REACTIVE POWER COMPENSATION
The invention relates to reactive power compensation and in particular an apparatus and method for achieving reactive power compensation. Reactive power, measured in volt-amperes reactive or vars is a well known phenomenon in electrical power systems. In an AC circuit including inductive and/or capacitative components (such as any electrical power supply system), an active current is taken by the resistive component and gives rise to a corresponding power component. In addition a reactive current is taken by the inductive and capacitative components of the system, 90° out of phase with the active current and giving rise to an out of phase reactive power component. Reactive power is discussed further below in terms of vars.
In an electrical power system the var demand varies in dependence upon, inter alia, the load conditions. For example loads requiring magnetising currents such as transformers and induction motors "import" vars. On the other hand capacitative loads export vars. Accordingly var compensators are commonly introduced into electrical power systems to generate or absorb vars dependent on the demand on the system. One suitable scheme, known as static compensation, includes capacitors used when the var load demand is heavy and inductors used when the load demand is low. Typically these are manually switched and hence very slow. Another known scheme uses synchronous compensators, for example synchronous motors with no mechanical output. This allows rapid, continuous compensation from vars export at the rated level to vars import at approximately half the rated level. Synchronous compensators of this type suffer from various problems, however, including high maintenance requirements.
In a paper entitled "Widening Applications for Var Compensation" by
Dr. D.F. Binns and Dr. R. Ghazi in Power Engineering Journal January 1991, further discussion of known var compensators is set out. However the paper is directed at identifying additional potential uses for var compensators and the problems discussed above are not addressed. Further discussion of static var compensators (SVC's) is found in an article entitled " Transmission SVC Design" by Heinz K. Tyll, presented as a Summary of Contribution to Discussion Meeting on Static Var Compensation at IEE, London, Savoyplace, March 13, 1996.
According to the invention there is provided a reactive power compensator apparatus for a power supply system comprising an active component, a voltage supply providing a voltage across the active component, a plurality of taps to respective voltages of the voltage supply and fast-acting, tap dedicated switches for switching between the taps to change the voltage from the voltage supply across the active component. As a result fast and accurate var compensation is achieved with reliable and non-complex equipment.
Preferably the apparatus includes a controller arranged to detect a voltage of the power supply system and control the switches dependent on the detected voltage. The apparatus preferably incorporates an "anti-hunting" feature to minimise the switching duty on the fast acting tap dedicated switches. The switches may be thyristors or vacuum-switches, may be arranged to switch in single and/or multiple steps, and may be arranged to switch in less than 80 milliseconds. Preferably the voltage supply comprises the secondary winding of a transformer and the active component is a capacitor.
The compensation apparatus hence acts as a capacitor bank. The vars supplied by a capacitor bank varies in proportion to the capacitance and in proportion to the square of the voltage upon it. Thus it is that by varying, by means of fast acting, tap dedicated switches, the voltage across the capacitor, a
variable var output can be effected.
Preferably the reactive power compensator comprises: a pair of first main switches capable of bearing the electrical load along respective first electrical path; a pair of diverter switches each for diverting current in the respective first paths along a respective second path during change over; and an auxiliary circuit comprising a transformer having a primary winding connected in a common portion of the first path and a secondary winding, an auxiliary switch and an impedance both connected across the secondary winding of the transformer.
According to the invention there is further provided a method of providing reactive power compensation in an electrical power supply in which a reactive power compensator includes a plurality of taps to respective voltages of a voltage supply and fast-acting tap dedicated switches for switching between the taps, comprising the steps of switching between the taps to achieve a desired voltage from the voltage supply across an active component.
Preferably switching takes place in less than 80 milliseconds.
Preferably the method comprises actuating one of the pair of diverter switches associated with the one of the main switches that is closed; opening the auxiliary switch so that current in the first path is diverted along the second path associated with the said one diverter switch; opening the said one closed main switch; closing the other diverter switch; closing the auxiliary switch so that current is diverted to the second path associated with the said other diverter switch; closing the other main switch; and opening the said other diverter switch so that current is diverted to the
first path associated with the said other switch.
