A Microgrid Test System for Protection Coordination Studies

Abstract

This paper proposes a comprehensive 26-bus microgrid (MG) test system designed to validate or propose new protection coordination schemes. The proposed MG test system comprises various components facilitating the simulation of diverse configurations and operating modes. It also presents a predefined overcurrent protection scheme which is validated with and without Distributed Generation (DG), offering detailed specifications and settings for the test network. The MG test system outlined in this paper incorporates multiple DG sources in different buses, multiple voltage levels, and two operating modes (with and without DG). Additionally, the overcurrent protection scheme considers multiple functions and devices simultaneously. The detailed features of the proposed protection coordination scheme enable the simulation of various fault types and contingencies. The paper also presents specific details of the MG test system under different single-phase and three-phase faults, reporting bus voltages in nodes, transformer and line loadability, short-circuit currents, and operating times. The MG test system is meant to serve as a valuable tool for researchers to conduct comparative studies and develop new strategies regarding the operation and planning of MGs.

1. Introduction

Protection coordination is crucial for the design and operation of electric power systems, ensuring reliable and safe functioning. While conventional systems facilitate protection coordination via unidirectional power flows, the recent integration of Distributed Energy Resources (DERs) in electric power systems brings along new challenges, including unbalanced conditions, low inertia, variations in short-circuit levels, and the management of bidirectional power flows. Consequently, conventional protection schemes, employed in DER integration, must prove their effectiveness. Overcurrent relays (OCRs) serve as the commercial solution for fault detection in power systems; however, under DER integration, systematic methodologies for protection coordination in various microgrid (MG) topologies and operation modes remain a focal point of current research [1,2,3].

Proper planning and implementation of a coordinated protection system is essential for efficiently managing and ensuring the reliable operation of DERs in MGs [4,5,6]. Effective coordination among various protection devices not only enhances reliability but also strengthens the system’s security [7,8]. To advance the formulation of protection coordination methodologies in MGs, it is crucial to address in detail the challenges they face, such as the high penetration of DG [9,10,11,12], bidirectional power flows [13,14], and variation in the fault current concerning the mode of operation [15,16]. Addressing these challenges is essential for developing effective protection coordination schemes and ensuring the reliability and security of renewable energy systems. However, due to the multitude of MG topologies, addressing these challenges is not an easy task. Also, the possibility of comparing results between researchers requires the availability of standard MG test systems.

Test networks are very common in electrical engineering, especially in the area of power systems where IEEE test systems stand out. However, after a thorough review, it was found that there are no test networks for MGs that provide full data availability and a variety of devices, such as conventional and non-conventional generators, storage, etc. Different types of MGs have been used in the technical literature to validate overcurrent protection coordination schemes. These MG test systems feature different characteristics and protection coordination schemes. A summary of the main characteristics of these MGs is presented in

Table 1. Characteristics of MG test systems in previous research work.

Reference	Authors	DERs Considered				OCR Functions		Data Availability
		Batteries	Wind	PV	Conv. DG	Single	Multip.	Partial	Total
[1]	(Zeineldin et al., 2005)				x	x		x
[14]	(Elsamahy et al., 2022)				x			x
[15]	(Singh et al., 2016)				x			x
[16]	(Sortomme et al., 2008)		x	x	x			x
[17]	(Saleh et al., 2017)				x	x		x
[18]	(Baghaee et al., 2018)		x	x	x	x		x
[19,20]	(Alam et al., 2020; Zhang et al., 2019)	x	x	x	x	x		x
[21]	(Lin et al., 2019)		x		x	x		x
[22,23]	(Alam et al., 2019; Tiwari et al., 2022)		x	x	x	x		x
[24,25]	(Plongkrathok et al., 2022; Abeid et al., 2022)		x		x	x		x
[26]	(Sampaio et al., 2022)				x	x		x
[27]	(Biswal et al., 2020)		x	x		x		x
[28]	(Alam et al., 2021)		x	x	x	x		x
[29]	(Usta et al., 2019)	x	x	x		x		x
[30]	(Seo et al., 2019)				x	x		x
[31]	(Altaf et al., 2020)		x	x		x			x
[32]	(Chae et al., 2021)	x			x	x		x
[33]	(Bakkar et al., 2022)					x		x
Proposed		x	x	x	x		x		x

. Note that the proposed MG test system stands out by the fact that it provides a variety of DERs, multiple OCR functions, and total data availability.

One of the main contributions of this work is to propose an MG test system containing a detailed overcurrent protection scheme that facilitates the study and validation of different protection coordination schemes. Among the main characteristics of the proposed MG test system are different DG sources, different voltage levels, different operation modes, and a detailed model of each of the components. It is also sought that the overcurrent protection scheme is as complete as possible. For this reason, it considers the following functions and devices simultaneously:

• Phase and neutral inverse-time overcurrent functions.
• Phase and neutral inverse-time directional overcurrent functions.
• Phase and neutral instantaneous overcurrent functions.
• Phase and neutral definite-time overcurrent functions.
• Phase and neutral instantaneous directional overcurrent functions.
• Phase and neutral definite-time directional overcurrent functions.
• Fuses.

The detailed characteristics of the protection coordination scheme proposed for the test microgrid allow us to perform different types of faults considering the multiple functions of the relays. It also integrates different types of DERs. This represents an advantage when compared to the MGs presented in Table 1. In this case, several limitations were identified in the consulted works, which include the unavailability of data, the use of only one type of function in the OCRs, and the limitation in the DERs modeled in the MG.

Benchmark test systems such as the one proposed in this paper play a crucial role in electrical engineering for several reasons. These systems provide a standardized and controlled environment for evaluating the performance of electrical devices and systems, offering several advantages such as standardization, transparency, and comparability of results.

