FIELD OF INVENTION

The present invention generally relates to high voltage power systems. More particularly the present invention relates to a switching device for a high voltage power system and an arrangement comprising such a switching device.

BACKGROUND

Switching devices are known to be used in various high voltage

applications. A switching device may as an example be used as part of or as the main breaker of a hybrid direct current (DC) circuit breaker, using both mechanical and electric switches. The switching device may also be provided in a valve of a voltage source converter (VSC). A VSC may in this case convert between alternating current (AC) and DC and be provided in a converter station that is an interface between a DC high voltage power system and an AC high voltage power system. A VSC may also be provided as a reactive power compensating device in an AC system such as a Static VAR Compensator (SVC). These are just a few examples of arrangements where HV switching devices may be used. A switching device in the above mentioned examples is required to be able to withstand high voltages. A switching device will thus have to be able to avoid breaking down below the withstand voltage. In common for all these systems is that a switching devices is normally nowadays realized through the use of semiconductor switching elements, such as Insulated Gate Bipolar Transistors (IGBTs) or Integrated Gate-Commutated Thyristors (IGCTs). In high voltage applications, the switching elements have relatively low voltage blocking capabilities, typically in the range of a few kVs. Therefore, for very high voltage applications such as in Grid systems, a series connection of such elements is needed to reach 10 s to 100 s of kVs. This requires a higher reliability of the individual devices, but also adds to the complexity, as well as, the size requirement for such a device. It is known for lower voltage ranges, that a combination of two different types of switching elements can be used to provide the higher withstand voltage of one of the devices together with the lower switching capabilities for the second one. A few suggestions have been made for providing combinations of different types of semiconductor switching elements in a switching device. Such a switching device is often termed a "cascode device " if these different types of switching elements are connected in series. A "cascode device ' 'is thus a series connected hybrid device and the expression will be used in the following with this meaning. The main reason for this approach is to combine certain advantageous characteristics of each element for a more optimum overall performance and trade-off relationship.

One such known combination is a SiC Junction Field Effect Transistor (J FET) and a Silicon Metal Oxide Semiconductor Field Effect Transistor (MOSFET) cascode, for instance in order to replace an IGBT. This combination provides a normally-off device since the normally-on JFET is not desirable in many applications. Furthermore, the MOSFET gate drive control is preferred in most applications where IGBTs are to be replaced by the JFET cascode.

Another example is a Gate Turn-Off Thyristor (GTO) or integrated Gate- Commutated Thyristor (GCT) in series with a MOSFET, referred to as the Emitter Turn-Off thyristor (ETO). This has been demonstrated to provide a low on-state Thyristor structure with a voltage controlled gate drive and saturation/ short circuit capability. It is important to note that also the IGCT concept employs a cascode structure by having a MOSFET connected in series with the IGCT gate. However, the MOSFET only commutates the turn-off current when compared to the ETO.

But these examples, combining two types of semiconductors can only reach withstand voltages below a few 10 kV and are therefore not suitable for typical HV applications, unless several of them are put in series again.

Many high voltage power systems are used for power transmission. In these systems it is important that the efficiency is high. As much as possible of the power delivered into a power transmission system has to leave the power transmission system. The losses in the power

transmissions system have to be low, especially in order to reduce the heating due to them.

However, the problem with series connected semiconductor switching elements is that each such element has a conduction loss. The conduction loss of a switching device made up of semiconductor switching elements will thus be the sum of the conduction losses of the semiconductor switching elements. The conduction losses of a switching device may thus have a significant influence on the efficiency of a high voltage power system. In addition the reliability of the individual devices has to be high if they are combined in series. They are also more complex, e.g., as each device needs to be switched using a gate unit. Finally the space

requirement is rather high for these systems.

It would because of this be of interest to obtain a switching device, where it is possible to obtain a reduction of the conduction losses, especially by the use of a single device with high voltage withstand capability and comparable lower losses.

The present invention is provided for addressing this problem.

"Electron tubes", that is, vacuum and especially gas-filled tubes have been used in the time before the widespread availability of power semiconductors to provide switching functionality in high voltage applications. These electron tubes are based on the flow of electrons in vacuum or in a plasma at low pressures. They have been shown to be able to withstand high voltages up to 135 kV. Therefore they have the potential to be used as single devices instead of series connection of a number of devices.

There exist a large number of designs of such electron tubes using different physical mechanism to provide different functionalities. They can be equipped with high-current as well as with switching-on and switching-off capabilities. Especially low-pressure gas-filled tubes have been developed for power applications.

