RU2072604C1 - Method for suppression of arc short circuits to earth in load windings and supplying transformers of three- phase supply line with non-grounded neutral wire - Google Patents

Abstract

FIELD: electric engineering. SUBSTANCE: method involves measuring instantaneous levels of voltage in phases of supply line and neutral bias voltage. If arc short circuit to earth is detected, compensation of capacitance and active constituents of current of single-phase short circuit to earth is controlled. . This involves measuring amplitude and phase of electromotive force which exist between earth short circuit and neutral wire of supply line, and keeping amplitude and phase of neutral bias voltage at level which is equal to these values. Device provides possibility to suppress arc short circuits to earth in windings of electric motors, load transformers, power supply transformers regardless to method of their connection (star or delta connection) and to position of short circuit in destroyed winding. EFFECT: increased field of application. 5 cl, 7 dwg

Description

 The method relates to the electric power industry and can be used in three-phase high-voltage networks with resonant neutral grounding, mainly in networks with a large number of transformer and motor loads.

 There is a method of suppressing single-phase arc faults on earth (OZNZ) [1] which consists in measuring phase voltages and neutral bias voltage, recognizing OZNZ, recognizing a damaged phase, compensating the reactive component of the current of a single-phase fault using an arc suppression reactor and controlling the compensation of the capacitive component (CES) by projection of the voltage vector of the damaged phase onto a vector orthogonal to the neutral bias voltage vector.

 The disadvantage of this method is the inability to completely suppress the arc OZNZ, since it does not provide compensation for the active component of the current OZNZ.

 A known method [2] of suppressing arc OZNZ, which consists in recognizing the mode of OZNZ, recognition of the damaged phase and artificial short circuit (bypass) of the damaged phase to the ground.

Another method [3] uses instead of shunting the inclusion between the network neutral and the ground of the source with an EMF equal to the EMF of the damaged phase. Both of these methods lead to a short circuit in the event that an earth fault (ZN) does not occur in the phase conductor of the network, but in the winding of the motor or transformer (supply or load), and this only aggravates the accident if the voltage on the arc gap is greater than the voltage U _about arc break.

 Closest to the proposed is a method of suppressing arc faults between the phases of the network and ground [4] This method consists in measuring the instantaneous values of the phase voltage of the network and the neutral bias voltage, recognition of the arc OZZZ and the damaged phase, compensation of the capacitive component (by means of a Bauch transformer in which the reactor the circuit of an open triangle of the secondary windings is switched by a thyristor switch) and the active component (by means of an additional inductor switched by another thyristor lyuchom and being connected parallel to one of the phase windings of said transformer secondary Bauch) and managing these components compensated by projections neutral displacement voltage vector on the EMF vector of the faulty phase (for active component) and an orthogonal vector it emf between intact phases (for capacitive component). The prototype method is able to completely suppress the arc processes during short circuits between the phase conductors of the network and the ground (that is, OZNZ). It does not lead to short circuits in cases where the ZNZ occurred in the load windings or the transformers supplying the network.

 However, this method is ineffective with the above types of ZNZ. The reason for the disadvantage is the fact that the voltage across the arc gap in this case is determined not by the difference between the neutral bias voltage and the phase EMF of the network (as in the case of an OZNZ in a phase conductor), but by the difference between the neutral and EMF bias voltage acting between the point of the short-circuit current in the winding and neutral network. The indicated difference, if the ZNZ point is anyway significantly removed from the phase output of the damaged winding, is significantly different (in amplitude, and if the winding is connected to the line voltage of the network also in phase) from the phase EMF of the network. At the same time, the prototype method minimizes precisely the phase voltage (in that phase, which is considered damaged) and therefore can bring the voltage across the arc gap in the damaged winding to a significant value, at which self-elimination of the arc process may not be possible. If the dielectric strength of the insulation after the first breakdown in the winding has been restored due to the protective properties of the resonant neutral grounding, that the introduction of active component compensation (CAS) by the prototype method will lead to an increase in voltage between the damage site and the ground, which can cause new arc breakdowns.

 The purpose of the invention is the suppression of arc faults that occurred in the windings of motors or load transformers or supply transformers, regardless of how they are connected (star or delta) and the location of the fault point in a damaged winding.

 The goal is achieved in that in a method that includes measuring the instantaneous values of the phase voltage of the network and the neutral bias voltage, recognition of an arc fault to earth, compensation of the capacitive and active components of the current of a single-phase earth fault and automatic control of this compensation, the amplitude and phase values are additionally determined EMF acting between the point of earth fault and the neutral of the network, and automatically control the compensation of capacitive and active components in this way so that the amplitude and phase of the neutral bias voltage are maintained equal to the amplitude and phase of the EMF mentioned above, acting between the earth fault point and the network neutral.

напряжения смещения нейтрали в момент t₁ первого дугового пробоя, а фазу β этой ЭДС равной мгновенной фазе wt₁ момента первого дугового пробоя, взятой с противоположным знаком; если же мгновенное значение е(t₁) напряжения смещения нейтрали в момент первого дугового пробоя было отрицательным, то значение фазы β увеличивают на p..In addition, to determine the amplitude of E _o and phase β of the EMF acting between the earth fault point and the neutral of the network, arc breakdowns are detected at the earth fault point, the instantaneous phase wt is recorded for the moment of the first arc breakdown and the instantaneous voltage value e (t ₁ ) neutral displacements at this moment and further consider the amplitude E _{o of the} said EMF equal to the absolute value

neutral bias voltage at time t _{1 of the} first arc breakdown, and phase β of this EMF equal to the instantaneous phase wt _{1 of the} moment of the first arc breakdown, taken with the opposite sign; if the instantaneous value e (t ₁ ) of the neutral bias voltage at the time of the first arc breakdown was negative, then the value of phase β is increased by p ..

а если выявлено более одного дугового пробоя, то величины E_i и β определяют по следующим формулам:

где ω круговая частота сети, а момент t_i-1 предшествует моменту t_i,
в том случае, если величина w(t_i-1-t_i) не кратна π; в противном случае величины амплитуды Е_o и фазы b ЭДС, действующей между точкой замыкания на землю и нейтралью сети, считают неопределенными.In addition, to determine the amplitude of E _o and phase β of the EMF acting between the point of earth fault and the neutral of the network, arc breakdowns are detected at the point of earth fault, for each of them the instant phase wt _i and the instantaneous value e (t ₁ ) neutral bias voltage at the time t _{i of the} breakdown, and if only one (first) breakdown is detected, then the steps of claim 2 are performed, that is, the values of E _i and β are determined by the following formulas:

and if more than one arc breakdown is detected, then the values of E _i and β are determined by the following formulas:

where ω is the circular frequency of the network, and the moment t _i-1 precedes the moment t _i ,
if the quantity w (t _i-1 -t _i ) is not a multiple of π; otherwise, the magnitudes of the amplitude E _o and the phase b of the EMF acting between the point of earth fault and the neutral of the network are considered undefined.

 In addition, the values of the amplitude and phase of the EMF acting between the earth fault point and the neutral of the network, if one arc breakdown is detected, is determined according to (1) (that is, according to claim 2), and in that case if more than one arc breakdown is recorded as the average values of the corresponding values obtained according to (1) (i.e., according to claim 2) for the first arc breakdown and according to (2), (3) (i.e., according to claim 3 claims) for each subsequent arc breakdown.

отклонения φ значения фазы b ЭДС, действующей между точкой замыкания на землю и нейтралью сети, от значения a_i фазы фазной ЭДС не превышает π/3; а затем, если удовлетворяется условие

где E'_o корректируемое значение амплитуды ЭДС, действующей между точкой замыкания на землю и нейтралью сети;
E_m амплитуда фазной ЭДС сети,
то скорректированную величину Е_o амплитуды ЭДС, действующей между точкой замыкания на землю и нейтралью сети, полагают равной амплитуде Е_m фазной ЭДС, а скорректированную величину β фазы ЭДС, действующей между точкой замыкания на землю и нейтралью сети, считают равной фазе a_i упомянутой фазной ЭДС; в противном случае, т.е. если вышеуказанное условие не удовлетворяется, то проверяют выполнение дpугого условия:

и в том случае, если данное условие выполняется, то скорректированную величину E_o амплитуды ЭДС, действующей между точкой замыкания на землю и нейтралью сети, определяют по формуле:

а скорректированную величину β фазы ЭДС, действующей между точкой замыкания на землю и нейтралью сети, считают равной фазе a_i упомянутой фазной ЭДС; в противном же случае скорректированные величины E_o амплитуды и β фазы ЭДС, действующей между точкой замыкания на землю и нейтралью сети, определяют по следующим формулам:

Кроме того, для компенсации активной составляющей, осуществляемой при помощи тока искусственной несимметрии, фазу упомянутого тока устанавливают в соответствии со значением фазы ЭДС, действующей между точкой замыкания на землю и нейтралью сети.In addition, the values of the amplitude and phase of the EMF acting between the earth fault point and the neutral of the network, obtained as described above (that is, obtained according to paragraphs 2-4 of the claims), are corrected by relating the place of a single-phase fault to a winding connected to one from phase or line voltage of the network or to one of the phase conductors of the network, for which first determine the phase (A, B or C) of the network for which the absolute value

the deviation φ of the value of the phase b EMF, acting between the point of earth fault and the neutral of the network, from the value a _{i of the} phase phase EMF does not exceed π / 3; and then if the condition is satisfied

where E ' _{o is the} adjusted value of the amplitude of the EMF acting between the point of earth fault and the neutral of the network;
E _{m the} amplitude of the phase EMF network,
then the corrected value E _{o of the} amplitude of the EMF acting between the earth fault point and the neutral of the network is assumed to be equal to the amplitude E _{m of the} phase EMF, and the adjusted value β of the phase of the EMF acting between the earth fault point and the neutral of the network is considered equal to the phase a _{i of the} said phase Emf; otherwise, i.e. if the above condition is not satisfied, then check the fulfillment of another condition:

and if this condition is met, then the adjusted value of E _{o the} amplitude of the EMF acting between the point of earth fault and the neutral of the network, is determined by the formula:

and the adjusted value β of the phase of the EMF acting between the point of earth fault and the neutral of the network is considered equal to the phase a _{i of the} said phase EMF; otherwise, the adjusted values of E _{o the} amplitude and β phase of the EMF, acting between the point of earth fault and the neutral of the network, is determined by the following formulas:

In addition, to compensate for the active component, carried out using an artificial asymmetry current, the phase of the current is set in accordance with the value of the EMF phase acting between the point of earth fault and the neutral of the network.

 In addition, the compensation of the capacitive component is controlled by the deviation of the phase of the neutral bias voltage from the EMF phase, acting between the point of earth fault and the network neutral.

 In addition, the compensation of the capacitive component is controlled by the projection of the EMF vector acting between the point of earth fault and the neutral of the network, on a vector orthogonal to the neutral bias voltage.

 In addition, at each moment of time, the sign of the neutral bias voltage is determined and the compensation of the capacitive component is controlled by synchronously detecting the aforementioned sign with a reference signal orthogonal to the EMF acting between the earth fault point and the network neutral.

 In addition, to perform the operation of synchronously detecting the sign of the neutral voltage by the reference signal orthogonal to the EMF acting between the earth fault point and the network neutral, the instantaneous phases of the moments of the change in the sign of the neutral bias voltage are recorded and then, for each period of the reference signal, the product of the cosines of the differences mentioned instantaneous phases of moments of change in the sign of the voltage of the neutral bias and the phase of the EMF acting between the point of earth fault and the neutral of the network, on the voltage signs neutral displacements occurring immediately after the change of sign.

 In addition, the compensation of the active component is controlled by the deviation of the amplitude of the neutral bias voltage from the EMF amplitude acting between the earth fault point and the network neutral.