This paper aims to introduce an MG test system for assessing various protection coordination schemes and supporting researchers in this field. Apart from presenting the MG test system, an initial protection coordination scheme is also suggested and validated using diverse protective equipment. This scheme can serve as a reference for other researchers to propose modifications, facilitating comparisons and evaluations of their own protection schemes against the suggested one. In this sense, the main contributions of this paper are as follows:

• A complete MG test system is proposed validating existing protection schemes and proposing new ones.
• The proposed MG test system features multiple elements for simulating various configurations and operational modes.
• Pre-set overcurrent protection schemes for both DG and non-DG operation, where detailed parameters and setting explanations are presented.

The proposed MG test system includes diverse DG sources, multiple voltage levels, different operating modes, and detailed models of each component. Additionally, the overcurrent protection scheme aims to be as comprehensive as possible, considering several functions and devices simultaneously. The detailed characteristics of the proposed protection coordination scheme for the MG test system facilitate the simulation of various fault types and the analysis of multiple considerations. This represents a notable advancement compared to other schemes documented in the technical literature. The tests of the MG are carried out via simulation in Digsilent Power Factory, a power system analysis software application widely known and used in the electrical industry [34].

This paper is organized as follows: Section 2 outlines the MG test system data, including a detailed description of the test MG with its corresponding parameters and the protection coordination scheme of the test systems. Section 3 presents the results, incorporating details concerning the coordination protection scheme with and without DG. Finally, Section 4 concludes and summarizes the most relevant findings.

2. Microgrid Test System Data

The proposed MG test system is not a physical network, but rather a simulated network that can be used for research purposes to perform tests and validate different protection coordination methodologies. The simulation of the MG was carried out using Digsilent Power Factory [34]. Nonetheless, other specialized computer software may be used for this purpose. The use of computer software allows for the creation of an approximate model of the MG, enabling experiments that simulate its real-world behavior to a reasonable extent. Based on the simulation model and the experiments conducted, design, planning, and operational decisions can be extrapolated and adapted to real MGs. Furthermore, researchers may use the proposed MG test system to test, validate, and compare hypotheses or approaches regarding protection coordination methodologies.

2.1. Parameters

Figure 1 illustrates the one-line diagram of the proposed MG test system. In this case, two operative scenarios are considered. In scenario 1, the DG units are turned off and the demand is fed only via the external grid. In scenario 2, the DG units are turned on to feed the demand along with the external grid. For both operative scenarios, single-phase and three-phase faults are considered at each bus to show their maximum short-circuit currents. These are calculated using the full or superposition method, in which the short-circuit fault currents are determined by superimposing a normal load flow condition prior to short-circuit initiation where all voltage supplies are set to zero and the negative operating voltage is switched on at the fault location [34]. The proposed MG test system includes DG units, distribution, storage, and protection assets. The characteristics of these elements are described in this section to provide enough information for a recreation of this same microgrid by other researchers.

The test network consists of 26 nodes, the nominal voltages of which are specified in

Table 2. Nominal voltage magnitudes of the proposed MG test system.

Bus	Label	Voltage [kV]	Bus	Label	Voltage [kV]
1	B1-34.5	34.50	14	B14-4.16	4.16
2	B2-0.48	0.48	15	B15-0.48	0.48
3	B3-0.48	0.48	16	B16-0.48	0.48
4	B4-0.48	0.48	17	B17-0.48	0.48
5	B5-34.5	34.50	18	B18-0.48	0.48
6	B6-34.5	34.50	19	B19-0.21	0.21
7	B7-34.5	34.50	20	B20-13.2	13.2
8	B8-0.48	0.48	21	B21-13.2	13.2
9	B9-0.48	0.48	22	B22-0.48	0.48
10	B10-0.48	0.48	23	B23-0.48	0.48
11	B11-0.48	0.48	24	B24-0.48	0.48
12	B12-34.5	34.50	25	B25-0.48	0.48
13	B13-4.16	4.16	26	B26-0.48	0.48

. There are several voltage levels that range from 0.214 to 34.5 kV. Each branch of the system, as well as the external network, is connected to bus 1; the DG sources and demands are connected to the remaining buses.

The proposed external network emulates an interconnected system to which the test network can be connected. The values of the parameters of the external network are detailed in

Table 3. External grid data of the proposed MG test system.

Max. Values		Min. Values
Short-Circuit Power Sk″max	356.3 MVA	Short-Circuit Power Sk″min	318.1 MVA
Short-Circuit CurrentIk″max	5.962605 kA	Short-Circuit Current Ik″min	5.323337 kA
c-Factor (max.)	1.1	c-Factor (min.)	1
R/X (max.)	0.2	R/X (min.)	0.233
Impedance Ratio		Impedance Ratio
Z2/Z1 max.	1.164126	Z2/Z1 min.	1.163282
X0/X1 max.	1.058075	X0/X1 min.	1.069984
R0/X0 max.	0.1332985	R0/X0 min.	0.16607

.

Transformers are common elements in distribution systems; they are used to lower voltage levels and allow for easy use of electrical energy by the demand side. The proposed MG test system has 19 transformers with capacities ranging from 0.15 MVA to 3.0 MVA. These transformers have high voltage connections up to 34.5 kV and low voltage connections up to 0.21 kV.

Table 4. Transformer data of the MG test system.

Transformer	Connection	Size [MVA]	High Voltage [kV]	Low Voltage [kV]	Impedance Z [%]
T1	Dyn5	3.00	34.50	0.48	10.15
T2	Dyn5	2.00	34.50	0.48	10.18
T3	Dyn5	3.00	34.50	0.48	10.15
T4	Dyn5	2.00	34.50	0.21	4.88
T5	Dyn5	3.00	36.00	0.48	7.15
T6	Dyn5	3.00	36.00	0.48	7.15
T7	Dyn5	0.15	0.48	0.22	4.00
T8	Dyn5	3.00	13.20	0.48	5.75
T9	Dyn5	2.00	34.50	0.48	11.27
T10	Dyn5	2.00	34.50	0.48	11.23
T11	Dyn5	2.00	34.50	0.48	13.08
T12	Dyn5	1.50	34.50	0.48	4.00
T13	Dyn5	3.00	34.50	0.48	10.97
T14	Dyn5	3.00	34.50	0.48	13.01
T15	Dyn5	1.00	34.50	4.16	7.83
T16	Dyn5	1.00	34.50	4.16	7.83
T17	Dyn5	1.00	34.50	0.48	7.25
T18	Dyn5	1.00	34.50	0.48	7.25
T19	Dyn5	3.00	13.20	0.48	5.75

describes the main characteristics of the 19 three-phase two-winding transformers contained in the test network.

