RU2199838C2 - Power converter for supplying electric arc furnace with dc current - Google Patents

Abstract

FIELD: secondary power supplies for electric arc furnaces. SUBSTANCE: converter has at least one transformer incorporating primary winding supplied with three-phase current and at least one secondary winding passing three-phase current from its output to rectifying facilities wherefrom DC voltage and current are supplied to load, mentioned rectifying facilities being provided with controlled semiconductor elements connected to each secondary winding. Rectifying facilities have circuit arrangement in the form of wheel of free-running clutch type that functions as wheel with free-running clutch and mentioned controlled semiconductor elements are triggered at firing (turn-on) angles actually varying so that reducing time of conducting state of main triggered semiconductor elements accordingly increases conducting state duration of circuit arrangement in the form of wheel of free-running clutch type, and vice versa, which makes it possible to feed load with actually DC active or reactive power irrespective of variations in load impedance. EFFECT: minimized reactive-power variations. 10 cl, 15 dwg

Description

 The invention relates to a power converter for supplying direct current to an electric arc furnace.

 Such electric arc furnaces are used, in particular, for melting and subsequent refining of scrap metal.

 The invention is applicable more precisely for supplying direct current to such a charge, the electrical parameters of which are often and substantially change.

 The most widely known method for feeding electric arc furnaces is shown in FIG. 1.

 It consists in the fact that the electrodes are directly powered by three-phase alternating current: the first step-down transformer 1 feeds the bus system 2, distributing the current to one or more individual adjustable transformers 3, lowering the voltage to the desired level and directly supplying the furnace 4. The latter consists of a housing 5, in the lower part of which scrap metal 6 is placed, equipped with three movable electrodes 7, each of which is connected to one of the phases of the secondary circuit of a step-down transformer 3, which, thus, yvaet 8 arcing between the electrodes 7 and the scrap metal 6, causing melting, and then refining (cleaning) of the latter.

FIG. 2 and 3 illustrate the operational characteristics of such an AC furnace, respectively, the current-voltage characteristic: voltage (U _arcs ) / current (I _arcs ) and the ratio reactive power (Q) / active power (P) for various points (moments) of the furnace ( SS is the point corresponding to short-circuit operation, F is the mode corresponding to the melting of the scrap metal and A is its refining).

 This type of furnace, powered by alternating current, provides substantially constant power output (ΔP is very small), but is characterized by significant fluctuations in reactive power (ΔQ) and current magnitude (ΔI), while outside the short circuit area the furnace operates essentially with constant active power, however, significant fluctuations in reactive power can cause significant voltage deviations in the supply circuit (input circuit from the side of the power source), introducing into it electrical noise, in particular in the form of a “flicker”, t o there are voltage fluctuations in the 0-30 Hz band leading to “ripple” or flickering of the lighting.

 In recent years, there has been a tendency to replace such AC power with DC power with the introduction of an AC to DC converter between the output transformer and the electrodes.

DC power does provide two decisive benefits, from which all other benefits come from, namely that
- the positive pole of the arc is the scrap metal itself, solid or in the molten state, and the movable electrode (single or multiple) serves as the negative pole. Since the energy dissipation along the arc is nonlinear and larger from the positive pole, the scrap metal heats up more than the electrode;
- current overloads during short circuits, very frequent during the start of melting and during the melting process, are limited by the converter.

 Due to this, the electrode consumption is reduced, the re-injection of electrical noise into the input network is reduced, that is, the flicker phenomenon, and the furnace productivity is also increased.

 The general type of DC / DC converters used to date is shown schematically in FIG. 4.

 The primary winding 9 of the output (feed furnace) transformer feeds one or more secondary windings 10, 11, each of which is connected to the corresponding rectifier bridge 12 of the type of a classic Gretz bridge circuit on controlled semiconductor elements (thyristors). One output of each bridge is connected to a common movable electrode 7, and the other, usually through a smoothing reactor 13, is connected to the corresponding hearth electrode 14, which is in direct contact with scrap metal 6.

 The structure of the Gretz Bridge, shown schematically in the figures under 12, is illustrated in more detail in FIG. 5.

