added production of VRES (wind and PV) expected in the future in the macro region m in terms of MWh, and PV_lifetime the service lifetime of PV power plants in years. 2.1.4 Adequacy costs The adequacy costs, which we consider in this analysis, are issues related to the system reliability and security of supply due to the increase of VRES penetration, are evaluated similarly to the transmission costs starting from the TSO investments aimed to guarantee the quality of the service when they are coupled with the RES integration objective. Also in this case, similar steady-state and dynamic studies are necessary for a reasonable estimation of these costs; like in the previous case, such analysis can only be carried out by a TSO and it is beyond the scope of this paper. The Equation (6) is applied in this case as follows (6)where InvTSO,Q&S (m) is the total investment made by the TSO for quality and security of the grid in the macro region m in €. 2.1.5 Curtailment costs The curtailment costs are introduced in this analysis as indirect costs to evaluate the economic losses due to the curtailed energy to prevent grid instability. This cost component is evaluated as a reduction of PV production directly in the power plant costs formulation. The reduction is represented by a percentage of PV curtailed for each macro region in respect of the macro regional PV production. The PV curtailment is estimated by using an energy system model that performs an hourly energy balance of the available power plants and the possible overgeneration that will appear in the energy system nodes characterized by high RES penetration. The PV overgeneration in each node is the macro regional curtailment used in this cost component calculation. The energy model used in this analysis is described more in detail in the Subsection 3.2. 2.1.6 Balancing costs The balancing costs are introduced to include the impacts of VRES on the existing fossil fuel power plants, which will be increasingly exploited to cover the peak demand arising when VRES are no more available or not enough. The balancing costs include in this analysis the decay of efficiency and the start-up costs. They are estimated by enlarging the energy system model to take into account the time-dependency of transient operations of fossil fuel power plants when the hourly dispatch optimization of the available energy sources is performed, as explained in Reference 14. The added time-dependent constraints are the start-up costs and ramp constraints as a function of the downtime hours of the plant and the type of technology, and the decay of efficiency at partial load that happens when the power plant is not working at nominal conditions. The latter is implemented as additional fuel consumption in respect of the nominal condition and the resulting decay of efficiency costs are calculated with the following Equation (7). (7)where ΔFadditional,u(t) is the additional fuel consumption of unit u at time t due to decay of efficiency in MW, Cfuel the specific fuel cost in €/MWh, and Ptot the total electricity generated by fossil fuel power plants in MWh. The start-up costs, instead, are estimated with the Equation (8). (8)where Cstart-up,Δt is the specific start-up cost depending on downtime Δt in €/MWh, Pnom the nominal power in MW and X(t) is a Boolean variable that returns 1 if the power plant at time t has been down for an interval of time equal or higher than the downtime Δt, otherwise it returns 0. The balancing costs calculated in this way are strictly related to the operational limits of fossil fuel power plants. It is also possible to directly use the prices coming from the balancing markets, but they might be influenced by the economic strategies adopted by the market participants. For this reason, it is interesting to compare the balancing costs calculated as shown above with the balancing market prices. 3 METHODOLOGY APPLICATION The main purpose of this work is the discussion of the system LCOE and the procedure for its calculation. However, the relevance of this parameter and the difference with the standard LCOE definition must be demonstrated by applying the methodology to a real case. This section describes how to determine the system LCOE and all the different parameters to a real case. The selected real case corresponds to the Italian one as the authors have access to more data. It is important to stress that the procedure is general and can be applied to any other region or country once the input data are available. 3.1 Case study In response to the Paris Agreement, the European Union fixed targets for each Member State in terms of CO2 emissions reduction and RES penetration within the years 2030 and 2050. Within this framework, Italy introduced further incentives for RES power plants with the Decree FER 115 in July 2019 and approved the National Energy and Climate Plan (PNIEC acronym from Italian Piano Nazionale Integrato per l'Energia ed il Clima, PNIEC) in the end of the year 2018, the legislative commitment that sets the Italian targets to be achieved within the year 2030 as established by the European Union. The PNIEC establishes (a) the 40% reduction of the CO2 emissions by 2030 with respect to the emissions registered in the year 1990 and (b) more than 30% of the overall gross energy demand covered by RES with different shares in the three major sectors: 55.4% in the power, 33% in the heat and 21.6% in the transport sectors respectively.16 The RES coverage in the power sector must be achieved considering an increase of around 5% of the annual electricity demand in the year 2030. According to the PNIEC projection is, the PV technology will give the most significant contribution by producing around 200% more than what is producing nowadays. Wind technology is also planned to increase significantly its share rising its production from the current 17.7 to 40.1 TWh in the year 2030. The combined cycle gas turbine (CCGT) power plants will still have a significant role in the energy mix since they must partially replace the coal power stations that are planned to be completely phased-out within 2025. The comparison of the Italian energy mix in the years 2017, taken as the baseline scenario, and 2030, used as reference in this analysis to estimate the future generation costs of PV, is shown in Table 1. TABLE 1. Comparison of the Italian energy mix nowadays (Baseline 2017) and that expected for the year 2030 (PNIEC 20