GB2375902A

GB2375902A - A hybrid fault current limiting and interrupting device

Info

Publication number: GB2375902A
Application number: GB0203685A
Authority: GB
Inventors: Ghanim Alesha Putrus; Mohamed Mostafa Ramadan Ahmed; Li Ran; Lejun Xiao; Roger Penlington; Harry Stanley May
Original assignee: NORTHERN ELECTRIC DISTRIB Ltd; Northumbria University
Current assignee: NORTHERN ELECTRIC DISTRIB Ltd; Northumbria University
Priority date: 2001-02-15
Filing date: 2002-02-15
Publication date: 2002-11-27
Anticipated expiration: 2022-02-15
Also published as: GB0103748D0; GB2375902B; GB0203685D0

Abstract

A hybrid fault current limiting and interrupting device for protecting an electric power line comprises a power circuit including a solid-state switching component (1), a voltage clamping element (2) connected in parallel with the switching component (1), a circuit breaker (4) connected in parallel with the switching component (1) and with the voltage clamping element (2), an isolating switch (3) to the supply side of the switching component (1), and a current sensor (5) the device further comprising a control circuit including a fault current detector and estimator (6) and a switching pattern controller (7), the arrangement being such that, on detection of a fault current by the detector (6), the circuit breaker (4) is opened and the fault current is transferred to the switching component (1) which is turned on during the final stage of current interruption by the circuit breaker (4), and, when the fault current reaches a predetermined maximum value, the switching component (1) is turned off and the current is diverted to the voltage clamping element (2) which decreases the current to a predetermined minimum value at which the switching component (1) is turned on again, the switching of the switching component (1) being repeated for a predetermined time period until the current is fully interrupted (fault cleared) by either operation of the device itself or by operation of downstream protection devices of the power line.

Description

A HYBRID FAULT CURRENT LIMITING AND INTERRUPTING DEVICE This invention relates to a hybrid fault current limiting and interrupting device hereinafter referred to as HFCLID.

The fault current level of a power system represents the maximum current that system causes to flow through a network when a fault occurs. Fault current limiting devices (FCLD), based upon different operating principles, have been employed to prevent the short-circuit current rating of a substation and associated plant from being exceeded.

Conventional devices that are presently in use may be categorised into two types: single shot devices and multi-operation devices. The former, such as current limiting fuses, have the disadvantage that, following operation, supply is lost until either a second device is switched into the circuit or the fuse unit is replaced. The second, which utilise tuned circuits that cause an increase of impedance at the sensing of fault currents, can be costly, bulky, and some have a relatively slow response.

Modem power networks exhibit continuous growth and integration. Consequently safety, reliability and quality of supply are becoming of utmost importance. These developments draw attention to the need for a new generation of fault current limiting devices that have fast response, multiple and flexible operation, long life span, and are cost effective. Advances in solid-state and super-conductor technologies show promise in meeting these requirements.

Superconducting fault current limiters are at the present time restricted by the technical and economical limitations of producing high-Tc superconducting materials.

Fault current limiters employing solid-state switching devices offer a good alternative, particularly at low voltage levels. Possible candidates for the switching devices are the Insulated-Gate Bipolar Transistor (IGBT), the Gate Turn-Off thyristor (GTO) and the Integrated Gate-Commutated Thyristor (IGCT).

High temperature superconducting fault current limiters have been developed and successfully implemented in a few practical applications. These limiters are, however, relatively costly, bulky and need a considerable re-set time subsequent to the clearing of a

fault. Also, they require an active refrigeration system to remove the heat generated by eddy current losses and to compensate for the heat leakage into the cryogenic container.

A fast Fault Current Interrupting Device (FCID) has been proposed which consists of a semiconductor a. c. switch (two thyristors connected in inverse parallel) and a varistor, as shown in Fig. 1. The device is connected in series with each phase of the power line to be protected. During normal operation, the semiconductor switch is continuously on. When a fault occurs, the switch is turned off and the fault current is completely interrupted within a few milli-seconds. The switch is capable of turning off the fault current quickly before it reaches the first peak of the prospective fault current.

The FCID suffers from the main disadvantage that, being very fast, its operation cannot be co-ordinated with other protection devices. Therefore, its implementation in existing power distribution networks is very limited.

The structure of a solid-state Fault Current Limiting Device (FCLD) is similar to that of the FCID except that there is a parallel current limiting impedance across the switch, as shown in Fig. 2. When a fault occurs, the normally conducting switch is turned off and the fault current is diverted, within few micro-seconds, to the current limiting impedance which limits the fault current. The current limiting impedance can be an inductor or a high power resistor. The snubber circuit and the varistor protect the semiconductor switch from high dv/dt and high voltage, respectively. A combination of a solid-state switch and either a series or a shunt circuit-breaker, operating as a FCLD or a FCID, respectively have also been proposed.

