TECHNICAL FIELD

The present invention relates to cycloconvertor equipment for transmision of power between a three-phase electric alternating voltage network with a first operating frequency and a single-phase alternating voltage network with a second operating frequency which is lower than the first frequency, and comprising a cycloconvertor for conversion of the voltage of the three-phase network into an alternating voltage with the second frequency as well as a three-phase alternating voltage terminal for connection to the three- phase network and a single-phase alternating voltage ter¬ minal for connection to the single-phase network.

BACKGROUND ART

It is previously known to use cycloconvertors for trans¬ mission of electric power between a three-phase network and a single-phase network. The three-phase network usually consists of a three-phase power network with the operating frequency 50 Hz or 60 Hz. The single-phase network may, for example, consist of a railway power. supply system with a operating frequency of, for example, 16 2/3 Hz or 25 Hz. Alternatively, the single-phase network may consist of a single single-phase load, for example an electrical induction f rnace.

In the above applications it is previously known to use as convertor a so-called cycloconvertor. Such a cycloconvertor may consist of two "direct-voltage" antiparallel-connected three-phase thyristor bridges, the alternating voltage terminals of which are connected to the three-phase network and the "direct voltage terminals" of which are connected to the single-phase network. The "direct voltage" of the bridges is controlled in accordance with a sinusoidal reference, and the cycloconvertor is in this way caused to

SUBSTITUTESHEET

generate on its single-phase side a sinusoidal alternating voltage. In a manner known per se, this voltage may be controlled with respect to magnitude, phase position __ffl_i frequency. Both active and reactive power may flow in arbitrary directions between the cycloconvertor and the single-phase network.

A cycloconvertor of the above known kind exhibits several disadvantages. The single-phase load in the secondary network gives rise to power pulsations in the three-phase primary network with twice the frequency of the secondary network. These power pulsations may have a high amplitude and give rise to harmful voltage variations in the primary network.

A cycloconvertor of the above-mentioned kind further generates harmonics with the frequencies | fp ± 2kfs | , where fp is the operating frequency of the primary network, fs the operating frequency of the secondary network and k = 1, 2, 3

Especially those of these harmonics which have the lowest frequency have large amplitude relative to the fundamental tone. Therefore, they exert a harmful influence on the primary network and because of their relatively low frequency they are difficult to suppress effectively with the aid of filter equipment.

In a cycloconvertor of the kind referred to here, a possible reactive power, consumed by the secondary network, is delivered from the primary network via the cycloconvertor. This reactive power loads the primary network and gives rise to an undesirable voltage drop therein. Further, the reactive power flowing through the cycloconvertor makes it necessary to dimension the cycloconvertor for a higher voltage than what would otherwise have been necessary. In normal operation, therefore, the cycloconvertor voltage is reduced in relation the maximally possible voltage, which causes the cycloconvertor to consume reactive power, which also loads the primary network in an undesirable way.

From EP-A1 0 026 374 and DE-C2 31 50 385, convertors of the kind described in the introductory part of this description are previously known, in which three-phase cycloconvertors are used and in which the secondary network is connected to two of the phase terminals of the cycloconvertor. Further, these prior art systems are provided with controllable reactive power members for symmetrization of the load. In this way, at least some of the above-mentioned disadvantages are eliminated. However, this is achieved at the cost of considerably increased complexity, since in these known systems the cycloconvertor consists of three complete single-phase convertors.

SUMMARY OF THE INVENTION

The present invention aims to provide cycloconvertor equipment of the kind described in the introductory part of the description, in which

no power pulsations with twice the secondary network frequency are transmitted to the primary network,

a considerable increase of the primary side power factor of the equipment can be attained,

a considerable reduction of subharmonics and harmonics in the currents on the primary side can be attained, and in which

these advantages can be attained with considerably simpler and hence more advantageous equipment, from a practical and economical point of view, than in the prior art cycloconvertors .

The above advantages are obtained, according to the inven¬ tion, by providing the cycloconvertor with two phases on the secondary network side, that is, designing the cyclo-