 In addition, the compensation of the active component is controlled by the deviation of the product of the EMF amplitude acting between the earth fault point and the network neutral, by the secant of the phase difference of the aforementioned EMF, acting between the earth fault point and the network neutral, and the artificial asymmetry current, from the product of the bias voltage amplitude neutral for the secant of the phase difference of the neutral bias voltage and the current of artificial asymmetry.

In FIG. 1 shows vector diagrams; figure 2 enlarged block diagram of the proposed device; in FIG. 3 correction of the parameters E _o and β EMF, acting between the point of the SC and the neutral of the network, by assigning it to one of the windings connected to the phase or linear voltage of the network; figure 4 is a diagram of the formation of controls for compensation of the capacitive component (CES) and the active component; Figures 5 and 6 are two versions of examples of the execution of the arc suppressing apparatus (DGA) and the compensator of the active component of the UAN (positions 21 and 22 in Fig. 2); Fig.7 is a functional diagram of an example implementation of the block (BUK) control compensation (position 23 in figure 2).

соответственно фазных ЭДС Е_A(t), E_B(t), E_C(t) фаз А, В и С сети; 4, 5, 6 векторы

соответственно линейных, ЭДС E_AB(t), E_BC(t), E_CA(t); 7 вектор

ЭДС Е(t), действующей между нейтралью N сети и точкой G замыкания на землю (ЗНЗ). Предполагается, что ЗНЗ произошло в секции обмотки нагрузки (соединенной в треугольник), включенной между фазами А и С сети. Кроме того, на фиг.1 обозначено: 8 пример вектора

напряжения e(t) смещения нейтрали, 9 пример вектора

напряжения U(t) между точкой G замыкания в нагрузке и землей О; 10, 11, 12 годографы векторов

напряжения е(t) смещения нейтрали при постоянной величине КАС, соответствующей полной компенсации токов ОЗНЗ в одной из фаз и при изменении величины КЕС для тех случаев, когда КАС вводится по аналогии с [4] и подготовлена к работе при ОЗНЗ, соответственно, в фазах А, В или С; 13, 14, 15 точки, разделяющие режимы КАС, осуществляемой по аналогии с [4] (под режимами подразумеваются фазы тока искусственной несимметрии); 16 пример вектора

17 годограф вектора 16 в случае реализации КАС при помощи отрицательного сопротивления в нейтрали сети при точной настройке КАС и расстройке КЕС.In FIG. 1 shows: 1,2,3 vectors

respectively phase EMF E _A (t), E _B (t), E _C (t) phases A, B and C of the network; 4, 5, 6 vectors

respectively linear, EMF E _AB (t), E _BC (t), E _CA (t); 7 vector

EMF E (t), acting between the neutral N of the network and the point G of the earth fault (ZNZ). It is assumed that the ZNZ occurred in the load winding section (connected in a triangle) connected between phases A and C of the network. In addition, figure 1 denotes: 8 example vector

neutral bias voltage e (t), 9 example vector

voltage U (t) between the point G of the circuit in the load and ground O; 10, 11, 12 hodographs of vectors

voltage e (t) of the neutral bias at a constant value of CAS corresponding to the complete compensation of the currents of OZNZ in one of the phases and when the value of CES is changed for those cases when the CAS is introduced by analogy with [4] and is prepared for operation with OZNZ, respectively, in phases A, B or C; 13, 14, 15 points separating the CAS modes, carried out by analogy with [4] (under the modes are meant phases of the current of artificial asymmetry); 16 example vector

17 hodograph of vector 16 in the case of the implementation of CAS using negative resistance in the neutral of the network with fine tuning of CAS and detuning CES.

фазных ЭДС E_A(t), E_B(t), E_C(t); 28, 29, 30 векторы

линейных ЭДС E_AB(t), E_BC(t), E_CA(t); 31-42 линии раздела зон отнесения ЗНЗ либо к одному из фазных проводников сети, либо к обмотке нагрузки или питающего сеть трансформатора, включенной на одно из фазных или на одно из линейных напряжений; 43 и 44 примеры нескорректированного

и скорректированного

векторов ЭДС Е(t), действующей между нейтралью сети и точкой ЗНЗ в обмотке, включенной на линейное напряжение; 45 и 46 аналогичные примеры при ЗНЗ в обмотке, включенной на фазное напряжение.In FIG. 2 shows: a three-phase network with phase EMFs E _A (t), E _B (t), E _C (t) and the total ground capacitance С С _A + C _B + C _C ; load (transformer or motor) with windings 18, 19, 20, connected, for example, in a triangle, and having ZNZ at point G of the winding 20; arc suppression apparatus 21 (DGA), which is, for example, a combination of a connecting transformer and an extinguishing reactor or a Bauch transformer with a thyristor key connected in series with a reactor in a secondary circuit (by analogy with [5] or [6]). 2 further indicated: a compensator 22 of the active component of the UAN, which, for example, is a single-phase dependent inverter connected in series with an arc suppression reactor [6] and playing the role of negative resistance in the neutral of the network, or an additional inductor connected in series with the thyristor key and connected by means of switching devices parallel to one of the phase secondary windings of the Bauch transformer [4] as well as the compensation control unit 23 (BPC) connected to the network via a three-phase KSR Control voltage transformer 24. Fig. 3 shows: 25-27 vectors

phase EMF E _A (t), E _B (t), E _C (t); 28, 29, 30 vectors

linear EMF E _AB (t), E _BC (t), E _CA (t); 31-42 dividing lines of zones of assignment of ZNZ to either one of the phase conductors of the network, or to the load winding or the supply network of the transformer connected to one of the phase or one of the line voltages; 43 and 44 examples of uncorrected

and adjusted

EMF vectors E (t), acting between the neutral of the network and the point of ZNZ in the winding connected to the line voltage; 45 and 46 are similar examples with ZNZ in a winding connected to phase voltage.

тока q(t) искусственной несимметрии для случая КАС при помощи включения регулируемой индуктивности параллельно обмотке фазы С трансформатора Бауха; 51-53 годографы вектора

напряжения смещения нейтрали при изменении величины КЕС в условиях постоянной (соответственно) недокомпенсации, точной компенсации и перекомпенсации активной составляющей для случая, когда вектор

ЭДС, действующей между точкой ЗНЗ и нейтралью сети, занимает положение 54; 55-57 векторы

напряжения е(t) смещения нейтрали при постоянной перекомпенсации емкостной составляющей и (соответственно) недокомпенсации, точной компенсации и перекомпенсации активной составляющей; 58 вектор

, ортогональный векторам

55-57; 59 проекция вектора 54

на вектор 58

; 60 вектор

напряжения е(t) смещения нейтрали при недокомпенсации емкостной составляющей и точной компенсации активной составляющей; 61 вектор

, ортогональный вектору

60; 62 проекция вектора 54

на вектор 61

.Figure 4 shows: 47-49 phase EMF, respectively, phases A, B and C; 50 - vector

current q (t) of artificial asymmetry for the case of CAS by switching on an adjustable inductance parallel to the phase C winding of a Bauch transformer; 51-53 hodographs of the vector

neutral bias voltage when changing the value of CES under conditions of constant (respectively) undercompensation, accurate compensation and overcompensation of the active component for the case when the vector

EMF, acting between the point of ZNZ and the neutral of the network, occupies position 54; 55-57 vectors

the neutral bias voltage e (t) with constant over-compensation of the capacitive component and (respectively) under-compensation, accurate compensation and over-compensation of the active component; 58 vector

orthogonal to vectors

55-57; 59 projection of vector 54

on vector 58

; 60 vector

voltage e (t) of the neutral bias during the undercompensation of the capacitive component and accurate compensation of the active component; 61 vector

orthogonal to the vector

60; 62 vector projection 54

by vector 61

.

 In FIG. 5 and 6 are indicated: 63 Bauch transformer with reactor 64 in the open-loop circuit of the secondary windings 65-67; thyristor switch TK1 68 with a sensor DZ1 69 for locking its thyristors, connected in series with the reactor 64; additional inductor 70; thyristor switch TK2 71 with a sensor DZ2 72 for locking its thyristors, connected in series with an additional inductor 70 and connected to a switching device 73. This switching device 73 in FIG. 5 is a signal "P" mode "block BUK (position 23 in figure 2 ) connects a circuit of elements 70-72 parallel to one of the phase secondary windings 65-67 of the Bauch transformer 63, depending on the required phase of the current of artificial asymmetry. Similar to the purpose of the switching element 73 in Fig.6 on the signal "p" "mode" unit BUK connects a circuit of elements 70-72 to one of the linear (low voltage, up to 1140 V) supply voltages applied between points A ', B', C '(Fig.6), in phase with the primary line voltage applied between points A, B, C. Points A, B and C in Figures 5 and 6 are connected to phases A, B and C of the network (Figure 2) . Elements 63, 68, 69 in FIG. 5 and 6 form an arc suppression apparatus DGA (position 21 in figure 2), and elements 70-73 compensator of the active component of the CAS (position 22 in figure 2).

Figure 7 shows: 74 a processor with read-only memory and RAM devices; 75 bus data, addresses and control signals; 76-78 programmable timers T1, T2, T3 with counting inputs "c" and triggering inputs "E"; 79 interrupt controller KP; 80 input (receiving information) port; 81 output (output information) port; 82 - zero comparator of the line voltage signal; 83, 84 pulse shapers (F1 and F2), eliminating the bounce, respectively, of the spod and the front of the output signal of the comparator 82; 85 arc breakdown sensor; 86 biased comparator (with a bias value corresponding, for example, 20% of the amplitude of the phase EMF network). 7 also shows a circuit 87 of converting into a time interval the instantaneous value of the neutral bias voltage e (t ₀ + δ) at time t ₀ + δ (where t _{o is the} moment of a sudden change in the neutral bias voltage, for example, due to an arc breakdown at the ZNZ site , and δ is a small quantity). Circuit 87 consists of an integrator formed by an operational amplifier 88 and a capacitor 89, an analog switch 90, resistors 91 and 92, which turn the integrator into an inertial unit with a small time constant (with the switch position 90 opposite to that shown in Fig. 7), a conversion circuit resistor 93 the initial conditions of the integrator are the time interval and relay link 94 with a small (about 10-30 mV) hysteresis. In addition, figure 7 shows the logic element 95 "VARIABILITY", a neutral neutral bias voltage comparator 96 and pulse shapers 97 and 98 (FI), which supply, respectively, short pulses to the inputs Ir7 and Ir4 of the interrupt controller 79 each time the signal changes , respectively, at the outputs of the comparators 96 and 86.

 The method is as follows.

In the event of the occurrence (in the phase conductor of the network, in the load winding or in the winding of the supply transformer) of an arc ZNZ (for example, in the load winding 18, Fig. 2) during the existence of an ionized conducting gap between the point G and the ground, the potential of the point G becomes close to earth potential due to the small voltage drop across the arc gap. The voltage e (t) of the neutral bias N at this moment becomes equal to the total EMF E (t) acting between the point G of the SC and the neutral N of the network (in Fig. 2 this, for example, is the sum of the phase EMF E _A (t) and the counter-EMF in the winding 20 between points A and G). After the overcharge of the capacitors C _A , C _B , C _C caused by the arc breakdown, the current at the ZNZ site, as a rule, significantly decreases, and the arc breaks. Further, in the circuit of the zero sequence of the network (KNPS), an oscillatory transient takes place with the natural frequency w _{C of the} circuit under the initial conditions corresponding to t _{o the} extinction of the arc, that is, e (t _o ) = E (t _o ). If the condition for the resonant adjustment of the KNPS ω _C = ω is fulfilled, then the neutral bias voltage e (t) compensates for the total EMF E (t) acting between the point G of the SC and the neutral N of the network, so that the voltage U (t) between the point G and ground is initially It turns out to be close to zero and slowly increases with the decay of free oscillations in the CNPS (if CAS is not carried out). When the voltage U (t) between the point G and the ground reaches the breakdown voltage of the arc gap (which will probably require several periods to several tens of periods of the network frequency after the first arc breakdown), a new arc breakdown occurs and the process repeats. By introducing the UAN into the SSC and thus compensating for the energy loss of free vibrations, it is possible to prevent their attenuation [4, 6] If, in this case, the neutral bias voltage e (t) has a frequency of the network and is maintained equal in amplitude and phase to the EMF E (t), acting between the point G of the SCZ and the neutral of the N network, the first harmonic of the voltage U (t) between the point G of the SCZ and the ground will be zero and the arc process at the site of the SCZ will stop.