The solid-state limiting or interrupting devices mentioned above do not utilise the full capabilities of the solid-state switch employed. That is, on the detection of a shortcircuit, the solid-state switch (which is normally conducting) is turned off and the current is instantly interrupted (or limited, and hence determined, mainly by the parallel highpower impedance). For this reason, there is no flexibility in the control of the resultant fault current. Moreover, a circuit breaker (not an isolator) is required in series with the FCLD to completely interrupt the fault current.

It would be desirable to be able to provide a hybrid device which combines the functions of both a fault current limiting device and a fault current interrupting device in a

manner which enabled flexibility of control and improved performance in an economic product.

According to the present invention there is provided a hybrid fault current limiting and interrupting device for protecting an electric power line, the device comprising a power circuit including a solid-state switching component, a voltage clamping element connected in parallel with the switching component, a circuit breaker connected in parallel with the switching component and with the voltage clamping element, an isolating switch to the supply side of the switching component, and a current sensor the device further comprising a control circuit including a fault current detector and estimator and a switching pattern controller, the arrangement being such that, on detection of a fault current by the detector, the circuit breaker is opened and the fault current is transferred to the switching component which is turned on during the final stage of current interruption by the circuit breaker, and, when the fault current reaches a predetermined maximum value, the switching component is turned off and the current is diverted to the voltage clamping element which decreases the current to a predetermined minimum value at which the switching component is turned on again, the switching of the switching component being repeated for a predetermined time period until the current is fully interrupted either by operation of the device itself or by operation of downstream protection devices of the power line.

It will be appreciated that such a device employs a solid-state switching component, preferably an Insulated-Gate Bipolar Transistor (IGBT), a voltage clamping element and a circuit breaker, preferably a vacuum circuit breaker, as the main power components. The device is primarily designed to be used in low voltage (400 to llkV) electrical power distribution networks in order to limit and/or interrupt the over-current resulting from a fault or any disturbance condition. However the structure of the device facilitates its design and use at higher voltage levels. The device of the invention offers many advantages that will assist in achieving the desired features of modem power distribution networks with regard to improved safety, reliability and quality of supply.

With such a device, the solid-state switch is controlled such that it contributes to the current limiting action as well as being able to fully interrupt the fault current after a desired delay-time following fault inception. Therefore, it will be appreciated that this

device makes full use of the available solid-state switching component and voltage clamping element, preferably a varistor, to limit the fault current, instead of using an additional fixed impedance as heretofore.

The vacuum circuit breaker connected in parallel with the switching component is normally closed in order to reduce power losses during normal operation. On detection of the fault current by the detector, the circuit-breaker is opened and current interruption starts. The fault current is transferred to the switching component which is turned on during the final stage of current interruption by the circuit breaker. The fault current now flows through the switching component and continues to increase. When the fault current reaches a predetermined maximum value (Imax), the switching component is turned off and the current is diverted to the varistor.

The energy trapped in the system inductance starts to dissipate as heat in the varistor and, hence, the voltage across it starts to decrease. Due to the non-linear characteristics of the element, the voltage across it remains almost constant (the clamping voltage). As long as the clamping voltage is greater than the instantaneous value of the supply voltage, the current through the varistor decays rapidly. When the current falls to a predetermined minimum value (Imm) ? the switching component is turned on again, and the current starts to increase with a rate determined by system impedance and circuit initial conditions. As the current reaches the pre-set value (Imax), the switching component is turned off again. This process is repeated for a pre-set period of time, during which the control strategy is the same for both the positive and negative half cycles of the fault current. If the fault persists for longer than the pre-set time (an indication of a permanent fault), the switching component is kept off. After this, the circuit current is normally very small and can be completely interrupted by the isolating switch.

If the short-circuit is very near to the position of the device and the short-circuit current exceeds the rating of the solid-state switching component, the vacuum circuitbreaker completely interrupts the current while the switching component is not operated.

An additional fuse may be added in series with the isolating switch (at the supply side) to interrupt the fault current if it exceeds the vacuum circuit breaker interrupting capacity.

Overheating of the switching component and the varistor may be prevented by the use of over-temperature protection units which reset the controller before the temperature

limits are exceeded. The temperature of the IGBT and that of the varistor are indirectly monitored using the voltage across and the current through each one. This technique could be upgraded to enable the final temperature to be estimated before it is actually reached.

