RU2819134C1 - Optical multichannel current meter for high-voltage networks - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: invention relates to measuring instruments for measuring current in high-voltage networks, and more specifically to current meters in which the optical Faraday effect is used. Proposed multichannel optical current meter for high-voltage networks consists of accurate and coarse optical channels in the form of two Faraday cells and five electronic channels. In each optical channel the second polarisers are made in the form of Wollaston prisms, polarization planes of linearly polarized beams separated by them make angles of ±45° with transmission planes of first polarisers, in focal planes of light-collecting lenses there are ends of additional fibre light guides, which transmit light to additional photodetectors. Magnetosensitive element of the coarse channel is also made in the form of a glass prism BP-180°, but from glass with a smaller Verdet constant compared to the glass of the prism of the fine channel. In the electronic unit of the coarse channel there is an adder of signals of photodetectors to form a constant component of the signal. Adder inputs are connected directly to coarse channel preamplifiers, and its output is connected to inputs of an additional microprocessor connected to a clock pulse generator to control the moment and number of current measurements during the period of AC oscillation of the network.

EFFECT: significant increase in measurement accuracy, due to which the coarse channel can be used as a backup for commercial metering, as well as for accurate pulse measurements of impact current and aperiodic component at short circuit; in addition, use of Wollaston prisms as second polarisers 12 and 22 made it possible to have two electronic channels for processing signals of photodetectors in a precise optical channel and four electrical channels in a coarse optical channel; proposed device will find wide application in electric power industry, especially in networks of 110 kV and higher.

1 cl, 5 dwg

Description

The invention relates to a group of measuring instruments designed to measure current in high-voltage networks, and more precisely to current meters that use the optical Faraday effect.

Until now, the most common current meters in high-voltage networks are electromagnetic measuring current transformers (ITCTs) [1], which contain a primary winding of one or several turns of a fragment of a high-voltage line conductor, a magnetic core made of transformer iron, one secondary winding with a large number of turns for accurate current measurements and several secondary windings for rough current measurements, providing information on the maximum relay protection current (overcurrent) and short-circuit surge current (TO cut-off current).

The secondary windings are necessarily grounded and loaded with known complex resistances.

Significant disadvantages of ITFC are:

- high fire hazard due to possible electrical breakdown of insulation between the windings;

- saturation of the magnetic circuit with the aperiodic component at short circuit currents;

- inevitable consumption of electricity in secondary windings (secondary circuits);

- influence on the accuracy of current measurement of the load size and the number of measured current recorders connected to the secondary windings;

- analogue form of the signal characterizing the measured current;

- large dimensions and current;

- high costs for maintenance and routine checks of the insulation condition.

As an alternative to ITFCs operating on the principle of electromagnetic transformation of current, work is being intensively carried out to create optical current meters (OCMs), the operating principle of which is based on the phenomenon of rotation of the polarization plane of linearly polarized light in a magnetically sensitive isotropic substance (for example, glass) located in the longitudinal (coaxial) magnetic field (Faraday effect).

The essence of the Faraday effect is as follows [2,3]. Linearly polarized light can be represented as the sum of two circularly polarized components of equal amplitude. Under the influence of a longitudinal (relative to the direction of light propagation) magnetic field in substances such as glass, birefringence occurs for the circularly polarized components of left and right circulation and a phase difference arises between these components

Where - phase velocities of propagation of left and right waves;

- refractive indices, respectively, for the left and right circularly polarized components;

- the path traveled by polarized light in a substance along the magnetic field strength;

- wavelength of light.

If the magnetically sensitive glass is isotropic, that is, does not have linear birefringence [2], then at the exit from the glass the circular polarization of both components is preserved and when they are added, linearly polarized light is obtained again, but with the polarization azimuth changed by an angle

Where: - the strength of the longitudinal magnetic field acting on the glass;

- Verdet constant of glass;

- path traveled by polarized light in glass;

- the angle between the direction of light propagation and the direction of the magnetic field lines;

- number of turns of a conductor fragment;

- current flowing through a fragment of a conductor;

- design factor that takes into account the distance from the glass to the current-carrying conductor and the averaging of the magnetic field strength at different points of the glass.

From formulas (1, 2) it is clear that to measure the current i flowing through the conductor of a high-voltage line, two methods can be used:

- a method for measuring the phase difference δ ₀ using an interferometer [4];

- a method based on measuring the azimuth of linear polarization α using conventional polarimeters [2].

The second method is simpler in implementation, in which linearly polarized light is directed onto a magnetically sensitive glass element located in a longitudinal magnetic field and, at the exit from the glass, the change in the azimuth of the linear polarization of the light is measured, which is proportional to the magnetic field strength, and, therefore, to the measured current i.

This method is implemented, for example, in a well-known optical current meter [5], which contains a light source installed in series, a first multimode fiber light guide, a collimating lens, a first linear polarizer, and a magnetically sensitive element in the form of a glass prism of the BP-180° type, located in longitudinal magnetic field of a fragment of a high-voltage line conductor, a second polarizer, the transmission plane of which makes an angle of ±45° with the transmission plane of the first polarizer, a collecting lens, a second multimode fiber light guide, a photodetector and an electronic unit.

The known device [5] works as follows.