суммарной ЭДС E(t), действующей между точкой G ЗНЗ и нейтралью N сети. В случае, если обмотка с повреждением включена на линейное напряжение, конец этого вектора лежит на прямой, соединяющей точки С и А, А и В, В и С, если повреждение находится в обмотках СА, АВ или ВС соответственно (для фиг.1 это прямая СА). Если же указанная обмотка включена на фазное напряжение, то конец вектора

лежит на прямой, соединяющей точку N с точками А, В или С на фиг. 2, в зависимости от того, в какой из фазных обмоток произошло повреждение. Вектор

напряжения U(t) между точкой G замыкания в нагрузке и землей О (фиг.2), равный разности векторов

и

(напряжения е(t) смещения нейтрали), обращается в нуль только в том случае, если векторы

и

равны между собой, то есть, когда напряжение е(t) смещения нейтрали и ЭДС, действующая между точкой G ЗНЗ и нейтралью N сети, совпадают по амплитуде и фазе. Регулировать амплитуду и фазу напряжения e(t) смещения нейтрали возможно путем соответствующего управления КАС и КЕС.The aforesaid can be explained on the vector diagram (figure 1). A short circuit at point G of the load winding connected in a triangle corresponds to a vector

the total EMF E (t), acting between the point G ZNZ and the neutral N of the network. If the winding with damage is switched on to a linear voltage, the end of this vector lies on a straight line connecting points C and A, A and B, B and C, if the damage is in the windings CA, AB or BC, respectively (for figure 1 this direct CA). If the specified winding is turned on for phase voltage, then the end of the vector

lies on a line connecting point N with points A, B or C in FIG. 2, depending on which phase winding has been damaged. Vector

voltage U (t) between the point G of the circuit in the load and ground O (figure 2), equal to the difference of the vectors

and

(voltage e (t) of the neutral bias), vanishes only if the vectors

and

are equal to each other, that is, when the voltage e (t) of the neutral bias and the EMF between the point G ZNZ and the neutral N of the network coincide in amplitude and phase. It is possible to adjust the amplitude and phase of the neutral bias voltage e (t) by appropriate control of CAS and CES.

 Thus, in order to minimize and then hold the voltage U (t) between the damage point G and the ground close to zero after the occurrence of an arc fault, it is necessary, having recognized the arc fault, first, to determine the amplitude and phase of the emf E (t) (which depend on the connection scheme of the damaged winding and on the location of the fault in the winding) and, secondly, automatically control the CAS and CES so that the neutral bias voltage and EMF E (t) acting between point G ZNZ and neutral N of a network, coincidence Ali in amplitude and phase.

. Иными словами, для величины E_o и β справедливы соотношения:

Из последнего уравнения находим:
β = -ωt₀+0,5π[1-sign e(t₀)],
считая, что мгновенная фаза ωt₀ приведена к отрезку [0,2π].. Таким образом, для нахождения величин E_o и β достаточно произвести действия, описанные в п.2 формулы изобретения.To solve the first problem, it is proposed to take advantage of the fact that the instantaneous voltage e (t), shortly after the arc breakdown, becomes almost equal to the instantaneous EMF value E (t) at the same time (due to the small voltage drop across the ionized arc gap) . By fixing this voltage, as well as the moment of time t _o when the first arc breakdown occurred, it is possible to determine the amplitude E _o and phase β EMF
E (t) = E ₀ cos (ωt + β), (4)
acting between the point G of the ZNZ and the neutral of the N network, based on the most probable assumption that the breakdown occurred at the maximum of the absolute value of the EMF E (t), which occurs when

. In other words, for the values of E _o and β the following relations are true:

From the last equation we find:
β = -ωt ₀ + 0.5π [1-sign e (t ₀ )],
assuming that the instantaneous phase ωt _{0 is} reduced to the interval [0.2π] .. Thus, to find the values of E _o and β it is enough to perform the steps described in paragraph 2 of the claims.

которая получена подстановкой величин t_i-1, t_i, е(t_i-1) и е(t_i) в (4). Данная система уравнений совместна в случае, если
sin(ωt_i-1-ωt_i) ≠ 0,
то есть, если величина ω(t_i-1-t_i) не кратна π.. Ее решениями являются следующие выражения:

где ω- круговая частота сети.To recognize the fact of the occurrence of arc ZNZ and determine the moments of arc breakdowns, you can use one of the known methods, for example, filtering the neutral bias voltage (in order to remove or attenuate components at frequencies close to the industrial one) and comparing the filtering result with some given threshold [7]
If more than one arc breakdown has been registered since the occurrence of a fault, it is possible to increase the reliability of determining the EMF parameters E (t) by obtaining additional information from these breakdowns and refusing the assumption that breakdowns are involved in extreme values of E (t). To extract additional information about the EMF parameters E (t), it is necessary to perform the steps described in paragraph 3 of the claims, the meaning of which is as follows. If we fix the instantaneous phases wt _i-1 , ωt _i and the instantaneous values e (t _i-1 ), e (t _i ) of the neutral bias voltage e (t) for two arc breakdowns that occurred at time moments t _i-1 , t ₁ (t _{i-1 is} preceded by t _i ), then the parameters E _o and β can be found from the following system of equations:

which is obtained by substituting the quantities t _i-1 , t _i , е (t _i-1 ) and е (t _i ) in (4). This system of equations is compatible if
sin (ωt _i-1 -ωt _i ) ≠ 0,
that is, if the quantity ω (t _i-1 -t _i ) is not a multiple of π .. Its solutions are the following expressions:

where ω is the circular frequency of the network.

Expressions (7), (8) coincide with the expressions given in paragraph 3 of the claims. Note that for the implementation of this paragraph it is enough to have the values of ωt _i and е (t _i ) only for the last (at a given time) and the breakdowns preceding it, and therefore it is enough to fix and remember only 4 quantities: ωt _i-1 , ωt _i , e (t _i-1 ) and e (t _i ).

If during ZNZ there was a series of several arc breakdowns, then the values of E _o and ω can be determined repeatedly by following steps 3 of the claims for each pair of breakdowns (for example, for the 1st and 2nd, for the 2nd and 3rd, for 3rd and 4th, etc.). In paragraph 4 of the claims, it is proposed to use this information in order to increase the accuracy of determining the parameters E (t) by determining their values obtained by processing measurements made on each of the arc breakdowns.

When calculating the average values, it is also advisable to use the values of E _o and b obtained after the first breakdown according to claim 2. Averaging can be carried out in various ways, of which the simplest and most effective are the determination of the arithmetic average of all arc breakdowns that have occurred at a given point in time, or the determination of a moving average from a certain predetermined number of arc breakdowns.

ЭДС Е(t) (фиг. 1) должен располагаться либо на одной из прямых АВ, ВС или СА, если ЗНЗ произошло в обмотке, включенной на линейное напряжение (то есть в одной из обмоток, соединенных в "треугольник"), либо на одной из прямых NA, NB или NC, если ЗНЗ произошло в обмотке, включенной на фазное напряжение (то есть в одной из обмоток, соединенных в "звезду", п.5 формулы изобретения). Данное требование объясняется тем обстоятельством, что ЭДС Е(t) всегда формируется суммированием одной из фазных ЭДС E_A(t), E_B(t), E_C(t), сети (фиг.2) и (возможно) противоЭДС в обмотке нагрузки. ПротивоЭДС в любой точке обмотки пропорциональна (и, следовательно, синфазна) одной из линейных ЭДС сети, если обмотки соединены в "треугольник". Если же обмотки соединены в "звезду", то противоЭДС пропорциональна (и поэтому синфазна) одной из фазных ЭДС сети (в наиболее распространенном случае симметричной нагрузки). Привлечение указанной априорной информации означает, что, получив (в результате выполнения п.п.1-4 формулы изобретения) величины Е_o и β, характеризующие вектор

, следует скорректировать данные величины, отнеся конец вектора

к ближайшему из отрезков прямых AN, BN, CN, AB, BC или CA (фиг.1). Для этого необходимо: во-первых, определить упомянутую ближайшую прямую; а во-вторых, скорректировать положение вектора

спроецировав на нее конец указанного вектора. Решение перечисленных задач иллюстрируется на фиг.3, на котором векторная диаграмма фазных и линейных ЭДС разбита отрезками 31-42 на зоны максимальной близости конца вектора

ЭДС, параметры которой определены согласно п.п.1-4 формулы изобретения, к одному из перечисленных выше отрезков AN, BN, CN, AB, BC, CA. Так, положение нескорректированного вектора

в позиции 43 на фиг.3 означает, что ближайшим к нему является отрезок СА, следовательно, ЗНЗ произошло, по-видимому, в обмотке, включенной на линейное напряжение между фазами С и А, и проецировать вектор

следует на прямую СА. Положение же нескорректированного вектора

в позиции 45 на фиг.3 означает, что ближайшим к нему является отрезок BN, и следовательно, ЗНЗ произошло в обмотке, включенной на фазное напряжение фазы В. В этом случае вектор

необходимо проецировать на прямую BN.An additional increase in the accuracy of determining the parameters of the indicated EMF is achieved by using a priori information that the end of the vector

EMF E (t) (Fig. 1) should be located either on one of the straight lines AB, BC or CA, if the ZZZ occurred in a winding connected to a linear voltage (that is, in one of the windings connected in a "triangle"), or one of the direct NA, NB or NC, if the ZNZ occurred in the winding connected to the phase voltage (that is, in one of the windings connected to the "star", paragraph 5 of the claims). This requirement is explained by the fact that the EMF E (t) is always formed by summing one of the phase EMFs E _A (t), E _B (t), E _C (t), the network (Fig. 2) and (possibly) the counter-EMF in the winding load. The back-EMF at any point of the winding is proportional (and, therefore, in phase) to one of the linear EMF of the network, if the windings are connected in a "triangle". If the windings are connected in a "star", then the back-EMF is proportional (and therefore in-phase) to one of the phase EMF networks (in the most common case of a symmetrical load). The involvement of the specified a priori information means that, having received (as a result of paragraphs 1-4 of the claims) the values of E _o and β characterizing the vector

, you should adjust these values, referring the end of the vector

to the nearest line segment AN, BN, CN, AB, BC or CA (Fig. 1). To do this, it is necessary: firstly, to determine the mentioned nearest line; and secondly, adjust the position of the vector

projecting the end of the specified vector onto it. The solution of these problems is illustrated in figure 3, in which the vector diagram of the phase and linear EMF is divided by segments 31-42 into zones of maximum proximity to the end of the vector

EMF, the parameters of which are determined according to claims 1 to 4 of the claims, to one of the above segments AN, BN, CN, AB, BC, CA. So, the position of the uncorrected vector

in position 43 in figure 3 means that the segment CA is closest to it, therefore, the ZNZ occurred, apparently, in the winding connected to the linear voltage between phases C and A, and to project the vector

follows on direct CA. The position of the uncorrected vector

at position 45 in figure 3 means that the closest to it is the segment BN, and therefore, the ZNZ occurred in the winding connected to the phase voltage of phase B. In this case, the vector

need to project onto direct BN.