According to the invention there is yet further provided a reactive power compensator including an active component, a plurality of taps arranged to connect the active component to respective voltages of a voltage supply and fast acting tap-dedicated switches for switching between the taps.
An embodiment of the invention will now be described, by way of example, with reference to the accompanying drawings of which:
Fig. 1 shows schematically a typical electrical power system;
Fig. 2 shows a var compensator according to the present invention; Fig. 3 shows the compensator of Fig. 2 in more detail;
Fig. 4 is a circuit diagram of a gate turn-off thyristor circuit for use in the compensator of Fig. 3;
Fig. 5 is a circuit diagram of a double gate turn-off thyristor circuit for use in a modified form of the compensator of Fig. 3; and Fig. 6 is a modified version of the compensator of Fig. 3.
Fig. 7 shows a compensator as part of a schematic electricity supply network;
Fig. 8 shows the arrangement of Fig. 7 geographically.
Referring to Fig. 1 an electrical power system is designated generally at 1. The system can be represented as including a voltage source 2, a transmission line impedance 4 of value Li and a load impedance 6 of value L2+R. A var compensator arrangement is shown generally at 8 connected in parallel with the load impedance 6.
Referring now to Fig. 2 the var compensator arrangement 8 includes a primary, high voltage winding 10 in parallel with the load impedance 6 forming part of a transformer shown generally as 36. An active component in the form of a capacitance 32 is provided across the secondary winding 34 of the
transformer 36. A var compensator switch 30 is provided in series with the capacitance 32 and is arranged to switch between a range of voltage values tapped from the secondary winding 34 to provide a corresponding voltage across the capacitance 32. The var compensator arrangement accordingly effectively comprises a capacitor bank. The vars supplied by a capacitor bank varies in proportion to the capacitance and in proportion to the square of the voltage upon the capacitance. Accordingly the vars supplied by the var compensator arrangement can be varied by varying the voltage tapped by var compensator switch 30.
Returning to Fig. 1 the system further includes a control module 12 connected between the transmission line and the compensator. The control module 12 includes a sensor for sensing the voltage at the point of connection. For example the sensor can be a conventional voltage transformer. Upon sensing a drop in voltage at the point of connection the controller sends a switching signal to the var compensator switch 30 to tap a higher voltage from the secondary winding 34 hence increasing the voltage across the capacitor 32. As a result the vars supplied to the system increases raising the voltage at the point of connection; in this way the voltage at the point of connection is maintained more constant that would otherwise by the case.
The controller 12 comprises standard control circuitry that will be apparent to the skilled person. For example the product sold by ABB under the name MACH Control System can be used in a conventional manner. The specific manner of switching the var compensator switch will be better understood with reference to the discussion below of Figs. 3 to 6. Preferably the controller incorporates an "anti-hunting" feature to minimise the switching duty on the fast- acting tap dedicated switches.
It will be further appreciated that the var compensator of the present invention can be placed at various points in electricity supply networks, for example near large variable inductive loads, at substations in load centres where there is little or nor local generation, and at the end of heavily loaded transmission lines.
Figs. 7 and 8 show possible configuration of the arrangement. Electricity generation occurs at a point 50 which can be geographically distant - for example 10's or even 100's of miles - from a load or loads 52. The compensation is then provided at a further point 54. The var compensator switch 30 is shown in more detail in Fig. 3.
The var compensator switch 30 comprises for example a series of 19 tap vacuum circuit breakers VBl-19 between the high voltage and neutral terminals of the secondary winding. The skilled person will be aware of the vacuum breakers commonly used in power transformers. For example, they are described in the Article 'Load Tap Changing with Vacuum Interrupters', in IEEE Transactions on Power Apparatus and Systems, Vol. PAS-86, N04, April 1967. In this particular example the vacuum breakers used are type V504E manufactured by Vacuum Interrupters Limited of London N3, England.