SUMMARY OF THE INVENTION

The present invention addresses the problem of obtaining a switching device that has low conduction losses, as well as, the possibility to use only two devices to provide high voltage withstand capability. This object is according to a first aspect of the present invention achieved through a switching device for a high voltage power system, the switching device comprising:

a first semiconductor switching element capable of being turned off and having a first gate and a first and a second current conduction terminal, and

a second switching element capable of being turned on and comprising an electron tube with a second gate and a first and second electrode, wherein the first switching element and the second switching element are connected in series with each other so that the first electrode of the second switching element is galvanically connected to the second current conduction terminal of the first switching element with the first current conduction terminal and the second electrode being provided for connection to the power system, and the switching elements being jointly operable for breaking or making a current path between the second electrode and the first current conduction terminal.

The object is according to a second aspect of the invention achieved through an arrangement in a high voltage power system comprising a switching device according to the first aspect.

The jointly operable switching elements may involve them being sequentially operable in order to obtain the common goal of making or breaking the current path.

The present invention has a number of advantages. The switching device has low conduction losses. Furthermore the number of switching elements in the switching device is low, thereby reducing the cost, size requirement as well as allowing for a high reliability.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will in the following be described with reference being made to the accompanying drawings, where

Fig. 1 schematically shows a DC power transmission system connected to two AC systems, where the DC system comprises two voltage source converters and a hybrid DC breaker and one of the AC systems comprises a reactive power compensating device,

fig. 2 schematically shows a first variation of a switching device according to the invention,

fig. 3 schematically shows a second variation of a switching device according to the invention,

fig. 4 schematically shows a third variation of a switching device according to the invention,

fig. 5 schematically shows a fourth variation of a switching device according to the invention, fig. 6 schematically shows a first type of voltage source converter where the switching device is used,

fig. 7 shows a second type of voltage source converter comprising cells, fig. 8 shows a cell where cell switches are realized through switching devices, and

fig. 9 schematically shows a hybrid DC breaker comprising a main breaker realized using at least one switching device.

DETAILED DESCRIPTION OF THE INVENTION

In the following, a detailed description of preferred embodiments of the invention will be given.

Fig. 1 shows a simplified Direct Current (DC) power transmission system comprising a first converter station 10 and a second converter station 12. The two converter stations 10 and 12 are interconnected by a DC link 18 , which DC link 18 comprises a hybrid HVDC circuit breaker 22 using both mechanical and electric switches. The first converter station 10 comprises a first converter 14 connected to an Alternating Current (AC) power transmission system via a first transformer Tl, and the second converter station 12 comprises a second converter 16 connected to a second AC power transmission system via a second transformer T2. None of the AC power transmission systems are shown in any detail. However there is provided a reactive power compensating device 20 in the first AC power transmission system , which reactive power compensating device may be a so-called Static VAR Compensator (SVC). The DC system and the AC systems are all examples of high voltage power systems, and in this case also examples of high voltage power transmission systems. The converters 14 and 16 may both be voltage source converters (VSCs) and may be either two-level converters or multilevel converters comprising cells, i.e. voltage source converters employing cells for forming multiple voltage levels. The conversion is in this example furthermore made between DC and three-phase AC. Therefore, both converters have three phase legs, one for each phase. In examples given later only one phase leg will be shown and described. However, it is well known that all phase legs have the same realization. It should also be realized that there are other types of voltage source converters, such as neutral-point clamped three- level converters and various n-level converters.

The converters 14 and 16, the reactive power compensating device 20 and the hybrid DC breaker 22 are all examples of arrangements in high voltage power systems where switching devices are used. It should be realized that these are just a few examples of arrangements in high voltage power systems, for instance power transmission systems, where switching devices may be used. As mentioned earlier, the traditional way of providing a switching device is through series-connecting a large number of semiconductor switching elements, i.e. switching elements realized through the use of

semiconductors, such as Insulated-Gate Bipolar Transistors (IGBTs) or Integrated Gate-Commutated Thyristors (IGCTs).

As was also mentioned earlier, the use of several such elements in series raises the conduction losses of the resulting switching device, which in many cases has a negative effect on the efficiency of the high voltage power system in which the switching device is used.

It may also increase the probability of the switching device being faulty as well as the cost and size of the switching device.

The above-mentioned problems are addressed through the introduction of a new switching device.