попадает в один из незаштрихованных секторов, ограниченных на фиг.3 линиями 40, 41 или 42, то его невозможно спроецировать ни на один из отрезков АВ, ВС или СА, то есть ЗНЗ в этом случае невозможно отнести ни к одной из обмоток, соединенных в "треугольник". Наиболее вероятным в данном случае будет предположение о том, что ЗНЗ произошло в фазном проводнике сети (фазы А, В или С для секторов, ограниченных на фиг.3 линиями, соответственно, 40, 41 или 42), и следовательно, вектор

должен совпадать с вектором

соответствующей фазной ЭДС.The dividing lines 31-36 are the bisectors of the angles, respectively, CAN, BAN, ABN, CBN, BCN and ACN, and the dividing lines 37-39 are the bisectors of the angles, respectively, CNA, ANB and BNC and therefore form angles with the vectors of the nearest phase EMF, equal, respectively, 15 and 60 ^o . Lines 40-42 are perpendicular to the vectors of the corresponding phase EMF. If the end of the vector

falls into one of the unshaded sectors, limited by lines 40, 41 or 42 in FIG. 3, then it cannot be projected onto any of the segments AB, BC or CA, that is, ZNZ in this case cannot be attributed to any of the windings connected to "triangle". The most probable in this case will be the assumption that the ZNZ occurred in the phase conductor of the network (phases A, B or C for the sectors limited by lines in FIG. 3, respectively, 40, 41 or 42), and therefore, the vector

must match the vector

corresponding phase EMF.

) из вышеперечисленных отрезков. В связи с симметричным расположением линий раздела на векторной диаграмме (фиг. 3) целесообразно определить сначала, в каком из трех равнозначных секторов MNP, PNK или KNM (разграничиваемых линиями раздела 37, 38 и 39) находится проецируемый вектор

. Данным сектором, как видно из фиг.3, оказывается тот из них, в котором лежит вектор

или

(позиции 25, 26 и 27 на фиг. 3), для которого угол ψ между вектором

и вектором фазной ЭДС

или

не превышает (по абсолютной величине) 60^o (что равносильно условию определения фазы, сформированному в п.5 формулы изобретения). Так, например, если вектор

занимает положение 43 (фиг.3), то его следует отнести к сектору MNP, в котором лежит вектор

(25) фазной ЭДС фазы А, а если вектор

занимает положение 45 (фиг.3), то его следует отнести к сектору PNK, в котором лежит вектор E_B (26) фазной ЭДС фазы В. В пределах каждого сектора положение вектора

однозначно определяется углом φ между эти вектором и вектором соответствующей фазной ЭДС

(или, что то же самое, отклонением Φ значения фазы b ЭДС, действующей между точкой ЗНЗ и нейтралью сети, от значения фазы a_i соответствующей фазной ЭДС). Угол φ лежит в диапазоне [-60^o, 60^o] что существенно упрощает дальнейшие рассуждения.Let us consider in more detail the actions that should be taken to recognize the nearest (to the end of the vector

) from the above segments. In connection with the symmetrical arrangement of the dividing lines in the vector diagram (Fig. 3), it is advisable to first determine in which of the three equivalent sectors MNP, PNK or KNM (delimited by dividing lines 37, 38 and 39) the projected vector

. This sector, as can be seen from figure 3, is the one in which the vector lies

or

(positions 25, 26 and 27 in Fig. 3), for which the angle ψ between the vector

and phase emf vector

or

does not exceed (in absolute value) 60 ^o (which is equivalent to the condition for determining the phase formed in paragraph 5 of the claims). So, for example, if the vector

occupies position 43 (figure 3), it should be attributed to the sector MNP, in which the vector

(25) phase EMF phase A, and if the vector

occupies position 45 (figure 3), it should be attributed to the PNK sector, in which lies the vector E _B (26) of the phase EMF of phase B. Within each sector, the position of the vector

uniquely determined by the angle φ between these vectors and the vector of the corresponding phase EMF

(or, which is the same thing, by the deviation Φ of the value of the phase b EMF, acting between the point of the SC and the neutral of the network, from the value of phase a _i corresponding to the phase EMF). The angle φ lies in the range [-60 ^o , 60 ^o ] which greatly simplifies further considerations.

к одной из упомянутых прямых (AN, BN, CN, AB, BC или CA на фиг.3) следует отнести вектор

(если это возможно) или же к какому из фазных проводников (А, В или С) следует отнести ОЗНЗ, если отнесение к упомянутым прямым на фиг.3 невозможно (то есть, если конец вектора

попадает в один из незаштрихованных секторов, ограниченных на фиг.3 линиями 40, 41 или 42). Благодаря симметрии секторов MNP, PNK и KNM относительно векторов фазных ЭДС, соответственно,

(25),

(26) и

(27), уравнения линий 40-42 и 31-36 (в полярной системе координат с полюсом в точке N) имеют следующий вид:

и

где радиус-вектор М и полярный угол λ полярные координаты точек линий раздела 31-36, 40-42. Поэтому, если удовлетворяются условия:

где

амплитуда корректируемой ЭДС

(t),
то вектор

находится в зоне максимальной близости к отрезкам AN, BN или CN и, следовательно, ЗНЗ произошло в обмотке, включенной на фазное напряжение (то есть в одной из обмоток, соединенных в "звезду"). Если не выполняется условие (9) и выполняется условие (10), то вектор

находится в зоне максимальной близости к отрезкам АВ, ВС или СА, причем конкретно зона определяется знаком угла Φ: в cекторе MNP v > 0 означает зону СА, а Φ < 0 зону АВ; в секторе PNK Φ > 0 означает зону АВ, Φ < 0 зону ВС; в секторе KNM Φ > 0 означает зону ВС, Φ < 0 зону СА (фиг.3). Таким образом, если не выполняется условие (9) и выполняется условие (10), то можно считать, что ЗНЗ произошло в обмотке, включенной на линейное напряжение (то есть в одной из обмоток, соединенных в "треугольник"). Если же не выполняется условие (10), то, как указывалось выше, ЗНЗ следует отнести к фазному проводнику А, В или С (в зависимости от выбранного сектора MNP, PNK или KNM соответственно).Next, it is necessary to determine (within the selected sector) which of the zones of maximum proximity of the end of the vector

one of the mentioned lines (AN, BN, CN, AB, BC or CA in figure 3) should include the vector

(if possible) or to which of the phase conductors (A, B or C) should be attributed to OZNZ, if the assignment to the mentioned direct in Fig.3 is impossible (that is, if the end of the vector

falls into one of the unshaded sectors, limited in Fig. 3 by lines 40, 41 or 42). Due to the symmetry of the MNP, PNK, and KNM sectors with respect to the phase EMF vectors, respectively,

(25)

(26) and

(27), the equations of lines 40-42 and 31-36 (in the polar coordinate system with a pole at point N) have the following form:

and

where the radius vector M and the polar angle λ are the polar coordinates of the points of the dividing lines 31-36, 40-42. Therefore, if the conditions are satisfied:

Where

amplitude of the corrected EMF

(t)
then vector

It is located in the zone of maximum proximity to the segments AN, BN or CN and, therefore, the short-circuit fault occurred in the winding connected to the phase voltage (that is, in one of the windings connected to the "star"). If condition (9) is not satisfied and condition (10) is satisfied, then the vector

is in the zone of maximum proximity to the segments AB, BC or CA, and more specifically, the zone is determined by the sign of the angle Φ: in the sector MNP v> 0 means the zone CA, and Φ <0 the zone AB; in the PNK sector, Φ> 0 means the zone AB, Φ <0 the zone BC; in the KNM sector, Φ> 0 means the BC zone, Φ <0 the CA zone (Fig. 3). Thus, if condition (9) is not satisfied and condition (10) is fulfilled, then it can be considered that the short-circuit protection device occurred in a winding connected to a linear voltage (that is, in one of the windings connected in a "triangle"). If condition (10) is not fulfilled, then, as indicated above, the ZNZ should be assigned to the phase conductor A, B, or C (depending on the selected sector, MNP, PNK, or KNM, respectively).

корректируют, спроецировав данный вектор на вектор

или

соответствующего сектора. При этом скорректированную величину E_o амплитуды данной ЭДС определяют по формуле:

а скорректированную величину фазы β данной ЭДС считают равной фазе a_i этой же фазной ЭДС, как и указано в п.5 формулы изобретения. Во втором случае (когда условие (9) не выполняется, а условие (10) выполняется и, следовательно, ЗНЗ произошло в обмотке, включенной на линейное напряжение, параметры вектора

корректируют, спроецировав данный вектор на соответствующую прямую АВ, ВС или СА (фиг.3) соответствующего сектора. При этом, благодаря симметрии перечисленных прямых относительно векторов

фазных ЭДС, выражение для скорректированной величиной Е_o амплитуды данной ЭДС не зависит от того, на какую именно прямую (из перечисленных выше) в данном секторе производится проецирование. Данное выражение имеет следующий вид:

Кроме того, в связи с упомянутой симметрией, выражения для скорректированной величины фазы β,, полученные для случаев проецирования на каждую из двух прямых АВ, ВС или СА, относящихся к данному сектору, совпадают с точностью до знака φ:

В третьем случае (когда условие (10) не выполняется и, следовательно, имеет место ОЗНЗ в фазе А, В или С) амплитуду Е_o и фазу β следует принять равными амплитуде E_m и фазе a_i соответствующей фазной ЭДС сети, что и отражено в п.5 формулы изобретения.In the first case (when conditions (9) and (10) are satisfied and, therefore, the short-circuit fault occurred in the winding connected to the phase voltage, the parameters of the vector

correct by projecting this vector onto a vector

or

relevant sector. In this case, the adjusted value of E _{o the} amplitude of this EMF is determined by the formula:

and the adjusted value of phase β of this EMF is considered equal to phase a _{i of} the same phase EMF, as indicated in paragraph 5 of the claims. In the second case (when condition (9) is not satisfied, and condition (10) is satisfied, and therefore, the short-circuit protection occurs in a winding connected to a linear voltage, the parameters of the vector

correct by projecting this vector on the corresponding line AB, BC or CA (figure 3) of the corresponding sector. Moreover, due to the symmetry of these lines with respect to vectors

phase EMF, the expression for the adjusted value of E _{o the} amplitude of this EMF does not depend on which line (from the above) in this sector is projected. This expression has the following form:

In addition, due to the mentioned symmetry, the expressions for the adjusted value of the phase β obtained for the cases of projecting onto each of the two straight lines AB, BC or CA related to this sector coincide up to the sign of φ:

In the third case (when condition (10) is not satisfied and, therefore, there is an SCD in phase A, B or C), the amplitude E _o and phase β should be taken equal to the amplitude E _m and phase a _{i of the} corresponding phase EMF network, which is reflected in claim 5 of the claims.

