Enhancing System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System

1 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. ENHANCING SYSTEM STABILITY WITH GFM (E-)STATCOMS: A CASE STUDY OF THE DK2 2020 INCIDENT IN THE DANISH POWER SYSTEM Liang Lu1*, Yicheng Liao1, Hong Gong1, Jun Bum Kwon1, Nan Qin1, Mats Larsson2, Hongyang Zhang3, Rasool Heydari4 1Power System Stability, Energinet, Fredericia, Denmark 2Hitachi Energy Research, Baden, Switzerland 3Grid Integration, Hitachi Energy, Västerås, Sweden 4Hitachi Energy Research, Västerås, Sweden *[email protected]k Keywords: GFM, STATCOM, SYSTEM STABILITY, VOLTAGE CONTROL CAPABILITY, DAMPING Abstract Denmark’s goal of carbon neutrality by 2050 requires a power system largely based on renewable energy, leading to the gradual replacement of synchronous machines with grid-following inverter-based resources. This transition reduces system strength, creating challenges for maintaining voltage and frequency stability. Grid-forming (GFM) control has emerged as one of the promising solutions to address the challenges. In 2023, the Danish TSO Energinet launched the “Deployment of Grid Forming Technology in the Danish Power System” project to evaluate commercial GFM products. Within this project, the GFM capabilities of different technologies were assessed through system-level EMT studies of the Danish grid. This paper presents a detailed study of an incident that happened in eastern Danish power grid (DK2) in 2020, where sequential faults caused a network split at 400 kV backbone, synchronous condenser (SynCon) tripping, and HVDC blocking. The incident is revisited by replacing the SynCon with either a GFM STATCOM or an E-STATCOM, evaluating their effectiveness in mitigating the instability. Comparative results indicate that both devices provide better voltage control compared with the SynCon, with the E-STATCOM demonstrating further improvements in damping capability. These findings underline the potential of GFM (E-)STATCOMs to ensure security and resilience in future renewable-dominated power systems. 1. Introduction The rapid growth of renewable energy integration into power systems is bringing profound challenges to system stability and security. As conventional synchronous generation is gradually replaced by grid-following (GFL) inverter-based resources, power systems lose essential characteristics such as inertia, voltage stiffness, and fault current contribution. This results in reduced damping capability, weaker voltage and frequency support, and an overall higher risk of system instability. To address these challenges, GFM control has emerged as one of the promising solutions. Unlike GFL control, which adjusts converter currents to an existing grid, GFM control enables converters to establish voltage and frequency references autonomously, thereby supporting system stability in weak and renewable-dominated grids. To accelerate the deployment of GFM technology in realworld power systems, the Danish TSO Energinet launched the “Deployment of Grid Forming Technology in the Danish Power System” project [1]. The initiative is intended to break the “chicken-and-egg” dilemma often faced in adopting new technologies: the TSO seeks to better understand their technical characteristics and thereby establish well-founded market entry requirements. Therefore, the project involves close collaboration with OEMs and developers, on a voluntary basis, who provide detailed EMT models of their GFM devices, including battery energy storage systems (BESS), wind power plants, PV power plants, and STATCOMs. Energinet conducts both single-machine infinite-bus (SMIB) tests and systemlevel evaluations using these models, complemented by extensive technical discussions with the industry. Through this process, the project aims to gain a comprehensive understanding of the market maturity and technical limitations of different products and technologies, thereby laying the foundation for the development of future technical standards. In this context, TSOs rely on dedicated devices to maintain system stability, such as reactive power compensators. Among them, STATCOMs represent a key technology, as they can rapidly inject or absorb reactive power, offering significantly faster and more flexible control compared with traditional passive solutions like capacitor banks or reactors. Their capability to provide dynamic voltage support makes them particularly attractive as stabilizing devices in systems with high renewable penetration. Moreover, when equipped with supercapacitors that enable temporary active power exchange, STATCOMs are enhanced into so-called E-STATCOMs, further broadening their stabilizing functionality. In this paper, the term (E-)STATCOM is used to collectively refer to both STATCOM and E-STATCOM. GFM technology enables (E-)STATCOMs to actively shape the grid voltage, thereby reducing their dependence on the grid’s pre-existing voltage and frequency conditions [2] [3] [4]. GFM E-STATCOMs can emulate synchronous generator inertia, providing synthetic inertia and enhancing frequency regulation. Their real-time adaptability further strengthens grid support, making both GFM (E-)STATCOMs well-suited for weak grids with high renewable penetration. 