KR0173492B1 - Signal handling method and circuit of interference type optical fiber sensor - Google Patents

Abstract

The present invention discloses a signal processing method and circuit of an interference type optical fiber sensor. The method is a signal processing method of an interference type optical fiber sensor having a length difference of L, wherein the sensor signal is sampled at a first, second or first, second, or third time after a laser modulation pulse is applied. Sample the sensor output obtained by transferring the optical pulse generated by applying 2-3 kinds of laser modulation pulses with narrow pulse width and specific amplitude difference to the interferometer, and using the sampling data at two time points Performing a step of increasing / decreasing the phase of the light source, setting a reference value of the light intensity, and counting the number of half fringes. Obtaining the sampling data R _a ', R _b ', R _c 'from, increase or decrease the coefficient according to the increase or decrease of the size of the sampling data R _a ', R _b ', R _c ' same Subdividing the curve in Lindsay and consists of a step, the amount of change in the quantity to be measured to measure the change amount of the measured amount in step to obtain a measurement signal of the interferometer by dividing by the sensitivity of the sensor. Therefore, optical sensors using FP interferometers with relatively short pore lengths, as well as optical sensors using other types of interferometers, can be put to practical use by minimizing the number of expensive optical components.

Description

Signal Processing Method and Circuit of Coherent Fiber Optic Sensor

1 shows a sensor using a Mach Zehnder (MZ) type interferometer.

2 shows a sensor using a Fabry-Perot (FP) interferometer.

3 is a graph showing the response of the interferometer to the phase shift of light induced by the measured amount.

4 shows a mirror-mounted optical fiber FP interferometer of the present invention.

5 is a schematic block diagram of a signal processing circuit of the interference type optical fiber sensor of the present invention.

6 is a graph showing the change in laser frequency with time after the start of the modulation pulse of the laser diode.

7 is a graph showing the change in reflectance of the interferometer with time after the start of the modulation pulse of the laser diode.

8 shows a characteristic graph of the sample output of the normalized sensor signal.

FIG. 9 is a graph showing sample data of one of a signal string reflected from two interferometers, a sampling pulse string, and a data line of an analog / digital converter as an optical signal at an input terminal of the analog / digital converter shown in FIG.

10 is a graph showing the relationship between the temperature and the number of half fringes.

The present invention relates to a signal processing method and circuit of an optical fiber sensor, and more particularly, to a signal processing method and circuit of a sensor using an optical fiber interferometer.

The interference type optical fiber sensor is a sensor that applies the principle of detecting the change of optical path by the interferometer after changing the measured amount into the change of the length of the optical fiber or the change of the refractive index. Mach-Zehnder (MZ), Michelson , Fabry-Perot (FP) or Sagnac type are mainly used. Using the wave interference phenomenon, it is possible to measure the phase shift by the optical path length difference (OPD) of ~ 1μrad and the advantage that can be obtained by using light, especially low loss, multiplexing ability, electromagnetic interference and Use resistance against corrosion.

1 shows a sensor using an MZ-type interferometer.

The MZ type interferometer shown in FIG. 1 is composed of a laser diode 1, optical couplers 2 and 3, a photodetector 4, a sensing fiber 5, and a reference fiber 6.

In the case of the MZ type interferometer, the signal optical fiber is exposed to the object to be measured to induce a change in the refractive index of the optical fiber or an expansion / contraction of the length due to the change of the measured amount, and thus the The whole section acts as a sensing section. At the same time, in order to isolate the reference optical fiber from the influence of the measured quantity, it is necessary to connect the length of the measurement system and the sensor element with two long optical fibers, in which case the lead portion and the sensing portion for guiding the sensing unit to the measurement target area There is a problem that the limit of the interval is not clear.

2 shows a sensor using a Fabric-Perot (FP) interferometer.

The FP interferometer shown in FIG. 2 is composed of a laser diode 10, an optical coupler 20, a mirror 30, and an optical fiber 40.

In the Fabry-Perot (FP) interferometer, the sensing section is clearly divided into the section between the mirrors 30, and when used in the reflection mode, the connection between the measurement system and the sensor element can be connected with one optical fiber. It has the advantage of using the minimum number, simple and small size, and especially has the advantage that it is not affected by the lead wire for the light phase and polarization effect. Therefore, when the cavity length is sufficiently short, the signal weakening due to polarization may not be considered.

The response of the interferometer to the phase shift of the light induced by the measurand is given in the form of a sine wave as shown in FIG. 3, and the sense of sensitivity is also not given in a constant form and has a sinusoidal periodicity.

In order to measure with high sensitivity and wide equivalence range, the sensitivity of sensor element should be kept constant to avoid measurement error caused by signal weakening.