This structure contains two rows, each of three thyristors, 15 16 each, connected in a “double parallel” (or “three-phase two-half-wave”) circuit with a common cathode and common anode, respectively, through the other electrode of which come in the same pattern (star or triangle) multiphase voltages; each row of thyristors is started with a common ignition angle, respectively α ₁ or α ₂ .
Usually, to limit the harmonics injected into the network, several six-half-period Gretz bridges are used, each of which is powered from phase-shifted secondary windings. So in FIG. Figure 4 shows two secondary windings 10, 11, phase shifted by 30 ^o by their inclusion in the star-delta circuit, feeding two Gretz bridges, but this solution can be expanded, it is valid for a circuit with three Gretz bridges (with a phase shift - 20 ^o , 0 ^o , +20 ^o ), with four Gretz bridges (phase displacements: 0 ^o , +15 ^o , +30 ^o and +45 ^o ), etc., while each Gretz bridge is a six-half elementary converter.

 Thus, although in the future we will consider a circuit with two Gretz bridges, it is obvious that the invention can be true for a larger number of rectifiers powered from phase shifted secondary windings.

 In FIG. 6 and 7 show the same as in FIG. 2 and 3, but for the case of a DC device.

 As you can see, the device operates with direct current, but in contrast to the technique of using alternating current, with significant fluctuations (changes) in active power. We can also note though smaller, but still noticeable fluctuations in reactive power Q. In addition, on average, reactive power consumption remains high, which usually necessitates the use of a load regulator 17 for the output transformer, and also suggests the presence of a rather significant compensation battery 18 In this part, the DC device has almost no noticeable improvement over the AC device.

 Fluctuations (Q) of reactive power during the transition from one operating mode to another, however, are smaller than in the case of AC power supply, hence lower voltage fluctuations in the input circuit (in the network from the power supply side) and a decrease in the flicker phenomenon. However, in insufficiently correctly (with insufficient margin) calculated power supply networks, these voltage fluctuations, proportional to the consumed reactive power and inversely proportional to the short circuit power of the network, often remain too large, in particular due to flicker, which forces the use of additional and very expensive corrective means ( devices called “anti-flicker” or “TCPs”; Thyriston Control Reactor), as is most the case with AC power.

 At the same time, if we consider the instantaneous values of the consumed reactive power, we can see that the peak values are significantly reduced in the case of direct current supply, since the current regulation, which is possible thanks to the converter, is usually quick enough to limit the current overload during short circuit to almost negligible magnitude.

 The closest to the claimed invention solution is a device for powering an arc DC furnace according to A.S. USSR 1702546, class H 05 B, 7/144, 1991

 In the specified device, as in the claimed one, the rectifying means comprise a wheel type circuit with an overrunning clutch made in the form of a return diode connected in parallel with the load.

 One of the objectives of the invention is to eliminate the disadvantages inherent in both AC power systems and DC power systems, by creating a new structure of a DC converter, which would significantly reduce the reactive power consumption and significantly improve the current-voltage characteristic to maintain a DC arc while reducing the total cost of the converter by simplifying the output transformer (since the regulator is not needed here) and significantly reduce the size ompensatsionnoy battery (typically twice).

 In particular, an AC-to-DC converter allows operating with a substantially constant active power, as in the case of supplying the furnace directly with alternating current, but with a low consumed reactive power and without resorting to an increase in the size of either the converter or the associated transformer.

 According to the invention, it is possible to very simply influence the characteristic - reactive power / active power (Q / P) ratio in such a way as to optimize this characteristic depending on the conditions of use of the furnace.

 It is possible, for example, to establish a characteristic that would minimize the amount of reactive power consumed for a given operating mode of the furnace, or, alternatively, a characteristic that would minimize fluctuations in reactive power relative to its average value, in particular for the case of small sizes of the supply network, reduction of flicker of such networks is one of the main requirements both in terms of efficiency (minimizing interference re-injected into the network) and in terms of installation cost (due to neniya devices "antiflikker").

 You can also choose the characteristics of the furnace so that the magnitude of the consumed reactive power remains essentially constant, despite possible fluctuations in current, in particular in the case of fluctuations in the impedance of the load.

 The device that is proposed in this application is a device of the general type mentioned above, that is, it contains at least one transformer, the primary winding of which receives a three-phase alternating current and at least one secondary winding of which produces a three-phase current supplied to rectifier means, from the output of which rectified current and voltage are supplied to the load, and these rectifier means contain controlled semiconductor elements for each secondary winding of the transformer.

 According to the invention, the device is characterized in that the rectifying means comprise a circuit operating on the principle of a wheel with an overrunning clutch, and these controlled semiconductor elements (valves) are started with substantially varying ignition (unlocking) angles, the ignition angles being changed so as to increase the duration conducting (open) state of the valves in the circuit as a wheel with a freewheel, with a corresponding reduction in the opening time (open state) of the running rectifier valves bridge, and vice versa, thus allowing to deliver to the load a substantially constant active power or reactive power despite fluctuations in the load impedance.

