Evaluation of the latest Spanish grid code requirements from a PV power plant perspective

Energy Reports 8 (2022) 8589–8604 Contents lists available at ScienceDirect Energy Reports journal homepage: www.elsevier.com/locate/egyr Research paper Evaluation of the latest Spanish grid code requirements from a PV power plant perspective Miguel Martínez-Lavín a, Raquel Villena-Ruiz b,c,∗, Andrés Honrubia-Escribano b, Jesús C. Hernández d, Emilio Gómez-Lázaro b aCertification Entity for Renewable Energies (CERE), 28906 Getafe (Madrid), Spain bRenewable Energy Research Institute and DIEEAC-ETSII-AB, Universidad de Castilla-La Mancha, 02071 Albacete, Spain cDepartment of Electrical Engineering, Universidad Politécnica de Cartagena, 30202 Cartagena, Spain dDepartment of Electrical Engineering, Universidad de Jaén, 23071 Jaén, Spain article info Article history: Received 14 October 2021 Received in revised form 4 February 2022 Accepted 24 June 2022 Available online 2 July 2022 Keywords: Grid code NTS Solar PV power plant Validation abstract The integration of new renewable power capacity into the grid constitutes an important challenge. Control issues become more complex, and the behavior of the new installations must be carefully assessed. In this sense, countries are establishing strict technical requirements when integrating new generation facilities into networks. Within this framework, given the great importance of photovoltaic solar energy as a clean electricity generation technology that is experiencing an unstoppable increase, the present paper evaluates the compliance of a real solar photovoltaic power plant with a number of technical requirements established in the new Spanish grid code. The analyses are performed by means of a photovoltaic power plant simulation model representing the actual facility. Thus, the main contribution of the present work consists of mitigating the lack of information about the performance of real experiences on the commissioning process of new renewable power plants, as well as demonstrating the usefulness of simulation models towards the modernization of grid codes. The results reveal the compliance of the renewable facility with the requirements analyzed, showing a maximum deviation of 0.47% in the case of the power–frequency requirements, and very accurate responses of the power plant under the reactive power control requisites. The results also reveal the facility fully covers the reactive power capability requirements. This paper has served not only to deepen the process of compliance with grid codes, but also to obtain approval from the Spanish transmission system operator, Red Eléctrica de España, to commission the power plant under consideration, this being a contribution of special interest to industry. ©2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction 1.1. Motivation and incitement The total cumulative installed Photovoltaic (PV) power capacity worldwide was more than 760 GW at the end of 2020 (International,2021). In Spain, 2020 was also a very positive year in terms of new PV capacity installed, since it was the renewable source of energy with the largest growth, increasing by 29.5% compared to 2019. Moreover, 2020 was also a record year in terms of electricity generation from renewables, since these ∗Corresponding author at: Renewable Energy Research Institute and DIEEAC-ETSII-AB, Universidad de Castilla-La Mancha, 02071 Albacete, Spain. E-mail addresses: [email protected] (M. Martínez-Lavín), [email protected] (R. Villena-Ruiz), [email protected] (A. Honrubia-Escribano), [email protected] (J.C. Hernández), [email protected] (E. Gómez-Lázaro). sources of energy accounted for 44% of Spanish total electricity generation, their highest share ever (Red Eléctrica,2020a). Despite this meaningful achievement, countries need to make an effort to multiply these numbers, so that the net-zero 2050 initiative, which pursues a zero-carbon economy by 2050, can be ensured. However, this goal will not be achieved only by promoting the installation of renewables, but bold actions are also needed across all sectors, including the building industry, which is the sector that contributes most to greenhouse gas emissions in the European Union. Therefore, in view of the above, not only short-term, but also long-term milestones must be established to decarbonize economies. This entails implementing broad policy packages that allow real changes to be triggered, affecting both societal behaviors and companies’ investments (Fernández-González et al., 2021). Research and development will need to be strongly supported, notably in relation to the designing of smart cities, new public and efficient transportation, digitalization of power grids https://doi.org/10.1016/j.egyr.2022.06.078 2352-4847/©2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 (Noussan and Tagliapietra,2020), electrification of residential demand, and in general, energy efficiency (Zsiborács et al.,2021). All these actions will immediately affect people’s lives, since pollution will be considerably reduced. Society as a whole must undoubtedly be involved in this transition towards a more sustainable future, and both individual and collective behavior will affect every step of the process. As mentioned, one of the most noteworthy, urgent actions is the decarbonization of countries’ energy mixes. Transition to emission-free energy models inevitably involves replacing fossil fuels and the commission of renewable energy power plants. Thus, decarbonization of power generation systems requires investing in ambitious renewable projects, such as large-scale solar PV and Wind Power Plants (WPP). Under this framework, the intermittent and stochastic nature of renewable resources is a challenge that must be carefully evaluated and considered. The absence of conventional synchronous generators significantly decreases power system inertia (Hernández et al.,2017), which leads to control issues being more complex. Fault-Ride Through (FRT) capability (Villena-Ruiz et al.,2020a), power quality (Mahela et al.,2020), generation reserve, active power regulation and frequency control (Akram et al.,2020;Zhang et al.,2020) are other issues to which attention must be paid. Therefore, the integration