Patent Grant
10302411
The invention relates to a method for measuring the interference contrast in an interference-based, closed-loop, phase-modulating optical sensor device. The invention also relates to such an interference-based, closed-loop, phase-modulating optical sensor device.
In this context, an “interference-based, closed-loop, phase-modulating optical sensor device” is a sensor device that is adapted to measure a measurand by means of the optical interference between a first and a second wave, at least one of which has passed a sensing element. It comprises a phase modulator for changing the phase shift between the two waves as well as a light sensor for determining a measured signal from the interference of the waves. The feedback controller maintains the measured signal at a given reference value by controlling the phase modulator.
A key issue for wide-range optical DC voltage sensing based on interferometric measurements is how to uniquely determine the voltage-induced electro-optic phase shift across many 2π periods. To that end, we have demonstrated in a previous patent application [1] an interference-contrast-based method, which works well for a voltage range beyond ±500 kV.
Modulation phase detection (MPD) [2] is a powerful interrogation technique to measure phase shift in an optical system. With this technique, a high-frequency phase modulation waveform is superposed onto the phase shift to be measured. This phase shift can then be derived from the measurement of the polarimetric optical response waveform. A great advantage of the MPD technique is that, because the phase shift of interest is calculated from the shape of the polarimetric response waveform rather than an absolute signal amplitude, the measurement (and particularly its zero-point stability) is in principle not affected by optical power fluctuation or loss variation. Compared to traditional polarimetric phase measurement techniques, MPD enjoys superior accuracy, DC stability and robustness.
Typically, MPD sensors are implemented in a reciprocal configuration, either in the form of a Sagnac interferometer or in a reflective form, in order to cancel phase shifts from additional birefringent elements in the system (such as PM fibers or the phase modulator crystal), which may drift slowly, e.g. with temperature change or mechanical disturbance. Such reciprocal MPD schemes are usually termed “non-reciprocal phase modulation”, because the non-reciprocal imposed phase modulation and the phase shift to be measured are the only phase shifts that remain in the measured signal. The reciprocal MPD schemes have seen wide adoption in fiber-optic gyroscopes [2], and later in fiber-optic current sensors (FOCS) [3, 4]. Recently, we have shown that a non-reciprocal form of MPD, namely differential MPD [5], can also be used to achieve similar performance, which is of great benefit for applications requiring a bulk optic sensing element.
The MPD technique can be utilized for optical voltage sensing [6], where the differential electro-optic phase shift between two orthogonally polarized light waves in a Pockels crystal is measured to determine the applied voltage. In cases where the required voltage range is larger than the n-voltage of the sensing crystal, we have shown that one may disambiguate the phase shift periods by measuring the interference contrast in addition [1], which varies as a function of the group delay (and equivalently the phase shift) between the two polarized light waves within the coherence range of a low-coherence light source. The combination of the phase shift principal value and the interference contrast allows the unambiguous determination of the electro-optic phase shift in many 2π periods, covering a voltage range >±500 kV. As an example [1], we have shown that in an open-loop sinusoidal-wave MPD scheme, the interference contrast may be determined by measuring the DC level of the optical response.
Most commercial MPD sensors use square-wave modulation with closed-loop control and the scheme taught by [1] is not applicable to such systems.
Therefore, the problem to be solved is to provide a method and device of the closed-loop type mentioned above that allow to measure the interference contrast or to measure a value depending on this interference contrast.
This problem is solved by the method and device of the independent claims. Embodiments are given by dependent claims, their combinations and the description together with the drawings.
In particular, the method comprises the steps of:
and is so adjusted in order to keep the feedback control system operating in the said optimal condition.
Further, the method comprises the step of using said gain G for determining the interference contrast of said waves.
In another aspect, the invention relates to an interference-based closed-loop, phase-modulating optical sensor device comprising:
Such a feedback controller strives to keep the error signal zero with the gain adjusted for optimal stability and responsiveness.
Further, the control unit is adapted and structured to use the optimal gain G for determining the interference contrast of the waves.
The invention is based on the understanding that the optimal gain G is a function of the interference contrast of the waves at the light sensor, i.e. the interference contrast can be determined from the optimal gain G.
In this context, the expression “determining an interference contrast” is to be understood such that the interference contrast is determined either explicitly, or that a quantity depending on the interference contrast is determined.