The novel switching device is a "cascode device" comprising a series connection of two different switching elements, a semiconductor switching element and a switching element that is based on an electron tube, such as a gas-filled tube or a vacuum tube with very high voltage blocking capability. Fig. 2 shows a first variation of the new switching device 24. The switching device 24 comprises a first semiconductor switching element 26 capable of being at least turned off and possibly also turned on and having a first gate Gl and a first and a second current conduction terminal CCT1 and CCT2. The switching device 24 also comprises a second switching element 28 capable of being at least turned on and possibly also turned off and comprising an electron tube with a second gate G2 and a first and second electrode El and E2, where the gate G2 may be configured to control a current flow, with advantage a unidirectional current flow, between the electrodes El and E2. The first electrode El is the cathode, while the second electrode E2 is the anode.

As can be seen in fig. 2, the first switching element 26 and the second switching element 28 are connected in series with each other so that the first electrode El of the second switching element 28 is galvanically connected to the second current conduction terminal CCT2 of the first switching element 26. The second electrode E2 and the first current conduction terminal CCT1 are here the connection terminals of the switching device 24, which means that they are provided for connection to other parts of a high voltage power system . If the switching device 24 is a part of an arrangement such as the converter 14, the SVC 20 or the DC circuit breaker 22, then the terminals CCT1 and E2 may be connected to other parts of the converter 14, the SVC 20 or the DC circuit breaker 22. The second switching element 28 may be a gas-filled tube. As an alternative it may be a vacuum tube. The first switching element 26 may be a Thyristor based switching element, such as an Integrated Gate

Commutating Thyristor (GTO, IGCT). As an alternative it may be a transistor, such as an Insulated Gate Bipolar Transistor (IGBT) or a Junction Field Effect Transistor JFET. In the case of an IGCT, the first current conduction terminal CCT1 may be a cathode and the second current conduction terminal CCT2 an anode. In the case of an IGBT, the first current conduction terminal CCT1 may be an emitter and the second current conductor terminal CCT2 may be a collector.

The switching device 24 may be a very high voltage gas or hard- vacuum tube element as second switching element in series with a relatively lower voltage semiconductor element as first switching element. As the second switching element a number of option exists based on different concepts such as hard vacuum tubes like triodes, tetrodes etc., gas-filled tubes such as Thyratrons, crossed-field plasma discharge switches, or specific elements such as the Crossatron and hollow-electrode elements such as the Hollowtron or the pseudo-spark-switch to name a few. These elements have a wide range of voltage and current ratings and different

functionalities. However, they mainly differ from power semiconductors in their very high voltage withstand capability for a single device, which can be between lOkV- 135 kV, therefore making them suitable for very high voltage systems such as HVDC. Some of those elements also have very attractive loss performances for such high voltage devices as compared with an equivalent series connected power element configuration to achieve the same voltage ratings. Depending on element type they also provide only turn-on capability such as the Thyratron or turn-on and -off functionalities such as for the Crossatron and Hollowtron.

It is possible that only the second switching element is capable of being turned on. The first switching element on the other hand requires at least turn-off control. This means that if the second switching element has a gate with the capability of being turned on, then it is possible that the first switching element 26 does not have the capability of being turned-on, but is only a turn-off type of switching element, such as a JFET. Furthermore, the first switching element 26 has a first voltage withstand capability and the second switching element 28 has a second voltage withstand capability and the voltage withstand capability of the second switching element 28 may be considerably higher than the voltage withstand capability of the first switching element 26. It may for instance be at least ten times higher than the voltage withstand capability of the first switching element. Alternatively it may be at least 2o times higher or at least 25 times higher. Such a combination could as an example be an IGCT able to withstand

4500 V connected in series to a Crossatron or Thyratron able to withstand up to 135 kV. As can be seen this would lead to the second withstand voltage being 80/ 3 times higher than the first withstand voltage. In fig. 2 it can also be seen that the second electrode E2 has a higher potential than the first current conduction terminal CCTl, which is shown through the second electrode having a positive potential (+) and the first current conduction terminal having a negative potential (-). This means that the technical current flow or conventional current flow is from the second electrode E2 to the first current conduction terminal CCTl. The flow of electrons would be in the opposite direction.

The switching elements are furthermore jointly operable for breaking or making a current path between the second electrode E2 and the first current conduction terminal CCTl. In the example shown in fig.2 this joint operation is obtained employing a gate control unit 30 that is a part of the switching device 24. The joint operation is thus a concerted operation. The joint operation may also be a sequential operation. It should however be realized that such joint operation is not necessarily obtained through such a gate control unit 30 , but may be provided as a part of the control of the arrangement in which the switching device 24 is included. For instance, if the switching device 24 is provided as a part of a VSC, such as the first converter 14, then the gate control functionality may be provided as a part of the overall control of the switching devices of the VSC, for instance in a part that is used for forming an AC wave shape.

