Abstract
The increasing penetration of variable renewable energy (VRE) generation along with the decommissioning of conventional power plants in Chile, has raised several operational challenges in the Chilean National Power Grid (NPG), including transmission congestion and VRE curtailment. To mitigate these limitations, an innovative virtual transmission solution based on battery energy storage systems (BESSs), known as grid booster (GB), has been proposed to increase the capacity of the main 500 kV corridor of the NPG. This paper analyzes the dynamic performance of the GB using a wide-area electromagnetic transient (EMT) model of the NPG. The GB project, composed of two 500 MVA BESS units at each extreme of the 500 kV corridor, allows increasing the transmission capacity for 15 min during contingencies, overcoming transmission limitations under normal operation conditions while maintaining system stability during faults. The dynamic behavior of the GB is also analyzed to control power flow as well as voltage stability. The results show that the GB is an effective solution to allow greater penetration of VRE generation while maintaining system stability in the NPG.
THE development and operation of the Chilean National Power Grid (NPG) are undergoing significant changes due to the increasing penetration of variable renewable energy (VRE) generation, creating numerous challenges for Coordinador Eléctrico Nacional (CEN), the Chilean electric system operator [
CEN is committed to facilitating the energy transition and the goal of reaching carbon neutrality by 2050, contributing to the development of a sustainable power grid in Chile. As part of this vision, a roadmap has been developed to phase out all coal-based power plants and achieve a 100% share of renewable generation by 2030 [
The GB, also referred to as virtual transmission solution based on battery energy storage systems (BESSs), has the potential to increase power export capability without necessitating the construction of new transmission lines [

