Overcurrent Relay Advances for Modern Electricity Networks

By Arturo Conde Enriquez

Overcurrent Relay Advances for Modern Electricity Networks explores how to optimize protection and improve system stability and resilience by implementing advanced overcurrent relays in highly dynamic renewable heavy power systems. This guide provides a foundation in relay functions and behaviors in current modern networks, particularly regarding renewable power sources and new electrical network structures such as microgrids. The work discusses the design and creation of protection schemes in smart grids and analyzes their impact on performance and security in protection systems. This practical book also presents a critical new coordination method for online applications.

• Reviews performance considerations and application challenges in optimizing overcurrent relays in future networks
• Provides mathematical and computational modeling scenarios for relays geared for application in future commercial equipment designs
• Describes how to adopt online protection systems by means of optimization algorithms for the adjustment and coordination of relays
• Includes pseudocodes of routines designed to support readers who are implementing or analyzing these systems
• Outlines a demonstrative virtual relay to execute programming operation and optimize coordination of relays

• Language: English
• Publisher: Academic Press
• Release date: Dec 5, 2022
• ISBN: 9780323972543

Arturo Conde Enriquez

Arturo Conde is Professor of Electrical Engineering at Autonomous University of Nuevo Leon, Mexico. He is interested in the protection and power quality of power systems, particularly as relates to relay design, applications of adaptive protection in transmission and distribution systems, and on-line applications through distributed management systems of smart grids. He is the author or coauthor of 33 indexed articles and has authored over 50 papers in national and international congress. He is the author of one book, two patents and three book chapters.

Book preview

Chapter 1: Mathematical model and application of overcurrent relays

Abstract

Historically, overcurrent relays (OCRs) were one of the first forms of protection for power electrical systems, and today they are the most widely used protection devices in electrical networks. These relays are applied by means of an arrangement of devices that are capable of acting in the event of faults or disturbances in any section of the electrical system, and of acting automatically after the disconnection of devices and/or the emission of alarm signals. Phase OCR are assigned to each of the phases for three-phase and phase-to-phase faults, whereas for the protection of ground faults, a relay connected to neutral is used. A relay protection system may consist of one or more relays assigned to operate both instantaneously and with time delays.

Although overcurrent protections are the simplest protections, they are the most difficult to apply in a power system, as they only use current as a signal to discriminate between a permitted operating state and a fault. Compromises must be made to discriminate between a minimum short-circuit condition and a maximum load condition, since in long or high-impedance lines these current values may be comparable. Thus the protection must be set above the maximum load current to be insensitive to minimum faults, since they are not eliminated, and may involve other phases or elements of the system. This setting criterion desensitizes and reduces the reach of protection against load conditions below the maximum load, which represent the most frequent operating state. The basic requirements of overcurrent protection are (a) the detection of any fault present in the protected line or the adjacent one, (b) selective operation, and (c) secure operation under normal load conditions and faults outside the protected area.

In this chapter, the background information and the fundamental mode of operation of overcurrent relays are presented as a basis for the subsequent chapters. The parameters that determine the operation of the relay and the procedure for the coordination of relays are specified. The application of directional overcurrent relays (DOCRs) in meshed networks has certain aspects that must be verified, such as curve crossing and relay sensitivity; these considerations are explained in detail. The three functional stages of operation of an OCR are digital processing of the current signal to obtain the fundamental frequency phasor, the formation of time curves to establish an operating time function to be inversely proportional to the magnitude of the current, and the integration of the input current of the electrical system to determine its actuation time. For this reason, the structure of the relay is described by means of functions that represent the operating state, in both steady and dynamic states. Likewise, the application criteria and their coordination with other protection devices such as fuses, reclosers, and sectionalizers are reviewed.

This chapter is presented as follows: Section 1.1 presents a statement of the problem involving OCR protection. In Section 1.2, the methodology of OCR coordination for radial lines is described. Considerations relating to the application of DOCRs and their coordination in meshed networks are introduced in Section 1.3. Section 1.4 discusses general considerations related to the lack of relay sensitivity. The structure of digital overcurrent relays through functions is evaluated in Section 1.5. In Section 1.6, the coordination between different overcurrent devices is presented.

Keywords

Overcurrent relays; Relay coordination; Time curves; Dynamic performance

1.1: Overcurrent relays: Overview

Overcurrent relays (OCRs) have operating times that have made them convenient for the protection of electrical networks for over a century. The operation times of these relays are determined by the operating conditions, such as temporary overload, damage curves, and the presence of other protection devices, rather than functional limitations. OCR can be applied at practically all voltage levels; they are used in distribution circuits as primary protection; in industrial systems, for the protection of equipment such as generators, transformers, and motors; and as backup protection for transmission lines in the form of ground protection or as backup for phase protection. Under operating conditions with fault current supply from both ends, directional operation of the relay is required, and a DOCR therefore has a wider range of applications. Recently, its applications have been evaluated owing to the increase in distributed energy resources (DERs) in the network, which cause bidirectional flow in networks with a radial nature and a reduction in the magnitude of fault currents, causing a lack of sensitivity.