Light from the source is transmitted through a fiber light guide to the focal plane of a collimating lens. The diverging beam of light emerging from the first optical fiber is converted by a lens into a collimated beam of diameter D. Then the light passes through the first polarizer, becomes linearly polarized, passes through a quadrangular prism of the BP-180° type four times, passes through the second polarizer and is collected by the second lens at the end of the second optical fiber. Next, the light hits the photodetector. The signal from the photodetector enters the electronic unit.

If no current flows through the conductor and there is no magnetic field around the conductor, then as light passes through the BP-180° prism, its polarization state does not change. Since the azimuth of the transmission plane of the second polarizer differs from the transmission plane of the first polarizer by an angle of ±45°, the photodetector perceives light intensity

Where - light intensity of source 1;

- overall reflection coefficient;

- the total transmittance of all optical elements.

If alternating current flows through a fragment of a conductor network frequencies then the photodetector perceives light intensity

Where: - maximum amplitude of the angle of rotation of the plane of polarization of light by the prism BP-180°;

- Verdet constant of the prism material;

- the total length of the light path in the prism;

- the angle between the direction of light propagation and the magnetic field lines.

The photodetector operates in linear mode, so the luminous flux I is converted by the photodetector into an electrical signal

Where - constant component of the electrical signal of the photodetector.

The electronic unit calculates the ratio Q of the variable component

to the constant component U ₀

and then the desired current i flowing through the conductor is calculated using the formula

Where - coefficient characterizing the efficiency of using the magnetic field.

The results of alternating current measurements are displayed on the digital display of the electronic unit and, using the interface, are transmitted to other external devices.

This known device [5] makes it possible to measure current in high-voltage networks with high accuracy (no worse than 0.2%) for commercial electricity metering, but cannot ensure the normal operation of relay protection and automation for the following reasons.

Firstly, the known device is configured so that at rated current in the network, the angle of rotation of the plane of polarization was within that is, the linear section of the curve so that, according to dependencies (6) and (7), proportionality between as well as Q and α.

However, during current surges in the network, for example, as a result of a short circuit, the current in the network can increase tens of times, for example In this case, the rotation angle will reach values which will lead to a violation of the principle of operation of the device, since according to formula (6) the angle α should only be within ±45° (the function U=f(α) is cyclic).

Secondly, the short-circuit shock current (the aperiodic component) appears 0.01 seconds after the short circuit and quickly fades away. To register this rapidly occurring process, an additional coarse channel and a special electrical circuit with high performance when measuring current are required, which is not present in this known device.

A two-channel optical current meter is known [6], which is a prototype of the proposed device. A two-channel optical current meter contains two independent fine and coarse optical channels in the form of Faraday cells, installed inside a common solenoid formed from one or several turns of a fragment of a high-voltage line conductor. Each of the channels contains a light source installed in series, a first polarizer, a magnetically sensitive element, for example, in the form of a glass prism, a second polarizer, a light-collecting lens, a second fiber light guide, a photodetector connected to an electronic unit with level shapers for the constant component and the alternating component of the photodetector signal, microprocessor, current measurement results indicator, interface board.

A glass prism of type BP-180° is installed as a magnetosensitive element in the fine channel, and in the coarse channel there is an ordinary glass prism with an additional prism of type BR-180° glued on.

The electronic unit also has two channels. The precise measurement channel contains a photodetector, a preamplifier, a signal processing unit, a microprocessor, an indicator of measurement results, an interface, and a power supply. A trigger-type threshold device with a power amplifier, which is connected to a microprocessor and with a relay protection input, is connected to the rough channel preamplifier.

The known two-channel optical current meter [6] operates as follows.

Light from the precision channel source is transmitted through the first optical fiber to the focal plane of the collimating lens. The diverging beam of light emerging from the first optical fiber is converted by the lens into a collimated beam of light with a diameter of D. Then the light passes through the first polarizer, becomes linearly polarized, passes through a quadrangular prism of the BP-180° type four times, passes through the second polarizer and is collected by a collecting lens at the end of the second optical fiber , then the light hits the photodetector.

If an alternating current of frequency ω (50 Hz) flows through a fragment of a conductor, for example, in the form of a full turn of a bus,

then the photodetector perceives light intensity

Where: - light intensity of the light source;

- reflection coefficient of the mirror surfaces of the prism BP-180°;

- light transmittance of the optical path.

Light flow is converted into an electrical signal by a photodetector

Where: - maximum angle of rotation of the light polarization plane by the prism BP-180°;

- the path traveled by linearly polarized light in the BP-180° prism.

After pre-amplification by the amplifier, the signal processing unit separates the DC component of the signal

and variable component

)

The DC component is fed to one input of the microprocessor, and the AC component is detected, smoothed and fed to the second input of the microprocessor.

The microprocessor calculates the ratio Q of the signal, proportional to the variable component (13) to the constant component

)

then calculates the desired current i flowing through the conductor fragment according to formula (8).

where K is the proportionality coefficient.

The AC current measurement results are displayed on a digital indicator and, using an interface board, are transmitted to external devices for recording and control.

At the same time the light intensity from the source of the second coarse channel is transmitted through the first multimode fiber of the coarse channel to the focal plane of the collimating lens. The diverging beam of light emerging from the first optical fiber is converted by the lens into a collimated beam of light. Next, the light of the coarse channel passes through the first linear polarizer, becomes linearly polarized, passes once through a simple quadrangular additional prism, the second polarizer, is reflected twice from the polished surfaces of the BR-180° prism, is collected by the lens at the end of the second fiber light guide and is transmitted to the photodetector.