 It is advantageous to carry out the overrunning clutch circuit on the diodes so that the converter is non-reversible.

It is very advantageous for a transformer having at least two secondary windings with corresponding rectifier means fed by these secondary windings to be designed so that they are interconnected according to a “bias” pattern so that the corresponding ignition angles α ₁ and α ₂ were unequal among themselves: α ₁ ≠ α ₂ .
In this case, in the first embodiment of the invention, the opening (ignition) control angles α ₁ and α ₂ for a given point of the melting course were selected based on the condition of minimizing the amount of consumed reactive power, that is, so that its resulting average value per cycle of operation was minimal.

In the second embodiment, on the contrary, the opening control angles α ₁ and α ₂ are selected for a given rectified current so as to minimize fluctuations, namely in the form of a flicker, reactive power relative to its average value.

In addition, the opening control angles α ₁ and α ₂ can be advantageously selected such that in the event of a rectified current deviation, the average value of the reactive power consumed remains substantially constant.

 According to a preferred embodiment, the offset is "parallel." It is preferable to provide means for cyclic cross-switching the direction of displacement of the respective rectifier means and / or means for regulating the currents of two respective rectifier means, with the aim of keeping these currents equal.

 According to a preferred embodiment of the invention, said controllable semiconductor elements of the rectifier means of each of the secondary windings are connected according to the Gretz bridge circuit, one of the terminals of such a bridge being connected to a negative load electrode, namely to a movable electrode of the furnace, and the other terminal is connected to the corresponding positive an electrode, namely, with a bottom electrode, while the "wheel with overrunning clutch" circuit is connected between the output of the corresponding secondary winding of the transformer and responsible of the winding bridge of Graetz.

 You can also group several rectifiers in one unit, such as shown, the corresponding transformers of which will be recorded with a phase offset.

The invention is further explained in the detailed description of examples of its implementation with reference to the accompanying drawings, in which:
figure 1 - depicts a diagram of an electric arc furnace with direct AC power according to the previously known technical solution;
FIG. 2 and 3, respectively, the current-voltage characteristic and the characteristic reactive power / active power according to a known solution with AC power;
FIG. 4 - an electric arc furnace powered by direct current through a converter according to the prior art;
5 is a known structure of the Gretz bridge, shown in more detail;
FIG. 6 and 7, respectively, the voltage / current and reactive power / active power characteristics of such a known direct current furnace.

FIG. 8 is a configuration of a DC electric arc furnace through a converter according to the invention;
FIG. 9-11 are known configurations for including several transducers of the same structure with an offset;
FIG. 12 and 13, respectively, the voltage / current and reactive power / active power characteristics of the furnace fed with direct current through a converter made according to the invention, the lower part of FIG. 13 the law of changing the angles of opening of the valves of the Gretz bridge is given;
Fig - various characteristics of reactive power / active power, which can be obtained by appropriate selection of the laws of variation of the opening / ignition angles / and which allow you to adapt the course of the furnace to various conditions of its use;
FIG. 15 is a family of characteristics of reactive power / active power obtained by changes in the rectified current caused by a decrease in the load impedance.

 The main idea of the invention is to connect the classical structure of the thyristor converter with a wheel-type device with an overrunning clutch and establish a thyristor control law corresponding to essentially constant active power, the value of which is set corresponding, on the one hand, to the characteristics of the arc (to increase the efficiency of the furnace), and on the other hand, the inherent capabilities of the converter.

 A freewheel type device, as you know, is a device that locks (blocks) in one direction and conducts in the other direction, with a bias that allows you to give electrical energy accumulated by inductive elements during the conducting period of the thyristor converter to the load during the period when thyristors are locked following the period during which the thyristors were in a conductive state.

 The overrunning clutch device according to the invention permits a change in the pulse duty factor for the pulses of the conducting state of the main thyristors (depending on the ignition angle), and the duration of the conducting state in the device, like a wheel with the overrunning clutch, increases as it decreases for the main thyristors.

 According to the invention, a device of the type of a wheel with an overrunning clutch of this kind is used to increase the amplitude of the direct current with decreasing voltage (or vice versa), while the ignition angle of the thyristors increases, and the pulse filling coefficient of the pulses of the conducting state decreases accordingly (and vice versa). Thus, we get work with essentially constant active power without resorting to an excessive safety margin of either thyristors or transformers.