The vacuum breaker has contacts sealed in an evacuated enclosure. During contact separation, a plasma created by the vaporisation of the contact material provides a way for the continuation of current flow. The charge carriers making up the plasma disperse very rapidly in the high vacuum and recombine on the metal surfaces of the contacts. The metal ions leaving the vacuum arc in this way are continuously replaced by new charge carriers generated by the vaporising contact material at its root. At current zero the generation of the charge carriers stops, but their recombination continues. Therefore the contact zone is rapidly deionised and the current is broken.
Vacuum circuit breakers are also reliable particularly when they are constructed so that the only moving part is a single movable contact. This also has a relatively long service life and low maintenance requirements relative to switches immersed in transformer oil. The fire risk is also improved using vacuum circuit breakers.
Each tap breaker VI- 19 is connected, at one terminal, to a point in the transformer winding which divides the winding into a set of, say, eighteen constituent winding parts LI- 18. Similarly, not all the taps and winding parts are specifically illustrated. The other terminals of every other tap breaker VB1, 3, 5- 19 and VB2, 4-18 are respectively commonly connected to inductors La and Lb. While the inductors La and Lb are indicated as discrete components, there is sufficient leakage inductance inherent in the tap windings in many circumstances. The opposite ends of the inductors are connected through two serially connected main vacuum circuit breakers VBA and VBB similar to those used for the tap breakers VBl-19. Two serially connected gate turn-off thyristor (GTO) switches GTOA and GTOB are connected in parallel across the main breakers VBA and VBB between the opposite ends of the inductors La and Lb.
An inductor Lc is connected between the GTO switches GTOA and GTOB and to the neutral terminal of the transformer winding. An auxiliary circuit is associated with the changeover breakers VBA and
VBB. The primary winding of, in this example, a 1:20 ratio auxiliary transformer T is connected between the changeover breakers and the neutral terminal of the high voltage transformer. A varistor VR or other constant voltage device, is connected across the secondary winding and an auxiliary GTO thyristor switch GTOC is connected in parallel with the varistor VR across the auxiliary transformer T.
In this particular example, the GTO thyristors GTOA and GTOB used are sold under device reference DG 758BX45 by GEC Plessey Semiconductors. More detail of the GTO thyristor switches are shown in Figure 4. Each switch comprises an anti-parallel diode bridge arrangement although other rectifying circuits can be used. The diodes used are sold under reference DFB55 by GEC Plessey Semiconductors. The GTO thyristor is connected in circuit between respective pairs of diodes D1/D2 and D3/D4 arranged in the anti-parallel bridge configuration. The GTO thyristor is centrally connected between oppositely conducting diodes in conventional manner. The GTO thyristor is actuated by opto-isolated (or magnetically isolated) signals driving a floating power supply and gate drive. The thyristor is force commutated. This is illustrated in Figure 5 which shows in more detail the GTO-based switch for GTOA and GTOB. The same principle of construction applies equally to GTOC.
The GTO thyristors GTOA, GTOB and GTOC are each supplemented by a turn-off snubber circuit which comprises a resistor/capacitor pair R2/C2 in series connected across the GTO and a varistor VR3 connected across the resistor/capacitor pair. A diode D5 is connected across the GTO. When the GTO is turned off, by removing the actuating signal from its gate, load current diverts onto the snubber capacitor C2 through the resistor R2. This limits the rate of rise of voltage across the GTO.
Although in a multi-phased power distribution system the power factor is kept close to unity and steady state, under some conditions the phase difference between current and voltage can be anywhere between +/-1800. Thus, a fast response tap changer must be capable of working over the full range of power factors. Thyristors would have difficulty in commutating. Although it is possible to devise circuitry in which ordinary thyristors could be used. GTO thyristors are more suited for this application because standard thyristors create a temporary tap-
to-tap short circuit when going, for example, from a high voltage to a lower voltage tap at leading power factors. GTO thyristors are turned off from the gate terminal and do not suffer from this.
The present invention circumvents the need to take into account power factor considerations by using the auxiliary circuit to transfer smoothly load current from the vacuum circuit breaker to the parallel diverter GTO switches. Referring again to Figure 3, in steady state the switch GTOC is closed. Consequently, the transformer secondary winding is short-circuited. When full load current (typically 1KA) flows through the primary of the transformer T, due to the turns ratio of 1 :20, 50A rms flows through the switch GTOC. This is within the capacity of the large GTO thyristors available.