Therefore, the device can be operated for a pre-determined period or until a safe maximum temperature is reached, whichever occurs first. Note that the period of operation of the device is relatively short (less than I second), as the fault will eventually be cleared by either operation of downstream protection devices of the power line or by operation of the device itself.

By way of example only, the device of the invention will now be described in greater detail with reference to the accompanying drawings of which: Fig. 1 illustrates a conventional fault current interrupting device (FCID); Fig. 2 illustrates a conventional fault current limiting device (FCLD); Fig. 3 shows a block diagram of the device of the invention; Fig. 4 shows the line current waveform before, during and after fault occurrence; Fig. 5 shows the current and voltage waveforms during the commutation period; Fig. 6 gives the circuit diagram of the bi-directional solid-state switching component and associated auxiliary components of the device of the invention; Fig. 7 presents a schematic diagram of the over-temperature protection unit used to protect the solid-state switching component of Fig. 3 from over heating; Fig. 8 presents a schematic diagram of the over-temperature protection unit to protect the voltage-clamping element of Fig. 3 from over heating; Fig. 9 shows a circuit diagram showing the connection of the varistors incorporated in the embodiment of the voltage-clamping element shown in Fig. 3; Fig. 10 shows a block diagram of the fault current detector and fault current level estimator shown in Fig. 3; Fig. 11 shows a block diagram of the switching pattern controller shown in Fig. 3; Figs. 12 to 14 show typical results during the operation of the device of the invention, and

Fig. 15 shows a single-line diagram of a typical distribution network with various possible locations of the device, of the invention.

The specifications of a single-phase device according to the invention are: 230/400 V, able to limit the fault current to a peak value of 120 A when the prospective short-circuit symmetrical current is up to 1-2 kA (r. m. s. ) ; pre-fault rated current downstream of the HFCLID is 50 A (r. m. s. ).

As shown in Fig. 3, the solid-state switching component I is connected in parallel with a vacuum circuit breaker 4 and a varistor 2. An isolator 3 is used to completely turnoff the circuit (stop the leakage current) following the permanent turn-off of the switching component. A current sensor 5 is used to measure the network current and provide an input signal to the fault current detector and estimator 6. The latter is used to detect the fault conditions and provide information to the vacuum circuit-breaker and the switching pattern controller 7. The pattern controller determines the switching pattern employed and sets the maximum fault current Imax at which switching component 1 should turn off and the minimum value Im, at which it is turned on after each current interruption. Overtemperature protection units 8 and 9 are incorporated to protect the switching component I and the varistor 2 from overheating.

The circuit of the switching component I consists mainly of two IGBT modules 10 and 11 connected as shown in Fig. 6. Each IGBT conducts the fault current in one direction, i. e. half a cycle. Each of the diodes 12 and 13 provides a path for the current during the conduction of the relevant IGBT 11 and 10, respectively. A fast operating fuse 14 is connected between the IGBT modules to protect the healthy module if the other fails short-circuit. A snubber circuit consisting of a capacitor 18, a resistor 16 and a diode 17 (or 21,19 and 20, respectively) is used in order to reduce the rate of rise of the voltage (dv/dt) during turn-off of the corresponding IGBT. A small inductor 24 may be included to reduce the rate of rise of the current (di/dt) when either IGBT is turned on. In order to increase the power capability of the HFCLID, multiple IGBTs connected in series and/or parallel can be used.

Current sensor 15 and voltage sensors 22 and 23 are used to measure the current through, and the voltage across, each IGBT. The measured values are processed by the over-temperature protection unit 8 to estimate the temperature rise of the IGBTs in order

to protect them from overheating. A schematic diagram showing the over-temperature protection unit is shown in Fig. 7. An analogue multiplier 29 is used to determine the instantaneous power dissipation in the IGBT. The junction temperature of the IGBT is then estimated by modelling the transient thermal impedance 30 of the solid-state switch and considering the initial temperature of the heat sink. The value of the junction temperature obtained is compared with a pre-set value by using a comparator 31. If this value is higher than a pre-set value, a signal is sent to the firing-control unit 42 to turn off the HFCLID completely.

An over-temperature protection unit for the varistor 2 is shown in Fig. 8. The current flowing in the varistor can be calculated by subtracting the IGBT current (current sensor 15) from the HFCLID current (current sensor 5). The voltage across the varistor when it is conducting is almost constant (equal to the clamping voltage). Therefore, power dissipation in the varistor can be calculated by multiplying a constant (clamping voltage) by the current using an analogue multiplier 32. The total energy consumed by the varistor is then calculated using an integrator 33. The estimated energy is compared with a pre-set value (Emax proportional to the maximum energy per pulse given in the varistor data sheets) by a comparator 34. If the estimated energy is larger than Emax, a signal is sent to the firing-control unit 42 to turn off the HFCLID completely.