converter such that it generates a two-phase voltage with the frequency of the secondary network. On the secondary side the cycloconvertor has two phase terminals as well as one terminal which constitutes the neutral point of the cycloconvertor. The single-phase secondary network is connected to two of the terminals of the cycloconvertor. Between the third teminal of the cycloconvertor and each one of the two first-mentioned terminals, controllable reactive power members are connected for symmetrization of the load of the cycloconvertor. In those cases where the single- phase network consumes reactive power, a third controllable reactive power member is preferably connected in parallel with the terminals to the single-phase network to compensate for the reactive power thereof. By controlling these controllable reactive power members in a suitable way, an essentially complete symmetrization of the load of the cycloconvertor can be obtained. In this way, a complete elimination of the above-mentioned power pulsations with twice the secondary network frequency can be obtained. . Further, a considerable improvement of the power factor on the primary side is obtained compared with prior art cycloconvertors . Also, a considerable reduction of the subharmonics and the harmonics in the currents on the primary side is obtained. In a cycloconvertor according to the invention, these advantages can be achieved with the aid of considerable simpler equipment than what has hitherto been possible. In this way, a considerable reduction of the price and space requirement of the equipment, as well as an increase of its reliability, can be obtained.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following the invention will be described in greater detail with reference to the accompanying Figures 1-5. Figure 1 shows an example of cycloconvertor equipment according to the invention. Figure 2 shows the configuration of the main circuits of a phase of the cycloconvertor shown in Figure 1. Figure 3 illustrates in

the form of a vector diagram the mode of operation of the equipment according to Figure 1. Figure 4a shows equipment according to the invention with the single-phase network connected between a phase and the neutral point of the cycloconvertor, and Figure 4b illustrates the mode of operation of the equipment in the form of a vector diagram. Figure 5a shows alternative equipment and Figure 5b shows in vector form the mode of operation of the equipment.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Figure 1 shows an embodiment of cycloconvertor equipment according to the invention for transmission of power from a three-phase power network Nl to a single-phase network N2. As mentioned above, the latter may consist of a network in the proper sense of the word, for example a network for power supply of a railway system, or alternatively of a single consumer of single-phase power, for example an induction furnace or similar equipment. The equipment has a three-phase alternating voltage terminal TE1 for connection to the three-phase network and a single-phase alternating voltage terminal TE2 for connection to the single-phase network. The equipment comprises a cyclo¬ convertor SC with two mutually identically constructed cycloconvertor phases SCA and SCB. Each cycloconvertor phase consists of (see further Figure 2) two series- connected controllable double bridges . A cycloconvertor transformer TR1 has a Y-connected primary winding Wl connected to the three-phase network and four secondary windings with a D-connected secondary winding (e.g. WAl) and a Y-connected secondary winding (e.g. WA2) for each cycloconvertor phase (A) . On its low-frequency side the cycloconvertor has two phase terminals A and B as well as a terminal NP which constitutes the neutral point for the two- phase voltage generated by the cycloconvertor. Each cycloconvertor phase is connected between the neutral point NP and one of the phase terminals . In a manner known per se, the cycloconvertor is controlled such that at its phase

terminals A, B a symmetrical two-phase alternating voltage system with the desired frequency, for example 16 2/3 Hz or 25 Hz, is obtained. The alternating voltage terminals TE2 are connected to the phase terminals A and B via a transformer TR2.

Between the terminals A and NP a controllable reactive power member is connected, which consists of an inductor LAC, the current and thus the consumed reactive power of which can be varied with the aid of a phase-angle controlled thyristor switch TAC, which comprises two antiparallel-connected thyristor valves .

Between the terminals B and NP a second controllable reactive power member is connected, which comprises a capacitor bank CBC and an inductor LBCl which is phase-angle controlled with the aid of a thyristor switch TBC. An inductor LBC2 is connected in series with the capacitor bank CBC. Together with the capacitor bank the inductor is tuned to the third tone of the operating frequency of the secondary network for reduction of the harmonics generated by the thyristor switch TBC.

Between the phase terminals A and B, that is parallel to the single-phase network, a third controllable reactive power member is connected. In the same way as the recently described member, this member comprises a capacitor bank CAB as well as a phase-angle controlled inductor LABI, TAB. An inductor LAB2 is connected in series with the capacitor bank CAB and together with the capacitor bank tuned to the third tone of the secondary frequency.

For suppression of the remaining harmonics, a harmonic filter is connected between the terminals A and B and consists of a capacitor bank CF2 as well as the parallel connection of an inductor LF2 and a resistor RF2.

On the primary network side an additional harmonic filter, consisting of a capacitor bank CF1 and an inductor LF1, is connected to the network.

A network switch SW is arranged between the cycloconvertor SC and the network Nl .

A control device CD is adapted to control the controllable reactive power members. From an instrument transformer UM the control device is supplied with a signal U2 corre¬ sponding to the single-phase voltage, and from a current transformer IM the control device is supplied with a signal 12 which is a measure of the current flowing between the cycloconvertor and the single-phase network. A power measuring device PC calculates, on the basis of the signals U2 and 12, the active component P2 and the reactive component Q2 of the power flowing from the cycloconvertor to the secondary network. The measuring device delivers a signal P2/V2 proportional to the active component a well as a signal Q2 proportional to the reactive component. The latter signal is inverted in an inverter IN2 and supplied to a control pulse device CP3 which delivers control pulses SP3 to the thyristor switch TAB. To the control pulse device there is also fed a constant quantity -QfAB, where QfAB is the reactive power of the filter circuit CAB-LAB2.