The proposed MG test system has five distribution lines used to distribute power between different buses. These lines range in length from 300 to 1900 m. In this case, the line labeled as L5 supports a voltage of 15 kV, while the rest of the lines are at 34.5 kV.

Table 5. Line data of the proposed MG test system.

Line	Voltage [kV]	Length [km]	Rated Current [kA]	R1 [Ohm/km]	R0 [Ohm/km]	X1 [Ohm/km]	X0 [Ohm/km]
L0	34.500	0.400	0.29290	0.17440	0.56132	0.39780	1.37544
L1	34.500	0.500	0.19610	0.26973	0.44219	0.44252	1.86004
L2	34.500	0.150	0.29290	0.17384	0.34612	0.39850	1.84224
L3	34.500	1.900	0.19610	0.26973	0.44219	0.44252	1.86004
L4	15.000	0.300	0.34790	0.06400	0.24004	0.04100	1.80540
L5	34.500	0.300	0.29290	0.17384	0.34612	0.39850	1.84224

presents the data of the distribution lines of the proposed MG test system.

There are different DG technologies that include the use of not only renewable but also conventional resources. The conventional generation considered in the MG test system consists of seven synchronous generators, of which three are smooth cylindrical pole generators (G1, G2, and G3) that may represent small diesel thermal plants, while the other four are salient pole generators (G4, G5, G6, and G7) that may represent small hydropower plants or biomass electric generators, among others. All seven generators are considered to have a rated power of 1.0 MVA.

The generators have different parameters indicated in Table 6, Table 7 and Table 8. These parameters are listed below:

• S: Nominal apparent power.
• Vn: Voltage Nominal.
• Fp: Power Factor.
• Rstr: Stator resistance.
• Xl: Stator leakage reactance.
• Xd: Direct-axis reactance.
• Xq: Quadrature-axis reactance.
• H: Time constant of inertia.
• Td0’: Transient direct-axis time constant.
• Td0”: Subtransient direct-axis time constant.
• Tq0’: Transient quadrature-axis time constant.
• Tq0”: Subtransient quadrature-axis time constant.
• Xd’: Transient direct-axis reactance.
• Xd”: Subtransient direct-axis reactance.
• Xq’: Transient quadrature-axis reactance.
• Xq”: Subtransient quadrature-axis reactance.
• X0: Zero-sequence reactance.
• R0: Zero-sequence resistance.
• X2: Negative-sequence reactance.
• R2: Negative-sequence resistance.

Table 6. Parameters of synchronous generators (Part 1).

Generator	S	Vn	Fp	Rstr	Xl	Xd	Xq	Rotor Type	H
G1–G3	1.00	0.480	0.800	0.00963	0.03000	2.990	1.93010	Round Rotor	4
G4–G7	1.00	0.480	0.980	0.00000	0.10310	1.056	0.56190	Salient pole	3.99960

Table 6. Parameters of synchronous generators (Part 1).

Generator	S	Vn	Fp	Rstr	Xl	Xd	Xq	Rotor Type	H
G1–G3	1.00	0.480	0.800	0.00963	0.03000	2.990	1.93010	Round Rotor	4
G4–G7	1.00	0.480	0.980	0.00000	0.10310	1.056	0.56190	Salient pole	3.99960

Table 7. Parameters of synchronous generators (Part 2).

Generator	Td0’	Td0”	Tq0’	Tq0”	Xd’	Xd”	Xq’	Xq”
G1–G3	2.24250	0.01385	1.00005	0.35741	0.18000	0.13000	1.93000	0.27000
G4–G7	5.49965	0.04383	0.00000	0.16275	0.32180	0.18730	0.56190	0.28095

Table 7. Parameters of synchronous generators (Part 2).

Generator	Td0’	Td0”	Tq0’	Tq0”	Xd’	Xd”	Xq’	Xq”
G1–G3	2.24250	0.01385	1.00005	0.35741	0.18000	0.13000	1.93000	0.27000
G4–G7	5.49965	0.04383	0.00000	0.16275	0.32180	0.18730	0.56190	0.28095

Table 8. Parameters of synchronous generators (Part 3).

Generator	X0	R0	X2	R2
G1–G3	0.02000	0.00963	0.19000	0.00963
G4–G7	0.65710	0.00000	0.18480	0.00000

Table 8. Parameters of synchronous generators (Part 3).

Generator	X0	R0	X2	R2
G1–G3	0.02000	0.00963	0.19000	0.00963
G4–G7	0.65710	0.00000	0.18480	0.00000

The placement of traditional or renewable DG in MG involves criteria such as resource availability and technical compatibility, as well as regulatory and policy compliance [35,36]. In this research, the DG units were strategically positioned near primary loads and at the end of certain feeders to offer voltage support, as is the case in many real grids. Furthermore, it is assumed that renewable DG has resource availability in the locations where it is installed and that all DG units, regardless of their type, meet the regulatory requirements.

In terms of renewable energy sources, the MG test system has four solar photovoltaic units of 1.0 MVA each and a wind farm of equal power. There is also an energy storage system represented by a set of batteries of 0.15 MVA.

Table 9. Non-conventional renewable energy sources of the MG test system.