(16 на фиг.1) по окружности 17 (фиг.1), а управление КАС радиус этой окружности. Очевидно, что каких-либо изменений режима КАС в зависимости от местоположения точки G ОЗЗ в данном случае не требуется, однако необходимо учитывать интегральный характер связи управления по КЕС с фазой е(t) (так как фаза напряжения е(t) является интегралом его частоты). Другим способом введения КАС, который можно использовать в составе предлагаемого изобретения, является введение в КНПС регулируемого по амплитуде тока естественной несимметрии, например, путем подключения регулируемой (углом отпирания тиристорного ключа) индуктивности параллельно одной из фазных вторичных обмоток трансформатора Бауха [4] то есть введением регулируемой индуктивности в одну из фаз сети. В этом случае регулирование КЕС приводит к перемещению вектора

напряжения е(t) смещения нейтрали по одной из окружностей 10, 11 или 12 в зависимости от того, параллельно какой именно из вторичных обмоток трансформатора Бауха включена регулируемая индуктивность. Управление же КАС приводит к изменению диаметра соответствующей окружности 10, 11 или 12 (фиг.1). Очевидно, что при данном способе введения КАС местоположение точки G должно определять режим КАС в отношении выбора фазы вводимой искусственной несимметрии, так как диапазон фазовых сдвигов, привносимых КНПС в фазу напряжения е(t) смещения нейтрали при различных расстройствах КЕС, здесь принципиально ограничен интервалом (-90^o, 90^o) (п.2 формулы изобретения).The solution to the second problem (which is reduced to the automatic control of CAS and CES in such a way that the neutral bias voltage e (t) coincides in amplitude and phase with the EMF E (t) acting between the GZ point GZ and the network neutral (Fig. 2), parameters E _o and β of which were determined in accordance with the foregoing) is determined by the selected method of compensation of the active component. If for KAS the introduction of negative resistance into the KNPS is used, for example, in the form of a single-phase dependent inverter connected in series with an arc suppression reactor [6] or in the form of parametric pumping of energy in the KNPS [8], then the control of the KES determines the frequency of self-oscillations in the KNPS, that is, vector diagram language vector rotation speed

(16 in FIG. 1) around the circumference 17 (FIG. 1), and the CAS control is the radius of this circle. Obviously, in this case, no changes in the UAN mode depending on the location of the G point G of the OZZ are required, however, it is necessary to take into account the integral nature of the connection between the CES control and phase e (t) (since the voltage phase e (t) is an integral of its frequency ) Another way of introducing CAS, which can be used as part of the present invention, is to introduce natural asymmetry in amplitude regulated current in the KNPS, for example, by connecting an adjustable (unlocking angle of the thyristor key) inductance parallel to one of the phase secondary windings of a Bauch transformer [4] that is, by introducing adjustable inductance in one of the phases of the network. In this case, the regulation of the CES leads to the displacement of the vector

voltage e (t) of the neutral bias along one of the circles 10, 11 or 12, depending on which parallel from which secondary windings of the Bauch transformer an adjustable inductance is connected. The control of CAS leads to a change in the diameter of the corresponding circle 10, 11 or 12 (figure 1). Obviously, with this method of introducing CAS, the location of the G point should determine the CAS mode with respect to the choice of the phase of the introduced artificial asymmetry, since the range of phase shifts introduced by the FSCC into the phase of the neutral bias voltage e (t) for various CES disorders is fundamentally limited by the interval ( -90 ^o , 90 ^o ) (claim 2).

8 напряжения е(t) смещения нейтрали (фиг.1) будет располагаться практически в любой точке круга, ограниченного окружностью 10, проходящей через начало N координат. Центр О этого круга лежит на пересечении вектора

тока искусственной несимметрии (который отстает на 90^o от фазной ЭДС фазы С) с отрезком 6 СА, изображающим вектор линейного напряжения E_CA. Если предположить теперь, что регулируемая индуктивность включена параллельно вторичной обмотке фазы В трансформатора Бауха, то вектор

будет располагаться в круге, ограниченном окружностью 12 на фиг.1. Исходя из сказанного и приписав (условно) нулевое значение фазы линейной ЭДС E_BC(t) (на фиг.1 ей соответствует вектор 5), приходим к выводу, что для значений фазы β ЭДС E(t), действующей между точкой G ЗНЗ и нейтралью N сети, лежащих в интервале (0^o, 120^o) (окружность 12 на фиг. 1), необходимо включать регулируемую индуктивность параллельно вторичной обмотке фазы В трансформатора Бауха для значений фазы b, лежащих в интервале (120^o, 240^o) (окружность 11 на фиг.1), необходимо включать регулируемую индуктивность параллельно вторичной обмотке фазы А трансформатора Бауха, а для значений фазы β, лежащих в интервале (240^o, 360^o (окружность 10 на фиг.1), необходимой включать регулируемую индуктивность параллельно вторичной обмотке фазы С указанного трансформатора. Из того факта, что отрезки NA, BN, CN, AB, DC и CA, на которых может располагаться конец вектора

при всевозможных локализациях ЗНЗ в обмотках, целиком находятся в кругах, ограниченных окружностями 10, 11 и 12 (фиг.1), следует также, что для подавления ЗНЗ в любой точке обмоток нагрузок и питающих сеть трансформаторов требуется величина КАС, не превышающая той ее величины, которая необходима для подавления дуговых замыканий между фазными проводниками сети и землей. Поэтому для реализации заявляемого способа можно использовать исполнительные органы компенсаторов активной составляющей, реализующих известные способы [4] [6]
Рассмотрим далее предлагаемые способы автоматической настройки КАС и КЕС при подавлении дуговых процессов в обмотках нагрузки и питающих сеть трансформаторов. Каждый из указанных способов в качестве необходимого свойства должен обеспечить соответствие знака управления по регулируемой (емкостной или активной) составляющей компенсации знаку расстройки сети по этой составляющей в условиях точной настройки по другой (соответственно, активной или емкостной) составляющей, и нулевые значения управлений по обеим составляющим в условиях полной компенсации токов ЗНЗ. Желательна также (для улучшения динамических характеристик процессов автонастройки) взаимная независимость управлений, то есть чтобы управление по емкостной составляющей не зависело от расстройки по активной составляющей и наоборот.If we conditionally limit the CAS to the value necessary for the exact compensation of the current in the given arrester, and assume that the adjustable inductance is connected in parallel with the secondary winding of phase B of the Bauch transformer, then by controlling the CAS and CEC, we can achieve the end of the vector

8, the neutral bias voltage e (t) (Fig. 1) will be located at almost any point in the circle bounded by a circle 10 passing through the origin N of coordinates. The center O of this circle lies at the intersection of the vector

the current of artificial asymmetry (which is 90 ^° behind the phase EMF of phase C) with a segment of 6 CA, depicting the linear voltage vector E _CA. If we now assume that the adjustable inductance is connected in parallel with the secondary winding of phase B of the Bauch transformer, then the vector

will be located in a circle bounded by a circle 12 in figure 1. Based on the foregoing and attributing (conditionally) a zero value of the phase of the linear EMF E _BC (t) (vector 5 corresponds to it in Fig. 1), we conclude that for the values of the phase β of the EMF E (t) acting between the point G ZNZ and neutral N of the network lying in the interval (0 ^o , 120 ^o ) (circle 12 in Fig. 1), it is necessary to include an adjustable inductance parallel to the secondary winding of phase B of the Bauch transformer for phase b values lying in the interval (120 ^o , 240 ^o ) ( circle 11 in figure 1), it is necessary to include an adjustable inductance parallel to the secondary winding of phase A transformer pa Bauch, and for β phase values lying in the range (240 ^o, 360 ^o (circle 10 in Figure 1), necessary to include an adjustable inductance across the secondary phase winding of said transformer. From the fact that the segments NA, BN, CN , AB, DC and CA, on which the end of the vector can be located

with all kinds of ZNZ localizations in the windings, they are entirely in circles bounded by circles 10, 11 and 12 (Fig. 1), it also follows that in order to suppress the ZNZ at any point of the load windings and the transformers supplying the network, the KAS value does not exceed that value , which is necessary to suppress arc faults between the phase conductors of the network and ground. Therefore, to implement the proposed method, you can use the executive bodies of the compensators of the active component that implement the known methods [4] [6]
Further, we consider the proposed methods for automatic tuning of CAS and CES when suppressing arc processes in load windings and transformers supplying a network. Each of these methods, as a necessary property, must ensure that the control sign corresponds to the adjustable (capacitive or active) component of compensation for the sign of network detuning according to this component under conditions of fine tuning to the other (respectively, active or capacitive) component, and zero control values for both components under conditions of full compensation of ZNZ currents. Mutual independence of the controls is also desirable (to improve the dynamic characteristics of auto-tuning processes), that is, that the control over the capacitive component does not depend on the detuning over the active component and vice versa.

When negative resistance is used for KAS in KNPS (according to [6]), the above requirements are met if the KES control is formed proportional to the deviation of phase v of voltage e (t) of the neutral bias from phase b of the emf phase E (t) acting between the point and network neutral. The specified phase b is determined according to claims 1 to 5 of the claims. CAS control is formed in this case in proportion to the deviation of the amplitude e _{o of the} voltage e (t) of the neutral offset from the amplitude E _{o of the} aforementioned EMF (as reflected in paragraph 11 of the claims). This circumstance is due to the fact that when the active component of a single-phase dependent inverter is used as a compensator [6], the amplitude and phase of the neutral bias voltage are self-oscillation parameters in KNPS (at its natural frequency), and the amplitude of these self-oscillations is almost completely determined (in addition to the network parameters) CAS value, and the frequency (associated with the phase of the integral dependence) determines the value of CES.

ЭДС Е(t), действующей между точкой ЗНЗ и нейтралью сети, на вектор

, ортогональный вектору

напряжения е(t) смещения нейтрали (как это предлагается в п.8 формулы изобретения), а управление U_α по КАС определяется по формуле:
U_α~ E₀sec(β-ν)-e₀sec(Φ-ν), (11)
где Е_о и β соответственно, амплитуда и фаза ЭДС Е(t), действующей между точкой ЗНЗ и нейтралью сети;
е₀ и v соответственно, амплитуда и фаза напряжения е(t) смещения нейтрали;
n фаза тока q(t) искусственной несимметрии,
как это предлагается в п.12 формулы изобретения. Проекция ЭДС Е(t) на вектор

, ортогональный вектору

напряжения е(t) смещения нейтрали, определяется, например, при помощи синхронного детектирования сигнала Е(t) опорным сигналом постоянной амплитуды, ортогональным напряжению е(t) смещения нейтрали например, меандром промышленной частоты, сдвинутым на 90^o по отношению к сигналу е(t), или же вычисляется по известным амплитуде Е_o и фазе β ЭДС Е(t), а также фазе v напряжения смещения нейтрали по формуле:

Поясним сказанное при помощи векторных диаграмм, показанных на фиг.4. Пусть в сети имеет место ЗНЗ в обмотке, подключенной к линейному напряжению СА, и вектор

ЭДС Е(t), действующей между точкой G ЗНЗ и нейтралью N сети, занимает положение, показанное на фиг.4 в позиции 54. Поскольку угол между векторами 54

по абсолютной величине меньше 60^o, то в соответствии с вышеизложенным для осуществления КАС регулируемая индуктивность подключается параллельно вторичной обмотке фазы G трансформатора Бауха, создавая ток индуктивной несимметрии, вектор

которого показан на фиг.4 в позиции 50. Годографы вектора

напряжения смещения нейтрали при постоянной КАС и меняющейся КЕС представляют собой окружности 51-53, проходящие через начало координат, центры которых располагаются на векторе 50

. Увеличение КАС приводит к увеличению диаметра окружности, увеличение КЕС к перемещению вектора

по часовой стрелке, причем постоянной расстройке по КЕС соответствует постоянный угол Φ-ν между векторами

(50). Как видно из фиг. 4, перекомпенсации по емкостной составляющей соответствует положительная проекция 59 U $\genfrac{}{}{0ex}{}{\text{+}}{\text{r}}$ вектора 54

на вектор 58

, ортогональный векторам 55-57

а недокомпенсации отрицательная проекция 62 U $\genfrac{}{}{0ex}{}{\text{-}}{\text{r}}$ вектора 54

на вектор

61, ортогональный вектору 60

напряжения смещения нейтрали. При точной настройке КЕС вектор

совпадает с вектором

, и проекция последнего на вектор

, ортогональный вектору

будет равна 0. Из фиг.4 следует также, что проекция 59 U $\genfrac{}{}{0ex}{}{\text{+}}{\text{r}}$ не зависит от величины КАС, поскольку векторы 55-57 совпадают по направлению. Кроме того, зависимость проекции вектора

на вектор

, ортогональный вектору 60

, от расстройки КЕС носит монотонный характер. Таким образом, при формировании управления КЕС в соответствии с п.8 формулы изобретения соблюдаются все сформулированные выше требования.In another case, namely, when the CAS is carried out by introducing an artificial asymmetry current into the KNPS, for example, by means of an adjustable inductance connected in parallel to one of the phase secondary windings of the arc suppression device made in the form of a Bauch transformer [4], the listed properties are ensured in the case if the CES control is formed proportional to the projection of the vector