Fig. 1 Chilean NPG segmented by voltage control areas (VCAs).
Historically, root mean square (RMS) or phasor domain transient (PDT) simulations have been the standard methods to study the dynamic behavior of power systems composed primarily of synchronous generators (SGs) [
The EMT-NPG database encompassed the modeling of all transmission lines in the bulk grid (220 kV and 500 kV), generation plants above 20 MW, and relevant dynamic voltage control devices. Furthermore, given their significance in voltage stability, the metal oxide varistor (MOV) systems associated with the series capacitors along the primary 500 kV corridors were also incorporated into the model.
The main contribution of this paper is the detailed modeling and validation of the design and dynamic performance of GB project in a large-scale EMT-NPG. This project is the first real-world application of this type and scale. In addition, the full GB model, including BESS and control schemes, was developed and integrated into a wide-area EMT database for accurate and detailed stability analysis.
The rest of this paper is organized as follows. Section II describes the main characteristics of the Chilean NPG and its EMT representation. Section III presents the GB project configuration, main parameters, and EMT model. Section IV presents the simulation results. Finally, the conclusions are presented in Section V.
The Chilean NPG is characterized by its extended and islanded topology representing a complex operational configuration as we transition to a 100% renewable scenario, mostly dominated by IBRs. The peak demand in 2022 reached 11.9 GW, which was supplied by an installed capacity of 33 GW, encompassing a diverse mix of thermal (41%) and renewable energy (59%) generation sources. The Chilean NPG is characterized by five VCAs, corresponding to distinct geographical regions: Big North, Small North, Center, Center South, and South, as shown in
VCA | Demand (GW) | Installed capacity (GW) | |||
---|---|---|---|---|---|
Solar | Wind | Hydro | Thermal | ||
Big North | 2.7 | 3.3 | 0.9 | <0.1 | 4.7 |
Small North | 1.1 | 2.5 | 1.6 | <0.1 | 0.8 |
Center | 3.5 | 0.8 | 1.0 | 3.0 | |
Center South | 2.5 | 0.9 | 0.9 | 5.6 | 1.0 |
South | 0.7 | 1.2 | 0.6 | ||
Total | 10.5 | 7.5 | 4.6 | 7.3 | 9.5 |
In recent years, the Chilean NPG has significantly evolved, facing multiple changes including:
1) Closure of thermal generators: closure of several thermal generators throughout the system.
2) Construction of VRE: construction of large-scale solar and wind power plants in the northern region.
3) Congestion in transmission corridors: congestion of the main transmission lines connecting the North (Big North and Small North) area, where solar and wind generation systems are concentrated, and the Center area, where a significant portion of demand is located. This issue has resulted in curtailment of VRE generation in both the Big North and Small North areas.
All these changes in the Chilean NPG have raised a need for innovative solutions to increase transmission capacity in a cost-efficient manner while maintaining grid stability. Despite the long-term strategies being considered, urgent short-term measures are required to address immediate challenges and constraints in the grid. The capacity of the Parinas-Lo Aguirre 500 kV corridor, a 1100 km heavily compensated transmission system, is 1700 MVA, which is primarily driven by voltage stability constraints. Thus, the need to increase power flow transfers in this corridor has become critical to reduce VRE curtailments. Innovative solutions such as the GB project aim at alleviating this constraint by temporally increasing the transmission capacity during faults while maintaining system stability.
The complexities of Chilean NPG, combined with the country’s ambitious goal for VRE integration, demand a holistic and forward-thinking approach to grid development and management of operations. The intricate interactions between VRE, conventional generation technologies, and transmission infrastructure necessitate advanced simulation tools and novel solutions like the GB project.
This study considers two daytime-peak operation scenarios projected for the year 2025, under conditions of the maximum instantaneous VRE penetration, covering a capacity of 400 MW or 500 MW for the BESS units due to the requirements of the project.
Parameter | Scenario 1 | Scenario 2 |
---|---|---|
Total generation | 10437.0 MW | 10522.4 MW |
VRE generation | 6787.2 MW | 6962.6 MW |
Total load | 9991.9 MW | 10136.9 MW |
Loss | 445.51 MW | 385.4 MW |
The maximum power transfer (NPA-Polpaico) | 2444.2 MW | 2377.2 MW |
BESS/VRE ratio | 7.37% | 7.18% |
The EMT-NPG used in this paper corresponds to the first version of an EMT representation of the Chilean NPG developed in EMT
The EMT-NPG is composed of 62 SGs, which are modeled considering salient-pole single-mass for the thermal units, and round rotors for the hydro units. All machines are modeled with their excitation, power stabilizers, and governor’s systems [
The model is composed of 32 wind parks and 59 solar plants. VRE generators have been reproduced from the models available in the phasor-domain tool used by CEN, and Generic Western Electricity Coordinating Council (WECC) models [
The model includes 678 transmission lines consisting of 500 kV, 220 kV, and 110 kV lines. Lower voltage levels are represented by aggregated equivalent loads. Lines longer than 10 km are represented by distributed parameter models with constant parameters [
The GB project is located between the Parinas and Lo Aguirre 500 kV substations. Its primary aim is to increase the power transfers along the 500 kV corridor between the Parinas and Lo Aguirre substations, to allow exceeding the normal capacity in this corridor by 400-500 MW. This is achieved through dynamic active power compensation, delivered by two 500 MVA BESS units installed at each substation. Thus, in case of losing one line circuit, the BESS units will simultaneously absorb (and inject) energy at the two extremes of the corridor to compensate for the excess of power transfer to avoid exceeding the limit of the healthy circuit. The general specifications of the GB project (size and duration) analyzed in this paper, are defined based on the current level of VRE curtailments due to transmission constraints in the 500 kV corridor of the NPG.
The BESS units shall be coordinated utilizing a redundant high-speed communication and control system, which incorporates a central control system (CCS) and a power plan controller (PCC) at each BESS site [
1) Normal operation mode: during normal operation mode, considering the power transfers from the North to the South, the systems maintain an operational setting of the minimum state of charge (SoC) for the BESS unit at Parinas substation (BESS 1) and the maximum SoC for the BESS unit at Lo Aguirre substation (BESS 2), as shown in

Fig. 2 Schematic diagram of normal operation mode.
2) Contingency operation mode: if a fault occurs at one of the circuit sections along the 500 kV corridor, causing the loss of the faulted line, both BESS units shall operate in a coordinated manner to maintain the pre-fault power flow balance without overloading the parallel healthy line. This is achieved by simultaneously charging (absorbing energy) BESS 1 and discharging (injecting energy) BESS 2, minimizing the impact on the grid without violating grid code requirements. This operation mode is illustrated in

Fig. 3 Schematic diagram of contingency operation.
A simplified diagram of the BESS modeled in the EMT-NPG is presented in