 In the illustrated example regarding the electric arc furnace, the overrunning clutch device as shown at 19 in FIG. 8 is preferably performed on diodes 20 connected between the midpoint of the secondary winding 10 of the transformers and each output terminal of the Gretz bridge 12, while the diode obviously has an offset in the direction opposite to the direction of current flow in the diodes of the Gretz bridge.

 The use of diodes instead of thyristors in the overrunning clutch device makes the converter non-reversible in power.

 At first glance, this characteristic may seem like a drawback, but in the application under consideration it actually provides remarkable advantages.

 Indeed, in the classical Gretz bridge, the converter is made “reversible in two quadrants” (unidirectional direct current, but bidirectional constant voltage), which makes it especially effective in case of current regulation in the case of arc shorting, when this short circuit is sought to be eliminated by melting scrap metal as soon as possible, which caused a short circuit (in a solution with direct AC power, current overload following a short circuit, on the contrary, helped melt this etallolom, but due to the appearance of reactive power peak, is highly undesirable).

 The invention allows, by making the converter irreversible in power, it is possible to successfully combine these two sides: indeed, at the time of a short circuit, the current increased due to the irreversible (irreversible) nature of the structure allows the molten metal to be melted faster, which caused a short circuit, but this does not cause a peak in reactive power in network from the power supply side, since such an increase in current will be perceived only by the diodes of the device as a wheel with an overrunning clutch. Moreover, such an increase in current can be controlled and optimized depending on the intrinsic characteristics of the electrodes under the condition of their least wear.

 In accordance with another aspect of the invention, it is advantageous to design a transformer with several secondary windings, such as 10 and 11, connected in a circuit called “biased”, namely in the type of “parallel bias” (in phase), a combination of “overrunning clutch” effects and displacement allows you to get a significant reduction in consumed reactive power.

 More specifically, the phase displacement technique, known per se and illustrated in FIG. 9-11, consists in connecting at least two converters of the same structure and such a phase shift of their ignition control to affect the consumed reactive power; in this case, they speak of “bias control” or “sequential control”. When using this method, two bridges are usually connected in series from the rectified current side, as shown in FIG. 9 (a circuit called a “sequential bias circuit”).

 In this case, since the constant voltage supplied to the arc is relatively small compared with the capabilities of very powerful thyristors, a “sequential bias” type circuit would lead to poor technological use of the thyristors. Therefore, a “parallel offset” type circuit is preferred, two possible variants of which are shown in FIG. 10 and 11.

 In the "parallel offset" circuits of FIG. 10 and 11 connect two Gretz bridges, each of which actually consists of two half-bridges displaced from each other, due to which even harmonics are formed (unlike the classical Gretz bridge), but with a cross internal displacement of both bridges, which allows a total of at the output, suppress these even harmonics.

 However, the classical “parallel displacement” circuit (that is, without a wheel-type device with an overrunning clutch), shown in Figs. 10 and 11, has a significant drawback - it creates the risk of "tripping" of the thyristors, in particular, at a constant voltage close to to zero. To limit this risk, they were previously forced to limit the range of deviation of the ignition angles, thereby significantly reducing the reactive power gain.

 Significant advantages provided by the claimed combination of displacement and the principle of a wheel with a freewheel clutch include the complete elimination of the risk of thyristors re-switching (triggering) if the device is used as a wheel with a freewheel clutch of the above type, which allows changing the pulse duty cycle for conductive pulses of the main thyristors . In this case, the total reactive power gain is fully used with a very high reliability of the device.

 In FIG. 8 shows the complete proposed circuit thus obtained, including a wheel-type circuit with an overrunning clutch with neutral with diodes and a “parallel displacement” type of inclusion.

 In the first embodiment, the invention seeks to use the appropriate control law (switching thyristors) to minimize the consumed reactive power Q and, consequently, the reactive energy at each point in time (that is, for a given point in the furnace stroke). This case, shown in the drawing, corresponds to a situation in which the reduction in average reactive power is the priority optimized parameter.

To achieve this, the control parameters α ₁ and α _{2 are} affected in such a way as to obtain the maximum bias, that is, the maximum absolute value of the difference / α ₁ - α ₂ /. Another type of regulation (since it is possible to affect both parameters α ₁ and α ₂ separately, it is possible to regulate two dependencies) allows you to adjust the direct current (or active power under load.