EMF E (t), acting between the point of ZNZ and the neutral of the network, on the vector

orthogonal to the vector

voltage e (t) of the neutral bias (as proposed in paragraph 8 of the claims), and the control U _α by CAS is determined by the formula:
U _α ~ E ₀ sec (β-ν) -e ₀ sec (Φ-ν), (11)
where E _about and β, respectively, the amplitude and phase of the EMF E (t), acting between the point of ZNZ and the neutral of the network;
e ₀ and v, respectively, the amplitude and phase of the voltage e (t) of the neutral bias;
n phase of the current q (t) of artificial asymmetry,
as proposed in paragraph 12 of the claims. The projection of the EMF E (t) on the vector

orthogonal to the vector

the neutral bias voltage e (t) is determined, for example, by synchronously detecting the signal E (t) with a constant amplitude reference signal orthogonal to the neutral bias voltage e (t), for example, a power frequency meander shifted 90 ^° with respect to the signal e ( t), or is calculated by the known amplitude E _o and phase β EMF E (t), as well as phase v of the neutral bias voltage according to the formula:

Let us explain what was said using the vector diagrams shown in figure 4. Let in the network there is a ZNZ in the winding connected to the line voltage SA, and the vector

EMF E (t), acting between the point G ZNZ and the neutral N of the network, occupies the position shown in figure 4 at position 54. Since the angle between the vectors 54

the absolute value is less than 60 ^o , then, in accordance with the foregoing, for the implementation of CAS, the adjustable inductance is connected parallel to the secondary winding of phase G of the Bauch transformer, creating an inductive asymmetry current, vector

which is shown in figure 4 at position 50. Hodographs of the vector

neutral bias voltages at a constant CAS and a changing CES are circles 51-53 passing through the origin, the centers of which are located on the vector 50

. An increase in CAS leads to an increase in the diameter of the circle, an increase in CES to the displacement of the vector

clockwise, and constant detuning according to CES corresponds to a constant angle Φ-ν between vectors

(fifty). As can be seen from FIG. 4, overcompensation in the capacitive component corresponds to a positive projection of 59 U $\genfrac{}{}{0ex}{}{\text{+}}{\text{<figure-callout id="54" label="r vector" filenames="00000162.png" state="{{state}}">r </figure-callout>}}$ vector 54

on vector 58

orthogonal to vectors 55-57

while undercompensation negative projection 62 U $\genfrac{}{}{0ex}{}{\text{-}}{\text{<figure-callout id="54" label="r vector" filenames="00000162.png" state="{{state}}">r </figure-callout>}}$ vector 54

on vector

61 orthogonal to vector 60

neutral bias voltage. With fine tuning CES vector

matches the vector

, and the projection of the latter onto the vector

orthogonal to the vector

will be equal to 0. From figure 4 it also follows that the projection 59 U $\genfrac{}{}{0ex}{}{\text{+}}{\text{r}}$ does not depend on the value of CAS, since the vectors 55-57 coincide in direction. In addition, the dependence of the projection of the vector

on vector

orthogonal to vector 60

, from detuning CES is monotonous. Thus, when forming the CES control in accordance with paragraph 8 of the claims, all the requirements formulated above are observed.

где черта вверху означает осреднение по времени на периоде опорного сигнала. При гармонической форме напряжения е(t) смещения нейтрали и точной настройке КЕС сигнал е(t) синфазен с ЭДС, действующей между точкой ЗНЗ и нейтралью сети: Φ = β.. Подстановка выражения
e(t) = e₀cos(ωt+Φ)
в (13) показывают, что при указанных условиях U_r 0. Из (13) следует также, что расстройка КЕС в ту или другую сторону приводит к появлению ненулевой величины U_r соответствующего знака. Таким образом выполняется сформулированное выше необходимое требование к управлению по КЕС. Для случая гармонического сигнала е(t) выполняются и другие требования к управлению по КЕС (монотонность и инвариантность к расстройке КАС). При наличии частых дуговых пробоев свойства управления (13) по КЕС сохраняются благодаря высокой помехоустойчивости операции синхронного детектирования. Помехоустойчивость в свою очередь объясняется гармонической формой опорного сигнала, которая обусловливает переход в колебательные составляющие и подавление фильтрацией всех компонентов спектра сигнала е(t) за исключением составляющих, близких к промышленной частоте ω (то есть частоте опорного сигнала).The CES auto-tuning method described in clause 7 requires information on the phase of the neutral bias voltage, and the method described in clause 8 requires additional information on the amplitude of the specified voltage. With significant detuning of the CES or with a low dielectric strength of the insulation at the site of damage, arc breakdowns can follow very often (more than one breakdown per period). The neutral bias voltage has a shape that is far from harmonic, and it becomes impossible to determine its amplitude and phase. In these cases, the CES auto-tuning is proposed to be carried out according to claim 9 of the claims, that is, by synchronously detecting the sign e (t) of the voltage t (e) of the neutral offset by the reference signal E ^* (t) orthogonal to the EMF E (t) acting between ZNZ point and network neutral. The result U _{r of} synchronous detection is described by the formula:

where the bar at the top means time averaging over the period of the reference signal. With the harmonic shape of the neutral bias voltage e (t) and fine-tuning the CES, the e (t) signal is in phase with the EMF between the point of the short-circuit protection and the network neutral: Φ = β .. Substitution of the expression
e (t) = e ₀ cos (ωt + Φ)
in (13) show that under the indicated conditions U _r 0. It also follows from (13) that detuning the CES in one direction or another leads to the appearance of a nonzero quantity U _{r of the} corresponding sign. In this way, the necessary requirement for CES management stated above is fulfilled. For the case of a harmonic signal e (t), other CES control requirements (monotonicity and invariance to CAS detuning) are also fulfilled. In the presence of frequent arc breakdowns, the CES control properties (13) are preserved due to the high noise immunity of the synchronous detection operation. The noise immunity, in turn, is explained by the harmonic shape of the reference signal, which causes the transition to vibrational components and filtering suppression of all components of the spectrum of the signal e (t) with the exception of components close to the industrial frequency ω (i.e., the frequency of the reference signal).

Если изменения знака напряжения Е(t) на периоде [β,2π+β] опорного сигнала происходили при мгновенных фазах ωt₁,ωt₂,...ωt_N,, то (14) можно представить в следующем виде:

где δ некоторая малая величина, дающая возможность определить направление изменения знака е(t) в момент t_i (с "+" на "-" или наоборот). Действия, сформулированные в п. 10 формулы изобретения, как раз и приводят к получению управления по КЕС согласно выражению (15).The operation of synchronous detection with microprocessor implementation of the proposed method, it is advisable to perform in accordance with paragraph 10 of the claims. Let us explain what was said. Expression 10 can be rewritten as follows:

If changes in the sign of the voltage E (t) over the period [β, 2π + β] of the reference signal occurred at instantaneous phases ωt ₁ , ωt ₂ , ... ωt _N ,, then (14) can be represented as follows:

where δ is a small quantity that makes it possible to determine the direction of the sign of e (t) at the moment t _i (from "+" to "-" or vice versa). The actions formulated in paragraph 10 of the claims just lead to obtaining control over the CES according to expression (15).

, пропорционален амплитуде тока θ(t) искусственной несимметрии, которая в свою очередь пропорциональна величине КАС, то перечисленные выше требования применительно к управлению КАС будет удовлетворены, если сформировать указанное управление пропорциональным отклонению диаметра годографа вектора

от диаметра данного годографа при точной компенсации КАС (то есть от диаметра окружности 52). Диаметр D_o окружности 52 можно определить, воспользовавшись тем обстоятельством, что угол SGN (фиг.4) является прямым, как вписанный угол, опирающийся на диаметр, по следующей формуле:

Величины Е_o (амплитуда ЭДС Е(t), действующей между точкой G ЗНЗ и нейтралью N сети), β (фаза данной ЭДС) и n (фаза тока q(t) искусственной несимметрии, вводимого в КНПС для КАС), фигурирующие в данном выражении, известны. Диаметр D годографа вектора

(55-57) можно также выразить через амплитуду е_o и фазу Φ напряжения е(t) смещения нейтрали (которые легко доступны для измерения), используя тот факт, что углы S₁Q₁N, S₂Q₂N, S₃Q₃N (фиг.4) прямые как вписанные углы, опирающиеся на диаметр. Выражение для данной величины имеет следующий вид:

Сопоставив (16) и (17), приходим к выводу, что выражение для управления U_α КАС
U_α~ D₀-D
совпадает с выражением (11), на основе которого построен п.8 формулы изобретения.Since the diameter of the circle (51-53 in figure 4) depicting the hodograph of the vector

is proportional to the amplitude of the current θ (t) of artificial asymmetry, which in turn is proportional to the CAS value, then the above requirements with respect to CAS control will be satisfied if the specified control is proportional to the deviation of the diameter of the vector hodograph

from the diameter of this hodograph with accurate compensation of the CAS (that is, from the diameter of the circle 52). The diameter D _{o of the} circle 52 can be determined by taking advantage of the fact that the angle SGN (FIG. 4) is straight, as the inscribed angle based on the diameter, according to the following formula:

The values of E _o (the amplitude of the EMF E (t) acting between the point G of the SC and the neutral of the N network), β (the phase of this EMF) and n (the phase of the current q (t) of the artificial asymmetry introduced into the FNC for UAN) expression, known. Diameter D of the hodograph of the vector

(55-57) can also be expressed in terms of the amplitude e _o and phase Φ of the voltage e (t) of the neutral bias (which are easily accessible for measurement), using the fact that the angles S ₁ Q ₁ N, S ₂ Q ₂ N, S ₃ Q ₃ N (FIG. 4) straight as inscribed angles based on diameter. The expression for this quantity has the following form:

Comparing (16) and (17), we conclude that the expression for the control U _α CAS
U _α ~ D ₀ -D
coincides with expression (11), on the basis of which p. 8 of the claims is built.

напряжения смещения нейтрали не зависит от конкретного местоположения вектора

на нем, то сформированное подобным образом управление КАС не зависит от расстройки по КЕС. Оно имеет монотонный характер благодаря пропорциональности диаметра годографа вектора

амплитуде тока θ(t) искусственной несимметрии. Таким образом, при формировании управления КАС в соответствии с п.12 формулы изобретения также соблюдаются все сформулированные выше требования.Due to the fact that the diameter of the hodograph of the vector

neutral bias voltage does not depend on the specific location of the vector

on it, then the CAS control formed in this way does not depend on the CES detuning. It has a monotonic character due to the proportionality of the diameter of the hodograph of the vector

the amplitude of the current θ (t) of artificial asymmetry. Thus, when forming the UAN control in accordance with paragraph 12 of the claims, all the requirements formulated above are also observed.

The obtained signals U _r and U _α can be used for integral, proportional-integral or proportional-integral-differential control, respectively, CES and CAS.

Next, we consider the action of the proposed method on the example of a device that implements this method and shown in figure 2 under the assumption that the blocks 21 (DGA), 22 (CAS) and 23 (BUK) are made in accordance with figure 5-7. In this case, the case covered by paragraphs 1-6, 9, 10, 12 of the claims is realized. Its use is most expedient in networks with small capacitive currents of OZNZ, for example, in networks of own needs of power plants, in career and rural distribution networks. Compensation of capacitive currents ZNZ is carried out here with the help of a 63 Bauch transformer. The equivalent inductance introduced by this transformer into the neutral of the network (i.e., the value of CES) is controlled by changing the time delay t _r between the time of the next locking of the thyristor switch 68 TK1 (which is detected by the sensor 69 DZ1 of locking the thyristors of the key 68) and issuing the next unlocking key 68 momentum σ ₁ . A similar CES method is proposed and described in detail in [5] Compensation of the active component of the current of the SCZ within the framework of the considered example can be carried out either as shown in Fig. 5, that is, by analogy with [4] or in accordance with Fig. 6.