This specification makes use of the convention that con¬ sumption of both active and reactive power is positive. Thus, the reactive power to an inductor is positive and to a capacitor negative. The control pulse device CP3 controls the thyristor switch such that the reactive power consumed by the inductor LABI is equal to the total input signal -QfAB - Q2 to the control pulse device.

As an example, it may be assumed that the single-phase load (including filter LF2, RF2, CF2) consumes the reactive power 10 Mvar, that is, Q2 = 10 Mvar. In the filter mentioned above, CAB-LAB2, the capacitor CAB dominates and the filter

may, for example, be assumed to generate a reactive power which is 25 Mvar. The result is that QfAB = -25 Mvar. The input signal to the control pulse device CP3 then becomes - 10 - (-25) = 15 Mvar. The control pulse device controls the current through the inductor LABI such that the inductor consumes 15 Mvar, which means that the total reactive power load from single-phase load plus compensator becomes zero, that is, only the active load remains.

In a corresponding way, the control pulse device CP1, which delivers control signals SP1 to the thyristor switch TAC, controls the current through the inductor LAC so that this inductor consumes the reactive power Q = P2/ 2.

The signal P2/V2 from the power measuring device PC is supplied via an inverter INI to a control pulse device CP2 for the thyristor switch TBC. To the control pulse device there is also supplied a constant quantity -QfBC which corresponds to the reactive power consumed by the circuit CBC-LBC2. Since the capacitor dominates, the consumed power is negative, that is the circuit generates reactive power. The control pulse device SP2 controls the current through the inductor LBCl so that the inductor consumes a reactive power Q = -P2/V2 - QfBC. This means that the compensating circuit connected between the phase terminals B and NP consumes a total reactive power Q = -P2/V2.

Figure 2 shows the configuration of the main circuits of one of the two phases of the cycloconvertor, namely, the phase which is connected to the phase terminal A. The cyclo¬ convertor phase is built up from two six-phase controllable double bridges SCAl and SCA2, respectively. The double bridge SCAl consists of the two six-pulse bridges BRAll and BRA12 and the double bridge SCA2 of the six-pulse bridges BRA21 and BRA22. The first two bridges are connected to the secondary winding WAl of the cycloconvertor transformer, and the last two bridges to the secondary winding WA2. Since one of these two windings is D-connected and the other

winding is Y-connected, the two double bridges operate with a 30° phase shift. The impact on the power system by the cycloconvertor phase will thus be a twelve-pulse impact, that is, the harmonics with the lowest frequency and the highest amplitude are eliminated and the remaining harmonics with low amplitude and high frequency can be damped, in a simple manner, to the desired degree.

Each one of the four bridges of the cycloconvertor phase comprises six controllable valves, in Figure 2 shown as conventional thyristor valves. In a known manner, each valve may comprise one single thyristor, or alternatively an arbitrary number of series-connected thyristors, parallel- connected thyristors or series-parallel-connected thyristors. Instead of thyristors, other controllable elements with a corresponding function may be used.

In those cases where an increased impact on the power system can be accepted, each cycloconvertor phase may be simplified to consist of one single double bridge.

Figure 3 shows in vector form the phase voltages UA and UB and the principal voltage UA-B on the secondary side of the cycloconvertor. The two cycloconvertor phases are con¬ trolled so as to generate alternating voltages UA and UB between the phase terminals A and B, respectively, and the neutral point NP, which have the same amplitude and are displaced in phase 90° in relation to each other. The current 12 flowing to the single-phase network has the active component Ip and the reactive component I . The latter is compensated for as described above with the aid of the members TAB, LABI, CAB and LAB2.

The reactive power member TAC, LAC is controlled such that its current I = Ip/^2. The reactive power member TBC, LBCl, LBC2, CBC is controlled such that its current IQO has the same amount. The two phase currents of the cyclo¬ convertor are:

IA = IP + 1_

that is, both I and Is will have the amplitude Ip/\2 and are in phase with the associated phase voltage.

Thus, the single-phase load is transformed with the aid of the compensator into a symmetrical and purely resistive two- phase load. Since the load on the cycloconvertor is a symmetrical three-phase load, the power pulsations with twice the secondary network frequency occurring in the single-phase network are not transferred to the primary network. Further, the load on the cycloconvertor is purely active. This, per se, entails a reduction of the reactive power load of the primary network. Further, this fact enables a narrower dimensioning of the cycloconvertor with respect to voltage than what has previously been possible, whereby the cycloconvertor is able to operate with a smaller degree of voltage reduction and therefore with a lower intrinsic reactive power consumption than what has previously been possible. In the cycloconvertor the primary network tones with the frequencies fp ± 2ks, where k = 1, 3, 5 ..., are eliminated, that is, the primary network tones fp ±2ks, where k = 2, 4, 6 ..., remain. These tones have low amplitudes and high frequencies and are therefore easier to damp, where necessary, to the desired degree through filters .