Element	Technology	Power	Power Factor
PV1–PV4	Photovoltaic	1.00	0.80
WP1	Wind	1.00	0.80
BESS	Battery	0.15	0.80

describes these renewable energy sources. PV2, PV3, and PV4 photovoltaic plants have the same characteristics as PV1.

The test system also features different demands that represent residential or industrial loads. These demands absorb active and reactive power as shown in

Table 10. Demand data of the proposed MG test system.

Demand	Active Power [MW]	Reactive Power [MVAr]
C1	1.00000	0.10000
C2	0.97014	0.24254
C3	1.00000	0.20000
C4	1.00000	0.10000
C5	1.00000	0.30000
C6	0.09500	0.03123
C7	0.09500	0.03123
C8	0.09500	0.03123
C9	0.05142	0.07024
C10	0.04410	0.00895
C11	2.50000	0.50000
C12	1.00000	0.20000
C13	1.00000	0.20000
C14	1.00000	0.10000
C15	0.80000	0.10000
C16	1.00000	0.05000
C17	1.00000	0.10000
C18	0.47500	0.15612
C19	0.28500	0.09367
C20	0.02250	0.01090
C21	0.10000	0.09367
C22	0.02250	0.01090
C23	0.07600	0.02498
C24	0.02250	0.01090
C25	0.93931	0.19073
C26	0.07600	0.02498
C27	0.02250	0.01090
C28	0.02250	0.01090

.

2.2. Protection Coordination Scheme of the MG Test System

The current transformers (CT) and potential transformers (PT) that supply current and voltage measurements to the protection system were designed with the transformation ratios detailed in

Table 11. Transformation ratios of CTs and PTs.

Relays	CT [A/A]	PT [V/V]
RL1–RL13,RL25,RL26	300/5	34,500/115
RL14–RL22,RL24,RL27	800/5	480/115
RL23	300/5	4160/115

.

The settings of each relay and fuse of the protection system are indicated in this section.

Table 12. Function 51 settings in relays.

Relay	Current Setting [p.u.]	Time Dial	Tripping Dir	Curve
RL1	1.29	0.47	None	IEC-VI
RL2	0.22	0.42	Forward	IEC-VI
RL3	0.15	0.42	Forward	IEC-VI
RL4	0.22	0.42	Forward	IEC-VI
RL5	0.37	0.42	Forward	IEC-VI
RL6	0.77	0.42	Forward	IEC-VI
RL7	0.22	0.70	Forward	IEC-VI
RL8	0.22	0.21	Forward	IEC-VI
RL9	0.30	0.42	Forward	IEC-VI
RL10	0.08	0.42	Forward	IEC-VI
RL11	0.08	0.42	Forward	IEC-VI
RL12	0.08	0.42	Forward	IEC-VI
RL13	0.08	0.42	Forward	IEC-VI
RL14	1.96	0.03	Forward	IEC-VI
RL15	1.96	0.03	Forward	IEC-VI
RL16	1.96	0.03	Forward	IEC-VI
RL17	1.96	0.025	Forward	IEC-VI
RL18	1.96	0.025	Forward	IEC-VI
RL19	1.96	0.025	Forward	IEC-VI
RL20	1.96	0.025	Forward	IEC-VI
RL21	1.96	0.025	Forward	IEC-VI
RL22	1.51	0.025	Forward	IEC-VI
RL23	0.61	0.085	Forward	IEC-VI
RL24	1.51	0.025	Forward	IEC-VI
RL25	0.06	0.025	Forward	IEC-VI
RL26	0.08	0.14	Forward	IEC-VI
RL27	0.30	0.14	Forward	IEC-VI

,

Table 13. Function 51N settings in relays.

Relay	Current Setting [p.u.]	Time Dial	Tripping Dir	Curve
RL1	0.391	0.700	None	IEC-VI
RL2	0.067	0.420	Forward	IEC-VI
RL3	0.045	0.420	Forward	IEC-VI
RL4	0.070	0.420	Forward	IEC-VI
RL5	0.112	0.420	Forward	IEC-VI
RL6	0.235	0.420	Forward	IEC-VI
RL7	0.067	0.420	Forward	IEC-VI
RL8	0.067	0.420	Forward	IEC-VI
RL9	0.09	0.420	Forward	IEC-VI
RL10	0.023	0.140	Forward	IEC-VI
RL11	0.023	0.140	Forward	IEC-VI
RL12	0.023	0.140	Forward	IEC-VI
RL13	0.023	0.140	Forward	IEC-VI
RL14	0.602	0.140	Forward	IEC-VI
RL15	0.602	0.140	Forward	IEC-VI
RL16	0.602	0.140	Forward	IEC-VI
RL17	0.602	0.025	Forward	IEC-VI
RL18	0.602	0.025	Forward	IEC-VI
RL19	0.602	0.025	Forward	IEC-VI
RL20	0.602	0.025	Forward	IEC-VI
RL21	0.602	0.025	Forward	IEC-VI

and

Table 14. Function 50 settings in relays.

Relay	Pickup Current [p.u.]	Time Setting [s]	Tripping Dir
RL2	2.26	0.10	Forward
RL3	1.48	0.10	Forward
RL4	2.19	0.10	Forward
RL7	2.88	0.10	Forward
RL8	2.79	0.10	Forward
RL10	0.99	0.10	Forward
RL11	0.99	0.10	Forward
RL12	0.99	0.10	Forward
RL13	0.99	0.10	Forward

present the initial settings of the relays. In this case, RL1 is connected to the line that links the external network to Bus 1. This is the backup relay for relays RL2–RL13, so it has different settings than the others. All relays were set to function 51 and 51N, while function 50 was set to relays RL2–RL4, RL7, RL8, and RL11–RL13, since instantaneous protection is not usually set on the lines as it makes it difficult to ensure selectivity with other protection devices downstream of the system.

The following criteria were used to set the phase-timing overcurrent protection (51):

• The starting current was set to 130% of the nominal current of the element connected via the cable.
• The very inverse IEC inverse time curve was used.