Assuming the main breaker VBA is initially closed and the circuit through the high voltage transformer follows its path through, for example, the tap breaker VB2 which is also closed, to begin a tap change the auxiliary switch GTOC in the auxiliary diverter switch circuit is turned off, just after the main breaker GTOA, is turned on. The current in the auxiliary transformer secondary winding now flows through the varistor VR, creating a secondary square-wave voltage of IkV and a primary square-wave voltage of 50 volts. The primary square-wave voltage is sufficient to divert the load current from the main breaker VBA to its associated diverter switch GTOA. The rate of transfer of current from the main breaker VBA to the diverter switch GTOA is governed by the primary square-wave voltage of 50 volts and the size of the inductor Lc. The rate of rise of current in the switch GTOA must be limited in its capacity.
Having transferred the load current to the switch GTOA, the vacuum switch VBA can be opened without a substantial current and therefore little arcing. To complete the tap change to, for example, LI + L2 ( which is a tap change down) the tap isolator VB3 will have to be closed in preparation and the
tap isolator VB2 opened. At a current zero the diverter switch GTOA is turned off and the diverter switch GTOB, associated with the main breaker VBB, is turned on. Thereafter the load current, now following the diverted path through the diverter switch GTOB, can be transferred to the main path by closing the main breaker VBB and then the auxiliary switch GTOC to remove the reflected impedance from the primary of the auxiliary transformer by shorting across the varistor VR.
The two main breakers VBA and VBB, due to the presence of the auxiliary circuit, never have to make or break a heavy current. The only current will be the leakage current of the auxiliary switch GTOC referred to the primary of the auxiliary transformer T and the magnetising current of the transformer T. This is likely to be in the region of about 3 A and will result in negligible contact wear.
For design reliability it is considered necessary to operate GTOs at about 70% to 80% of the recommended rated voltage. In many higher voltage applications, such as the electricity distribution networks, the currently available GTOs may be inadequate to achieve this. For example, the DG758BX45 GTO previously mentioned has a voltage rating of 4500V and a current rating of 1365 A for halfwave rectification. Thus, it may be necessary to use two GTO's in parallel for the diverter circuits associated with the main breakers VBA and VBB. A double GTO anti-parallel bridge arrangement in place of the circuits GTOA and GTOB is illustrated in Figure 5. Again, suitable snubber circuitry is connected around two GTO thyristors connected between the anti-parallel diodes.
It has previously been necessary to effect a tap change in a sequence of steps between adjacent winding parts on alternate legs of the high voltage paths. The present arrangement allows larger steps to be taken between non- neighbouring taps. For example, a gate turn-off thyristor may have a breakdown
voltage of about 4.5kV or more. With a typical voltage drop across a tap of IkV it is possible to change 3 taps in one step.
It is necessary for the diverter switches GTOA and GTOB to switch at or near a current zero to ensure low switching losses. Therefore a circuit is needed to enable the GTO's to be correctly timed. A clock signal can be derived from the main a.c. current. For example, Figure 6 illustrates a clock pulse generating circuit for one phase. It will be appreciated that a multi-phase supply will require separate synchronisation for each phase. A current transducer CT isolates the main a.c. circuit from the control logic and produces a signal proportional to main current. This signal is buffered by an inverting amplifier Ul and then applied to the input of a second operational amplifier U2 which is arranged as a comparator. A diode D3 clamps the output voltage of the comparator to ensure compatibility with following logic circuitry. A conventional arrangement of NAND gates and associated resistor and capacitor components produces pulses in synchronisation with the current zeros at the output CLKA.
The arrangement described above therefore allows very fast switching for reactive power compensation. It will be appreciated that various alternative fast acting switches can be used including but not limited to thyristors and vacuum switches. It will be noted that switching can be achieved in a single and/or multiple steps and the present arrangement allows the changes to occur in less than 80 milliseconds.