of increasing wind and solar PV capacity requires proper evaluation (Amigue et al.,2021), so that stable operation during transients and fault situations can be ensured (Basu et al.,2020). In this sense, most countries have developed technical documents, known as grid codes, that specify the requirements an electricity generation facility connected to a particular national network must comply with to ensure the power system functions safely, securely and reliably (Al-Shetwi et al.,2020). For instance, one of the requirements established in most grid codes states that renewable energy power plants must remain connected to the grid during faults (Jiménez-Buendía et al.,2019;Villena-Ruiz et al.,2020b). In the case of Spain, a new grid code -based on the European regulation EU 2016/631 (European,2016a)- was released in July 2020, together with a guidelines document, called Technical Supervision Standard, ‘Norma Técnica de Supervisión’ (NTS), for commissioning and grid integration of Renewable Energy Sources (RES) - based power plants. The second edition of this document was published in November 2020 (Red Eléctrica,2020b). This new grid code replaced the previous one, formed by the set of documents known as Operation Procedures, ‘Procedimientos de Operación’ (PO). These procedures, whether technical or instrumental, were needed for adequate technical management of both the peninsular and non-peninsular electricity systems. It should be noted that the are significant differences between the previous Spanish grid code and the new one. The previous grid code only included the technical requirement for assessing power plants responses under voltage dips -or FRT capability, and only required the facility to remain connected to the network during the fault to avoid causing further damage and affect a wide area of the electric power system. However, the current normative document includes a significantly higher number of technical requirements, some of them listed as follows: (i) power regulation under over-frequency; (ii) power regulation under sub-frequency; (iii) power control in the event of both frequency reductions and increases; (iv) active power control; (v) reactive power control through different methods; (vi) inertia emulation; (vii) FRT capability; (viii) power response under transient over-voltages; and (ix) damping of power oscillations. On the other hand, one of the features already included in the previous Spanish grid code was the possibility of employing dynamic simulation models validated to certify the compliance of the power plants’ responses with certain technical requirements. In that case, field tests were conducted in the actual device under study before using the recorded measurements to validate the simulation model. Then, once the model was validated, it could be used to simulate the complete power plant (Villena-Ruiz et al.,2019). The use of dynamic simulation models marked an important step towards the modernization of grid codes. The use of simulation models was an intermediary step that allowed the wide range of actual electrical models in need of certification to be rapidly studied and analyzed using simulation software tools, without the need to always conduct field tests. As would be expected, the possibility of employing simulation models to evaluate the conformity assessment of each technical requirement is maintained in the NTS. Therefore, given the increasing importance of using dynamic models during the evaluation procedures for the compliance of power plants with the technical specifications of grid codes, this paper presents the responses of a solar PV power plant simulation model that represents an actual PV power plant in Spain. The different power– frequency regulation modes defined in the NTS – the new Spanish grid code – are studied, and the solar PV power plant behavior is thoroughly analyzed. 1.2. Literature review Regarding the scientific literature, a comprehensive search has been carried out to review all the works that analyzed the performance of solar PV power plants in compliance with, or based on, technical requirements established in grid codes. One of the first contributions found presented the applicable grid code in Germany in the year 2009, as an example of the new interconnection requirements that began to be defined to improve the integration of PV systems into the electric power system (Degner et al.,2009). A few years later, in 2014, a number of publications related to the compliance of PV installations with technical requirements defined in national normative documents were also released. In Bae and Kim (2014), a control strategy scheme for controlling the active and reactive power responses of a PV system, based on the German grid code in effect at the time, was defined, supporting the proposal with experimental results. In Hossain and Ali (2014), authors focus on enhancing the Low Voltage Ride Through (LVRT) capability of a PV power plant by proposing the use of a series dynamic braking resistor and bearing in mind certain grid code constraints, while Nanou and Papathanassiou (2014) also focuses on the LVRT capability requirement to comply with some grid codes to present a PV system model. Other contributions proposed modifications to these normative documents to deploy PV energy at wide-scale (Yang et al., 2015); characterized and analyzed requirements defined in grid codes of several countries (Honrubia-Escribano et al.,2015); presented an enhancement of this LVRT -FRTtechnical specification on the part of a PV power plant model (Hasanien,2015); or proposed a grid code itself for PV integration to mitigate voltage issues (Chirapongsananurak and Hoonchareon,2017). Regarding more recent publications, there are several contributions also addressing the FRT capability requirements of PV power plants. In Tarafdar Hagh and Khalili (2019), authors review, in more than thirty countries, the FRT technical requirement established in grid codes, while Buraimoh and Davidson (2020) analyzes the impact of PV power in the development process of these normative documents. The work in Al-Shetwi et al. (2019) describes, focusing on grid-connected PV power plants, the main methods that exist of FRT capability control, and