Advantageously, a phase modulation ϕm between two levels ϕ±=±ϕm0+ϕf is generated. In this case, ϕf is an offset controlled by a control signal. Advantageously, ϕm0=π/2.
In this case, the amplitude of the interference signal under modulation ΔI=I−−I+ is measured, wherein I+ and I− are the interference signal levels at modulated phase levels ϕ±, respectively. This signal ΔI is used as the error signal to be kept zero by the control system.
Advantageously, the sensor device comprises a voltage driver to the phase modulator, generating the above-mentioned phase modulation.
In an advantageous embodiment, the optimal gain G can be used for calculating said measurand.
In particular, in interferometric optical sensor devices the measured signal is often a quasi-periodic function of the measurand. In this context, “quasi-periodic” and “quasi-periodicity” are understood to describe that the measured signal is a periodic function of the measurand, with the exception that its oscillation amplitude (i.e. interference contrast) varies, in particular due to the limited coherence length of the light source. Hence, the optimal gain G can be used for a disambiguation of the quasi-periodicity of the measured signal, as e.g. taught in [1].
In a particularly advantageous embodiment, the invention can be used for sensing a voltage. In that case, the said measurand is said electrical voltage. The sensing element is located in an electrical field generated by the electrical voltage. And the at least one refractive index of the sensing element depends on the electrical field, i.e. the sensing element is an electro-optic sensing element, in particular a crystal exhibiting the Pockels effect wherein the linear birefringence of the material is a linear function of the applied electrical field.
The invention is particularly suited to measure high voltages of at least 100 kV, in particular of at least 500 kV. It can advantageously be used for DC voltages. But it can also be used for measuring AC voltages.
The invention is ideal for various HVDC applications, including HVDC Light, HVDC Classic, and offshore DC applications, e.g. in a direct-current gas-insulated substation (DC GIS).
In another advantageous embodiment, the invention can be used for sensing an electrical current. In that case, the measurand is said electrical current. The sensing element is located in a magnetic field generated by the electrical current. And at least one refractive index of the sensing element depends on the magnetic field, e.g, the sensing element is a magneto-optic sensing element, in particular a fiber wound around a conductor carrying the current, wherein the circular birefringence is a linear function of the magnetic field.
In yet another advantageous embodiment, the invention can be used for sensing acceleration in an optical gyroscope.
The gain can e.g. be determined by intermittently measuring a response of the measured interference signal amplitude to a variation of the control signal ϕf. The term “intermittently” is in this case advantageously to be understood as a periodic or non-periodic occurrence of this determination step, while a normal operation of the control loop takes place between the determination steps.
In another advantageous embodiment, the invention relates to a method, wherein the determined optimal gain G is used to correct systematic errors, in particular due to drift and/or component misalignment, in the sensor device, in particular in optical voltage sensing.
The invention shows that it is possible to use the feedback loop gain of e.g. a closed-loop square-wave MPD system to determine the interference contrast, without additional hardware. The ability to measure the interference contrast e.g. enables to extend the unambiguous phase shift measurement range beyond 2π, which is of vital importance for some applications, such as optical DC voltage measurement. The invention can also be used to monitor PER (polarization extinction ratio) variation and to correct for corresponding changes in the current measurement scale factor in a FOCS system.
The invention demonstrates that, without any additional hardware, the interference contrast can be readily determined in a MPD detection scheme from the loop gain, which is continuously measured in the current ABB system for operation regulation.
In conjunction with the interference-contrast-based period disambiguation method e.g. as described in [1], this invention provides the core technology enabling optical DC voltage sensing in a wide voltage range.
Particular advantages of the closed-loop square-wave modulation phase detection (MPD) scheme are the excellent scale factor and zero point stabilities, both of which are needed for the accurate measurement of a DC voltage. The invention demonstrates that, without any additional hardware, the interference contrast can be readily determined in a MPD detection scheme from the loop gain, which is continuously measured by the control system for operation regulation. In conjunction with the interference-contrast-based period disambiguation method, this invention provides the core technology enabling optical DC voltage sensing in a wide voltage range.
The measurement of the interference contrast with closed-loop MPD is also beneficial to the further improvement of sensing systems, where the measurement scale factor is sensitive to the polarization extinction ratio (PER) of the fiber link. PER changes, e.g. due to changes in fiber connector properties, are directly reflected in the interference contrast; therefore, monitoring the interference contrast provides a way to control and compensate such systematic errors of scale factor variations, which is important for maintaining high measurement accuracy over long time and a wide temperature range.