Overcurrent protection is used as primary protection for the own line and as backup protection for faults on adjacent lines, and both functions are performed based on the same time curve. DOCRs use short-circuit current as an indicator of a fault, but the variation in the Thevenin impedance and the fault type causes different current magnitudes for the same fault location. The relay has great uncertainty concerning the fault location and a highly dynamic reach; in view of this, the coordination of OCRs is carried out based on the operation time of a neighboring relay. Thus, for a short-circuit condition, both relays will have the same operating dynamics, and the coordination between the primary (R1) and backup (R2) relays is guaranteed by the coordination time interval (CTI), which is determined based on the maximum fault current of both relays (F1 in Fig. 1.1).

Fig. 1.1

Fig. 1.1 Fault location for overcurrent coordination.

1.2: Time curve

The inverse type of time curves is widely used for line protection, as the operating times are generally adequate for this application. The time curve of the OCR represents an operating time in inverse proportion to the current signal flowing in the relay and is represented on a bi-logarithmic scale to allow for better visualization for adjustment and coordination. Existing standards define the analytical expressions for the time curves of inverse time relays, and these time curves differ because the operating time decreases as the current increases. Fig. 1.2 shows these standardized curves [1]. In electromechanical relays, different types of curves are obtained by means of a magnetic flux saturation process that affects the induction disk. The saturation is controlled and different degrees of inversion of the curve are obtained by means of irregular areas on the disk. In digital relays, the coefficients of the analytical expression for the time curve are defined to obtain the desired family of curves.

Fig. 1.2

Fig. 1.2 Standardized operating time curves.

The time curve of an OCR is defined by international standards as follows [1,2]:

si1_e

  

(1.1)

si2_e    (1.2)

where TDS is the time dial setting used to set the time operation, Isc is the maximum short-circuit current measured by the coordinated relays, and Ipickup is the setting of the relay and the vertical asymptote of the time curve. These international standards are used to define the type of time curve based on their parameters, where A, B, and p are defined by IEEE Std (1.1) and IEC Std (1.2) defines A and p (see Table 1.1). Specifically, B is the horizontal asymptote of time curve, whereas A and p define the shape of the time curve.

Table 1.1

The relay operates when the pickup current (Ipickup) is greater than the maximum load current (Iload), where a plug setting multiplier (PSM) is used to avoid false operation, and a minimum Isc is applied to ensure adequate fault detection.

si3_e    (1.3)

PSM is a factor that prevents operation under normal conditions caused by temporary overloads, power transfer, and errors in current transducers. Typical values are 1.5–2.0 [3], although the specific value will be determined by the application. The variables used for relay coordination are TDS and PSM, and these are defined to set each pair of relays. For different coordination conditions, the time curves must be shifted to achieve the CTI, and this is why the asymptotes of the time curve and their dependence on the values of the time characteristic must be identified.

The asymptotes of the inverse time curve of the OCR are obtained from the settings for Ipickup and TDS. The design criteria for the time curves are established based on the electromechanical properties of the relays, where the torque exerted on the relay disk depends on the magnetic flux that affects it, and this is directly proportional to the ampere-turns. If the number of turns is increased, the current necessary to operate the relay decreases, and vice versa; this allows us to modify the value of Ipickup by varying the tap of the number of turns, and the digital relay reproduces this behavior through the modification of Ipickup. Fig. 1.3 shows the effect of the tap change, where the operating characteristics move from left to right with a change in Ipickup.

Fig. 1.3

Fig. 1.3 Inverse time curve for change in I pickup .

The inverse time curve has a disadvantage when it is necessary to use large magnitudes for the settings of Ipickup; this scenario occurs in electrical networks with high current demand and is very common in actual electrical networks. A high value for Ipickup results in a high value for the operating time, and the relay time increases for a fault current lower than the maximum current, as shown in Fig. 1.3. This increase in operating time is appreciable for any current Isc_x between similar time curves, and the time difference is due only to the difference in Ipickup.

For a particular type of time curve, the TDS setting defines different times and is therefore used to determine the operating time during relay coordination. In an analogous way to an electromechanical relay, the operating time depends on the start position of the disk, if the disk starts from further away, it takes longer to reach the contact position for the same current. This effect is used to change the operating time for the relay based on the TDS, thereby generating a family of curves. By means of this parameter, the operating time of the relay can be controlled, as shown in Fig. 1.4.