researchers in Rey-Boué et al. (2019) model the vector control for a grid-connected PV system with LVRT capability, according to the constraints defined in the Spanish grid code. Moreover, it has to be noted that few current publications address the assessment of technical 8590 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 requirements defined in grid codes – other than FRT capability requirements – when these are applied to specific PV power facilities. Thus, Cabrera-Tobar et al. (2019) analyzes the active and reactive power responses of a PV generator based on the technical requirements of several grid codes, and Al-Shetwi and Sujod (2018) reviews the requirements established in several countries for the integration of grid-connected PV power plants. Finally, the work in Ghaderi et al. (2021) proposes an inverter, the performance of which is investigated in real time, for grid code compatibility purposes, while in Bayrak et al. (2021) similar is done, examining the grid code requirements in Turkey and proposing a real-time energy management system for a PV generator. On the other hand, those works that do perform analyses based on scenarios or real PV generation power plants do not focus on the compliance of these facilities with grid codes, but rather aim to study the dynamic behavior of the power plant, forecast its production, or perform a techno-economic analysis to study its viability. Thus, authors develop in Farah and Boland (2021) a time series model, for two solar PV power plants located in Australia, to improve the forecasting of power output. The work performed in Bandaru et al. (2021) proposes a model based on a multi-criteria decision analysis, which allows the feasibility of a solar PV power plant project to be determined, applying the model to a real PV facility located in England. How the high PV energy penetration impacts voltage regulation and dynamic performance of the grid in Tunisia is investigated in Saidi (2020). Researchers in Yongchareon et al. (2021) and Das (2021) conduct forecasting analyses. In particular, Yongchareon et al. (2021) proposes a new methodology for forecasting energy production based on actual data from a PV facility in Thailand, whereas Das (2021) conducts short term forecasting of solar radiation and power output in India. Finally, authors conduct technoeconomic analyses in both Ahmed et al. (2021) and Agyekum (2021). Specifically, different climate zones of Pakistan are chosen to conduct techno-economic assessments of large-scale PV power plants in Ahmed et al. (2021), and the results obtained are compared between each other, while Agyekum (2021) does something similar, analyzing the techno-economic potential of two solar PV installations under different conditions in Ghana. 1.3. Contribution and paper organization Giving that the existing literature on the analysis of the compliance of renewable power installations with grid codes follow a similar structure, based on the analysis of the response of hypothetical power plant simulation models, or focused on the review and comparison of technical requirements between grid codes, it is necessary to perform a detailed analysis of a specific PV power plant, mitigating the lack of information about the performance of real experiences on the commissioning process of new renewable power plants. Moreover, a large number of the scientific works on the performance analysis of PV installations according to national grid codes were published a long time ago, having already been replaced, many of the normative documents mentioned above, by others with more updated and modern technical requirements. In view of all the above, the main contribution of the present work consists of the conduction of an in-deep analysis of the behavior of a newly installed PV power plant in Spain; therefore an analysis based on real data and a set of updated technical requirements -the Spanish grid code is, without a doubt, one of the most advanced grid codes in the world, and thus a work also aiming to contribute to the correct integration of the huge amount of solar PV capacity that will be installed in the near future to reach emission-free power systems. This comprehensive study shows how the new renewable power plants that will be installed must comply with stricter technical specifications, and demonstrates the capacity of the most modern inverters to adapt to the requirements demanded. Moreover, this paper has served not only to deepen the process of compliance with grid codes, but also to obtain approval from the Spanish Transmission System Operator (TSO), Red Eléctrica de España (REE), to commission the power plant under consideration, this being a contribution of special interest to industry. Furthermore, the results obtained provide the basis for the extensive use of these simulation models as part of the evaluation and certification processes of new electricity generation facilities. Moreover, it describes in detail the steps that must be followed to certify a power plant according to the new Spanish grid code, and shows, as an innovation, the compliance results of a dynamic PV power plant simulation model representing an actual PV power plant. The paper is structured as follows: Section 2presents the new Spanish grid code, which is based on recent European Regulations, as well as the technical document released to monitor the compliance of new power plants with the national technical requirements. Section 3reviews the technical requirements according to which the PV power plant under study is analyzed, while Section 4presents the results obtained when submitting the PV power plant to the technical requirements previously defined. Finally, Section 5summarizes the main conclusions obtained. 