The luminous flux I ₂ is converted into an electrical signal by a coarse channel photodetector

Where: - maximum angle of rotation of the plane of polarization of light by a simple prism of a rough channel;

- the path traveled by linearly polarized light in the prism of the rough channel.

After pre-amplification of the coarse channel by the amplifier, the signal processing unit separates the constant component of the signal (16) U ₁ = and the variable component

DC level the coarse channel is equal to or close to the level of the DC component of the fine (main) channel And the level (amplitude) of the variable component of the coarse channel (24) is several times less than the amplitude of the fine channel (13) due to the fact that, other things being equal, the path L2 traversed by light in the prism of the coarse channel is several times (for example, 5 times) less than the path L ₁ traversed in the BP-180° prism of the precise channel.

So, for example, if at the rated current in the network The angle of rotation of the plane of polarization α _1max in the fine channel is usually in the range from 5 to 10°, while in the coarse channel it is only 1°-2°.

Therefore, if a current surge occurs in the network, for example, as a result of a short circuit then in the exact channel there will be a violation of the operating principle and in the rough channel α _2max < 45° and there is no disruption to the operation of the optical current meter.

In addition, in the pre-amplifier of the coarse channel, the signal of the variable component increases to the level of operation of a trigger-type threshold device, which is amplified by the amplifier and supplied to relay protection to turn off the network.

The microprocessor is programmed so that at angle α _1max < 30°, when the ratio of the variable component to the constant component information about the measured current i in the high-voltage network comes from the first (fine) channel, and as the current i increases, information about the measured current i comes from the additional (coarse) channel.

So, for example, if a 10 × 100 mm ² bus in the form of one turn is used as a fragment of a high-voltage line conductor, then when and with current and Q≈0.87. That is, when the current in the network exceeds more than 3 times, the channels are automatically switched from the first to the second (rough) and the current measurement continues without loss of information. Of course, this takes into account the coefficient where L ₁ is the path traveled by light in the prism of the fine channel, and L ₂ is the path traveled by light in the prism of the coarse channel.

The same thing happens when an 8×60 mm ² bus in the form of two turns is used as a conductor fragment, when Yes, when and when And in this case, if the current exceeds about three times Channels are automatically switched, but the relay protection has not yet triggered. The relay protection will operate when the short circuit current reaches.

Thus, adding an additional magnetosensitive element to the known device [6] and organizing a second additional coarse channel makes it possible to measure current with high accuracy in a wide range of current measurements in a high-voltage network and ensure guaranteed operation of relay protection only when a short circuit occurs

This known device has a significant drawback, which manifests itself when operating in harsh climatic conditions. It is known that during sudden changes in ambient air temperature, temperature gradients can appear at various points in the glass prism of a Faraday cell, which leads to the appearance of thermal or mechanical loads in the prism glass and, as a consequence, to the occurrence of linear birefringence. That is, under the influence of a temperature gradient, the isotropic glass of the prism becomes anisotropic. In this case, the glass prism can be represented as a phase plate [7] with an arbitrary orientation of the main axes ϕ and introducing a phase difference δ _g between the mutually orthogonal components of polarized light [8]

Where - wavelength of light;

- photoelastic coefficient of prism glass [8];

- difference in tangential stresses[7];

- light path length in glass;

- the difference in refractive indices for the ordinary and extraordinary components of polarized light.

In this case, the glass prisms of both channels, without the influence of a magnetic field on them, are similar to phase plates with arbitrary orientations of the main axes ϕ, introducing a phase difference δ _g .

Matrices of glass prisms as equivalent phase plates can be represented in the form

If an alternating current with a frequency of ω=50 Hz flows through a fragment of a high-voltage network conductor, then with the advent of the Faraday effect, the glass prisms of both channels acquire the properties of a rotator [8], which is reflected by the matrix

where α is the angle of rotation of the plane of polarization of linearly polarized light according to formula (2);

- reflectance coefficient of the mirror surfaces of the prism 5 for horizontally polarized light.

In this case, the channel prism matrices can be represented as the product of matrix (19) and rotator matrix (20), that is, the matrix

Where

If for each channel we multiply the Stokes vectors characterizing unpolarized light incident on the first polarizer and the known matrices [8] of the first polarizer prisms BP-180° after passing the light four times in the fine channel and once passing in the coarse channel, the second polarizer then we can obtain the first Stokes parameters characterizing the light intensity in the exact channel and in the rough channel:

And the ratio of the variable components to the constant components of the signals in each channel will be characterized by the dependencies:

Comparing formula (14) and formulas (24), (25), we see that thermal or mechanical loads in the glass of channel prisms cause the appearance of linear birefringence (anisotropy) in it and, as a consequence, errors in current measurements in a high-voltage network.

The second significant drawback of the known device [6] is the lack of a high-speed electronic channel to provide selective current measurement (approximately 80 samples per network oscillation period of 50 Hz) in order to record the level of shock current during a short circuit in a high-voltage network by relay protection equipment [9].

A new multichannel optical current meter is proposed, free from the mentioned disadvantages.