In FIG. 12 and 13 show furnace operating characteristics obtained using the circuit of FIG. 8 in the case of the first embodiment of the invention (minimizing the average consumed reactive power)
In FIG. 12 shows a voltage / current characteristic that is very close to that obtained by direct supply with alternating current, that is, working with constant power (the shaded part illustrates the improvement compared to the classical AC to DC converter).

In FIG. 13 shows the characteristic reactive power / active power for a given furnace travel point, that is, for a constant arc current I _arcs (or a rectified current I _d ) of constant magnitude and equivalent varying arc resistance.

This characteristic shows that the reactive power consumed remains less than about a third of the maximum active power, which represents a significant improvement compared to the known solutions, both in the case of AC power and in the case of DC power (compare with Fig. 3 and 7) ; in addition, at the bottom of FIG. 3 illustrates the laws of variation of the ignition angles α ₁ and α ₂ depending on the power output.

The advantages of such a scheme are
- elimination of the regulating device (under load or at idle) of the transformer-converter,
- a significant reduction in the power of the compensation-filter battery (at least twice),
- a noticeable improvement in productivity in terms of duration and quantity,
- reducing voltage fluctuations in the network from the power source,
- reduction of flicker,
- reduction of converter losses, and
- reduction of harmonics and interference from the converter.

 These advantages are obtained by adding a device of the type of a wheel with an overrunning clutch on diodes and a special calculation of the connections (smoothing reactors, cables, electrodes) at the converter output, on the load side, providing a sufficient margin of safety with respect to current overloads acceptable according to the invention. In practice, this leads to lower capital costs for the electric arc furnace.

In the second embodiment of the invention, a control law is selected, that is, a law of variation of the ignition angles α ₁ and α ₂ depending on the power supplied in order to not reduce the average reactive power consumed, but to reduce the reactive power fluctuations relative to its average value, and therefore, reduce the flicker level although this consumes a slightly larger average reactive power.

 Thus, we obtain a converter capable of controlling active power, while consuming reactive power of a constant value, which can be compensated for by a simple capacitor bank of constant capacity (of the type shown under 18 in Fig. 4), which allows you to completely get rid of the need to use an expensive TCP or antiflicker device, even with a very low power supply.

More specifically, in this embodiment, the general arrangement of FIG. 8 (i.e., a circuit with a Gretz bridge and an overrunning clutch with displacement) with such a control law for the parameters α ₁ and α ₂ that, unlike the previous case, does not seek to maximize the displacement / α ₁ - α ₂ /.
In FIG. 14 shows a characteristic of reactive power Q / active power P for a given furnace travel point, that is, for arc current 1 of _arcs (or rectified current I _d ), constant in magnitude, and varying equivalent arc resistance. On this figure
- characteristic I corresponds to the classical converter circuit without a device according to the type of a wheel with a freewheel and without bias (that is, the characteristic of Fig. 7);
- characteristic II corresponds to a circuit with a device similar to a wheel with a freewheel, but without any displacement, that is, the circuit of FIG. 8, but with α ₁ = α ₂ ;
- characteristic III corresponds to the same scheme with maximum displacement, that is, the maximum absolute value / α ₁ - α ₂ / max (characteristic illustrated in Fig. 13).

 It is clear that between the displacement equal to zero and the maximum displacement, an infinite number of different characteristics can be obtained, and all of them will be located in the shaded area between characteristics II and III.

More precisely, these characteristics are determined by the following relationships:
P = R • I $\genfrac{}{}{0ex}{}{\text{2}}{\text{d}}$ = E _d • I _d = E _do • I _d • f _p (α ₁ • α ₂ );
Q = = E _do • I _d • f _p (α ₁ • α ₂ ),
where P is the active power,
Q is the reactive power,
R is the equivalent arc resistance,
I _d is the rectified current,
E _d is the rectified voltage under load,
E _do - rectified idle voltage at zero bias,
f _p , f _q are the ignition angle functions.

Thus, for some given operating point, that is, for some data, R and I _d P will be given, and Q will be adjustable in a certain range.

It is possible, for example, to set the constant Q, which corresponds to characteristic IV, or also to any other characteristic located in the shaded area. Having thus determined I _d , P and Q, we deduce directly from them α ₁ and α ₂ .
In addition, if we assume, as shown above, it is necessary to maintain a substantially constant active power by increasing the current I _d when the equivalent arc resistance decreases (and vice versa), then the operating point will move and you can get the characteristic offset in this diagram.