2. European regulations: New Spanish grid code Regulation EU 2016/631 (European,2016a) is an official European document establishing a grid code that defines the requirements for grid connection of electricity generation facilities, mainly synchronous power-generating modules, power park modules and offshore power park modules. Regulation EU 2016/631 is one of the three grid codes on which Spain based the development of its own legal texts in this area. These grid codes are known as Network Connection Codes, ‘Códigos de Red de Conexión’ (CRC). Thus, as a result of the domestic implementation of Regulation EU 2016/631 – together with the other two European grid codes, Regulation EU 2016/1388 (European,2016b) and Regulation EU 2016/1447 (European,2016c) – two new official Spanish documents were released in this regard. In 2020, Royal Decree (RD) 647/2020 of July 7 (Ministerio,2020a) and Ministerial Order (MO) TED/749/2020 of July 16 (Ministerio,2020b) were published in Spain’s official state gazette, ‘Boletín Oficial del Estado (BOE)’ -the official gazette dedicated to the publication of mandatory laws, regulations and other decrees. RD 647/2020 regulates aspects required for the implementation of the grid codes for the connection of certain electrical installations, while MO TED/749/2020 establishes the technical requirements for the network connection necessary for the implementation of the grid connection codes. These two official documents also drew on proposals submitted by the network operators and input from the working groups created for that purpose. Accordingly, in order to coordinate the process of compliance monitoring of the CRCs, as well as to ensure transparency and overcome efficiently different issues that may arise, two working groups were created. One of these groups was specifically created to monitor compliance with the technical requirements defined in Regulation 2016/631 for power – or electricity – generation modules, ‘Módulos de Generación de Electricidad’ (MGE). As a result, the new Spanish technical supervision standard, ‘Norma Técnica de Supervisión’ (NTS) was published in July 2019. The second and latest edition of the NTS was published in November 2020, and is entitled ‘‘Technical standard for monitoring the conformity of electricity generation modules according to EU Regulation 2016/631’’. 8591 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Fig. 1. General diagram of an MGE composed of several UGEs and a CAMGE. 2.1. Spanish technical supervision standard: NTS The NTS (Red Eléctrica,2020b) develops the different aspects gathered in Title IV ‘‘Compliance monitoring’’ of Regulation EU 2016/631, being those which require a greater level of detail to properly verify the compliance of the MGEs with the technical requirements defined in that Regulation. Thus, in order to follow and apply the NTS adequately, in addition to the previously defined MGEs, it is important to distinguish between the other main components that may be subject to assessment by this technical document: •Power generation units, ‘Unidades de Generación de Electricidad’ (UGE). The UGEs refer to the main power generation plant. •Additional MGE components, ‘Componentes Adicionales del MGE’ (CAMGE). The CAMGEs are devices that form part of the MGE — which are not a UGE, the response of which can affect the compliance of the MGE with the technical requirements established. For instance, active and/or reactive power control devices, Flexible Alternating Current Transmission Systems (FACTS), synchronous compensators, Power Plant Controllers (PPC) and batteries are devices considered CAMGEs. Fig. 1 shows a schematic representation of an MGE, which may consist of several UGEs and/or CAMGEs. At this point, it should be noted that MGEs can be divided into synchronous power generation modules, ‘Módulos de Generación de Electricidad Síncronos’ (MGES) and power park modules, ‘Módulos de Parque Eléctrico’ (MPE). Moreover, MGEs are divided into four types, depending on their maximum capacity and voltage at their connection point to the network, ‘Punto de Conexión con la Red’ (PCR), -also known as Point of Common Coupling (PCC)-: •Type A: MGEs the PCC of which is below 110 kV and its maximum capacity (Pmax) is 0.8 kW ≤Pmax ≤100 kW. •Type B: MGEs the PCC of which is below 110 kV and its maximum capacity is 100 kW <Pmax ≤5 MW. •Type C: MGEs the PCC of which is below 110 kV and its maximum capacity is 5 MW <Pmax ≤50 MW. •Type D: MGEs the PCC of which is equal to or greater than 110 kV or its maximum capacity is Pmax >50 MW. The power installation under study must be subjected to the conformity assessments corresponding to the technical requirements the installation must comply with. According to European (2016a), the conformity assessment of each requirement may be conducted via three different paths, and each of these may be required separately or together with one of the other two: •Compliance Testing, ‘Pruebas de conformidad’ (T). Compliance testings of the MGE for each of the technical requirements that must be evaluated will be carried out by an accredited entity, which will release a testing report and send the results to an authorized certification entity to be evaluated. Then, for each of the requirements, the compliance of the MGE or the UGE, as applicable, will be evaluated, and a certificate of compliance through testing of each requirement will be issued. •Compliance Simulations, ‘simulaciones de conformidad’ (S). An accredited entity will conduct the MGE compliance simulations for each technical requirement. The accredited entity will then prepare a simulations report and send the results to an authorized certification entity to be evaluated. For each of the requirements, the compliance of the MGE or the UGE, as applicable, will be evaluated, and a certificate of compliance through simulation of each requirement will be issued. •Equipment Certificates, ‘certificados de equipo’ (C). The MGE conformity assessment for each requirement can be conducted through equipment certificates — on the basis of the pertinent tests conducted in the UGEs and CAMGEs, which will be issued by an authorized certification entity. However, when following this third path, certain aspects must be taken into consideration. On the one hand, having obtained all the UGE and CAMGE certificates does not always imply, automatically, the conformity of the MGE with the technical requirements. This is because, among other aspects, having obtained the UGE and CAMGE certificates does not always ensure the compliance of the MGE with the requirements at the PCC. Thus, depending on the technical requirement under consideration, additional simulations may be required. On the other hand, the UGE and CAMGE