In the equipment described above, the secondary network is connected to the phase terminals of the cycloconvertor. Alternatively, the single-phase network may be connected between a phase terminal and the neutral point of the cycloconvertor. Figure 4a shows such equipment. Between the terminals A and NP, A and B and B and NP, controllable reactive power devices QCA, QCAB and QCβ, respectively, are connected. Each one of these may consist of a controllable inductor in parallel with a capacitor, that latter tuned to

the third tone with the aid of a inductor, in the same way as, for example, the reactive power member LABI, TAB, CAB, LAB2 in Figure 1. In those cases where a reactive power member only needs to serve as a consumer of reactive power, the capacitor branch may be omitted, if desired. Further, for the sake of simplicity, the control and measuring members and the high-frequency filter CF2, LF2, RF2 are eliminated in Figure 4a. The reactive power members are controlled such that QC _AB generates and QC _B consumes reactive power. QC _A is controlled so as to consume reactive power if II _Q | > | Ip|/2 and generates reactive power if |I _Q | < IIpI/2. Further, the control is performed such that their currents are:

UCABI = HplΛ'2

UCB I = Up 1 /2

where Ip and IQ are the active and reactive components of the secondary current 12. Figure 4b shows the different currents in the equipment in vector form. Since

IA = ICAB ICA ⁺ 12

IB = ICB ^~ lCAB

it can be simply shown that the two phase currents become purely active and equally great (= Ip/2), that is, a symmetrical two-phase load of the cycloonvertor is obtained.

Figure 5a shows equipment in which the single-phase network is instead connected between the terminals B and NP. The reactive power members are controlled in a corresponding way and such that QCA _B consumes and QC _A and QC _B generate reactive power. The vector diagram for voltages and currents is shewn in Figure 5b. Also in this case, a symmetrical and

purely active two-phase load is obtained, the amount of the phase currents being equal to Ip/2.

In the cycloconvertor equipment described above the secondary voltage has been assumed to have a constant amplitude. The quantities QfBC and QfAB used in the control equipment can therefore be approximated as constant quantities. In the case where the cycloconvertor operates with a varying single-phase voltage, measuring members may instead be arranged to determine these quantities by current and/or voltage measurement.

An open-loop control system has been described above, in which the control angles for the thyristor switches of the reactive power members are controlled in dependence on the measured active and reactive components of the single-phase load. Alternatively, of course, a closed-loop control sys¬ tem may be arranged for each reactive power member, which system then determines the reactive power of the reactive member by measurement, compares this reactive power with the desired one and influences the control angle of the thyris¬ tor switch so as to obtain agreement between the measured and the desired reactive power.

In the equipment described above, the active and reactive components of the single-phase -load are measured and used for controlling the reactive power members. Alternatively, the phase currents of the cycloconvertor can be measured and a closed-loop control system be adapted to control the reac¬ tive power members in dependence on these phase currents such that the currents form a symmetrical two-phase system which is in phase with the secondary voltages of the cyclo¬ convertor.

In the equipment shown in Figure 1, one of the reactive power members - TAC, LAC - can only consume reactive power. If desired, all the reactive power members may be designed such that each one of these members can be controlled to

consume as well as to generate reactive power. The cyclo¬ convertor equipment can then be caused to provide the function described above in case of arbitrary directions of the active and reactive power flows between the cyclo¬ convertor and the single-phase network.

Further, there has been described above how the reactive power members are controlled to provide complete symmetry of the load of the cycloconvertor and to completely eliminate the reactive power flow through the cycloconvertor. However, in practice, of course, neither the symmetry nor the elimination of reactive power need be complete, but certain deviations in this respect may be tolerated without any significant reduction of the good properties of the equipment . If desired, the reactive power members may even be controlled such that a symmetrical reactice three-phase power is generated (or possibly consumed) by the cyclo¬ convertor, in which case the cycloconvertor in relation to the primary network may be caused to function as a controlled phase compensator.

The members described above for compensation of the reactive component of the single-phase load may be omitted if the power factor of the load is near 1, which may be the case in certain applications, for example when using the cyclo¬ convertor equipment for power supply of a railway system.

The single-phase transformer TR2 may be omitted if the voltage in the single-phase network is of such a magnitude that the network can be directly connected to the cyclo¬ convertor equipment .