Suppose now that the solution of a system of two equations with two unknowns given above provides for a given operating point with constant current a characteristic such as that shown under V in FIG. 14, i.e., a monotonically increasing function Q = f (P). In this case, when the current I changes, the characteristic V moves, as shown in FIG. 15, thus defining a family of characteristics, each of which corresponds to a certain given point (current I _d ) of the furnace stroke. In this case, choose a monotonically decreasing law of change I _d = f (P), such that Q remains constant despite changes in I _d .

 In this latter case, a converter is proposed that provides constant reactive power Q and active power P. at the same time over the entire range of the furnace course (or at least most of it).

 Various enhancements may be added to the general scheme described above.

First of all, it is advantageous to provide for cyclic switching of the directions of displacement of both bridges ((α ₁ + α ₂ for one of the bridges and α ₂ + α ₁ for the other bridge) in order to balance the heating of semiconductor elements and limit the risk of the appearance of a constant component harmful to the magnetic induction of the transformer .

 The period of such cyclic switching can be calculated depending on the thermal time constant of the semiconductor elements and on the maximum permissible constant component for transformers. As for the switching moment in the period, it should be selected from the condition of minimizing the amplitude of the resulting transient switching current from the AC mains or from the DC side so that this transient current does not exceed such currents arising from natural arc oscillations.

 Secondly, a double current adjustment can be provided, namely, separate adjustment of the currents of each of the two Gretz bridges connected in parallel, or better, the simultaneous regulation of the sum of two currents and their equality among themselves (regulation of a type with several variables in the diagonal of the bridge) .

, имеющего целью свести четные гармоники к приемлемому уровню. Такой риск может иметь место, когда два моста Гретца не вполне идеально соединены в параллель, а именно, когда соединение осуществлено не на уровне выходных зажимов (со стороны нагрузки) сглаживающих реакторов, а на уровне двух подовых электродов, которые должны быть электрически связаны между собой через жидкий металл.In case of a mismatch between the two bridges, poor compensation of even harmonics appears and it may be necessary to limit this mismatch (limitation

with the goal of reducing even harmonics to an acceptable level. Such a risk can occur when two Gretz bridges are not perfectly connected in parallel, namely, when the connection is made not at the level of output clamps (on the load side) of the smoothing reactors, but at the level of two bottom electrodes, which must be electrically connected through liquid metal.

 Thirdly, it is possible to include between them several blocks of the same type, such as shown in FIG. 9, powered by phase-shifted transformers to reduce harmonics. The phase displacement of the transformers is preferably carried out at the level of the primary windings, for example, by connecting the windings in the "triangle" -zigzag or "triangle with an intermediate tap" configuration.

Claims (10)

1. A power converter for supplying direct current to an electric arc furnace, comprising at least one transformer, the primary winding of which is connected to a three-phase alternating current source, and at least one secondary winding to rectifier means configured to be fed from their output the load of the rectified voltage (U _arcs ) and current (I _arcs ), and these rectifying means contain in the circuit of each secondary winding controlled semiconductor elements, characterized in that semiconductor elements are configured to be triggered by variable ignition angles, selectively variable, within the operating mode of the arc furnace, the reactive power consumed by the converter with respect to a given average value (Q).  2. The Converter according to claim 1, characterized in that the ignition control angles for a given rectified current are selected so as to minimize oscillations, in particular in the form of a flicker, reactive power relative to its average value (Q).  3. The Converter according to claim 1, characterized in that the ignition control angles are selected with the ability to maintain a constant average value (Q) of the consumed reactive power when the rectified current changes.  4. The Converter according to claim 1, characterized in that the rectifying means are interconnected according to the scheme with a parallel phase displacement.  5. The Converter according to claim 1, characterized in that means are provided for cyclically switching the direction of displacement of both respective rectifier means.  6. The Converter according to claim 1, characterized in that means are provided for regulating the currents of two respective rectifier means to ensure the equality of these currents with each other.  7. The Converter according to claim 1, characterized in that the controlled semiconductor elements of the rectifier means corresponding to each secondary winding are connected according to the Gretz bridge circuit, one of the terminals of which is connected to the negative load electrode, namely to the movable electrode of the furnace, and the other terminal - with a positive electrode, namely with a bottom electrode.  8. A converter according to any one of claims 1 to 7, characterized in that the rectifying means include a wheel type circuit with an overrunning clutch.  9. The converter according to claim 8, characterized in that the wheel type circuit with an overrunning clutch is made on diodes.  10. The Converter according to claim 8 or 9, characterized in that the ignition angles are selected with the possibility of releasing constant active power on the load, regardless of changes in the rectified current.

Images