The criteria for setting the timed neutral overcurrent protection (51N) were:

• The starting current was adjusted to 40% of the load current.

The criteria for configuring instantaneous overcurrent protection (50) were as follows:

• This function is used to protect the transformers.
• The starting current was set at 150% of the short-circuit current sensed on the high side when the fault occurs on the low side.
• If the adjusted starting current is lower than the transformer’s inrush current, it should be recalculated as 150% of the inrush current.
• The very inverse IEC inverse time curve was used.

To obtain the settings shown in Table 12, Table 13 and Table 14, the following types of faults were performed: three-phase and two-phase for the settings of functions 51 and 50, while for function 51N, single-phase to ground and two-phase to ground faults were performed. In order to have the best protection in the worst-case scenarios, once the faults were executed, the setting that worked best for all types of faults was taken.

Table 12 and Table 13 show a column with the relay name, Current Setting, which is the percentage ratio of the pick-up current of the relay and the nominal secondary current of the CT. The Time Dial setting represents each of the possible time settings of the overcurrent relay curve, the Tripping Dir setting is the direction the current must go to be detected by the relay, and the Curve setting is the type of relay curve setting.

In Table 14, the first column refers to the relay name; the second column is the Pickup Current, i.e., from this current value the relay considers that there is a fault. The Time Setting is the time from which the instantaneous protection acts, and the Tripping Dir is the direction that the current flow must have to be detected by the relay.

Fuses are also elements that are part of the protection system. Three fuses were used in the test network to protect specific elements or loads or were used in the absence of an overcurrent relay.

Table 15. Fuse characteristics of the MG test system.

Fuse	Brand	Type	Voltage [V]	Current [A]
F1	HV-Fuse Links Size Standard	HHD	36,000.0	40.0
F2	HV-Fuse Links Size Standard	HHD	17,500.0	160.0
F3	HV-Fuse Links Size Standard	HHD	36,000.0	31.5

shows the characteristics and curves of the fuses used.

Fuses allow for the simultaneous definition of two key aspects of their operation: the minimum melt time-current curve (TCC), which indicates the moment when the fuse begins to melt, and the maximum clear TCC, which represents the moment when the fuse has completed its melting process.

Table 16. F1 fuse curve.

Min. Melt.: I [A]	Min. Melt.: t [s]	Total Clear: I [A]	Total Clear: t [s]
129.9006	9.9865	154.0072	15.1117
137.6387	5.1140	158.7731	9.9865
144.7772	3.0380	168.2164	5.1140
153.3936	1.8937	176.9520	3.0380
166.8041	1.0304	187.4766	1.8937
183.3859	0.5757	203.8726	1.0303
211.3557	0.2749	224.1353	0.5758
268.0909	0.0972	258.3283	0.2748
367.2151	0.0305	327.6598	0.0972
513.0226	0.0103	448.8239	0.0305

,

Table 17. F2 fuse curve.

Min. Melt.: I [A]	Min. Melt.: t [s]	Total Clear: I [A]	Total Clear: t [s]
579.0485	4.9354	660.0041	8.3986
654.7481	2.1969	707.7302	4.9354
741.4207	1.1203	800.2503	2.1969
864.3313	0.5854	906.1835	1.1203
1042.6310	0.2811	1056.3990	0.5853
1396.1350	0.0980	1274.3170	0.2810
2003.1280	0.0306	1706.3840	0.0980
2831.8190	0.0102	2448.2710	0.0306

and

Table 18. F3 fuse curve.

Min. Melt.: I [A]	Min. Melt.: t [s]	Total Clear: I [A]	Total Clear: t [s]
99.2086	22.3410	121.0026	23.1480
105.1097	10.0494	121.2522	22.3410
111.3726	5.1354	128.4662	10.0493
117.3136	3.0700	136.1164	5.1354
123.6731	1.9136	143.3843	3.0700
134.0065	1.0325	151.1546	1.9137
145.3057	0.5784	163.7763	1.0325
162.9285	0.2799	177.5962	0.5783
201.3958	0.0978	199.1389	0.2800
274.7829	0.0308	246.1509	0.0978
391.1513	0.0103	335.8471	0.0307

show a detailed breakdown of these two crucial characteristics. These tables present information on minimum fusing currents and maximum clearing currents, expressed in amperes, as well as the corresponding times in seconds. These data provides a comprehensive view of the properties and behavior of fuses in the electrical system, playing a key role in ensuring effective and reliable protection.

3. Electrical Analyses of the Proposed MG Test System

This section presents an electrical study of the microgrid test system connected to the external network with and without DG. This is performed to have actual results of the load flow and short-circuit currents that demonstrate proper operation of the network with and without contingencies. In this case, 13 contingencies were carried out as shown in

Table 19. Contingencies considered in the MG test system.

Contingency	Element	Contingency	Element
Cont. 1	L1	Cont. 8	T16
Cont. 2	L2	Cont. 9	T17
Cont. 3	L3	Cont. 10	T18
Cont. 4	L5	Cont. 11	T2
Cont. 5	Line to external network	Cont. 12	T3
Cont. 6	T1	Cont. 13	T5
Cont. 7	T15

. A detailed model of the MG was created using the Digsilent Power Factory software, version 2020 SP3. This program allows for the construction of a simulation model by designing each element of the electrical system. Among these elements, measurement devices such as current transformers and voltage transformers can be modeled and included, which simulates the behavior of real-world components, thereby providing voltage and current signals that are supplied to the protection relays.

3.1. Load Flow of the Test System without DG

Table 20. Voltage magnitudes of the microgrid test system without DG.