2 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. While recent research has demonstrated the potential of GFM control and highlighted the benefits of advanced devices such as STATCOMs and E-STATCOMs, important gaps remain. To date, EMT-based studies have rarely evaluated GFM (E-)STATCOMs in the context of a real power system under actual operating conditions, let alone in relation to a historical disturbance event. As a result, their practical performance in realistic grid scenarios remains insufficiently understood. Furthermore, previous studies have not provided a detailed comparison of the GFM capabilities of GFM STATCOMs, GFM E-STATCOMs, and SynCons. This paper addresses these gaps by revisiting the DK2 2020 incident through EMT simulations, thereby offering new insights into the relative strengths and limitations of different GFM devices under realworld system conditions. The contributions of this paper are threefold. First, the investigations demonstrate at the system level how GFM (E-)STATCOMs differ from SynCons in maintaining stability under severe disturbances. Second, the study identifies the key GFM capabilities that play a decisive role in this process. Finally, through both system-level evaluations and SMIB tests, the paper provides a comparative assessment of GFM STATCOM, GFM E-STATCOM, and SynCons with respect to these critical GFM capabilities. The paper is organized as follows: Section 2 introduces the incident and simulation setup; Section 3 presents the incident mitigation using GFM (E-)STATCOMs; Section 4 discusses detailed comparisons of different devices; Section 5 summarizes the conclusions; and Section 6 outlines directions for future work. Unless otherwise specified, (E-)STATCOMs mentioned hereafter are assumed to operate with GFM control. 2. DK2 2020 incident The Danish power system is divided into two synchronous areas: DK1 and DK2. DK1 primarily covers the Jutland Peninsula and the island of Funen, and it is interconnected with the German power system via AC transmission lines, operating synchronously with the Continental European grid. DK2, covering the island of Zealand, is interconnected with the Swedish power system through AC transmission lines and operates in synchrony with the Nordic grid. The two areas are linked by the Storebaelt (SB) LCC-HVDC interconnection, which enables power exchange between them. SB is connected at 400 kV busbar of HKS station in DK2. In addition, another LCC-HVDC link, KONTEK (KO), connects DK2 with the German grid and is terminated at the 400 kV busbar of the BJS station. DK2 is also equipped with two SynCons: a 270 MVA device at BJS station (SynCon@BJS) and a 200 MVA device at HKS station (SynCon@HKS). 2.1 Incident review Around noon on August 6, 2020, a single-phase-to-ground short-circuit fault (Fault 1) occurred on the 400 kV overhead transmission line BJS_400_ISH between the BJS and ISH stations. The faulted-phase breaker opened less than 100 ms after the fault occurred and reclosed 80 ms later. As the fault was permanent rather than temporary, all three-phase breakers were opened within 100 ms after reclosure, leading to the disconnection of the BJS_400_ISH line. Approximately 10 minutes later, a second single-phase-to-ground short-circuit fault (Fault 2) occurred on the parallel 400 kV overhead transmission line BJS_400_HVE between the BJS and HVE stations. Similarly, the faulted-phase breaker operated within 100 ms and was reclosed after 80 ms, but since the fault persisted, all three-phase breakers were subsequently opened, disconnecting the BJS_400_HVE line. Within 1–2 seconds of the disconnection of BJS_400_HVE, the synchronous condenser at BJS station (SynCon@BJS) tripped due to reverse active power protection, while the Storebaelt (SB) LCC-HVDC link at HKS station was blocked as a result of commutation failures. Subsequently, the KONTEK (KO) HVDC link was switched to manual control, and its active power setpoint was reduced. With the loss of these two lines, the 400 kV network in DK2 was split at BJS station. Consequently, the