The proposed device consists of two independent from each other fine and coarse optical channels in the form of Faraday cells, installed in a single housing on the upper flange of a high-voltage insulator inside a common solenoid formed from one or several turns of a fragment of a high-voltage line conductor. Each of the channels contains a light source, a fiber light guide, a collimator, a first polarizer, a magnetically sensitive element, for example, in the form of a glass prism of the BP-180° type, a second polarizer, a light collecting lens, a second fiber light guide, a photodetector connected to an electronic unit with level generators for the constant component and alternating component of the photodetector signal, microprocessor, indicator of current measurement results, interface board.

Unlike known devices, in each channel of the proposed multichannel optical current meter, the second polarizers are made in the form of Wollaston prisms, the polarization planes of the linearly polarized beams separated by them make angles of ±45° with the transmission planes of the first polarizers, additional fiber light guides are installed in the focal planes of the light-collecting lenses, at the output of which additional photodetectors, pre-amplifiers, level shapers of constant and variable components are installed. In the coarse channel, the magnetically sensitive element is made in the form of the same BP-180° prism, but made of glass with a lower Verdet constant compared to the glass of the fine channel prism. In the electronic block of the coarse channel, a device is installed for summing the signals of photodetectors to form their common constant component, the inputs of which are connected directly to the preamplifiers, and its output is connected to the inputs of an additional microprocessor connected to a clock pulse generator to control the torque and the number of current measurements during the period of alternating current oscillation networks.

Figure 1 shows a block diagram of the proposed multichannel optical current meter for high-voltage networks.

Figure 2 shows a structural electrical diagram of a precise channel.

Figure 3 shows a structural electrical diagram of a rough channel.

Figure 4 shows the waveforms of the first coarse channel photodetector before a network short circuit, during the action of a shock current and during a steady-state network short circuit.

Figure 5 shows the waveforms of the second photodetector of the coarse channel at the same stages of short circuit development.

The proposed multichannel optical current meter for high-voltage networks consists of two optical independent from each other precise (Fig. 1, projection A-A) and coarse (Fig. 1, projection BB) channels in the form of two Faraday cells 1,2 installed in a single housing 3 on the upper flange 4 of a high-voltage insulator 5 inside a common solenoid 6, formed from one or several turns of a fragment of a high-voltage line conductor.

The precise channel (Fig. 1, projection AA) contains a light source installed in series, for example, an LED 7, a multimode fiber light guide 8, a collimator in the form of a lens 9, a first polarizer 10, a magnetically sensitive element, for example, in the form of a glass prism 11 of type BP -180° with a high value of the Verdet constant, for example, from TF5 glass, a second polarizer 12, a light collecting lens 13, a second fiber light guide 14, a photodetector 15 connected to the electronic unit 16.

The rough channel (Fig. 1, projection BB) contains the same elements installed in series, namely a light source 17, a fiber light guide 18, a collimator 19, a first polarizer 20, a magnetically sensitive element in the form of a prism 21 type BP-180° made of glass with a low value of the Verdet constant, for example K8, a second polarizer 22, a light collecting lens 23, a second fiber light guide 24, a photodetector 25 connected to the electronic unit 26.

In each channel, the second polarizers 12 and 22 are made in the form of Wollaston prisms, the polarization planes of the linearly polarized beams separated by them make angles of ±45° with the transmission planes of the first polarizers 10 and 20.

In the focal planes of the light-collecting lenses 13, 23, the ends of additional fiber light guides 27 and 28 are installed.

After the additional light guide 27, an additional photodetector 29 is installed, connected to the electronic unit 16, and after the additional light guide 28, an additional photodetector 30 is installed, connected to the electronic unit 26.

Due to the fact that Wollaston prisms 12 and 22, in addition to dividing the collimated light beams into two, simultaneously perform the functions of polarizers with mutually orthagonal planes of polarization of the light beams emerging from them, both the fine and coarse optical channels should be considered separately in turn just like the air ducts.

In this regard, the electronic unit of the precise channel 16 (Fig. 2) also contains two identical electronic channels. One electronic channel contains a photodetector 15, a pre-amplifier 31, a resistive divider 32, a constant component signal generator of the photodetector 15, consisting of a resistor 33, a capacitance 34 and a resistive divider 35, which is connected to the first input of the microprocessor 36, as well as a level generator for the variable component of the photodetector signal 15, consisting of a capacitor 37, a resistor 38, a full-wave rectifier 39, a capacitance 40, which is connected to the second input of the microprocessor 36.

Another electronic channel of the electronic block of the precise channel 16 contains a photodetector 29, a pre-amplifier 41, a resistive divider 42, a constant component shaper of the photodetector signal 29, consisting of a resistor 43, a capacitance 44 of a resistive divider 45, which is connected to the third input of the microprocessor 36, as well as a level shaper variable component of the signal of the photodetector 29, consisting of a capacitor 46, a resistor 47, a full-wave rectifier 48, a capacitance 49, which is connected to the fourth input of the microprocessor 36. The microprocessor 36 is connected to the interface board 50, which is connected to the digital indicator 51. The electronic unit 16 contains a power source 52 and a thermal relay with a miniature heater 53.

The electronic block of the rough optical channel 26 (Fig. 3) also contains two identical electronic channels.