постороннего трехфазного низковольтного (220-1140 В) источника, линейные напряжения которого синфазны с линейными напряжениями U_AB(t), U_BC(t), U_CA(t) сети.In both cases, CAS is performed by introducing into the KNPS a current θ (t) of artificial asymmetry. However, when the CAS block 22 is implemented in accordance with FIG. 5, an artificial asymmetry current θ (t) is introduced by connecting (using a switching device 72 KU) an adjustable inductance parallel to one of the secondary phase windings of the Bauch transformer 63, which is equivalent (including the transformation coefficient) to turn on adjustable inductance between one of the network phases and ground. When realizing the CAS block 22 in accordance with FIG. 6, the current asymmetry θ (t) is introduced through the adjustable inductance and the secondary winding 65-67 of the Bauch transformer 63 from one of the line voltages

an external three-phase low-voltage (220-1140 V) source whose linear voltages are in phase with the linear voltages U _AB (t), U _BC (t), U _CA (t) networks.

The first implementation (Fig. 5) is more autonomous than the second (Fig. 6), since it does not require an extraneous source, however it is somewhat less effective for ZNZ in phase conductors and windings connected to a "star", since in the first case the phase ν current q (t) of artificial asymmetry lags 30 ^o from the nearest phase EMF (E _A (t), E _B (t), E _C (t), (figure 2) network, and in the second case, the current θ (t) Therefore, with the above types of OZNZ, the resonant KNPS tuning, which requires the minimum power of the active component compensator, coincides with the fine tuning K C only for the second implementation (Fig. 6) .In addition, the first implementation (Fig. 5) requires upgrading the finished Bauch transformer 63 if a serial sample is used (for example, TADTM or UDTM). As an adjustable inductance, through which current θ is introduced (t) artificial asymmetry, an additional inductor 70 is used, connected in series with the TK2 71 thyristor switch. The inductance is regulated, as before, by changing the time delay τ _a between the time of the next locking of the TK2 71 thyristor switch (which is detected sensor 72 DZ2 locking thyristors) and the issuance of the next unlocking pulse σ ₂ .
The indicated time delays and unlocking pulses supplied to the thyristor switches 68, 71 are formed by the block 23 of the ACU (Fig. 1), for which the signals ζ ₁ (t) and ζ ₂ (t) of the state of the thyristor switches 68 (TK1) are supplied to this block and 71 (TK2) with sensors 69 (DZ1) and 72 (DZ2).

 Due to the relative complexity of the actions provided by the claimed method, it is advisable microprocessor implementation of block 23 BUK, for example, in accordance with Fig.7. In this case, the basis of the block 23 BUK is a microprocessor 74 with permanent (ROM) and random access memory (RAM), which as follows interacts with other elements of block 23. The processor of block 74 reads the next program command from the ROM, decodes it, and then performs, while addressing, in accordance with the contents of the command, to RAM or ROM of block 74, and through bus 75 also to timers 76-78, to interrupt controller 79, to input port 80 or to output port 81. When a front arrives signal to one of the inputs Ir1 Ir7 to of the interrupt controller 79, the latter, interacting with the ROM unit 74, forces it to interrupt the execution of the current program, perform one of the interrupt processing procedures recorded in the ROM of the block 74, and then return to the interrupted program.

We now continue to consider the process of controlling thyristor switches 68 (TK1) and 71 (TK2) (Fig.5.6) block 23 BUK (Fig.7). The signals ξ ₁ and ξ ₂ , indicating the locking of the thyristors of the keys 68, 71, are fed to the triggering inputs "E" of programmable timers 76 (T1) and 77 (T2) (Fig.7). Timers consider (in the direction of decreasing codes) pulses of a high-frequency (for example, 2 MHz) clock signal supplied to the counting inputs "c" of the indicated timers (Fig. 7). Codes corresponding to the required time exits τ _r and τ _a are recorded in timers 76 and 77 by a processor 74 with read-only and read-only memory devices. When the code reaches any of the timers 76 and 77, the latter generates a pulse arriving at the input (Ir1 or Ir2) of the interrupt controller 79 and indicating the end of the next exposure (τ _r or τ _a ) time. As a result, the ROM unit 74 proceeds with the processing of the corresponding interrupt. These procedures provide for the supply (via port 81) of the unlocking pulse σ ₁ or σ ₂ to the corresponding thyristor switch 68 or 71 (Fig. 5,6) and restarting the timer (76 or 77) to hold τ _r and τ _a time equal to the duration of the unlocking pulse (e.g. 200 μs). After this exposure, a new interrupt follows at the same input. Block 74 PRZU in the process of processing the processing of the considered interrupt removes (via port 81) the unlocking pulse from the corresponding thyristor switch (68 or 71 in Fig.5,6) and sets the timer 76 or 77 again for the time delay τ _r or τ _a , which determines , ultimately, the value of CES or CAS, respectively. The regulation of CES and CAS is thus reduced to writing to the timers 76 and 77 new values of the shutter speeds τ _r and τ _a .

Let us further consider the process of general synchronization of the block of beech (Fig.7) network frequency. The signal U _AB (t), which is proportional to the line voltage of the network between phases A and B, is generated by a voltage measuring transformer 24 (Fig. 2), fed to the input of zero-comparators 82 of the BUK unit (Fig. 7), and then passes through the shapers 83 and 84 , eliminating bounce, respectively, on the decline and on the front of this signal (at the cost of delay, respectively, of its edge and decline). The front of the signal at the output of the shaper 83, which occurs almost simultaneously with the transition of the linear voltage U _AB through "0" (from negative to positive), restarts the timer 78, which starts counting clock pulses arriving at its counter input "s". The count begins with a code approximately corresponding to the half-period of the network. As a result, according to the code read from this timer of the blocks of the PRZU 74 at any time t time, it is possible to restore the instantaneous phase ωt of the reading time t. With the described synchronization method, the measurement of all phase shifts is carried out relative to the linear voltage U _AB (t), which is attributed to a zero phase shift. At the moment the timer code 78 passes the zero value, a pulse is generated at its output, which initiates (via the interrupt controller 79) a call by the RAM unit 74 to call the interrupt processing routine at the Ir3 input of the controller 79. This procedure involves polling (via port 80) the state of the comparator 82 at that moment time when the instantaneous phase ωt according to the readings of the timer 78 is equal to π. Since at this moment the signal U _AB (t) must pass through "0", according to the state of the comparator 82 (that is, by the signal at the output of the driver 84), it is possible to judge the correspondence of the timer delay 78 to the actual half-cycle duration at a given network frequency, as well as adjust it, that is, slightly increase this shutter speed if a logical "0" is present at the output of the shaper 84 (ie, U _AB <0), or reduce it otherwise (which is provided for in the interrupt processing routine at the Ir3 input of controller 79) . The tracking system thus formed, tracking changes in the network frequency, eliminates their influence on the functioning of the device.

In the normal mode of operation of the network, there is no abrupt change in the signal of the voltage of the neutral bias voltage e (t) (coming to the input of the AC unit 23 from the output of the voltage transformer 24, Fig. 2), and its amplitude e (t) does not exceed 20% of that amplitude e _m , which takes place with a deaf OZNZ. In this mode, there are no pulses at the inputs Ir4, Ir5 of the controller 79 interruptions, and the unit 87 is transferred (by signal z of port 81) to the state of the inertial link with a small time constant (for which an analog switch 90 connects the input of the operational amplifier 88 to the connection point of the resistors 91, 92 ), and therefore monitors the neutral bias voltage e (t) supplied to its input. The interrupt at the input Ir6 of the controller 79 is caused if the signal g at the output of port 81 does not correspond to the signal h at the output of the comparator 94 (as a result of which “1” appears at the output of the logic element 95 “DISCHARGE”, which causes this interrupt) . The procedure for processing this interrupt maps the output signal g of the port 81 with the output signal h of the comparator 94, as a result of which this correspondence is a normal state of the considered example of the device before the possible occurrence of the SCR mode. In the normal mode of network operation, the device can also carry out preliminary (resonant) tuning of the KNPS, realizing any of the known methods of auto-tuning, for example, [4]
When an arc breakdown of insulation occurs at time t _o at point G, for example, in the load winding, as shown in Fig. 2, the neutral bias voltage e (t) undergoes an abrupt change, which is, firstly, monitored by the inertial link with a small time constant (elements 88-92 in Fig. 7), and, secondly, are detected by the arc breakdown sensor 85, which generates a pulse at the input of Ir4 of the interrupt controller 79. The latter initiates a call to the corresponding interrupt processing procedure, during which the RAM unit 74 reads and stores the contents of the timer 78 in RAM, that is, it records information on the instantaneous phase ωt _{0 of} the arc breakdown moment t _o , and also reads and stores the value in RAM through port 80 signal h, that is, the sign of the instantaneous value e (t ₀ + δ) of the voltage e (t) of the neutral offset immediately after the breakdown. In addition, in the process of performing this procedure, the RAM block 73 changes the signal z at the output of port 81, as a result of which the block 87 switches to the conversion mode into the time interval of that instantaneous value e (t ₀ + δ) of the neutral bias voltage e (t) that was fixed immediately after the arc breakdown. In this case, the analog switch is set to the position shown in Fig. 7, and the output voltage of the integrator (at elements 88, 89, and 93) begins to change at a constant speed in the direction of decrease in absolute value (regardless of the sign of e (t _o )). After a certain period of time Δt (proportional to the instantaneous value e (t ₀ + δ) of the neutral bias voltage e (t) immediately after the arc breakdown), the signal at the output of the indicated integrator reaches zero, and then changes sign. At the time of changing the sign of the signal at the output of the operational amplifier 88, the correspondence between the signals h and g is violated, as a result of which a logical “1” appears at the output of the element 95 and the interrupt controller 79 initiates a call by the block 74 of the ROM of the interrupt processing routine at the input of Ir6. The specified procedure reads and stores in the RAM unit 74 the contents of the timer 78 and again brings the signals h and g into correspondence. The device is prepared to receive similar information during the next arc breakdown (if it occurs). In addition, with each change in the sign of the neutral bias voltage e (t), the null-comparator 96 is switched, as a result of which, through the pulse generator 97, the FI supplies short pulses to the input Ir7 of the interrupt controller 79, which initiates the call by the RAM unit 74 of the corresponding interrupt processing procedure. When this procedure is performed, the contents of the timer 78 and the state of the comparator 96 (through port 80) are read and stored in the RAM unit 74. These actions prepare the initial information for auto-tuning CES in accordance with paragraphs 9, 10 of the claims.

напряжения смещения нейтрали, имеющая место сразу после дугового пробоя, а по состоянию сигнала q (на выходе компаратора 95) в момент пробоя знак этого напряжения. По содержимому таймера 78 в момент t₀+δ определяется также мгновенная фаза ωt₀ момента пробоя. Выполняя п.2 формулы изобретения, полагают амплитуду Е_o ЭДС, действующей между точкой ЗНЗ и нейтралью сети, равной определенному выше значению

, а фазу β этой ЭДС величине wt₀=, взятой с противоположным знаком и увеличенной на π в случае, если e(t₀+δ)<0..Next, the block 74 PRZU processes the received information. Based on the difference in the contents of the timer 78, read at the moment of changing the sign of the signal at the integrator output (interrupt at the input of Ir6) and read at the moment t ₀ + δ of the arc breakdown (interrupt at the input of Ir4), the absolute value is determined

neutral bias voltage, which takes place immediately after the arc breakdown, and according to the state of the signal q (at the output of the comparator 95) at the time of breakdown, the sign of this voltage. The contents of the timer 78 at the time t ₀ + δ also determines the instantaneous phase ωt _{0 of the} moment of breakdown. Performing claim 2 of the claims, it is believed that the amplitude E _{o of the} EMF acting between the ZNZ point and the network neutral is equal to the value defined above

, and the phase β of this EMF is wt ₀ = taken with the opposite sign and increased by π if e (t ₀ + δ) <0 ..

When subsequent breakdowns occur, the device operates in a similar way, however, the information collected by the Ir4 and Ir6 interrupt processing procedures is processed differently, namely, in accordance with claim 3. In this case, the values of E _o and β are calculated by the RZU block 74 according to formulas (7) and (8), and with each new determination of the values of E _o and b, they are increased by 1 counter of determinations located in the RAM of block 73.