equipment certificates will be valid provided that the parameters potentially affecting control features, which in turn impact the compliance of these devices with the technical requirements, are not modified after having obtained those equipment certificates. According to the type of MGE, in order to obtain its final certificate, the technical requirements may be assessed following one, two, or three of the paths defined above, separately or simultaneously. For instance, the requirement ‘‘power–frequency limited over frequency regulation mode’’ must be assessed for all types of MGEs (from Types A to Types D), and may be assessed via S and T together, or via C. Moreover, this will apply to both MPEs and MGESs. The ‘‘reactive power capacity at maximum capacity’’ requirement will only apply to Types B, C and D MGEs, and of these, only to MPEs. This requirement may be assessed via S and T at the same time, or via C. Fig. 2 presents the general conformity assessment scheme, which can be divided into three main stages: (i) obtaining the UGE and/or CAMGE certificates of compliance of each requirement; (ii) obtaining the MGE final certificate and issuance of the Final Operational Notification (FON); (iii) commercial operation of the MGE. 2.2. Conformity assessment procedure via equipment certificates The present paper addresses the additional simulations that are required at MGE level to make it comply with the technical requirements established in the grid code. Therefore, the conformity assessment procedure via equipment certificates, explained in Section 2.1, is followed in this case. As is indicated in Section 4.1 of the NTS (Red Eléctrica,2020b), validated or certified simulation models must be used to conduct the additional simulations required to assess some of the technical requirements. Fig. 3 shows the conformity assessment procedure via equipment certificate for a specific technical requirement and for the specific case of a Type D MPE requiring additional simulations to obtain the final MGE certificate, as is the case in the present work. The complete flowchart showing all the possible paths is shown in Section 4.2 of the NTS (Red Eléctrica,2020b). 8592 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Fig. 2. Stages of the general conformity assessment scheme. Table 1 Characteristics of the PV power plant. N◦Rated power HV/MV transformers (132/30 kV) 1 50 MVA MV/LV transformers (30/0.8 kV) 7 6 MVA 3 2.8 MVA Inverters 465 105 kW 3. Technical requirements analyzed The power plant analyzed, being a PV power plant, is an MPE and has a nominal power of 42 MW. Moreover, it is a Type D because the voltage at the PCC where the power plant is connected is above 110 kV, in particular 132 kV. This configuration represents an actual solar PV power plant located in Spain. Table 1 summarizes the main characteristics of the power installation, including the number of each of the main elements that form it (transformers and inverters) and their nominal power. The simulation model of the PV power plant is studied according to the technical requirements defined in the following sections of the NTS (Red Eléctrica,2020b): •Section 5.1: Power–Frequency Limited Over-Frequency Regulation Mode (RMPFL-O). •Section 5.2: Power–Frequency Limited Sub-Frequency Regulation Mode (RMPFL-S). •Section 5.3: Power–Frequency Regulation Mode (RMPFL). •Section 5.7: Reactive Power Capability. •Section 5.8: Reactive Power Control. In particular, the complementary simulations defined in those sections are carried out at MPE level, with the MGE being MPE type in this case, with the objective of obtaining the MPE equipment certificate. The technical requirements analyzed are further detailed in Sections 3.1,3.2, and 3.3 of the present paper, and correspond, on the one hand, to the power–frequency regulation mode, and on the other hand, to the reactive power capability of the power plant and the reactive power control modes. The power system simulation software tool DIgSILENT PowerFactory, version 2019 SP3, was used to conduct the simulations. 3.1. Power–frequency regulation modes The different power–frequency regulation modes (Obaid et al., 2019) are studied at the PCC of the PV installation, which is a point located upstream the power plant transformer and has a rated voltage of 132 kV. It should be noted that, to study the power–frequency regulation modes, the droop, or statism, of each inverter was configured so that the increments defined by the NTS with respect to the power plant’s rated power, 42 MW, were obtained. An ideal voltage source was used to set the frequency at the PCC to the desired value. The following terms must be taken into account when studying the power–frequency regulation modes of the MPE: •∆f: deviation of frequency ffrom 50 Hz (∆f=f−50). •Initial active power (Pini): pre-disturbance active power of the MGE. This coincides with the test power. •Power deviation (∆P): difference with respect to the initial active power Pini. I.e., ∆Pis equal to the active power (P) minus the Pini (∆P=P-Pini). •Initial delay time (ta): activation time of the RMPFL-O. This value coincides with the time it takes a variation of 1% of the expected power response (∆P) against a deviation of frequency (∆f) to occur. •Response time (tr): time it takes 90% of the expected or measured power (∆P) against a deviation of frequency (∆f) to be reached, without including the initial delay time (ta). •Setting time (te). Time it takes the response to stabilize within a band of ±5% of the power deviation, (∆P), against a frequency deviation ∆f, without including the initial delay time (ta). Fig. 4 shows, graphically, all these terms. 3.1.1. Power–frequency limited over-frequency regulation mode The objective is to verify that the MGE is able to activate the supply of power–frequency regulation reserves as indicated in article 13.2 of Regulation EU 2016/631 (European,2016a) with regard to the limited frequency-over frequency sensitive model, and article 1.3 of MO TED/749/2020 of July 16 (Ministerio,2020b). According to Section 5.1 of the NTS (Red Eléctrica,2020b), in the RMPFL-O requirements, the allowable limit values of the previously described terms are as follows: •ta=30 s. In cases where ta≥2 s, the MGE owner must provide evidence to the TSO that justifies this value. •tr=2 s for reductions in P and tr=10 s for increases in P. •te=20 s for reductions in P and te=30 s for increases in P. To conduct the complementary simulations that will allow the MPE certificate to be obtained, the initial active power must coincide with the maximum active power, droop must be equal to 5%, and the trigger threshold of ∆fmust be equal to 0.2 Hz, among other specifications. Section 4.1.1 presents the results obtained once conducted the complementary RMPFL-O simulations. 