short-circuit capacity was drastically reduced—from approximately 8 GVA to 4.5 GVA at BJS station, and from 6.3 GVA to 4.1 GVA at HKS station. Prior to the incident, the active power flow in DK2 is illustrated in Figure 1. Following the loss of SB and the reduced import from KO, DK2 became highly dependent on active power imports from Sweden via the AC lines, which supplied nearly 75% of the total demand. Fig. 1 Active power flow before the incident (in MW) 2.2 Simulation reproduction The incident was reproduced using a full-system EMT model of the DK2 grid [5], with faults applied as shown in Fig. 2. Fig. 2 Fault implementation in simulations 3 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. For both Fault 1 and Fault 2, two consecutive single-phase-toground short circuits were modeled at the BJS end of each line, each lasting 100 ms, separated by an 80 ms interval. This interval represents the time between the initial breaker opening and its subsequent reclosure. During this interval, all three phases of the corresponding line remained connected in the simulation, whereas in reality only two phases were connected because the breaker of the faulted phase was open. This constitutes a minor discrepancy between the real incident and the simulation but is expected to have negligible influence on the results of this case study. The protection of SynCon@BJS and HVDC SB were not modeled; therefore, their tripping and blocking were not represented in the simulations. This omission made the instability phenomena more clearly visible in the results (Fig. 3) and enabled a clearer comparison with the mitigated stable case presented in Section 3. (a) (b) (c) (d) Fig. 3 Reproduction of the incident in simulations. (a) Voltage at 400 kV busbar of BJS station. (b) Powers of HVDC SB. (c) Powers of SynCon@BJS. (d) Powers of SynCon@HKS Fig. 3(a) shows the voltage at the 400 kV busbar of BJS station, which closely resembles the voltage at the 400 kV busbar of HKS station and explains the commutation failures in HVDC SB, as illustrated in Fig. 3(b). Fig. 3(c) and Fig. 3(d) present the dynamic responses of the two SynCons. During the second short circuit of Fault 2, SynCon@BJS injected around 350 Mvar of reactive power into the grid, while SynCon@HKS injected about 450 Mvar. Both devices exhibited significant active power swings following the disconnection of BJS_400_HVE, consistent with the real event in which SynCon@BJS tripped due to reverse active power protection. As observed from the results, the SynCons failed to damp the oscillations and restore system stability, which underscores their inherent limitations under severe contingencies. The system entered a sustained oscillatory state without converging to a steady operating point. Persistent oscillations in the power and voltage of different devices serve as clear indicators of system instability—phenomena that would not be expected in a stable case. Moreover, under unstable voltage conditions, the initial power exchanges at the interconnection boundaries, as illustrated in Fig. 1, cannot be sustained. In other words, the disturbed voltage profile affected the ability to transfer power from HVDC imports through the BJS station to eastern DK2 after the disconnection of the two lines. 2.3 Root cause investigation Further investigations were carried out to identify the root cause of the incident. The analysis confirmed that the most critical factor leading to the final system instability was the disconnection of BJS_400_HVE following Fault 2, rather than the short circuits themselves during the fault. This disconnection caused a significant change in system strength, as reflected by the variations in short-circuit capacity discussed in Section 2.1, which in turn prevented the system from maintaining stable voltages. Consequently, commutation failures occurred, leading to the blocking of HVDC SB. Therefore, the key to mitigating the incident lies in enhancing the voltage control capability of devices. In addition, damping also plays a significant role, particularly for SynCons. In the simulations, the tripping of SynCon@BJS and the blocking of HVDC SB were not modeled. Under these conditions the system could still maintain the initial power 4 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. exchanges at the interconnection boundaries shown in Fig. 1, provided it was stabilized after the two lines were disconnected. In such a scenario, HVDC imports could be successfully transferred to eastern DK2 via the BJS station through the remaining transmission paths. 