не превышает

После этого согласно п.6 формулы изобретения блок 74 ПРЗУ устанавливает сигналы ρ_A,ρ_B,ρ_C порта 81, определяющие режим компенсатора 22 активной составляющей (в смысле фазы ν тока искусственной несимметрии) в соответствии с фазой b.. Так, если компенсатор 22 выполнен в соответствии с фиг.5, то при

сигналами ρ_A,ρ_B,ρ_C передается команда коммутационному устройству 73 подключить цепь дополнительного дросселя 70 параллельно обмотке 67 фазы. С трансформатора 63 Бауха (фиг.5) при

данными сигналами передается команда подключить указанную цепь параллельно обмотке 65 фазы А трансформатора 63 (фиг. 5), а при

параллельно обмотке 66 фазы В этого трансформатора. Если же компенсатор 22 выполнен в соответствии с фиг.6, то при

сигналами ρ_A,ρ_B,ρ_C передается команда коммутационному устройству 73 подключить цепь дополнительного дросселя 70 к линейному напряжению между фазами B' и C' низковольтного источника питания (фиг.6), при

этими сигналами передается команда подключить указанную цепь к линейному напряжению между фазами C' и A' источника питания, а при

или при

к линейному напряжению между фазами B' и C' данного источника. Описанные действия устанавливают требуемую фазу ν вводимого в КНПС тока q(t) искусственной несимметрии.In accordance with paragraph 4 of the claims, the values of E _o and b obtained with each new arc breakdown are summed up in the RAM cells of the RAM block 74, and the sums obtained are divided by the value of the determination counter, that is, the arithmetic average of the indicated values is found. As a result, all the initial data for the execution (block 74 of the RAM) of the actions according to claim 5 of the claims are prepared. The result of these actions are refined values of E _o and β of the amplitude and phase of the EMF acting between the point of the SC and the neutral of the network, as well as the number of that phase (A, B or C) of the network for which the absolute value

does not exceed

After that, according to claim 6, the RZU unit 74 sets the signals ρ _A , ρ _B , ρ _{C of} port 81, which determine the mode of the compensator 22 of the active component (in the sense of phase ν of the current of artificial asymmetry) in accordance with phase b .. So, if the compensator 22 is made in accordance with figure 5, then when

the signals ρ _A , ρ _B , ρ _C send a command to the switching device 73 to connect the circuit of the additional inductor 70 parallel to the phase winding 67. With the transformer 63 Bauch (figure 5) when

These signals transmit the command to connect the specified circuit parallel to the winding 65 of phase A of the transformer 63 (Fig. 5), and when

parallel to the 66 phase B winding of this transformer. If the compensator 22 is made in accordance with Fig.6, then with

the signals ρ _A , ρ _B , ρ _C send a command to the switching device 73 to connect the circuit of the additional inductor 70 to the line voltage between the phases B 'and C' of the low-voltage power supply (Fig.6), when

with these signals, the command is transmitted to connect the specified circuit to the line voltage between the phases C 'and A' of the power source, and when

or at

to the line voltage between phases B 'and C' of a given source. The described actions establish the required phase ν of the current q (t) of artificial asymmetry introduced into the KNPS.

The values of the contents of the timer 78 recorded in the RAM of the block 74 of the RAM at the time t _i of the sign change with the voltage e (t) of the neutral offset are converted to the values ωt _{i of the} instantaneous phase of these moments and then, taking into account the states of the comparator 96 ( moments t _i + δ, the control U _r according to CES is calculated according to expression (15), that is, in accordance with claims 9, 10 of the claims. If the integral law of regulation of the CES is applied, then this control is summed (with the corresponding coefficient) with the previously set shutter speed τ _r , and the new shutter speed is recorded in the timer 76. Thus, the next step is taken to reduce the detuning of the CES, and the result of a sequence of similar steps is (if do not take into account CAS) a significant decrease in the frequency of arc breakdowns in the place of ZNZ.

где δ смещение компаратора 85.To carry out CAS tuning in accordance with paragraph 12 of the claims, additional information is required on the amplitude and phase of the neutral bias voltage e (t). The amplitude e _{o of the} neutral bias voltage e (t) is determined by the duty cycle S of the pulses at the output of the biased comparator 86. To determine the specified duty cycle, the interrupt processing routine at the input Ir5 of the interrupt controller 79 records the contents of the timer 78 in the RAM unit 73 of the RAM at the time of switching of the comparator 86. Then, the difference Δt of the codes read from the timer 78 at the neighboring time switching moments is found. This difference is associated with the mentioned duty cycle S, and therefore with the amplitude e _o unambiguous monotonic dependence

where δ is the comparator offset 85.

Таким образом, фаза Φ напряжения е(t) смещения нейтрали определяется блоком 74 ПРЗУ по формуле (19). Управление же U_A по КАС вычисляется блоком 74 по формуле (11), поскольку все необходимые исходные данные для подстановки в это выражение уже подготовлены. Полученное управление U_A имеет смысл только при гармонической форме напряжения е(t) смещения нейтрали и поэтому используется для изменения выдержки времени t_a (то есть для управления величиной КАС) только в том случае, если во время сбора исходных данных для вычисления U_α отсутствовали дуговые пробои в месте ЗНЗ. Информация о наличии дуговых пробоев передается программе управления КАС процедурой обработки прерывания по входу Ir4 контроллера 79 путем записи некоторого условного кода (установки флага) в определенной ячейке ОЗУ.The phase v of the neutral bias voltage e (t) is determined by the PRZU block 74 on the basis of the CES previously calculated (by expression (15)) control U _r according to the fact that, with the harmonic form of the voltage e (t), the neutral bias is used to control U _r expression (12) is also valid. From here we find:

Thus, the phase Φ of the neutral bias voltage e (t) is determined by the PRZU block 74 according to formula (19). The control U _A by CAS is calculated by block 74 according to formula (11), since all the necessary input data for substitution into this expression are already prepared. The obtained control U _A makes sense only with the harmonic form of the neutral bias voltage e (t) and therefore is used to change the time delay t _a (that is, to control the CAS value) only if, during the collection of the initial data, U _{α were} absent arc breakdowns in the place of ZNZ. Information about the presence of arc breakdowns is transmitted to the CAS control program by the interrupt processing routine at the input of Ir4 of controller 79 by writing some conditional code (setting a flag) in a specific RAM cell.

If the integral law is applied to control the UAN, then (in the absence of arc breakdowns during the collection of initial information), the obtained control U _{α is} summed (with the corresponding coefficient) with the previously set shutter speed t _a , and the new shutter speed is recorded in timer 76. Thus, the next a step in the direction of decreasing CAS detuning, and the result of a sequence of such steps is a complete compensation of ZNZ currents.

 As a result, the voltage between the GZZ point G and the ground becomes close to zero and the arc process stops, as a result of which the network can be in this state for an unlimited time.

It should be replaced that in conditions of fine-tuning the CAS and CES, a change in the ZZZ parameters, including the complete restoration of the dielectric strength of the insulation, does not affect the first harmonics of the voltage and current in the network. Therefore, it is obvious that a number of actions are required to recognize the fact of repairing the damage and restoring the normal operation mode of the network. In this case, you can use any known method of recognition and restoration of the normal mode of operation of the network, for example, described in [4]
The proposed method can be combined with other (known) methods of suppressing arc faults in those cases when a short circuit occurred between the phase conductor of the network and ground.

Claims (5)

 1. A method of suppressing arc earth faults in the load windings and supply transformers of a three-phase network with an earthed neutral, including measuring the instantaneous values of the phase voltage of the network and the neutral bias voltage, detecting an arc fault on the earth and compensating for the current of a single-phase earth fault, characterized in that after recognition of an arc earth fault determines the value of the amplitude and phase of the EMF acting between the earth fault point and the neutral of the network, after which they compensate Single-phase ground fault, for which the amplitude of the support and the neutral displacement voltage phase equal amplitude and phase of the EMF. 2. The method according to claim 1, characterized in that to determine the amplitude and phase of the EMF acting between the earth fault point and the neutral of the network, arc breakdowns are detected at the earth fault point and, at the time of the first breakdown, the instantaneous phase and instantaneous value of the bias voltage are recorded neutral and, if only one breakdown is detected, the EMF amplitude is taken equal to the absolute value of the instantaneous value of the amplitude of the neutral bias voltage, and the EMF phase is taken to be the value of the instantaneous phase of the neutral bias voltage, taken from false sign for a positive value of the instantaneous phase, and if the value is negative, increase the EMF phase by π, if more than one breakdown is detected, then at the time of each breakdown, the instantaneous phase and the instantaneous value of the neutral bias voltage are recorded and the amplitude and phase of the EMF for each breakdown are determined by the formulas : for w (t _t-1 - t) not multiple of π

where l (t _t ) is the instantaneous value of the neutral bias voltage;
E ₀ (t) EMF amplitude corresponding to the detected breakdown;
β _t phase EMF corresponding to the detected breakdown;
ω circular frequency of the network;
wt instantaneous phase of the neutral bias voltage;
t point in time: the moment t _{t - 1} precedes the moment t _t ,
then the values of the amplitude and phase of the EMF are found as the average value between the values of the amplitude and phase of the EMF for one breakdown and the values of the amplitudes and phases of the EMF for all subsequent breakdowns detected, after which the obtained values of the amplitude and phase of the EMF are corrected, for which the place of the single-phase fault is assigned to the winding, connected to one of the phase or linear voltages of the network or to one of the phase conductors of the network, for this the phase of the network is determined for which the absolute value of the deviation of the value of the phase of the EMF from the phase value of the phase EMF is not exceeds π / 3, and then, if the condition

Where

Adjustable value of the EMF amplitude;
E _{m the} amplitude of the phase EMF network;

absolute value of the deviation of the EMF phase value,
then the adjusted value of the EMF amplitude is taken equal to the amplitude of the phase EMF network, and the adjusted value of the EMF phase is taken equal to the phase of the phase EMF network, if the specified condition is not met, then check the fulfillment of another condition

and when this condition is met, the adjusted amplitude value of the EMF is determined by the formula

and the adjusted magnitude of the phase of the EMF is taken equal to the phase of the phase EMF, if another condition is not met, then the adjusted value of the amplitude and phase of the EMF is determined by the following formulas:

where α _i is the phase EMF phase value.  3. A method of suppressing arc faults on the ground in the load windings and supply transformers of a three-phase network with an earthed neutral, including measuring the instantaneous values of the phase voltage of the network and the neutral bias voltage, recognition of the arc fault to the earth and compensation of the current of a single-phase fault to the earth, characterized in that after recognition of an arc fault to earth determine the value of the amplitude and phase of the EMF acting between the point of earth fault and the neutral of the network, and compensate for the current of a single-phase closure ground, for which, to compensate for the active component of the point of a single-phase earth fault, an artificial asymmetry current is used, the phase of which is set in accordance with the value of the EMF phase, and compensation of the active component of the current of a single-phase fault is judged by the deviation of the product of the amplitude of the EMF by the secant of the difference in phase voltage of the EMF and the current of artificial asymmetry from the value of the product of the amplitude of the neutral bias voltage by the secant of the phase difference of the neutral bias voltage and the artificial asymmetry current, wherein the compensation of the capacitive component of the current of a single-phase earth fault is judged by the magnitude of the projection of the EMF vector onto the vector orthogonal to the neutral bias voltage vector.  4. The method according to claim 3, characterized in that the projection magnitude of the EMF vector onto a vector orthogonal to the neutral bias voltage vector is determined by synchronously detecting the EMF signal with a constant amplitude reference signal orthogonal to the neutral bias voltage. 5. The method according to claim 3, characterized in that the projection value of the EMF vector on a vector orthogonal to the neutral bias voltage vector is calculated by the formula

where U _{r is the} projection of the EMF vector on a vector orthogonal to the neutral bias voltage vector;
E ₀ EMF amplitude;
β phase EMF;
v phase of the neutral bias voltage.

Images