Fig. 1.4

Fig. 1.4 Family of curves created by changing the TDS setting.

1.3: OCR coordination

OCR coordination aims to establish a time sequence for the operation of the primary relay and its backup relays. If the primary relay fails to operate, the current contribution to the fault must be eliminated by the operation of the backup relays. To verify the backup function of these lines, fault detection during minimum generation must be guaranteed by considering the analysis of (n − 1) contingencies. The coordination criterion is based on compliance with Eq. (1.4) for each of the m pairs of relays. The breaker fault release time, uncertainties in the calculation of the fault current, and the errors in the instrument current transformers (CTs) are considered when determining the operating time of the backup relay (tb) based on the time of the primary relay (tp). The CTI must consider the total fault release time of the primary relay to avoid simultaneous tripping, and a time interval of between 0.2 and 0.5 s is usually applied [3].

si4_e    (1.4)

The backup relays must detect faults in the remote line-end, and to achieve this, minimum phase-to-phase faults (F2) are used to check the sensitivity of relay R2 for the adjacent lines, as illustrated in Fig. 1.5. Minimal generation and arc-phase resistances (usually 5 Ω) are typically used to obtain minimal fault currents. The backup time depends on the primary time, owing to the uncertainty in the relay reach; this causes the coordination of n relays to have a functional dependence, since the time setting of one relay depends on the time setting of another relay.

Fig. 1.5

Fig. 1.5 Fault location for overcurrent coordination.

In the early days of relay production, each manufacturer obtained a time curve as a result of the design of their relay, and protection engineers used acetates showing these time curves to achieve coordination. After the digital revolution, various organizations proposed standards [1,2] in which specific time curves were defined (and coordination software was developed). Traditionally, coordination is done via selection of a time curve from a catalog, meaning that the time curve determines the success of coordination. However, these standardized time curves were not found to solve all coordination problems, and due to the increase in coordination commitments in modern power systems with renewable sources (RESs), many manufacturers have defined their own curves. The process of relay design has therefore returned to the beginning.

Another point of view is that the coordination problem defines the time curve for enhancing the coordination via nonconventional curves (NCCs). These nonstandardized time curves can offer more solutions, since they are more flexible and can be adapted to different coordination problems. However, the curve selection process becomes more complicated, as it involves a greater number of time curves, and this increase in complexity is solved by optimization algorithms, which can obtain the settings for relays that comply with the restrictions on coordination.

Coordination of radial networks. The settings of the relays needed for coordination are based on Ipickup and TDS, since the type of curve and the constants (A, p, and in some cases B) must be previously defined. To ensure the protection of distribution and transmission networks, it is general practice that the same type of curve is selected; typically, the very inverse type is used due to its having tolerance times to adequate overload conditions for the operation of the electrical network. However, there is greater difficulty in selecting curves for industrial networks due to the presence of different protection devices such as fuses and low-voltage switches, and the need for protection of cables and transformers, and hence different types of curves are commonly selected. The value of Ipickup does not depend on the coordination process, and this is previously determined as shown in Eq. (1.3), using the maximum value of Iload from a contingency analysis.

The process of relay coordination is done with pairs of relays, where the relay on the downstream or load side is known as the primary relay, and the relay on the upstream or source side is the backup relay. The TDS settings for primary relays are known, and the objective is therefore to compute the TDS for the backup relays.

General procedure for coordinating relays. The processes of adjustment and relay coordination must be following the same steps for each pair of relays: after the first coordination is complete, the backup relay is considered the primary relay and the next upstream relay becomes the backup relay. The steps are as follows:

(1)Calculate Ipickup for all relays (Eq. 1.3).

(2)Determine the primary relay time, as TDS is known.

(3)Determine the backup relay time (Eq. 1.4).

(4)Based on Tbackup and Isc, determine the TDS for the backup relay.

Example.

DOCR coordination is performed for the 34.5-kV radial system shown in Fig. 1.6. In this problem, the number of pairs for coordination is m = 3.

Fig. 1.6

Fig. 1.6 Radial network.

Solution

:

Coordination example, input data: Thevenin impedances

si5_e

Short-circuit currents,

si6_e

Fault currents

Pickup currents:

si7_e

Relay coordination process (Fig. 1.7):

si8_esi9_esi10_esi11_esi12_esi13_esi14_eFig. 1.7

Fig. 1.7 Coordination chart.

For the same fault location, different topological configurations and network operation conditions can modify the fault current contributions. Fig. 1.8 shows the time curve for an OCR for a range of currents and operation times. Both relays (relays R3 and R4 in Fig. 1.16) see the same fault current, so current variations will result in higher CTIs; however, the proper sequence of operation is obtained only in the case where the primary relay does not operate.