3.1.2. Power–frequency limited sub-frequency regulation mode The objective is to verify that the MGE is able to activate the supply of power–frequency regulation reserves as indicated in article 15.2.c of Regulation EU 2016/631 (European,2016a) with regard to the limited frequency-under frequency sensitive mode, and article 1.7 of MO TED/749/2020 of July 16 (Ministerio,2020b). According to Section 5.2. of the NTS (Red Eléctrica,2020b), in the RMPFL-S requirements, the allowable limit values of the previously described terms are the same as those established for the RMPFL-O requirements: 8593 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Fig. 3. Conformity assessment procedure via equipment certificates: additional simulations at MGE level. Fig. 4. Example of power response. •ta=30 s. In cases where ta≥2 s, the MGE owner must provide evidence to the TSO that justifies this value. •tr=2 s for reductions in P and tr=10 s for increases in P. •te=20 s for reductions in P and te=30 s for increases in P. To conduct the complementary simulations that will allow the MPE certificate to be obtained, the initial active power must be 60% of the maximum active power, droop must be equal to 5%, and the trigger threshold of ∆fmust be equal to −0.2 Hz, among other specifications. Section 4.1.2 presents the results obtained once conducted the complementary RMPFL-S simulations. 3.1.3. Power–frequency regulation mode The objective is to verify that the MGE is able to activate the supply of power–frequency regulation reserves as indicated in article 15.2.d of Regulation EU 2016/631 (European,2016a) with regard to the situations where the frequency sensitive mode is operating, and Article 1.8 of MO TED/749/2020 of July 16 (Ministerio,2020b). This requirement is analyzed according to Section 5.3 of the NTS (Red Eléctrica,2020b). In this case, the allowable limit values are: •ta=500 ms. •te=30 s. To conduct the complementary simulations that will allow the MPE certificate to be obtained, a number of conditions, listed in Section 5.3.3. of the NTS (Red Eléctrica,2020b), must be complied Fig. 5. Capability curve used. with, such as setting the droop to 5%, or the initial active power to 80% of the maximum active power. Section 4.1.3 presents the results obtained once conducted the complementary RMPFL simulations. 3.2. Reactive power capability The objective of these analyses is to verify that the MGE is able to supply the reactive power required at the maximum capacity of the MGE and below the maximum capacity of the MGE, as indicated in Article 21.3 of Regulation EU 2016/631 (European, 2016a) with regard to voltage stability and hence with regard to reactive power capability, and article 2.3.2 of MO TED/749/2020 of July 16 (Ministerio,2020b). In order to study this requirement, different load flows analyzing the reactive power capability of the PV power plant are conducted under different active power levels, covering from no load conditions to nominal power, as well as under different voltage levels at the PCC: 0.90 pu, 0.95 pu, 1.00 pu, 1.05 pu and 1.10 pu. In order to set the voltage level at the PCC, an infinite electrical network is used. This is done according to Section 5.7 of the NTS (Red Eléctrica,2020b). The capability curve used is shown in Fig. 5. Section 4.2 presents the results obtained once conducted the additional reactive power capability simulations that will allow the MPE certificate to be obtained. 3.3. Reactive power control at MPE The objective of these analyses is to verify that the MGEs corresponding to the MPE type are able to control reactive power, 8594 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 as indicated in article 21.3d of Regulation EU 2016/631 (European, 2016a) with regard to reactive power control modes, and article 2.3.3 of MO TED/749/2020 of July 16 (Ministerio,2020b). That is, the dynamic behavior of the MPE will be studied according to its three different control modes: (i) voltage control mode (Section 3.3.1); (ii) reactive power control mode (Section 3.3.2); and (iii) power factor control mode (Section 3.3.3). This is done according to Section 5.8. of the NTS (Red Eléctrica,2020b). 3.3.1. Voltage control mode According to the NTS, parameter t1 is defined as the time it takes the reactive power measured to reach 90% of the expected response. This percentage is calculated on the basis of the difference between the initial reactive power value and the requirement value. Thus, it should be made clear that this value is not calculated on the basis of the requirement value, but its increase. Parameter t2 is defined as the time it lasts the reactive power to stabilize within the range defined in column ‘‘Q expected at PCC (Q/Pmax)’’ of Table 5 (see Section 4.3.1). Moreover, also in Table 5, when reference is made to a 7% slope, it means that, against a 7% voltage variation, the maximum reactive power must be injected (minimum 0.3Q/Pmax pu). The same applies to a 2% slope (see Table 6 of Section 4.3.1). The simulation of the voltage control mode was conducted by providing reactive power setpoints to the inverters based on the voltage at the PCC. In other words, the reactive power control function through setpoints is used. On the other hand, to set the desired voltage level at the PCC, a voltage source is employed. The reactive power control mode through setpoints is used because no PPC is available. Thus, each inverter dynamic model measures at the inverter terminals, not at the PPC. If the function of reactive power depending on voltage was used instead, the injection of reactive power would not be done according to the voltage at the PPC but according to the voltage at the inverter terminals. Thus, because the voltage at each terminal is slightly different, an unreal error would be added when using the function of reactive power depending on voltage. Therefore, we opted to control reactive power directly through setpoints. To conduct the complementary simulations that will allow the MPE certificate with regard to the voltage control mode to be obtained, the MPE will be injecting a value of active power of at least 80% the maximum active power, while control slopes will be equal to 2% and 7%. Voltage changes will be simulated as indicated in Section 4.3.1, which presents the results obtained once conducted the additional reactive power control-voltage control mode simulations that will allow the MPE certificate to be obtained. 