Fig. 4 Block diagram of BESS model.
The architecture of GB control system presented in

Fig. 5 Architecture of GB control system.
The active power output in the converter can be linearly ramped from 0 to the maximum power of 500 MW once the activation signal is received using a settable ramp rate in MW/s. The activation signal for the converters is generated 100 ms after fault clearance in the 500 kV corridor, represented by the communication delay . The main parameters of GB are detailed in
Parameter | Value | Parameter | Value |
---|---|---|---|
Capacity | 125 MWh | Power | 500 MVA |
25 kV | 1 | ||
0.95 p.u. | 0.05 p.u. |
The reference active power is defined as:
(1) |
where is set to be 500 MW, and the ramp rate can take different values depending on the specific scenario under consideration.
This section presents the simulation results for different contingencies. The goal is to assess the dynamic behavior and performance of the GB under fault conditions.
The analysis aims to evaluate the dynamic behavior of the GB and to define the specifications and technical requirements for key performance features. The following aspects of the GB control system are assessed.
1) Active power ramp rate to keep the grid stable.
2) Short-term overload.
3) Power factor range and dynamic reactive power compensation during and after faults.
The analysis is comprised of the following two stages.
1) Preliminary contingency case simulations: identify the worst case by simulating contingencies across the 500 kV corridor with different active power ramp rates in MW/s. Twelve two-phase-to-ground (2PG) faults, at each 500 kV line section, are simulated. An example of fault location is shown in
2) Worst-case simulations: for the worst case identified in the previous stage, simulations are performed considering different active power ramp rates in MW/s and FRT limits for the dynamic reactive power compensation devices near the 500 kV corridor. The goal is to identify the critical parameter and control setpoints required to keep the NPG stable in compliance with the performance requirements in the grid code.
All time-domain EMT simulations are initialized from the multiphase load-flow solver available in EMT
The preliminary contingency cases evaluated, considering fixed FRT limits of the reactive power compensator (, ), are summarized in
Case | Scenario | Active power ramp rate of BESS (MW/s) |
---|---|---|
1 | 1 | 50 |
2 | 1 | 500 |
3 | 2 | 50 |
4 | 2 | 500 |
Twelve faults at different locations in the 500 kV corridor were analyzed for each case, as shown in
Fault | Line | Location |
---|---|---|
F1 | Parinas-Cumbre | North |
F2 | Parinas-Cumbre | South |
F3 | Cumbre-Nueva Cardones | North |
F4 | Cumbre-Nueva Cardones | South |
F5 | Nueva Cardones-Nueva Maitencillo | North |
F6 | Nueva Cardones-Nueva Maitencillo | South |
F7 | Nueva Maitencillo-NPA | North |
F8 | Nueva Maitencillo-NPA | South |
F9 | NPA-Polpaico | North |
F10 | NPA-Polpaico | South |
F11 | Polpaico-Lo Aguirre | North |
F12 | Polpaico-Lo Aguirre | South |
Based on the preliminary contingency case simulations, fault F10 occurring within the 500 kV segment NPA-Polpaico, as shown in

Fig. 6 Worst case for all preliminary contingency cases.
Cases 1 and 2: both cases exhibit clear signs of instability. This is illustrated in

Fig. 7 Voltages and frequency for Case 1. (a) Voltages at Cumbre busbar. (b) Frequency at Polpaico busbar under fault F10.

Fig. 8 Voltages and frequency for Case 2. (a) Voltages at Cumbre busbar. (b) Frequency at Polpaico busbar under fault F10.
Cases 3 and 4: in these two cases, the simulations are fully stable and comply with the requirements of the grid code. As shown in

Fig. 9 Voltage and frequency for Case 3. (a) Voltages at Cumbre busbar. (b) Frequency at Polpaico busbar under fault F10.

Fig. 10 Voltages and frequency for Case 4. (a) Voltages at Cumbre busbar. (b) Frequency at Polpaico busbar under fault F10.
From

Fig. 11 GB power injection and active power transfer on NPA-Polpaico line for Case 3 under fault F10. (a) GB power injection. (b) Active power transfer on NPA-Polpaico line.