The gate control unit 3o is provided for making or breaking a current conduction path that goes between the second electrode E2 and the first current conduction terminal CCT1.

The gate control unit 30 may be configured to make the current path through using a gate control sequence comprising first applying a gate control signal to the first gate Gl for turning on the first switching element 26 and thereafter applying a gate control signal to the second gate G2 for turning on the second switching element 28. The second gate G2 may thus receive the gate control signal after the turning on of the first switching element 26. The order could also be the opposite.

Both the semiconductor switching element 26 and Gate tube element 28 may thus be turned on initially with a given sequence, which is not critical as long as the power semiconductor 26 is switched on first and the tube 28 still blocks the voltage. It may therefore be critical to have a switch-on of the electron tube, while a switch-on of the semiconductor may be optional.

One limitation of most types of electron tubes used as second switching element is that they can often switch-off only moderate currents as they rely on a low plasma density during that process. In contrast to this semiconductor elements can switch off rather high currents.

In order to address this the gate control unit 30 when breaking the current path may first apply a gate control signal to the first gate Gl for turning off the first switching element 26, thereby lowering the current in the current path from a first regular current level to a second lower turn-off current level in order to allow the second switching element 28 to be turned off at this second current level for breaking the current path. Depending on the type of second switching element 28 , the gate control unit 30 may furthermore apply a gate control signal to the second gate G2 for turning off the second switching element 28 when the current in the current path is at the second current level for breaking the current path.

In the example with IGCT and Crossatron or Thyratron given above, during turn-off, the IGCT is turned-off initially to reduce the current to a very low level before turning off the Gas tube 28 in the case of a

Crossatron. If the second switching element 28 is a Thyratron, no gate control signal may be needed because the Thyratron may switch itself off due to the low current flowing.

It can be seen that the use of the first switching element 26 allows to provide turn-off capabilities also to second switching elements 28 that normally do not provide them or alternatively enhance strongly the turn- off capabilities of them to higher nominal currents. In addition the first switching element 26 improves the reliability of the combined device.

Compared to the typical forward voltage of 10 V up to 1 kV of tubes, strongly depending on the chosen tube design, the additional voltage drop through the semiconductor is always only a small factor.

In principle the configuration of the two switching elements in the cascode device can be arbitrary. It is thus as an alternative possible that the first electrode El of the second switching element 28 is connected to the negative potential (-), the second electrode E2 galvanically connected to the first current conduction terminal CCT1 of the first switching element 26, the second current conduction terminal CCT2 of which is connected to the positive potential (+) . A second variation of the invention showing this realization but without a control unit is schematically shown in fig. 3.

Thereby there would be a current flow from the second current conduction terminal CCT2 to the first electrode El. However, it may be preferable to have the tube first in the technical current flow direction and the semiconductor second, just as in fig. 2. The switching device 24 may thus be designed for connection to the high voltage power system so that the current conduction direction through it is from the second electrode E2 to the first current conduction terminal CCT1. This is due to the asymmetric design of the tube, where the switching on (and off) is due to voltages with respect to the cathode. Typically electron tubes have a high withstand voltage between gate and anode and a low voltage required between gate and cathode for their control. In such a configuration both Gates Gl and G2 need to be controlled relative to the "-" voltage in Figure 2. In contrast with the two devices exchanged G2 would be controlled with respect to the "-" voltage, but Gl with respect to the intermediate voltage level. During the open state the voltage between Gl and G2 would then be of the order of the high voltage. In addition some gas tube elements require a commutation of a substantial part of the current to the gate often by the use of an additional power source. In the configuration shown in Figure 2 the semiconductor element is allowed to limit this commutated current, as it is part of the current path. In an alternative configuration with the positions of the

semiconductor and gas tube exchanged this is not the case, reducing the usefulness of the design.