3.3.2. Reactive power control mode In order to conduct the simulations under the reactive power control mode, voltage at the MPE terminals will coincide with the nominal voltage, and the reactive power setpoint will be null at the moment the simulation is started. Then, during the simulations, the reactive power setpoints shown in Table 7 will be established (see Section 4.3.2). The reactive power measured at the MPE terminals will be indicated, as well as the time it takes the new reactive power value to stabilize after setting a new setpoint. To conduct the complementary simulations that will allow the MPE certificate for the reactive power control mode to be obtained, the MPE will be injecting a value of active power of at least 80% the maximum active power. Initially, voltage at the verification point will coincide with the nominal voltage, and the reactive power setpoint will be null at the moment the simulation is started. 3.3.3. Power factor control mode In order to conduct the simulations under the power factor control mode, voltage at the MPE terminals will coincide with the nominal voltage. During the simulations, the power factor setpoints shown in Table 8 will be established (see Section 4.3.3). The reactive power measured at the MPE terminals will be indicated, as well as the active power injected by the MPE at the moment the reactive power is being measured, and also the time it takes the new power factor value to stabilize within the band after setting a new setpoint or occurring a variation in active power. To conduct the complementary simulations that will allow the MPE certificate for the power factor control model to be obtained, the MPE will be injecting a value of active power of at least 80% the maximum active power, and the different power factor setpoints, shown in Table 8, will be simulated. 4. Results Firstly, it should be noted that testing was performed by an accredited testing laboratory according to the NTS requirements. Moreover, measurement uncertainty was in accordance with the NTS requirements. Sections 4.1–4.3 present the results obtained when conducting, at MPE level, the additional simulations required to obtain the final MPE certificate for the following technical requirements, respectively: (i) power–frequency regulation modes; (ii) reactive power capability; and (iii) reactive power control at MPE. 4.1. Additional simulations: Power-frequency regulation modes Section 4.1.1 presents the results obtained when conducting the additional simulations of the requirement ‘‘power–frequency limited over-frequency regulation mode’’ to obtain the final MPE certificate. Section 4.1.2 presents the results obtained when conducting the additional simulations of the requirement ‘‘power– frequency limited sub-frequency regulation mode’’, while Section 4.1.3 presents the results obtained when conducting the additional simulations of the requirement ‘‘power–frequency regulation mode’’. 4.1.1. Additional simulations: Power-frequency limited overfrequency regulation mode (RMPFL-O) Frequency rises are common in highly interconnected networks since these have a fast response. Frequency rises are due to unbalances between generation and demand. The way to reduce these frequency rises is with traditional inertial power systems, which regulate power depending on network’s frequency. These traditional power systems are synchronous generators. Therefore, PV power plants cannot provide inertia to the power system. That is why the NTS defines this RMPFL-O requirement for MPEs, so that these power systems can guarantee reductions in P when frequency rises occur. This allows for a better integration of PV power plants into power systems. In Table 2, the permissible power ranges, in percentages, are indicated in the last column, while the deviation values obtained are shown in the fifth column. As can be observed, the deviation values are within the permissible limits in all cases, which means that the MPE complies with the requirements established for this ‘‘power–frequency limited over-frequency regulation mode’’ technical requisite. Fig. 6 shows the simulation response of the MPE model when conducting the additional simulations required to obtain the final MPE certificate for this technical requirement. In particular, Fig. 6(a) shows the frequency variations that allow the additional simulations in this case to be assessed, while Fig. 6(b) shows the active power response of the model. 8595 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Table 2 PV power plant response: Power–frequency limited over-frequency regulation mode (RMPFL-O). PV power plant model P vs. over freq.; 50.2 Hz; s =5%; Pini =Pmax Simulation point f (Hz) ∆P/Pmax P (%) registered Deviation (%) 90% P (%) registered tr (s) ta (s) te (s) Permissible power range (%) (%) expected (90% P (%) Registered) Error 1 50 0.00% 100.08% 0.08% – – – – ±5% 2 50.2 0.00% 100.07% 0.07% – – – – ±5% 3 50.4 −8.00% 92.23% 0.23% 92.80% 0.153 0.022 0.153 ±5% 4 50.6 −16.00% 84.34% 0.33% 84.80% 0.164 0.024 0.164 ±5% 5 50.8 −24.00% 76.41% 0.41% 76.80% 0.174 0.025 0.174 ±5% 6 51 −32.00% 68.46% 0.45% 68.80% 0.181 0.025 0.181 ±5% 7 51.2 −40.00% 60.47% 0.47% 60.80% 0.184 0.026 0.184 ±5% 8 51.4 −48.00% 52.46% 0.46% 52.80% 0.182 0.026 0.182 ±5% Fig. 6. Power–frequency limited over-frequency regulation mode: Simulation response of the MPE model under over-frequency variations. 