One electronic channel contains a photodetector 25, a pre-amplifier 54, a resistive divider 55, a constant component signal generator of the photodetector 25, consisting of a resistor 56, a capacitance 57 and a resistive divider 58. The divider 58 is connected to the first input of the microprocessor 59. The level generator of the variable component of the photodetector signal 25 consists of a capacitor 60, a resistor 61, a full-wave rectifier 62, a capacitance 63, which is connected to the second input of the microprocessor 59.

Another electronic channel of the electronic block of the coarse channel 26 contains a photodetector 30, a pre-amplifier 64, a resistive divider 65, a constant component generator of the photodetector signal 30, consisting of a resistor 66, a capacitance 67 and a resistive divider 68 connected to the third input of the microprocessor 59.

The level generator of the variable component of the signal of the photodetector 30 consists of a capacitance 69, a resistor 70, a full-wave rectifier 71, a capacitance 72, which is connected to the fourth input of the microprocessor 59. The microprocessor 59 is connected to the interface board 73 and to the digital indicator 74. The electronic unit 26 contains its own power source R ₅₅ and R ₆₅ , and the output of the adder 76 (capacitance C ₇₇ ) is connected to the inputs “input 2” and “input 4”. A clock generator 79 is connected to the microprocessor 78 to control the torque and the number of samples (the number of current measurements during the period of oscillations of the alternating current of the network). Interfaces 80 and 81 are connected to the outputs of microprocessor 78 for broadcasting measured currents using MEK 61850-9-2 communication protocols.

For closed substations and for other power plants located in rooms with normal climatic operating conditions, a design option is provided in which light sources 7, 17 (Fig. 1), photodetectors 15, 25 and 29, 30, as well as electronic units 16 and 26 located inside the common stand 82.

For open substations with harsh climatic operating conditions, a design option is provided in which long-length fiber light guides 8, 14, 27, 18, 24, 28 (Fig. 1) are used, and light sources 7, 17, photodetectors 15, 25 and 29, 30, as well as electronic units 16 and 26, are located in a separate cabinet (not shown in the drawings), inside which a temperature sensor, relay and heater are located.

A multichannel optical current meter for high-voltage networks works as follows.

Precise channel operation.

Light from the source 7 (Fig. 1), which is usually partially polarized, is supplied to the focal plane of the collimating lens 9 using a multimode fiber light guide 8.

Multimode optical fiber 8 for partially polarized light is an ideal depolarizer. Therefore, the diverging beam of light emerging from the multimode fiber light guide 8 can be represented by the Stokes vector, which characterizes unpolarized light

)

Where - light intensity of source 7;

- light transmission coefficient of the optical fiber taking into account vignetting.

The collimating lens 9 converts the diverging beam of light into a collimated one. After light passes through the polarizer 10, the light becomes linearly polarized with a polarization azimuth of 0°, as can be seen from the multiplication of known matrices [8]:

If no current flows through the fragment of conductor 6, and there are no mechanical or thermal loads on prism 11, then in terms of its effect on polarized light, prism 11 can be represented as a matrix of an isotropic substance [8]:

- the overall reflectance of the mirror surfaces of the prism is 11 for horizontally polarized light.

Therefore, after collimated linearly polarized light passes through prism 11 four times, the polarization state of the light will not change, which can be represented by the Stokes vector

Next, the collimated linearly polarized beam of light passes through the Wollaston prism 12 and is divided by it at an angle of 2β into two linearly polarized beams of light, which can be represented by Stokes vectors , according to the equations:

From equations (30) and (31) it is clear that ideally the light beams separated by a Wollaston prism 12 have the same intensities

and, accordingly, the azimuths of linear polarization are +45° and -45° with respect to the first polarizer 10. Lens 13 focuses light beams separated by Wollaston prism 12, respectively, onto the ends of fiber light guides 14 and 27, which depolarize the light and transmit it to photodetectors 15, 29, respectively. At the output of photodetectors 15, 29, with their equal sensitivity, we obtain electrical signals of equal amplitude, proportional to the constant component of light

However, in practice, when transmitting light in optical fibers 14, 27, additional and unequal vignetting may occur at their ends, and the sensitivities of photo receivers 15, 29 may differ from each other.

Therefore, during the adjustment process, using resistors R ₃₂ and R ₄₂ (Fig. 2), the signals (33) are equal.

If the prism 11 (Fig. 1) is not affected by temperature gradients or mechanical loads, and an alternating current frequency from the network flows through a fragment of the network conductor 6 then prism 11, located in the longitudinal magnetic field of conductor 6, in terms of its effect on linearly polarized light, can be represented by a rotator matrix [8]

Where - maximum amplitude of the angle of rotation of the plane of polarization of light;

- the total reflection coefficient of the mirror surfaces of the prism 11 for one horizontal component of linearly polarized light;

- network frequency (50 Hz).

In this case the light intensity can be found from the equations

Namely:

Light streams photodetectors 15 and 29 are converted into electrical signals, respectively

After the pre-amplifier 31 (Fig. 2), the electrical signal (39) is divided into a constant component and variable component The constant component shaper, using resistor R ₃₃ and capacitance C _34, integrates (smoothes) the signal of the photodetector 15 and, using a divider R _35, in the form of a potential supplied to “input 1” of microprocessor 36.

The variable component of the signal (39) is filtered using a differential circuit (capacitance C ₃₇ and resistor R ₃₈ ), detected by a full-wave detector 39, smoothed by capacitance C ₄₀ and in the form of a potential supplied to “input 2” of microprocessor 36.