The invention will be better understood and objects other than those set forth above will become apparent when consideration is given to the following detailed description thereof. This description makes reference to the annexed drawings, wherein:
The design of a reflective optical voltage sensor using the closed-loop MPD phase shift detection scheme is shown in
The device comprises an MPD optoelectronics module 1, which contains a low-coherence light source 2. The light from light source 2 is fed through a polarizer 3, a phase modulator 4 and into both polarization directions of a polarization maintaining (PM) fiber 5.
A collimator 6 sends these waves from PM fiber 5 through a 45° Faraday rotator 7 and into a first end of Pockels effect crystal 8, which is being exposed to the electrical field generated by the voltage to be measured. At the second end of the crystal, the waves are reflected by reflector 9 and sent back through the components 3-8.
A beam splitter 10 is arranged between light source 2 and polarizer 3 and sends at least part of the returning light into a light sensor 11.
The device of
Faraday rotator 6 rotates both linear polarizations from PM fiber 5 by 45° before they propagate along the electro-optic axes (principal refractive index axes) of sensing crystal 8. The reflected waves pass through Faraday rotator 6 again, further rotating the polarizations by 45° in the same direction, thereby making a combined 90° rotation from the input polarizations, which is equivalent to a swap between the two orthogonal linear polarizations.
Phase modulator 4 is operating at a frequency adapted to the round-trip time of the light waves traveling from modulator 4 to reflector 9 and back, such that the relative phase shifts induced by the phase modulator in the two passes of the waves through it have opposite signs and the phase modulations in the two passes are consequently added due to the swap in light polarizations.
The coherence length of light source 2 is advantageously between 5·λ0 and 100·λ0, with λ0 being the center wavelength of the light source, in order to obtain a good variation of interference contrast A when changing the phase between the two polarizations by a few multiples of 2π.
With such a low-coherence light source 1 and a properly selected group delay offset τ0 (e.g. by means of a birefringent element), the voltage measurement range can be set such that the interference contrast varies strongly and monotonically with the applied voltage (see
In the MPD system, a phase modulation ϕm(t) is added in phase modulator 4 onto a phase shift of interest (ϕ0), i.e. here the phase shift caused by the measurand, and the polarimetric optical response is measured in the returning waves after they have passed polarizer 3. This response is the intensity I of the interfering waves
where I0 is the total light power, and A is the interference contrast.
In a closed-loop square-wave modulation system, the modulation ϕm(t) is advantageously a square wave alternating between two levels ϕ±=±π/2+ϕf, where ϕf is a dynamically controlled feedback phase [2]. ϕf can be considered to be the control signal that feedback controller 12 uses for controlling phase modulator 4.
The corresponding measured polarimetric response at the two modulation levels are
The feedback-loop error signal is the difference between the two polarimetric response levels
ΔI=I−−I+=ΔI0 sin(ϕ0+ϕf)
The system is operated at a point that the error signal ΔI is maintained at 0, which corresponds to ϕ0+ϕf=0. Therefore, at the operating point, the phase shift offset to be measured ϕ0 is simply the opposite of the dynamically controlled feedback phase shift ϕf. The expected loop gain in the designed voltage range is plotted in
The optimal loop gain of the feedback loop in feedback controller 12 is found to be the derivative of the feedback phase shift with respect to the error signal at the set point, i.e.
As an important control parameter for the control loop, the optimal loop gain G is determined by control unit 13 e.g. in the following way. Intermittently, in particular periodically (but possibly also in non-periodic manner), the response of the measured signal, in particular the response of the error signal ΔI, to a variation of the control signal ϕf applied to the phase modulator 12 is measured.
For example, a phase modulation sequence [ϕ++, ϕ−+, ϕ+−, ϕ−−] is invoked, whereby small phase deviations ∈± (typically ∈+=∈−) are added to or subtracted from the normal phase modulation ϕ± (see
Here, ϕf0 corresponds to the control signal ϕf that presently sets the error signal ΔI to zero.
Correspondingly, the polarimetric responses at the four modulation levels are
From these, we can calculate two error signals corresponding to ∈± as
ΔI+=I−+−I++=AI0∈+
ΔI−=I−−−I−+=−AI0∈−
The optimal loop gain can then be calculated as
The calculated optimal loop gain G is then used for setting the amplitude of the feedback phase shift.