4.1.2. Additional simulations: Power-frequency limited subfrequency regulation mode (RMPFL-S) As in the previous case, frequency drops are common in highly interconnected networks, and are also due to unbalances between generation and demand. Since PV power plants are not synchronous generators that can provide inertia, the NTS defines the RMPFL-S requirement for MPEs, so that these power systems can guarantee increases in P when frequency drops occur. In Table 3, as in the previous case, the permissible power ranges are indicated in the last column, while the deviation values obtained are shown in the fifth column. As can be observed, the deviation values are within the permissible limits in all cases, which means that the MPE complies with the requirements established for this ‘‘power–frequency limited sub-frequency regulation mode’’ technical requisite. Fig. 7 shows the simulation response of the MPE model when conducting the additional simulations required to obtain the final MPE certificate for this technical requirement. In particular, Fig. 7(a) shows the frequency variations that allow the additional simulations in this case to be assessed, while Fig. 7(b) shows the active power response of the model. 4.1.3. Additional simulations: Power-frequency regulation mode (RMPFL) In Table 4, as in the previous cases of ‘‘power frequency regulation modes’’, the permissible power ranges are indicated in the last column, while the deviation values obtained are shown in the fifth column. As can be observed, the deviation values are within the permissible limits in all cases, which means that the MPE complies with the requirements established for this ‘‘power-frequency regulation mode’’ technical requisite. Fig. 8 shows the simulation response of the MPE model when conducting the additional simulations required to obtain the final MPE certificate for this technical requirement. In particular, Fig. 8(a) shows the frequency variations that allow the additional simulations in this case to be assessed, while Fig. 8(b) shows the active power response of the model. 4.2. Additional simulations: Reactive power capability With the introduction of a greater number of renewable power plants in networks and the reduction of conventional synchronous power systems with a high reactive power injection capability, it is necessary for renewable power plants to have the capability to 8596 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Table 3 PV power plant response: Power–frequency limited sub-frequency regulation mode (RMPFL-S). PV power plant model P vs. sub freq.; 49.8 Hz; s =5%; Pini =40%Pmax Simulation point f (Hz) ∆P/Pmax P (%) registered Deviation (%) 90% P (%) registered tr (s) ta (s) te (s) Permissible power range (%) (%) expected (90% P (%) registered) Error 1 50 0.00% 40.04% 0.04% – – – – ±5% 2 49.8 0.00% 40.04% 0.04% – – – – ±5% 3 49.6 8.00% 48.10% 0.10% 47.20% 0.135 0.024 0.135 ±5% 4 49.4 16.00% 56.12% 0.12% 55.20% 0.133 0.024 0.133 ±5% 5 49.2 24.00% 64.12% 0.12% 63.20% 0.133 0.023 0.133 ±5% 6 49 32.00% 72.08% 0.08% 71.20% 0.135 0.023 0.135 ±5% 7 48.8 40.00% 80.01% 0.01% 79.20% 0.139 0.024 0.139 ±5% 8 48.6 48.00% 87.91% −0.09% 87.20% 0.146 0.025 0.146 ±5% 9 48.4 56.00% 95.78% −0.22% 95.20% 0.158 0.026 0.158 ±5% Table 4 PV power plant response: Power–frequency regulation mode (RMPFL). PV power plant model Power response time in case of abnormal frequency 80%Pmax and s =5% Simulation point f (Hz) ∆P/Pmax P (%) registered Deviation (%) t1 (s) (Limit: 500 ms) t2 (s) (Limit: 30 s) Permissible power range (%) (%) expected Error 1 50 0% 80.02% 0.01% – – ±5% 2 50.02 −0.80% 79.22% 0.02% – 0.144 ±5% 3 50.10 −4.00% 76.05% 0.05% 0.027 0.135 ±5% 4 50.20 −8% 72.08% 0.07% 0.023 0.137 ±5% 5 50.30 −10%a70.09% 0.09% 0.023 0.040 ±5% 6 50 0.00% 80.03% 0.03% 0.024 0.138 ±5% 7 49.98 0.80% 80.81% 0.01% – 0.121 ±5% 8 49.90 4.00% 83.98% −0.03% 0.030 0.134 ±5% 9 49.80 8% 87.92% −0.08% 0.026 0.141 ±5% 10 49.70 10%a89.89% −0.11% 0.027 0.048 ±5% aLimitation of active power increase. Fig. 7. Power-frequency limited sub-frequency regulation mode: Simulation response of the MPE model under sub-frequency variations. inject reactive power as well, so that the absence of conventional generation can be compensated. The injection of reactive power into networks allows the risk of unbalances to be reduced, a phenomenon that could lead to the disconnection of generators and the appearance of faults in the distribution and transmission networks. Moreover, this capability to regulate reactive power 8597 M. Martínez-Lavín, R. Villena-Ruiz, A. Honrubia-Escribano et al. Energy Reports 8 (2022) 8589–8604 Tarafdar Hagh, M., Khalili, T., 2019. A review of fault ride through of PV and wind renewable energies in grid codes. Int. J. Energy Res. 43 (4), 1342–1356. Villena-Ruiz, R., Honrubia-Escribano, A., Fortmann, J., Gómez-Lázaro, E., 2020a. Field validation of a standard type 3 wind turbine model implemented in DIgSILENT-PowerFactory following IEC 61400-27-1 guidelines. Int. J. Electr. Power Energy Syst. 116, 105553. Villena-Ruiz, R., Honrubia-Escribano, A., Jiménez-Buendía, F., Molina-García, A., Gómez-Lázaro, E., 2020b. Requirements for validation of dynamic wind turbine models: An international grid code review. Electronics 9 (10), 1707. Villena-Ruiz, R., Jiménez-Buendía, F., Honrubia-Escribano, A., Molina-García, A., Gómez-Lázaro, E., 2019. Compliance of a generic type 3 WT model with the spanish grid code. Energies 12 (9), 1631. Yang, Y., Enjeti, P., Blaabjerg, F., Wang, H., 2015. Wide-scale adoption of photovoltaic energy: Grid code modifications are explored in the distribution grid. IEEE Ind. Appl. Mag. 21 (5), 21–31. Yongchareon, W., Thongthew, S., Juncharoen, P., 2021. Photovoltaic forecasting model in Thailand: case study solar farm at Nakhon Ratchasima province. In: IOP Conference Series: Materials Science and Engineering. 1137, (1), IOP Publishing, 012068. Zhang, Z., Du, E., Teng, F., Zhang, N., Kang, C., 2020. Modeling frequency dynamics in unit commitment with a high share of renewable energy. IEEE Trans. Power Syst. 35 (6), 4383–4395. Zsiborács, H., Pintér, G., Vincze, A., Birkner, Z., Baranyai, N.H., 2021. Grid balancing challenges illustrated by two European examples: Interactions of electric grids, photovoltaic power generation, energy storage and power generation forecasting. Energy Rep. 7, 3805–3818. 8604