As mentioned above, the switching device 24 may be used in various arrangements such as in VSCs. In such arrangements it is customary to use an anti-parallel current condition element, such as a diode. A switching device 24 may thus comprise an anti-parallel diode. One example of this is shown in fig. 4, which shows the first and second switching elements 26 and 28 from fig. 2. However, in addition there is here an anti-parallel diode D, having its cathode connected to the second electrode E2 and its anode to the first current conduction terminal CCT1. Please keep in mind that the diode in this case can either be based on a semiconductor or electron-tube principle. In this variation the gate control unit has been omitted. However, it is possible that one is included also here. Another example of a switching device comprising an anti-parallel unidirectional current conduction element is shown in fig. 5, which also shows the first and second switching elements 26 and 28 from fig. 2. However, in addition there is here a unidirectional current conduction element in device 24 comprising of an anti-parallel sealed electron tube 31 comprising a third gate G3 and a third and fourth electrode E3 and E4, where the gate G3 is configured to control a current flow, i.e. a

unidirectional current flow, between the electrodes E3 and E4. The third electrode E3 may be an anode, while the fourth electrode E4 may be a cathode. The third electrode E3 is connected to the first current conduction terminal CCT1 and the fourth electrode E4 to the second electrode E2 . Also here it is of course also possible to add the gate control unit, which may then also control the third gate G3 , for instance to be always on.

As mentioned above the switching device 24 may be used in a number of types of arrangements. Fig. 6 shows a phase leg of the first converter realized as a two-level converter 14A with two converter valves CV1 and CV2, where a first converter valve CV1 is connected between the first transformer Tl and a negative DC voltage - VDC of the DC link 18 and a second converter valve CV2 is connected between the first transformer Tl and a positive DC voltage +VDC of the DC link 18. There is also a DC link capacitor C connected across the two valves CV1 and CV2. In this arrangement each valve may be realized through a switching device, for instance as shown in fig. 2 and 3. If providing the second valve CV2, then the second electrode would be connected to the positive DC voltage +VDC while the first current conduction terminal would be connected to the first transformer Tl. If the switching device of fig. 2 or 3 were to be used, an anti-parallel unidirectional current conduction element may have to be added as is shown in fig. 4 and 5. Alternatively the switching device of fig. 4 and 5 could directly be used. Another example of an arrangement where the switching device 24 may be used is shown in fig. 7. In this case the first converter is a modular multilevel converter 14B using half-bridge cells, where fig. 7 shows a first phase leg comprising an upper and a lower phase arm joined to a phase leg midpoint via a corresponding phase arm reactor LA sand LB, respectively, where the phase leg midpoint is joined to the first transformer (not shown). As can be seen in fig. 7, each phase arm is made up of a number of cells 32, each being realized as a series connection of two switches in parallel with a cell capacitor. As is also shown in fig. 7 such switches are conventionally realized as IGBTs with anti-parallel diodes.

As can be seen in fig. 8 , each switch of a cell 32' may be replaced by a switching device 24, for instance the switching device of fig. 4 or 5.

Thereby a series-connection of two switching devices 24 is connected in parallel with a cell capacitor Ccell. In this case the second electrode of the upper switching device would be connected to the upper end of the cell capacitor Ccell, while the first current conduction terminal would be connected to the lower switching device and more particularly to the second electrode of the lower switching device, the second current conduction terminal of which would be connected to the lower end of the cell capacitor Ccell.

If the switching device of fig. 2 or 3 would be used instead, then a corresponding unidirectional current conduction element would of course have to be connected in parallel with each switching device 24 as is shown in fig. 4 and 5.

An SVC may also be realized as a VSC, for instance through a modular multilevel VSC, with three phase legs connected in a delta configuration. In this case each phase leg may be realized through one or more switching devices. Finally there is shown a hybrid HVDC breaker 22 in fig. 9. It comprises of a main breaker 35 in parallel with a series-connection of an ultrafast disconnector 36, for instance realized through a mechanical switch, and a load commutation switch 38. There is here also a first surge arrester SAl in parallel with the main breaker 35 and a second surge arrester SA2 in parallel with the load commutation switch 38.

It is in this case possible that the main breaker 35 is realized through one or more switching devices in series. It is also possible that the load commutation switch employs a switching device.

In case the AC system has an AC circuit breaker that is normally realized through a series-connection of semiconductor switching elements, then also such semiconductor switching elements may be replaced by a switching element.

A new switching device has thus been shown that can be used in a number of different arrangements in a high voltage power system, such as a high voltage power transmission system . The new switching device has a number of advantages including low conduction losses, can be realized through a limited number of elements and has a high reliability.

The gate control unit may be realized in the form of discrete components, such as a combination of logic circuits. It is also possible to use

programmable circuits such as Field-Programmable Gate Array (FPGA) or an Application Specific Integrated Circuit (ASIC). Yet another alternative is in the form of a processor with accompanying program memory comprising computer program code that performs the desired control functionality when being run on the processor.

From the foregoing discussion it is evident that the present invention can be varied in a multitude of ways. It shall consequently be realized that the present invention is only to be limited by the following claims.