3. Incident mitigation As discussed in Section 2.3, the system instability was mainly driven by insufficient voltage support following the disconnection of lines at BJS substation, and the SynCons in operation were unable to provide adequate control to prevent the incident. For Energinet, SynCons remain critical assets to enable stable operation of LCC-HVDC and to enhance system strength; however, their limitations in fast voltage control and damping capability highlight the need to investigate alternative technologies. GFM solutions, particularly GFM (E-)STATCOMs, offer promising capabilities to complement or substitute SynCons. Therefore, in this section, the potential of mitigating the incident through the integration of either a STATCOM or an E-STATCOM to the system is investigated. Three scenarios are investigated: (i) replacing SynCon@BJS with the device, (ii) replacing SynCon@HKS with the device, and (iii) adding the device at BJS station while retaining both SynCons in operation. For each scenario, the device’s capacity was varied to determine the minimum rating required to stabilize the system. 3.1 The parameters of the STATCOM and E-STATCOM The capacity of the STATCOM is characterized by a single parameter: the reactive power capacity (Q). In contrast, three parameters are relevant for defining the capacity of the ESTATCOM: reactive power capacity (Q), active power capacity (P), and time (T). Here, T denotes the duration for which the E-STATCOM can provide active power support— either injection or absorption—at its rated active power capacity P. The product of P and T represents the available energy stored in the supercapacitors of the E-STATCOM. This highlights the key distinction that, unlike the STATCOM, the E-STATCOM is also capable of providing active power support for a limited duration, thereby enhancing its effectiveness in system stabilization. 3.2 Replacing SynCon@BJS In the first scenario, SynCon@BJS at the 16 kV busbar was replaced by either a STATCOM or an E-STATCOM, both connected to the 400 kV busbar at the station. For the STATCOM case, its capacity was varied across a series of simulations. The system remained unstable at 200 MVA, whereas stability was achieved when the capacity was increased to 250 MVA. Thus, the required STATCOM capacity to stabilize the system lies between 200 and 250 MVA. Fig. 4 provides detailed results of a stable case in which an STATCOM rated at 270 MVA replaced SynCon@BJS. (a) (b) (c) (d) Fig. 4 Detailed testing results for STATCOM (270 MVA) replacing SynCon@BJS. (a) Voltage at 400 kV busbar of BJS station. (b) Powers of HVDC SB. (c) Powers of the STATCOM. (d) Powers of SynCon@HKS From these results, it can be observed that after the disconnection of line BJS_400_HVE, both voltage and power settle into new steady states. In the simulation reproduction of 5 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. the real incident (hereafter referred to as the original case), during the final short circuit period, SynCon@BJS injected around 350 Mvar of reactive power (Fig. 3(c)), whereas in the stable case the STATCOM injected only about 100 Mvar (Fig. 4 (c)). This demonstrates that the instability was not triggered directly by the short circuits, but rather by the subsequent disconnection of the BJS_400_HVE line. For the E-STATCOM case, its three parameters (Q, P, T) were varied, and a series of simulations were performed. An overview of the results is shown in Fig. 5. (a) (b) (c) Fig. 5 Overview of testing results for E-STATCOM replacing SynCon@BJS. (a) T = 0.5 s. (b) T = 1 s. (c) T = 1.5 s. (Green dot: E-STATCOM did not trip, and system was stable; Magenta dot: E-STATCOM did not trip, and system was unstable; Red dot: E-STATCOM tripped, and system was unstable; Gray area: operational zone of the E-STATCOM model.) In Fig. 5, each dot represents a single simulation run, with three different colours indicating three possible outcomes depending on whether the E-STATCOM tripped and whether the system eventually remained stable. Based on the results, the maximum E-STATCOM capacity that failed to stabilize the system was 224 MVA, while the minimum capacity that succeeded was 250 MVA. Thus, the required E-STATCOM capacity to stabilize the system lies between 224 and 250 MVA. The yellow dashed line indicates the approximate boundary between stable and unstable cases. In Fig. 5(a), one red dot appears on the stable side of this boundary. This occurs because the discussion in this section focuses on capacity (in terms of power) only, without considering the energy dimension. With the same power capacity but a larger T, the E-STATCOM is able to stabilize the system, as illustrated in Fig. 5(b) and Fig. 5(c). Therefore, that power capacity is considered as a