Bus	Voltage [p.u.]	Bus	Voltage [p.u.]
B1-34.5	0.998	B14-4.16	0.988
B2-0.48	0.988	B15-0.48	0.985
B3-0.48	0.958	B16-0.48	0.985
B4-0.48	0.958	B17-0.48	0.934
B5-34.5	0.998	B18-0.48	0.931
B6-34.5	0.998	B19-0.21	0.996
B7-34.5	0.997	B20-13.2	0.930
B8-0.48	0.917	B21-13.2	0.929
B9-0.48	0.977	B22-0.48	0.990
B10-0.48	0.971	B23-0.48	0.995
B11-0.48	0.976	B24-0.48	0.985
B12-34.5	0.998	B25-0.48	0.985
B13-4.16	0.988	B26-0.48	0.979

presents the voltage magnitudes of the system. Under normal operating conditions and without DG, the lowest voltage is presented at bus B8-0.48 with 0.917 p.u. This is a bus with a load far from the generation and the external network; to meet the demand of this bus (C11), the power flow passes through two transformers and a line that causes the voltage to drop. On the other hand, the highest voltage magnitudes are presented at buses B1-34.5 and B12-34.5, both with 0.998 p.u. These buses are electrically close to the external network.

Table 21. Loadability of assets without GD.

Asset	Loadability [% ]	Asset	Loadability [% ]
L0	94.67	T8	92.65
L1	17.67	T9	51.76
L2	25.05	T10	51.76
L3	21.84	T11	50.78
L4	34.94	T12	54.00
L5	48.61	T13	51.15
T1	14.04	T14	34.22
T2	61.21	T15	29.47
T3	55.91	T16	29.46
T4	50.46	T17	60.72
T5	68.25	T18	59.53
T6	72.99	T19	92.65
T7	68.12

shows the loadability level of the lines and transformers of the microgrid test system. The loadability levels of all the assets are below 100% under normal operating conditions and without DG; that is, there are no overloads, although there are elements above 90% of loadability such as L0, the line connecting the test network with the external network, and transformers T8 and T19 that serve the load C11 and are separated by line L4.

Subsequently, a load flow analysis is carried out under N-1 contingencies for each element of the microgrid. In this case, low voltages and elements with overloads are evidenced. In

Table 22. Most affected buses under contingency.

Contingency	Bus	Voltage [p.u.]
Cont. 6 T1	Bus 2-0.48kV	0.756
Cont. 12 T3	Bus 3-0.48kV	0.888
Cont. 12 T3	Bus 4-0.48kV	0.888
Cont. 12 T3	Bus 8-0.48kV	0.842

, elements with low voltages due to contingencies in specific elements are presented. With the failure of transformer T1, the voltage at bus B2-0.48 drops to 0.756 p.u. Since this bus is electrically away from the source, the line that feeds it is overloaded, as shown in

Table 23. Overloaded elements.

Contingency	Element	Loadability [%]	Contingency	Element	Loadability [%]
Cont. 6-T1	L5	163.81	Cont. 12-T3	T19	100.96
Cont. 9-T17	T18	122.39	Cont. 12-T3	T2	158.02
Cont. 10-T18	T17	122.12	Cont. 12-T3	T8	100.96

, and the voltage drops due to the radiality of the network. The contingency of transformer T3 causes the power flow to go entirely through the parallel transformer T2, and the new radial system causes the voltages to drop at buses B3-0.48, B4-0.48, and also in the downstream buses.

Under contingency conditions, situations are presented where some elements of the test network experience significant overloads. An example of this is transformer L5, which must supply power to loads at bus B2-0.48 when T1 is disconnected, resulting in an overload of 63.81%. Transformers T17 and T18 operate in parallel, which means that a contingency of any of these results in an overload on the other, with overload levels as high as 122% of their rated capacity. Furthermore, in the event of a contingency in T3, T2 is subject to a 58% overload, since it is connected in parallel and must supply all downstream loads. It is also important to note that transformers T8 and T19 experience overloads due to the increase in reactive power required to supply load C11.

A short-circuit analysis is performed by simulating faults at each bus and in some of the system’s lines.

Table 24. Fault currents of the microgrid test system without DG.

Failed Element	Single-Phase Failure [kA]	Three-Phase Failure [kA]	Failed Element	Single-Phase Failure [kA]	Three-Phase Failure [kA]
71%L2	5.012	56.087	B9-0.48	36.397	456.611
79%L3	3.727	47.078	B10-0.48	40.537	453.567
B1-34.5	5.173	57.019	B11-0.48	40.537	453.567
B2-0.48	25.235	289.534	B12-34.5	5.173	57.019
B3-0.48	51.659	447.391	B13-4.16	3.640	36.105
B4-0.48	51.659	447.391	B14-4.16	3.640	36.105
B5-34.5	4.646	53.601	B15-0.48	30.747	312.063
B6-34.5	4.976	56.041	B16-0.48	30.747	312.063
B7-34.5	3.523	45.134	B17-0.48	39.945	424.859
B8-0.48	24.004	181.684	B18-0.48	37.470	399.330

reports the values of both single-phase and three-phase currents obtained in the elements of the microgrid test system. In Table 24, it is evidenced that fault currents are higher in the bus whose voltage levels are lower and also closer to the source, which, in this case, is the external network.

3.2. Load Flow Analysis of the Test System with GD

In this section, the analysis is conducted for scenario 2, in which the DG is operating. In normal operating conditions, the voltage magnitudes remain within regulatory limits. The B8-0.48 bus exhibits improved voltage compared to the case without a distributed generation, as it increases from 0.917 p.u. to 0.937 p.u. In this scenario with a distributed generation, the lowest voltage is found in the B18-0.48 bus with 0.932 p.u., while the highest voltages are maintained in the B1-34.5 and B12-34.5 buses with 0.999 p.u., as indicated in

Table 25. Voltage in buses with DG.