After the preamplifier 41, the electrical signal (40) is also divided into a DC component and variable component

The constant component shaper, using resistor R ₄₃ and capacitor C _44, integrates (smoothes) the signal of the photodetector 29 and, using divider R _45, in the form of a potential supplied to “input 3” of microprocessor 36.

The variable component of the signal (40) is filtered using a differential circuit (capacitance C ₄₆ and resistor R ₄₇ ), detected by a full-wave detector 48, smoothed by capacitance C ₄₉ and in the form of a potential supplied to “input 4” of microprocessor 36.

Microprocessor 36 calculates the potential ratio

finds their arithmetic mean

calculates the angle of rotation of the plane of polarization

multiplies by a constant design factor K and finds the value of the measured current

In real operating conditions of the proposed device, glass prism 11 (Fig. 1), as a rule, experiences small thermal or mechanical loads, as a result of which birefringence occurs in the glass of prism 11 and prism 11 can be considered as an equivalent phase plate with an arbitrary orientation of the main axes ϕ , introducing a phase difference δ between the components of polarized light [8]. In the general case, this equivalent phase plate can be represented by a table matrix of a general form [8]. But considering that the photoelastic coefficient of glass is small and, as studies have shown, the phase difference introduced by loads is δ ≤ 5°, the glass prism matrix 11 can be represented in a simplified form

In this case, in the absence of current in solenoid 6, after horizontally polarized light passes four times through prism 11, at its output the light will be characterized by the Stokes vector

Consequently, in the absence of current in solenoid 6 after Wollaston prism 12, equality (32) is violated and the intensity , arriving at photodetectors 15, 29 will not be equal to each other.

Namely

Where - light transmission coefficients by fiber light guides are 14, 27, respectively.

If there is current in solenoid 6, there will be electrical signals at the outputs of pre-amplifiers 31 and 41 (Fig. 2)

Expressions (50), (51) can be represented in the form

Where

- potentials on capacitors C ₄₀ and C ₄₉ (Fig. 2). Microprocessor 36 calculates potential ratios which are proportional to the variable components, relatively constant components

and their arithmetic mean values

the magnitude of the rotation angle of the plane of polarization as well as the value of the measured current (45)

From expressions (48), (49) and (52), (53) it is clear that due to the normalization of the signals, the relative values of Q1 and Q ₂ do not depend on the level of light intensity and from the total light transmittance but significantly depend on the phase difference δ, and if Q1 decreases by the amount then Q ₂ increases by approximately the same amount.

Consequently, after calculating the arithmetic mean value of Q, the influence of the phase difference δ on the ratio Q of the variable component to the constant component of the signals of photo receivers 15, 29 will be practically compensated.

For example, with ϕ = ±22.5° (worst case) and δ = ±5° (real values), the errors in the values of Q _t and Q ₂ separately according to formulas (51), (52) will be

After finding the arithmetic mean

the error is only +0.0005168, that is, 45 times less.

For clarity, we express the errors under consideration in terms of currents.

Let us assume that at a current in the network i = 500 A and at an air temperature t = 20 ° C the value is measured then the design factor In this case, the error -0.0222255 in determining Q ₁ is proportional to the error Δα = 0.63676 in measuring the current Δi ₁ = 0.63676⋅78.553=50 A (10.0%), the error in determining Q ₂ is proportional to the current Δi ₂ = 0.66638⋅78.552=52.34 A (by about 10.5%). And after finding the arithmetic mean value Q = 0.5(Q ₁ +Q ₂ )

That is, after finding the arithmetic mean value Q, the error in measuring current i decreases by 45 times! The results of precise current measurements are transmitted to external devices in RS-485 or RS-422 code via interface 50 and indicated by a digital indicator 51. Thus, the precise channel of the proposed device provides accurate measurements of alternating current in high-voltage networks for commercial electricity metering no worse than ±0.2 % even when exposed to extreme temperatures.

The work of the rough channel.

Structurally, the optical-mechanical part of the coarse channel (Fig. 1, projection B-B) differs from the precise channel only in that the prism 21 (Fig. 1) is made, for example, of light crown, whose Verdet constant is approximately four times less, than, for example, TF5 glass prism 11 precise channel. Therefore, its operating principle is similar to that of a precision channel. Namely. Light from the source 17 is fed through a multimode fiber light guide 18 to the focal plane of the collimating lens 19. The divergent beam of light emerging from the fiber light guide 18 is converted into a collimated beam by the lens 19, passes through the polarizer 20 and becomes linearly polarized with a polarization azimuth of 0°.

If no current flows through the fragment of conductor 6, then after a collimated linearly polarized beam of light passes through prism 21 four times, the state of light polarization will not change (29), and the light beams separated by Wollaston prism 22 have the same intensities (32). Lens 23 focuses light beams separated by Wollaston prism 22 onto the ends of fiber light guides 24, 28, which transmit them to the corresponding photodetectors 25 and 30. In principle, with equal sensitivities of photodetectors 25 and 30 and the absence of light vignetting at the ends of fiber light guides 24, 28 at the outputs of photodetectors 25, 30 we obtain electrical signals of equal amplitude (33). But in practice, at the output of photodetectors 25, 30, their signals may turn out to be unequal. Therefore, during the adjustment process, using resistors R ₅₅ and R ₆₅ (Fig. 3), the potential equality of U ₁ and U ₂ (33) is established.