stable case. This also indicates that the parameter T directly affects the simulated stability of the E-STATCOM itself. In the case of Q=250 Mvar and P=50 MW, the E-STATCOM tripped when T was small, but remained in service when T was increased. This implies that T—and thus the energy storage capacity of its supercapacitors—has a critical impact on the observed performance. In practice, however, it is expected that an ESTATCOM would be able to ride through such faults, with reduced active power support when energy is exhausted, thereby converging toward the behaviour of a STATCOM with very limited storage. Hence, the impact of T observed here should be interpreted as a modelling artifact rather than a fundamental limitation of the technology, and further investigation is warranted. Moreover, a comparison among cases with the same reactive power capacity Q (200 Mvar) reveals that a higher active power capacity P improves system stability. It is also noteworthy that in the original case, SynCon@BJS, with a capacity of 270 MVA, was unable to stabilize the system. By contrast, both the STATCOM and E-STATCOM achieved system stability with a smaller overall capacity, highlighting their advantage over the SynCon in terms of stabilizing capability. 3.3 Summary of investigations In the second scenario, SynCon@HKS at the 16 kV busbar was replaced by either a STATCOM or an E-STATCOM, both connected to the 400 kV busbar at the station. In the third scenario, either a STATCOM or an E-STATCOM was integrated to the 400 kV busbar at the BJS station in addition to the existing SynCons. Table 1 provides an overview of the results from all scenarios, where the minimum capacities that can stabilize the system are listed for different solutions. For the E-STATCOM case in the third scenario, even the minimum configuration supported by the model (158 MVA, Q = 150 Mvar, P = 50 MW, T = 0.5 s) was sufficient to stabilize the system. 6 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. Table 1 Overview of results from all investigations Minimum capacity (MVA) Add @BJS Replace SynCon@BJS Replace SynCon@HKS STATCOM 150-200 200-250 250-300 E-STATCOM ≤ 158 224-250 292-304 The results indicate that when SynCon@BJS was replaced by either a STATCOM or an E-STATCOM, both devices demonstrated a clear advantage in terms of the capacity required to stabilize the system, compared with the original SynCon. It should be noted that the models applied in this study were standard OEM-provided versions without casespecific tuning, meaning that their performance could be further optimized for specific projects or system conditions. The dynamic behaviour of the STATCOM when it stabilized the system in these scenarios is broadly consistent with that observed in the first scenario (Fig. 4) and is therefore not repeated. The differences in the dynamic behaviour between STATCOM and E-STATCOM are analysed in Section 4.2. 4. Comparison between different devices In this section, investigations were conducted to highlight different behaviour and capabilities of various devices. 4.1 Detailed comparison of E-STATCOM and SynCon In previous sections, comparison could be made between the E-STATCOM and SynCon@BJS within the system-level study case. However, in that context, their behavior directly influenced the voltage at the 400 kV busbar of BJS station, while in turn the busbar voltage also affected the dynamic responses of the devices. This interaction is particularly evident in the original case, where unstable voltage conditions had a significant impact on the performance of SynCon@BJS. To better isolate and compare the inherent capabilities of the two devices, SMIB tests were carried out. The setup is illustrated in Fig. 6. At 3 s, a single-phase-to-ground short circuit of 100 ms duration was applied at the point of connection (PoC) of the device under test (DUT). At 3.1 s, the short-circuit ratio (SCR) at the PoC was reduced from 10 to 3 by increasing the grid impedance, emulating the effect of disconnecting line BJS_400_HVE in the system-level study. The E-STATCOM has the same overall capacity of 270 MVA as SynCon@BJS. The results are shown in Fig. 7. Fig. 6 SMIB test setup (a) (b) (c) Fig. 7 Comparison between E-STATCOM and SynCon@BJS in SMIB test. (a) PoC voltage. (b) Reactive power response from DUT. (c) Active power response from DUT. From the results, it can be observed that during the fault period (between 3.0 and 3.1 s), SynCon@BJS injects more reactive power compared to the E-STATCOM. This is consistent with the observations from the system-level evaluations, where SynCon@BJS injected around 350 Mvar in the original case while E-STATCOM