The varistor 2 is used not only to clamp the voltage but also to provide a current path when the switching component 1 is turned off. Thus, considerable energy is absorbed by the varistor during its operation. Analysis of ZnO varistors using an infrared imaging system showed that, when subjecting a varistor to repetitive pulses, it can handle a total energy absorption higher than the maximum energy per pulse usually specified in varistor data sheets. For high-power applications, one ZnO varistor may not have enough energy handling capability and hence shunt connection of two or more varistors is necessary.

Varistors connected in parallel do not share current equally unless they have identical V-I characteristics which is very difficult to achieve in practice (even within the same batch of varistors). In this invention, the current sharing has been significantly improved by connecting suitable small linear resistors 27 and 28 in series with the varistors, as shown in Fig. 9.

Fig. 10 is an implementation of the fault current detector and fault current level estimator 6. The fault current detector 35 includes a low-pass filter to prevent the HFCLID from nuisance tripping caused by switching or lightening transients. The fault current level estimator 36 (incorporating a high-pass filter to eliminate the dc offset) is used to estimate the value of the first peak of the fault current. If the estimated value is greater than Ihn t (see Fig. 10), i. e. over the interrupting capability of the IGBTs, a signal is sent to the firing-control unit 42 to block the firing of the IGBTs. This is to protect the IGBTs from excessive high short-circuit current which can only be interrupted by the vacuum circuitbreaker or an additional fuse with sufficient breaking capacity.

The maximum fault current limiting period is determined by the FCL period 41. If the fault persists for longer than the maximum time determined by the FCL period, the switching component 1 is kept off and the isolator 3 is opened.

A block diagram of the switching pattern controller 7 is shown in Fig. 11. On receiving a'fault'signal from the fault detector and estimator 6, the switching pattern generator 38 (in conjunction with two comparators 39 and 40) determines the switching pattern to be applied. To realize the HFCLID characteristics, two control variables are of interest. One is the maximum instantaneous fault current (Imax) at which the IGBT is turned off. The second is the minimum value (Imm) at which it is turned on after every current interruption. The maximum value should not exceed the maximum current interrupting capability of the switching component 1. Also, the value chosen should allow a sufficient time from the instant when a fault occurs to the instant when the fault current is detected and interrupted.

To set the minimum current (Imm) and maximum current (Imax), different techniques may be adopted. One technique is to set Imax and Im, and to let the switching frequency vary. Another is to set the average switching frequency and let Imm vary when the system impedance changes. Fig. 12 and 13 show the experimental results recorded with constant

Imax and Imm. These figures show the measured current waveforms and their corresponding frequency spectrum for Imin=O and Im, n=25 A, respectively. Fig. 14 shows the results from a computer simulation when Imax and Imm are set to vary in a sinusoidal way in order to reduce harmonics.

As may be noted from the current wavefonn shown in Fig. 12 to 14, operation of the HFCLID produces harmonic currents. Since the duration of current limiting operation is normally short, the level of harmonics generated is within the limits recommended by IEEE 519-1992 and G5/3 standards. The main concern would be the resonance with power-factor correction capacitors that would be caused under certain operating conditions. As shown in Figs. 12 to 14, choosing to switch-on at higher Imm or by setting Imax and Imin to vary in a sinusoidal way, the switching frequency can be controlled and harmonics can also be reduced. Therefore, resonance with capacitor banks can be avoided by selection of an appropriate switching pattern.

The present invention can find a wide range of applications in power distribution networks, particularly at low-voltage levels 400 V to 11 kV. Applications at higher voltage levels are also possible. It can be used in single-phase as well as in three-phase networks. Industrial plants using these voltage levels will find the HFCLID suitable to improve the performance of their networks and save them the cost of other expensive alternatives, for example upgrading circuit breakers.

The HFCLID may be implemented to improve the quality of the supply (e. g. voltage regulation, system stability against perturbations, etc. ) without increasing the overall short-circuit capacity. The use of the HFCLID can also help in mitigating voltage sags and improve system reliability.

Fig. 15 shows a single-line diagram of a typical distribution network with various possible locations of the HFCLID. The HFCLID can be implemented in transformers 44 and 45 paths to limit and interrupt short-circuit currents (positions 46 and 47). As a result, the series impedance of the transformer can be minimized in order to reduce losses and voltage drop during normal operation. In this position, the HFCLID can also provide backup protection for downstream devices. In addition, the stress on down stream equipment during a fault is reduced because of the limiting action of the HFCLID.