Bus	Voltage [p.u.]	Bus	Voltage [p.u.]
B1-34.5	0.999	B14-4.16	0.992
B2-0.48	0.996	B15-0.48	0.994
B3-0.48	0.977	B16-0.48	0.994
B4-0.48	0.977	B17-0.48	0.935
B5-34.5	0.999	B18-0.48	0.932
B6-34.5	0.999	B19-0.21	0.996
B7-34.5	0.998	B20-13.2	0.949
B8-0.48	0.937	B21-13.2	0.949
B9-0.48	0.980	B22-0.48	0.990
B10-0.48	0.976	B23-0.48	0.996
B11-0.48	0.982	B24-0.48	0.986
B12-34.5	0.999	B25-0.48	0.986
B13-4.16	0.992	B26-0.48	0.987

.

Under normal operating conditions, the loadability on the elements decreased due to the inclusion of DG (see

Table 26. Loadability of elements with DG.

Element	Loadability [%]	Element	Loadability [%]
L0	67.51	T8	90.67
L1	9.59	T9	51.73
L2	18.59	T10	51.73
L3	19.92	T11	50.75
L4	34.20	T12	53.97
L5	59.64	T13	47.49
T1	2.13	T14	30.34
T2	29.37	T15	18.87
T3	26.83	T16	18.86
T4	50.44	T17	33.75
T5	67.23	T18	33.09
T6	72.95	T19	90.67
T7	67.53

). However, there are still elements with a loadability above 90%, such as T8 and T19, for the same reasons mentioned in the scenario without DG. Nevertheless, it is important to note that the loadability of L0 decreased from 94.67% to 67.51% because the system no longer relies entirely on importing power from the external grid.

In the scenario with DG, there are no voltage issues or overloads, indicating that the voltages observed in the scenario without DG were improved due to the proximity of generators to the loads. However, fault currents in the buses increased due to the contribution of short-circuit currents from the DG units (see

Table 27. Fault currents with DG.

Failed Element	Single-Phase Failure [kA]	Three-Phase Failure [kA]	Failed Element	Single-Phase Failure [kA]	Three-Phase Failure [kA]
71%L2	7.725	89.312	B9-0.48	54.952	692.371
79%L3	5.667	73.929	B10-0.48	61.515	690.815
B1-34.5	7.966	90.689	B11-0.48	61.515	690.815
B2-0.48	77.774	711.800	B12-34.5	7.966	90.689
B3-0.48	97.231	1.020.720	B13-4.16	6.564	65.220
B4-0.48	97.231	1.020.720	B14-4.16	6.564	65.220
B5-34.5	7.185	85.646	B15-0.48	46.781	475.750
B6-34.5	7.668	89.244	B16-0.48	46.781	475.750
B7-34.5	5.346	70.663	B17-0.48	58.391	624.170
B8-0.48	34.334	268.966	B18-0.48	54.931	586.970

).

3.3. Protection Coordination Analysis

In this section, the locations of each of the simulated faults are presented to facilitate the effective coordination between main and back-up relays in each case. Additionally, a comparison is conducted between the curves and fault detection times of various relays in operational scenarios, both with and without DG. This evaluation encompasses aspects such as reliability, selectivity, speed, redundancy, and safety of the protection scheme. The following single-line diagram displays the faults that were simulated for coordinating the overcurrent relays and their respective backups.

3.3.1. Protection Coordination without DG

In the scenario without DG, and in the event of faults at the locations shown in Figure 2, the operating times of the main and backup relays were obtained. These times are listed in

Table 28. Operating times (in seconds) of relays without DG.

Fault	Main Relay	Backup Relay	Current [kA]	Fault	Main Relay	Backup Relay	Current [kA]
Fault 1	RL2	RL1	5.702	Fault 7	RL8	RL1	5.702
	0.120	0.462			0.120	0.462
Fault 2	RL3	RL1	5.702	Fault 8	RL9	RL1	5.702
	0.120	0.462			0.298	0.462
Fault 3	RL4	RL1	5.702	Fault 9	RL10	RL9	5.667
	0.120	0.462			0.120	0.298
Fault 4	RL5	RL1	5.702	Fault 10	RL11	RL9	5.672
	0.298	0.462			0.120	0.298
Fault 5	RL6	RL1	5.702	Fault 11	RL12	RL9	5.588
	0.298	0.462			0.120	0.298
Fault 6	RL7	RL1	5.702	Fault 12	RL13	RL9	5.592
	0.120	0.462			0.120	0.298

, with the adjustments shown in Table 12, Table 13 and Table 14.

The curves of the protection relays RL2 vs. RL1, RL4 vs. RL1, and RL9 vs. RL1, for faults 1, 3, and 8, respectively, are shown in Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8, Figure 9, Figure 10, Figure 11, Figure 12, Figure 13 and Figure 14. Figure 3 corresponds to the coordination of RL1 and RL2 for a three-phase fault.

As observed in Figure 3, for a three-phase fault at location 1 in Figure 2, a fault current of 5.7 kA is reached. Relay RL2 operates at 0.12 s, while RL1 operates at 0.47 s, demonstrating the effectiveness of the protection scheme for this fault.

Figure 4 corresponds to a single-phase fault at location 1 in Figure 2.

Similar to the three-phase fault scenario, in the case of a single-phase fault with a current of 5.1 kA, RL2 also initiates its operation before RL1, with response times of 0.12 s and 0.52 s, respectively.

For faults occurring at location 3 in Figure 2, relays RL4 and RL1 must coordinate.

When a three-phase fault occurs at location 3 in Figure 2, a current of 5.7 kA is established. In this case, RL4 operates at 0.12 s, while RL1 operates at 0.46 s. This can be evidenced in Figure 5.

A single-phase fault at location 3 is performed in Figure 6, establishing a current of 5.1 kA. RL4 operates at 0.12 s, while RL1 operates at 0.51 s.

For faults occurring at location 8 in Figure 2, relays RL9 and RL1 must coordinate.

When a three-phase fault occurs at location 8 in Figure 2, a current of 5.7 kA is established. For this case, RL9 operates at 0.298 s, while RL1 operates at 0.462 s; this can be evidenced in Figure 7.

A single-phase fault at location 8 is performed in Figure 8, establishing a current of 5.1 kA. RL9 operates at 0.298 s, while RL1 operates at 0.513 s.