injected around 50 Mvar in the first scenario in Section 3.2. Consequently, the PoC voltage is slightly higher during the fault in the SMIB test with SynCon@BJS than with E-STATCOM, as shown in Fig. 7(b). As concluded in Section 2.3, the decisive factor in the systemlevel case is not their performance during the fault, but rather their different behaviors in the post-fault period—specifically, whether they can stabilize the system after the trip of the second line. Two aspects are particularly noteworthy. The first is their voltage control capability, i.e., the ability to resist voltage deviations at the PoC and to regulate the voltage back to its reference value in a fast and stable manner. From Fig. 7(a), it can be seen that after 3.1 s, the E-STATCOM restores the PoC voltage more quickly than SynCon@BJS. Moreover, once restored, the E-STATCOM maintains the 7 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. voltage tightly at 1 pu, whereas in the case of SynCon@BJS there is a noticeable overshoot that lasts for more than 1 s before stabilization. These differences highlight the more advantageous voltage control capability of the E-STATCOM. The second aspect is their damping capability, i.e., the ability to attenuate oscillations in their power responses. The difference in damping performance can be clearly observed in Fig. 7(c). During the post-fault period, the E-STATCOM exhibits stronger damping of active power oscillations than SynCon@BJS, both in the higher-frequency range (50 Hz) during the first 150 ms and in the lower-frequency range (around 4.18 Hz) over the following 2 seconds. Due to these two advantages in voltage control and damping capability, the E-STATCOM demonstrates a more favorable overall stabilizing performance compared to SynCon@BJS in the incident study case. This also explains why, in the systemlevel investigations, the E-STATCOM was able to stabilize the system with a smaller capacity than the SynCon@BJS. It is worth noting that SynCon is a particular application of synchronous machines, whose dynamic behavior constitutes the foundation of the GFM control concept. However, the comparison presented here reveals that different GFM devices may exhibit distinct performance in their GFM capabilities. In general, power-electronics-based GFM devices offer relatively greater flexibility in parameter adjustment, while SynCons are more dependent on hardware characteristics, which may be less easily modified. 4.2 Comparison between E-STATCOM and STATCOM This section compares the performance of the E-STATCOM and the STATCOM in the incident study case, under the scenario where SynCon@BJS is replaced. Both devices are configured with roughly the same overall capacity of 270 MVA. Specifically, the parameters of the E-STATCOM are set to Q = 250 Mvar, P = 100 MW, and T = 0.5 s. They also share identical control settings, including the voltage control integrator gain and droop coefficient. The results are presented in Fig. 8. To highlight the key dynamics, only the period during and after Fault 2 is shown. (a) (b) (c) (d) Fig. 8 Comparison between E-STATCOM and STATCOM. (a) Voltage at 400 kV busbar of BJS station. (b) Active power of HVDC SB. (c) Reactive powers of (E-)STATCOMs. (d) Active powers of (E-)STATCOMs. The results indicate that while both devices are capable of stabilizing the system, they exhibit distinct dynamic behaviors. First, during and after the fault, the STATCOM injects more reactive power into the grid than the E-STATCOM. Consequently, the post-fault voltage at the 400 kV busbar of BJS station is stabilized at a slightly higher level when the STATCOM is used. However, in the post-fault period, the reactive power response of the STATCOM shows stronger oscillations compared to the E-STATCOM. This leads to more pronounced oscillations in the 400 kV busbar voltage at BJS and in the active power flow of HVDC SB before stabilization, when the STATCOM is applied. These observations show the more favorable damping capability of the E-STATCOM. 8 For citation: L. Lu, Y. Liao, H. Gong, J. B. Kwon, N. Qin, M. Larsson, H. Zhang, and R. Heydari. Enhanced System Stability with GFM (E-)STATCOMs: A Case Study of the DK2 2020 Incident in the Danish Power System. Wind & Solar Integration Workshop. 