From the results above, one sees that the loop gain value G measured in the MPD system is related to the interference contrast A and the total light power I0. Conversely, one can calculate the interference contrast A as
In the modulation scheme described above, the total light power I0 is just twice the optical response level I± at the operating point ϕ0+ϕf=0. Therefore, practically it can be measured as the DC offset of the optical response I±, while the AC oscillation amplitude ΔI is used as the error signal to control the feedback loop.
It should however be noted, that in some applications, such as period disambiguation of the phase shift (e.g. for optical DC current sensing), it is not necessary to determine the absolute value of the interference contrast A, but it would be sufficient to follow its relative change. In such cases, one may record the relative variations of the loop gain G and the total optical power I0, without determining their absolute scales. For example, one may set the values of the loop gain and the total optical power measured at a certain moment (e.g. at the onset of operation) to a constant, and evaluate all subsequent measurements accordingly. If the optical power is maintained at a constant level, its measurement may not be needed at all.
With the interference contrast A measured, one can then determine, in which period the phase shift is residing, and convert the measured phase shift principal value to a corresponding unique full-range value, following the procedure that is described in [1], the disclosure of which is herewith enclosed in its entirety. For an optical voltage sensor, an unambiguous voltage value can then be calculated from the full-range phase shift.
In a FOCS sensor, the interference contrast calculated from the loop gain can likewise be used to extend the measurement range, which is, without history tracking, limited to 2π phase shift in current products.
Furthermore in some FOCS sensors, the current measurement scale factor is influenced by the polarization extinction ratio of the fiber link between the sensor head and the optoelectronics, which in turn directly affects the interference contrast. Therefore, the measurement of the interference contrast can be used to monitor the variation of the polarization extinction ratio, and to compensate the corresponding variation of the current measurement scale factor. Hence, also in this case, the interference contrast can be used to determine the measurand in a more precise manner.
In general, the determined gain G can be used to correct systematic errors, in particular systematic errors due to drift and/or component misalignment, in the sensor device, in particular if the systematic errors affect the interference contrast.
Notes:
Electro-optic crystal 5 can be replaced by any other electro-optic element that exhibits a birefringence depending on the applied electrical field.
In another application for measuring the current in a conductor, the sensing element can e.g. be a fiber wound around the conductor. In that case, a quarter-wave retarder is used instead of Faraday rotator 7.
Advantageously, the optical sensor device has a reciprocal configuration, e.g. in the form of a Sagnac interferometer or in a reflective form, where the phase shifts, in particular from PM fiber 5 and/or sensing element 8, are canceled. In that case, the phase modulator 4 is advantageously operated, as mentioned, to apply opposing phase shifts between the two waves at a frequency corresponding to a round trip from the modulator 4 to reflector 9 and back.
When the device is a voltage sensor, the techniques shown herein allow the compensation of systematic errors for an optical DC voltage sensor with measurement range >±500 kV to achieve an accuracy <0.2%.
The techniques are ideal for applications in HVDC air-insulated systems, HVDC cables, and DC gas-insulated switching (GIS) systems. Such GIS may be filled with dielectric gas based on SF6 or alternative gases, such as fluoroketones or fluoronitriles, preferably in mixtures with a background gas, such as e.g. selected from: nitrogen, carbon dioxide and oxygen.
They allow to remove the periodwise measurement ambiguity in a wide range (>16 phase periods, or equivalently >±500 kV).
They provide a means for the measurement of interference contrast in a closed-loop modulation phase detection system for ambiguity removal.
And they make it also possible to enhance the accuracy of fiber-optic current sensors (monitor and control of polarization extinction ratio, PER) by correction of variations in scale factor during operation
While presently preferred embodiments of the invention are shown and described, it is to be distinctly understood that the invention is not limited thereto but may be otherwise variously embodied and practiced within the scope of the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 16187126 | Sep 2016 | EP | regional |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5477323 | Andrews | Dec 1995 | A |
| 5513003 | Morgan | Apr 1996 | A |
| 6078706 | Nau | Jun 2000 | A |
| Number | Date | Country |
|---|---|---|
| 102981136 | Mar 2013 | CN |
| 1231454 | Aug 2002 | EP |
| 2015124678 | Aug 2015 | WO |
| Entry |
|---|
| European Patent Office, Search Report in corresponding application 16187126.4, dated Feb. 14, 2017, 5 pages. |
| Number | Date | Country | |
|---|---|---|---|
| 20180066932 A1 | Mar 2018 | US |