Fig. 12 GB power injection and active power transfer on NPA-Polpaico line for Case 4 under fault F10. (a) GB power injection. (b) Active power transfer on NPA-Polpaico line.
Judging by the results of the preliminary contingency cases, the worst case is the fault F10, located at the end of Polpaico busbar. It is the only fault that always leads to instability in scenario 1. Therefore, it was used for the worst-case analysis.
This subsection presents an extended analysis of the worst case (fault F10) for scenario 1. Modifications will be made to the FRT limits of the reactive power compensator presented in the system. Additionally, variations will be introduced to the active power ramp rates in the response of the BESS. The specifics of these variations can be found in
Condition | Active power ramp rate of BESS (MW/s) | FRT limits (p.u.) | |||
---|---|---|---|---|---|
GB | Compensator | ||||
WC1 | 50 | 0.2 | 0.200 | 0.5 | 0.075 |
WC2 | 1000 | 0.2 | 0.075 | 0.1 | 0.075 |
WC3 | 500 | 0.2 | 0.075 | 0.1 | 0.075 |
1) Worst-case condition 1 (WC1): the system remains stable under this condition, as shown in

Fig. 13 Voltages and frequency for WC1. (a) Voltages at different busbars. (b) Frequency at Polpaico busbar.
Altering the FRT limits requires reactive power compensators to stay connected even under more extreme conditions beyond stable operation. This ensures that voltage levels remain within the range specified by the grid code, maintaining the system stability even when the active power ramp rate is 50 MW/s, as illustrated in

Fig. 14 GB power injection and active power transfer on NPA-Polpaico line for WC1. (a) GB power injection. (b) Active power transfer on NPA-Polpaico line.
2) Worst-case condition 2 (WC2): as illustrated in

Fig. 15 Voltages and frequency for WC2. (a) Voltages at different busbars. (b) Frequency at Polpaico busbar.
An important improvement is observed with the increase of the active power ramp rate, as shown in

Fig. 16 GB power injection and active power transfer on NPA-Polpaico line for WC2. (a) GB power injection. (b) Active power transfer on NPA-Polpaico line.
3) Worst-case condition (WC3): under this condition, a voltage dip below 0.8 p.u., at certain busbars, violates the grid code. Although frequency is not the primary issue, a slight over-frequency occurs when compared with WC2, as observed in

Fig. 17 Voltages and frequency for WC3. (a) Voltages at different busbars. (b) Frequency at Polpaico busbar.