Fig. 1.8

Fig. 1.8 Tridimensional chart of relay pair.

Fig. 1.16

Fig. 1.16 Coordination pairs.

The process of calculating the settings and coordination relays is carried out based on the maximum border values for both Isc and Iload. In this way, we ensure that under any other operating condition, coordination criterion (3) is not violated.

Sensitivity. To provide a backup function, minimal faults at the adjacent remote line-end must be detected. To overcome the sensitivity filter, each relay in the network is analyzed as follows:

si15_e    (1.5)

where Isc2ϕ is the minimum phase-to-phase fault current at the remote line-end with a fault resistance of 5 Ω. The backup relay (R2) must operate for fault F2 in Fig. 1.5.

A lower limit of 1.5 is considered in Eq. (1.5) since the operation times are large and are therefore useless for protection purposes for multiples of current (M) between 1 and 1.5. This range can be considered a dead zone for protection purposes (Fig. 1.9). The curve of M versus T is generally provided for coordination, and the protection engineer needs to verify that the active zone of the relay is used for the adjustment.

Fig. 1.9

Fig. 1.9 Multiples of current of time curve.

Security. The range of M was established as 1.5–20. A value of 20 was used to avoid a current of more than 100 A in the secondary and hence damage to the relay (Eq. 1.6). When a current of 5 A is used in the secondary CT, safety can be ensured when:

si16_e    (1.6)

1.4: DOCR coordination

DOCRs are designed to detect faulty contribution in the right current contribution and find applications in electrical networks with fault supply on both sides of the line. This contribution can be through other lines with a source connection or through the connection of distributed generation (DG) sources. The directionality is calculated based on the use of potential transformers (PTs) for the polarization signal and the current signal from the CT as the operation signal; the phase difference between these signals defines whether the fault is in front of or behind the relay [4]. When a neutral connection is available, the polarization can be by 3I0. For operation of the relay, the fault current must be detected based on Ipickup and the fault location must be in front of the relay and activated by the directional function, as illustrated in Fig. 1.10.

Fig. 1.10

Fig. 1.10 Logic scheme for tripping of a DOCR.

The polarization signal must not change phase during the fault. In contrast, the operation signal must shift by around 180 degrees for faults in front of and behind the relay, and this phase change is exploited by the directional function. The most common directional function of a DOCR is illustrated in Fig. 1.11. For relay protection with phase a, the polarizing quantity is Vbc. The terms maximum torque line and zero torque line are inherited from the electromechanical era and are also known as operating lines or thresholds in digital relays.

Fig. 1.11

Fig. 1.11 Ninety-degree type connection for DOCR.

DOCRs mainly find applications in mesh networks as both backup protection for phase and earth in transmission networks and primary protection for active subtransmission and distribution networks.

Meshed networks. In meshed networks, additional contributions from adjacent lines are frequent, and this infeed current effect causes the primary relay to see more fault current than the backup relay. Coordination is not affected because only the primary operation time is reduced, and although the CTI is greater, the coordination conditions are fulfilled. In addition, it is likely that the Ipickup for a primary relay (R1) is greater than for a backup relay (R2), and the crossing of these curves can make coordination difficult (Fig. 1.12). The use of different types of time curves, different short-circuit magnitudes, and different relays (Ipickup) can cause crossing of the curves. Relay R2 is sensitive to a detected fault F2 (Fig. 1.5), as shown in Fig. 1.12; however, if F2 is close to the intersection of the curves, coordination may be lost. It is therefore important to verify that the coordination of faults is fulfilled at both ends of the line for primary relays.

Fig. 1.12

Fig. 1.12 Fault location for overcurrent coordination.

The complexity of the coordination problem increases exponentially with the size of the power system. For example, the 14-bus IEEE test system has 40 relays, resulting in 50 coordination pairs, while the 30-bus IEEE test system has 82 relays with 120 coordination pairs. In case of updates to a relay due to a change in the topology or operation of the electrical network, recoordination of all the relays is necessary even if there is no change in the load or fault current of any relay. In the coordination solution provided by optimization methods, this complexity forces the optimization algorithm to perform a more exhaustive search to comply with the restrictions. In the case of nonpermissible operating times for some relays, due to the limitations of the protection principle (maximum demand, low short-circuit current) or a lack of sensitivity, this result may affect other relays, giving long times owing to the coordination requirement.

The coordination criterion establishes the operational sequence of the relays. When a fault occurs on the protected line, the primary relay must operate immediately; however, if it does not operate, the backup relay must be given enough time to provide backup to the adjacent line. The functional dependency between relays means that a modification of the setting for one relay has a domino effect, affecting the coordination of the entire network (Eq.