3.3.2. Protection Coordination of the MG Test System with DG

In the scenario with DG, the operation times of the relays for the faults depicted in Figure 2 are presented in

Table 29. Operating times (in seconds) of relays in scenario with DG.

Faults	Main Relay	Backup Relay	Current [kA]	Faults	Main Relay	Backup Relay	Current [kA]
Fault 1	RL2	RL1	8.918	Fault 7	RL8	RL1	8.918
	0.120	0.462			0.120	0.462
Fault 2	RL3	RL1	8.918	Fault 8	RL9	RL1	8.918
	0.120	0.462			0.298	0.462
Fault 3	RL4	RL1	8.918	Fault 9	RL10	RL9	8.860
	0.120	0.462			0.120	0.298
Fault 4	RL5	RL1	8.918	Fault 10	RL11	RL9	8.869
	0.298	0.462			0.120	0.298
Fault 5	RL6	RL1	8.918	Fault 11	RL12	RL9	8.727
	0.298	0.462			0.120	0.298
Fault 6	RL7	RL1	8.918	Fault 12	RL13	RL9	8.734
	0.120	0.462			0.120	0.298

. Despite the increasing magnitude of fault currents at these points, the settings detailed in Table 12, Table 13 and Table 14 remained effective in ensuring protection for this scenario.

In this section, faults are simulated in the same previously mentioned locations (1, 3, and 8), but also takes into account the contribution of DG.

Figure 9 corresponds to the coordination of RL1 and RL2 for a three-phase fault.

As observed in Figure 9, for a three-phase fault at location 1 in Figure 2, a fault current of 8.8 kA is achieved (greater than the current of the scenario without DG). Relay RL2 operates at 0.12 s, while RL1 operates at 0.47 s (these times are the same as the ones for the scenario without DG), demonstrating the effectiveness of the protection scheme for this fault.

Figure 10 corresponds to a single-phase fault at location 1 in Figure 2 incorporating DG. In this case, a fault current of 7.9 kA is established, which is greater than the fault current of the scenario without DG. RL2 initiates its operation before RL1, with response times of 0.12 s and 0.55 s, respectively (these times are similar to those obtained without DG).

For faults occurring at location 3 in Figure 2, relays RL4 and RL1 must coordinate.

When a three-phase fault takes place at location 3 in Figure 2, a current of 8.7 kA is established, which is greater than the fault current of the scenario without DG. For this case, RL4 operates at 0.12 s, while RL1 operates at 0.46 s (these times are the same as the ones without DG). This can be evidenced in Figure 11.

A single-phase fault at location 3 is performed in Figure 12. In this case, a current of 7.8 kA is established, which is greater than the fault current of the scenario without DG. RL4 operates at 0.12 s, while RL1 operates at 0.55 s; note that when DG is incorporated, the protection coordination is lost.

For faults occurring at location 8 in Figure 2, relays RL9 and RL1 must coordinate.

When a three-phase fault occurs at location 8 in Figure 2, a current of 8.7 kA is established, which is greater than the fault current of the scenario without DG. For this case, RL9 operates at 0.3 s, while RL1 operates at 0.46 s (these times are similar to the ones obtained in the scenario without DG); this can be evidenced in Figure 13.

Finally, a single-phase fault at location 8 is performed in Figure 14 and a current of 7.7 kA is established, which is greater that the fault current without DG. RL9 operates at 0.3 s, while RL1 operates at 0.54 s (similar times to those without DG).

The last two sections compare the cases of the test microgrid connected to the external grid without a distributed generation (DG) and with DG. In the second case, higher currents are evident due to the contributions of each DG. However, the incorporation of DG does not significantly affect the protection coordination system, as the coordination times remain similar for both scenarios. Therefore, it is not necessary to readjust the protections set when DG is incorporated.

3.4. Analysis of Current Paths

Figure 15 depicts the fault current paths of the three representative cases illustrated in the relay operation charts presented in Section 3.3.

The red line represents the fault current at the high voltage terminals of transformer T1 (red point 1). This current is primarily supplied from the external grid (red point 2). The coordination of relays RL2 and RL1 is carried out. Figure 3 and Figure 4 illustrate the relay coordination in an operation mode connected to the external grid without DG. Figure 9 and Figure 10 show the relay coordination when the test network is connected to the external grid and also has active DG.

The violet line represents the fault current at the high terminals of transformer T3 (violet point 1). In this case, the fault current is primarily supplied from the external grid (violet point 2). The coordination of relays RL4 and RL1 is carried out. Figure 5 and Figure 6 illustrate the relay coordination in an operation mode connected to the external grid without DG. Figure 11 and Figure 12 show the relay coordination when the MG test system is connected to the external grid and the DG units are operational.

The orange line represents the fault current at bus B12-34.5 (orange point 1). The fault current is mainly supplied from the external grid (orange point 2). The coordination of relays RL9 and RL1 is carried out. Figure 7 and Figure 8 illustrate the relay coordination in the operation mode connected to the external grid without DG. Figure 13 and Figure 14 show the relay coordination when the test network is connected to the external grid and also has active DG.

4. Conclusions

This paper introduced a comprehensive 26-bus microgrid test system for validating and proposing advanced protection coordination schemes. Detailed specifications regarding the topology and parameters were provided. The MG test system facilitated the simulation of diverse configurations and operating modes, utilizing a predefined overcurrent protection scheme validated for scenarios with and without DG. Both scenarios incorporated single- and three-phase faults for comparative purposes. The study demonstrated the significant impact of DG on improving voltage levels at distant buses and reducing the loadability on critical buses during N-1 contingencies. Notably, the inclusion of DG in this MG test system increased short-circuit levels, with no significant effect on the operating times of the main and backup relays. Overall, the results indicate that the inclusion of DG does not disrupt the protection scheme in this test network and no readjustment of the relay parameters is required.