If alternating current flows through a fragment of conductor 6 (Fig. 1) then prism 21 in terms of its effect on polarized light can be represented by a matrix

7)

Where - coefficient taking into account the reduction of the Faraday effect in the glass of the prism 21 compared to the glass of the prism 11 of the precision channel;

- Verdet constants of prism glasses 21 and 11, respectively;

- maximum amplitude of the angle of rotation of the plane of polarization of light by prism 11;

- the total reflection coefficient of the mirror surfaces of the prism 21 for one horizontal component of linearly polarized light;

- network frequency (50 Hz). In this case, the intensity of the separated light beams after the Wollaston prism 22 will change according to the laws

and electrical signals on resistors R ₅₅ and R ₆₅ , respectively, according to the laws

After the pre-amplifier 54 (Fig. 3), the electrical signal (60) is divided into a constant component U ₀ and an alternating component The constant component shaper, using a resistor R ₅₆ and a capacitor C ₅₇ , integrates (smoothes) the signal of the photodetector 25 and, using a divider R ₅₈ in the form of a potential U _0, is supplied to “input 1” of the microprocessor 59.

The variable component of the signal (60) is filtered using a differential chain (C ₆₀ , R ₆₁ ), detected by a full-wave detector 62, smoothed by capacitance C ₆₃ and in the form of a potential U _~ is supplied to “input 2” of the microprocessor 59.

After the pre-amplifier 64, the electrical signal (61) is also divided into a constant component U ₀ and an alternating component

The constant component shaper (resistor R ₆₆ , capacitance C _b7 ) integrates (smoothes) the signal of the photodetector 30 and, using a divider R _68, in the form of a potential supplied to “input 3” of microprocessor 59.

The variable component of the signal (58) is filtered using a differential circuit (capacitance C ₆₉ and resistor R ₇₀ ), detected by a full-wave detector 71 and in the form of a potential supplied to “input 4” of microprocessor 59.

Microprocessor 59 calculates the potential ratio

angle of rotation of the plane of polarization takes into account constant coefficients N, K and finds the value of the measured current

But, as already noted, when considering the operation of the fine channel in real operating conditions, the glass prism 21 (Fig. 1) of the coarse channel can also experience thermal or mechanical loads, as a result of which, in the presence of current in the solenoid 6 at the outputs of the pre-amplifiers 54, 64 ( Fig.3) electrical signals will operate

Where - phase difference between the components of polarized light induced by loads in the prism 21 (Fig. 1);

- the direction of the main axes of the equivalent phase plate, which is prism 21.

In prism 21 made of K8 glass can also be considered and ϕ is an arbitrary value. At worst

Therefore, expressions (65), (66) can be represented in the form

Where - signal level at capacitance C ₆₃ (at “input 2” of microprocessor 59) (Fig. 3), proportional to the magnitude of the network current;

- the limiting value of the influence of birefringence in the glass of prism 21.

Microprocessor 59 calculates the ratios

their arithmetic averages (53)

Calculates the angle of rotation of the plane of polarization as well as the value of the measured current (64)

For example, a reference current i = 529 A flows through solenoid 6 (Fig. 1), prism 21 is made of glass K8 of the same dimensions as prism 11 of the precision channel, then the microprocessor 29 calculates:

Comparing the measured current with reference current we find that due to the presence of the Wollaston prism 22, the accuracy of current measurement in the rough channel is also within ±0.2%, which corresponds to GOST 7746-2001 regarding the requirements for current meters for commercial metering.

The above example shows that at the rated current in the network, for example, when in the exact channel that is, at the limit of the linear section of the function I = ƒ (α) (37), (38), then in the rough channel there is only and current measurement accuracy of ±0.2% is maintained up to maximum current protection

If we take into account that for relay protection GOST 7746-2001 allows the use of current meters with an accuracy class of 5P and even 10P, then for measuring short circuit currents α short _circuit ≈ 30° is permissible and, accordingly (current ratio 18.5).

That is, the presence of a coarse channel allows you to measure current with an accuracy of ±0.2% and thus duplicate the precise channel, as well as measure current with accuracy class P10 in cases where the rated current is exceeded by 18.5 times, which is sufficient for overcurrent or short-circuit currents. With the onset of the transient process during a short circuit, when a shock current occurs and aperiodic component “input 1” and “input 3” of the additional microprocessor 78 (Fig. 3) receive electrical signals proportional to the network current i, respectively, from photodetectors 25 and 30, shown in Fig. 4 and Fig. 5.

Due to the fact that the Wollaston prism 22 (Fig. 1) is a polarizer, in which the transmission planes for the beams separated by it make up angles of ±45° with respect to the transmission plane of the first polarizer 20, the light intensities I ₁ and I ₂ of the separated beams will be change in antiphase. Consequently, the electrical signals on resistors R ₅₅ , R ₆₅ and on “input 1”, “input 3” of microprocessor 78 (Fig. 3) will also change in antiphase (65), (66). So, for example, if the moment of short circuit occurs with a positive half-cycle of current i and on resistor R ₅₅ the signal increases due to the aperiodic component to level (Fig.4), then quickly drops to the level then at the same time the signal on resistor R ₆₅ decreases to the level (Fig.5) relatively due to the aperiodic component U _a , and then quickly increases to the level .