2025. As expected, the disturbances also excite dynamics in the active power response of the E-STATCOM. Nevertheless, it is able to damp out these oscillations within 3 seconds. Furthermore, its active power response illustrates the expected GFM behavior, inherently responding to phase jumps in the grid, as measured at the PoC and shown in Fig. 9. Fig. 9 Phase changes at the PoC of E-STATCOM 5. Conclusion This paper investigated the dynamic performance and stabilizing capabilities of different grid support devices in the DK2 2020 incident study case. The analyses focused on replacing or supplementing the existing SynCons with (E-)STATCOMs, and on comparing their characteristics through both system-level simulations and dedicated SMIB tests. The main findings can be summarized as follows: Both STATCOM and E-STATCOM demonstrated more effective stabilizing capability compared to the SynCon. When replacing SynCon@BJS, the required capacity of the STATCOM or E-STATCOM to ensure stability was found to be lower than the original SynCon, indicating higher effectiveness in stabilizing the grid. While both devices can stabilize the system under the same overall capacity, the ESTATCOM shows better damping performance in reactive and active power responses than the STATCOM, reducing oscillations in the grid voltage and HVDC power flow. Comparative studies between E-STATCOM and SynCon in SMIB tests highlight that different GFM devices may exhibit distinct performance in their GFM capabilities. Powerelectronics-based GFM devices, such as (E-)STATCOMs, offer greater flexibility in parameter tuning, allowing their response to be tailored more effectively to system stability needs. In principle, similar dynamic support could be achieved with SynCons if their excitation, inertia, and damping were optimally configured, though such adjustments would require substantial hardware modifications and are constrained by physical limitations. By contrast, power-electronics-based GFM devices allow fast and wide-range parameter tuning through software control, enabling more efficient adaptation to evolving system needs with little or no hardware redesign. In summary, the investigations confirm that power-electronic GFM solutions, such as (E-)STATCOMs, can provide effective and practical capabilities for enhancing both voltage support and oscillation damping, thereby offering a promising complement to traditional SynCons in future power systems. The findings provide validation of the performance of GFM (E-)STATCOMs and support Energinet’s consideration of such devices as one of the promising solutions for addressing future stability challenges in a power system with high penetration of power-electronics-based renewable generation. 6. Future work GFM devices are generally considered as voltage-source-type resources. This category includes both traditional synchronous machines and power electronics devices operated with GFM control. This study suggests that different GFM devices may exhibit distinct performance in their inherent GFM capabilities. Therefore, a promising direction for future research and industrial deployment is to classify the family of GFM devices into different categories based on their expected GFM performance. Each category would correspond to differentiated specifications in technical requirements based on the performance in expected GFM capabilities. This performance-based classification also naturally respects the inherent constraints of different technologies, such as SynCons, wind turbines, PV panels, or BESS. For transmission system operators, an important next step is to quantify the system needs for GFM devices. This includes determining how much GFM resources are required to accommodate the increasing share of inverter-based renewable generation, or to support a future scenario with 100% powerelectronics-interfaced renewable generation. Developing a comprehensive roadmap for GFM deployment—addressing when, where, how many, and how large devices to install, as well as which technologies to adopt—will be essential to ensure stable, secure, and reliable system operation while aligning with long-term carbon neutrality goals. 7. Reference [1] Energinet, »Towards a stable and sustainable future with grid-forming technologies,« April 2025. [Online]. Available: https://en.energinet.dk/about-ourreports/reports/towards-a-stable-and-sustainable-futurewith-grid-forming-technologies/. [2] R. Heydari, »Grid-Forming Control for STATCOMs – a Robust Solution for Networks with a High Share of Converter-Based Resources,« CIGRE Paris Session, 2022. [3] R. Heydari, »Stability Enhancement of Weak Grids with High penetration of Renewables with Grid-Forming STATCOM/Enhanced STATCOM,« CIGRE Paris Session, 2024. [4] P. Mitra og e. al., »Simulations and Field Experience of Opladen STATCOM with Grid-Forming Behavior,« CIGRE Symposium Vienna, 2023. [5] Y. Liao, L. Lu, J. Kwon, N. Qin, D. Muthumuni, Y. Pipelzadeh og K. Dirks, »Enabling system-level EMT studies of Danish power systems,« CIGRE Paris Session, 2024.