Fig. 18 GB power injection and active power transfer on NPA-Polpaico line for WC3. (a) GB power injection. (b) Active power transfer on NPA-Polpaico line.
This paper provides an analysis of the impact of the GB project on the Chilean NPG using EMT simulations. Key findings of this study are as follows:
1) The GB increases the transmission capacity and maintains the system stability.
2) FRT limits of reactive power compensators are critical to grid stability. Large FRT limits ensure that voltages are kept within the requirements of the grid code, while lower limits require higher active power ramp rates.
3) This study introduces the first wide-area EMT model for the NPG, unlocking new opportunities for advanced and complex time-domain stability analyses and research.
These findings reinforce the effectiveness of the GB and contribute valuable guidelines for implementing innovative solutions.
Future work may consider the assessment and impact of incorporating grid-forming capabilities in the BESS to support and improve system stability, as well as supplementary features and capabilities such as damping oscillations, voltage unbalance and harmonics compensation, and resonance mitigation.
References
CEN. (2023, May). Chilean Independent System Operator. [Online]. Available: https://www.coordinador.cl [Baidu Scholar]
F. Arraño-Vargas and G. Konstantinou, “Longitudinal power systems for modern and future grid studies: a graph theory analysis,” Sustainable Energy, Grids and Networks, vol. 35, p. 101132, Sept. 2023. [Baidu Scholar]
CEN. (2022, Jun.). Roadmap for an accelerated energy transition: vision of the Chilean system operator. [Online]. Available: https://www.coordinador.cl [Baidu Scholar]
J. Porst, J. Richter, G. Mehlmann et al., “Operation of grid boosters in highly loaded transmission grids,” in Proceeding of 2022 IEEE International Conference on Power Systems Technology, Kuala Lumpur, Malaysia, Sept. 2022, pp. 1-6. [Baidu Scholar]
CNE. (2019, Jun.). NTSyCS, Chilean grid code. [Online]. Available: https://www.cne.cl [Baidu Scholar]
M. Lindner, D. Mende, A. Wasserrab et al., “Corrective congestion management in transmission grids using fast-responding generation, load and storage,” in Proceeding of 2021 IEEE Electrical Power and Energy Conference, Toronto, Canada, Oct. 2021, pp. 1-6. [Baidu Scholar]
T. A. Nguyen and R. H. Byrne, “Evaluation of energy storage providing virtual transmission capacity,” in Proceeding of 2021 IEEE PES General Meeting, Washington DC, USA, Jul. 2021, pp. 1-5. [Baidu Scholar]
CNE. (2021, Aug.). 2021 transmission expansion plan. [Online]. Available: https://www.cne.cl [Baidu Scholar]
P. Kundur, Power System Stability and Control. New York: McGraw-Hill, 1994. [Baidu Scholar]
F. Arrano-Vargas and G. Konstantinou, “Development of real-time benchmark models for integration studies of advanced energy conversion systems,” IEEE Transactions on Energy Conversion, vol. 35, no. 1, pp. 497-507, Mar. 2020. [Baidu Scholar]
F. Milano, F. Dorfler, G. Hug et al., “Foundations and challenges of low-inertia systems,” in Proceeding of 2018 Power Systems Computation Conference, Dublin, Ireland, Jun. 2018, pp. 1-25. [Baidu Scholar]
J. D. Lara, R. Henriquez-Auba, D. Ramasubramanian et al., “Revisiting power systems time-domain simulation methods and models,” IEEE Transactions on Power Systems. doi: 10.1109/TPWRS.2023. 3303291 [Baidu Scholar]
M. Ouafi, J. Mahseredjian, J. Peralta et al., “Parallelization of EMT simulations for integration of inverter-based resources,” Electric Power Systems Research, vol. 223, p. 109641, Oct. 2023. [Baidu Scholar]
J. Mahseredjian, V. Dinavahi, and J. A. Martinez, “Simulation tools for electromagnetic transients in power systems: overview and challenges,” IEEE Transactions on Power Delivery, vol. 24, no. 3, pp. 1657-1669, Jul. 2009. [Baidu Scholar]
U. Karaagac, J. Mahseredjian, and O. Saad, “An efficient synchronous machine model for electromagnetic transients,” IEEE Transactions on Power Delivery, vol. 26, no. 4, pp. 2456-2465, Oct. 2011. [Baidu Scholar]
P. Pourbeik. (2018, Jul.). Model user guide for generic renewable energy system models, EPRI. [Online]. Available: https://www.epri.com/research/products/000000003002014083 [Baidu Scholar]
P. Pourbeik, J. J. Sanchez-Gasca, J. Senthil et al., “Generic dynamic models for modeling wind power plants and other renewable technologies in large-scale power system studies,” IEEE Transactions on Energy Conversion, vol. 32, no. 3, pp. 1108-1116, Sept. 2017. [Baidu Scholar]
WECC Renewable Energy Modeling Task Force. (2014, Jan.). WECC wind plant dynamic modeling guidelines. [Online]. Available :https://www.esig.energy/wiki-main-page/wind-plant-power-flow-modeling-gui-de/ [Baidu Scholar]
H. Dommel, “Digital computer solution of electromagnetic transients in single-and multiphase networks,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-88, no. 4, pp. 388-399, Apr. 1969. [Baidu Scholar]
CEN. (2023, Jun.). BESS functional technical specifications. [Online]. Available: https://www.coordinador.cl [Baidu Scholar]
H. Saad, S. Dennetiere, J. Mahseredjian et al., “Modular multilevel converter models for electromagnetic transients,” IEEE Transactions on Power Delivery, vol. 29, no. 3, pp. 1481-1489, Jun. 2014. [Baidu Scholar]
A. Yazdani and R. Iravani, Voltage-sourced Converters in Power Systems. Hoboken: Wiley, 2010. [Baidu Scholar]
S. Golestan, J. M. Guerrero, and J. C. Vasquez, “Three-phase PLLs: a review of recent advances,” IEEE Transactions on Power Electronics, vol. 32, no. 3, pp. 1894-1907, Mar. 2017. [Baidu Scholar]