If a short circuit occurs at the moment of the negative half-cycle, then the polarity of the aperiodic component changes and the curves shown in Fig. 4 and Fig. 5 are swapped. On average, the short circuit transient process lasts no more than 0.08 seconds, and the rise time of the shock current is less than 0.01 seconds. Therefore, the measurement of the surge current and the aperiodic component is carried out at least 80 times during the period of current oscillations in the network, that is, at least 4000 times per second as follows.

Synchronizing pulses with a frequency of 4 kHz from the generator 79 (Fig. 3) are supplied to “input 5” of the microprocessor 78. With the arrival of each pulse, the microprocessor 78 calculates the ratio of Q ₁ and Q ₂ levels of the signals U ₁ (65) and U ₂ (66) received from resistors R ₅₅ and R ₆₅ , to their sum U ₀ = U ₁ + U ₂ , received from capacitance C ₇₇ , which serves as an integrator at the output of the adder 76, as well as the ratio Q ₁ Q ₂ , multiplied by a constant coefficient D. From the output of the microprocessor 78 “output 1” and “output 2” measured instantaneous current values of each sample

in RS485 code are sent to interface blocks 80, 81, converted to MEK 61850-9-2 code and transmitted to relay protection and automation devices.

The proposed device has a number of advantages compared to known devices that solve similar problems.

Firstly, the presence of Wollaston prisms in both the fine and coarse optical channels makes it possible to compensate for birefringence induced by thermal loads in glass prisms 11, 21 (Fig. 1), which significantly increases the accuracy of measurements, due to which the coarse channel can be used in as a backup for commercial metering, as well as for precise pulse measurements of surge current and aperiodic component during a short circuit.

The use of Wollaston prisms as second polarizers 12 and 22 made it possible to have two electronic channels for processing photodetector signals in the fine optical channel and four electrical channels in the coarse optical channel, two of which are the same as in the fine optical channel and two high-frequency pulsed ones for fixing the shock current at short circuit in the network. It turns out that each channel performs the functions of a separate device, which increases the reliability and accuracy of measurements.

Secondly, BP-180° glass prisms made from different types of glass, for example, TF4 and K8, but of the same geometric dimensions, are used as magnetically sensitive elements of the Faraday cell of the fine and coarse channels, which allows us to unify the production of two-channel Faraday cells.

Thirdly, in order to maintain the level of the constant component of the signals of photodetectors 25, 30 (Fig. 3) during the transient period of a network short circuit, the proposed device contains an adder 76, at the output of which a capacitance C ₇₇ is installed. This allows you to achieve two important goals:

- stabilize the reference signal U ₀ even under shock current conditions, thereby making accurate measurements of the aperiodic component and shock current;

- compensate for the influence of the temperature gradient when linear birefringence occurs in the glass of the prism 21 (Fig. 1).

The proposed device will find wide application in the power industry, especially in networks of 110 kV and above.

INFORMATION SOURCES

1. Great Soviet Encyclopedia, vol. 26, p. 166.

2. Godzhaev N.M. Optics. Textbook for universities. - M.: Mir, 1965, p. thirty.

3. Landsberg G.S. Optics: 5th ed., revised. and additional - M.: Nauka, 1976.

4. Gubin V.P., Starostin N.I., Prizhiyalovsky et al. Fiber-optic transformers of electric current // Photonics. - Vol. 12 No. 7 and No. 8, 2018.

5. RF patent (utility model) No. 171401, G01R 15/24.

6. RF patent No. 2752341, G01R 15/24.

7. Frokht M.M. Photoelasticity. Gosizdat. M., L., 1950.

8. Vasiliev B.I. Optics of polarizing devices: “Mechanical Engineering”, M. 1969.

9. GOST RMEC 61850-9-2.

Claims (1)

Multichannel optical current meter for high-voltage networks, consisting of two independent of each other, fine and coarse optical channels in the form of Faraday cells, installed in a single housing on the upper flange of a high-voltage insulator inside a common solenoid formed from one or several turns of a fragment of a high-voltage line conductor, each of the channels contains a light source, a fiber light guide, a collimator, a first polarizer, a magnetically sensitive element, for example, in the form of a glass prism BP-180°, a second polarizer, a light collecting lens, a second fiber light guide, a photodetector connected to an electronic unit with level shapers installed in series constant component and variable component of the photodetector signal, microprocessor, indicator of current measurement results, interface board, characterized in that in each channel the second polarizers are made in the form of Wollaston prisms, the polarization planes of the linearly polarized beams separated by them make angles of ±45° with the transmission planes of the first polarizers , additional fiber light guides are installed in the focal planes of the light-collecting lenses, at the output of which additional photodetectors, pre-amplifiers, level shapers of constant and variable components are installed; in the coarse channel, the magnetically sensitive element is also made in the form of a BP-180° prism, but from glass with a lower value Verdet constant in comparison with the glass prism of the fine channel, in the electronic unit of the coarse channel there is a device for summing the signals of photodetectors to form their common constant component, the inputs of which are connected directly to the preamplifiers, and its output is connected to the inputs of an additional microprocessor connected to a clock generator for control torque and number of current measurements during the period of oscillation of the alternating current of the network.

Images