Implementing the HFCLID in the bus-tie position 50 splits the main busbar into two sections 48 and 49 during a fault. If there is a fault on a feeder supplied from one bus section, the voltage sag on the healthy section can be reduced and sensitive loads supplied from the healthy section are not affected.

The HFCLID may be used as a transfer switch (positions 52 and 53) to supply critical loads 55. If there is a fault on one feeder (say from Bus 48), the load can be quickly transferred to the healthy Bus 49. The voltage sag created at the supply point of the critical load will be very small. The HFCLID can also be used as a fault current limiter and interrupter for a single feeder, as in positions 51 and 54.

The device of the invention offers the following features that make it better suited than conventional devices to the duty required. These features enable the development of a "smart"device which is capable of automatically adjusting its operation (e. g. switching pattern) to suit the device and the distribution network requirements (e. g. protection coordination).

(a) The configuration and control scheme of the device of the invention enables flexible and fully controllable operation for both functions, i. e. current limiting and interruption.

(b) The device of the invention allows for the effective value of the fault current to be controlled (via the switching pattern controller 7) to suit the system requirements. In addition, unlike other FCLD, the fault current level during the operation of the device of the invention is independent of the system impedance or the fault location. This can assist in setting protection co-ordination within the distribution network.

(c) The device of the invention offers improved performance, including very fast response, ability to limit the fault current to within safe limits and a long multi- operation life span. It also offers the advantage of not requiring active cooling. In addition, it offers the ability to switch the power line ON and OFF via a remote control signal. It can also be designed to do a self-testing routine periodically, thus its reliable operation can regularly be checked and maintained.

(d) The temperature for each of the main components of the device (the switching component and the varistor) is continuously monitored and an estimated value of the final temperature could be found before it is actually reached. Therefore, safe operation of the device can always be ensured.

(e) The period of operation of the device of the invention is controllable which makes the device immune to nuisance tripping caused by transformer or capacitor inrush current,

network switching, etc. This makes it easier to incorporate the device with other relay protection schemes.

(f) The deployment of varistors across the solid-state switching components facilitates series connection of IGBTs, which is necessary for high voltage applications, e. g. 11 kV.

Claims

1. A hybrid fault current limiting and interrupting device for protecting an electric power line, the device comprising a power circuit including a solid-state switching component, a voltage clamping element connected in parallel with the switching component, a circuit breaker connected in parallel with the switching component and with the voltage clamping element, an isolating switch to the supply side of the switching component, and a current sensor, the device further comprising, between the current sensor and the switching component, a control circuit including a fault current detector and estimator and a switching pattern controller, the arrangement being such that, on detection of a fault current by the detector, the circuit breaker is opened and the fault current is transferred to the switching component which is turned on during the final stage of current interruption by the circuit breaker, and, when the fault current reaches a predetermined maximum value, the switching component is turned off and the current is diverted to the voltage clamping element which decreases the current to a predetermined minimum value at which the switching component is turned on again, the switching of the switching component being repeated for a predetermined time period until the current is fully interrupted either by operation of the device itself or by operation of downstream protection devices of the power line.

2. A device as claimed in claim 1 in which the switching pattern controller enables different switching patterns to be used to control the switching component so as to change the harmonic frequency spectrum and/or reduce the level of generated harmonics.

3. A device as claimed in claim I or claim 2 in which the switching of the switching component is repeated for a pre-set period after which the current is fully interrupted.

4. A device as claimed in any one of claims 1 to 3 in which the voltage clamping element is a varistor, the current limiting action being carried out by both the switching component and the varistor.

5. A device as claimed in claim 4 in which the voltage clamping element is a ZnO varistor.

6. A device as claimed in claim 5 and including a plurality of ZnO varistors connected in parallel, a linear resistor being connected in series with each varistor.

7. A device as claimed in any one of claims 1 to 6 in which the switching component is provided with an over-temperature protection unit.

8. A device as claimed in claim 7 in which the over-temperature protection unit comprises a current and voltage sensor together with a thermal transient impedance model of the switching component to predict an estimate of the junction temperature.

9. A device as claimed in any one of claim 1 to 8 in which the voltage clamping element is provided with an over-temperature protection unit.

10. A device as claimed in claim 9 in which the over-temperature protection unit is used to predict an estimate of the voltage clamping element temperature.

11. A device as claimed in any one of claims 1 to 10 for use in power networks operating at any voltage level.

12. A device as claimed in any one of claims 1 to 10 for use in single phase or three-phase networks.

13. A hybrid fault current limiting and interrupting device substantially as described with reference to and as illustrated by Figs. 3 to 15 of the accompany drawings.