These signal processing methods can be broadly classified into a homodyne method and a heterodyne method. The heterodyne method requires a large volume and power consumption, and a phase modulator or frequency shifter, which causes many problems such as cost and complexity, including an optical component alignment problem. In addition, the homodyne method requires two light sources having different wavelengths or uses polarization maintaining optical fiber parts to use polarized light, and also requires special optical parts or electronic parts such as 3 × 3 combiners, In many cases, the fringe counting method is limited to the case where the phase shift amount is within 2π or very large than 2π, and the number of fringes is counted after increasing the amount of phase shift due to the change of the measured amount. I'm using.

Therefore, in the case of a sensor element having a short sensing period and a small amount of phase shift due to the measured amount, the sensitivity of the sensor is problematic, and at the same time, information on the change direction (increase or decrease) of the measured amount is not provided. In order to solve the problem of the Fringe Counting method, a signal processing method using a broadband light source such as an LED lamp or two light sources having different wavelengths has been proposed, but in the case of these signal processing methods, the sensitivity of the sensor and the dynamic range are limited or There are problems such as the price due to the use of a large number of expensive optical parts.

SUMMARY OF THE INVENTION An object of the present invention is to provide a signal processing method of an interference type optical fiber sensor which can avoid signal attenuation along with the change direction of the sensor signal by using the oscillation frequency modulation of the laser diode and sampling data of the interference signal from the sensor element. It is.

Another object of the present invention is a FP optical fiber sensor having a pore length of several millimeters (mm) to several meters (m), which is not easily processed by conventional interference sensor signal processing methods as well as general unbalanced interference optical fiber sensors. It is to provide a signal processing method of the interference type optical fiber sensor that can be used in.

Still another object of the present invention is to provide a signal processing circuit of an interference type optical fiber sensor using a digital homodyne detection method.

The signal processing method of the interference type optical fiber sensor using the digital homodyne detection method of the present invention for achieving the above object and other objects is a mirror-integrated interferometer optical fiber partitioned by two mirrors of length L optical fiber R ₁ , R ₂ In the signal processing method of the sensor, sampling is performed at the first, second or first, second, and third time points after the laser modulation pulses reflected by the mirror are applied, and sampling at two of them. Setting a phase of the sensor output signal, a reference value of the light intensity, and performing a coefficient of half fringe, using the data, taking into account the optical loss factor that is a change of the peripheral environment of the optical line in the sensor; Obtaining the sampling data R _a ′, R _b ′, R _c ′ at the second and third time points; increasing or decreasing the coefficient according to an increase or decrease in the size of the sampling data R _a ′, R _b ′, R _c ′, or On decrease And measuring the change amount of the measured amount by the Kim, and obtaining the measurement signal of the interferometer by dividing the change amount of the measured amount by the sensitivity of the sensor.

The signal processing circuit of the interference type optical fiber sensor using the digital homodyne detection method of the present invention for achieving the above another object is a mirror-integrated interferometer optical fiber sensor in which the length L optical fiber is partitioned by two mirrors having reflectivity R ₁ , R ₂ A signal processing circuit comprising: a laser diode for transmitting light to the optical fiber, a laser diode driving means for driving the laser diode, a crystal oscillator for generating a clock pulse signal of a predetermined frequency, and an output pulse signal of the crystal oscillator Pulse width modulation means for pulse width modulation, sampling pulse generation means for sampling the output pulse signal of the crystal oscillator, and a first, second time point or first time after the laser modulation pulse from the light detection means is applied. Sampling is performed at the second and third time points, and the sampling data at the two time points is By setting the reference value of the phase of the sensor output signal, the direction of increase and decrease of the light intensity, and counting the number of half fringes, and considering the optical loss factor that is the change of the peripheral environment of the optical line in the sensor. The sampling data of R _a ', R _b ', R _c 'is obtained by the following equation, and the coefficient is increased or decreased as the size of the sampling data R _a ', R _b ', R _c ' increases or decreases. And a signal processing means for measuring a change amount of the measured amount and obtaining a measurement signal by dividing the change amount of the measured amount by the sensitivity of the sensor.

The signal processing method and circuit of the interference type optical fiber sensor using the digital homodyne detection method of the present invention will be described with reference to the accompanying drawings.

4 shows a mirror-mounted optical fiber FP interferometer of the present invention.

Mirror embedded optical fiber FP interferometer length mirror, an optical fiber of the (L) been defined by the two mirror reflectivity (R _1, R ₂₎ built-FP interferometer as those coated with a dielectric (usually using a TiO ₂₎ in the diagnostic mode optical fiber section Mirrors produced by fusion splicing the uncoated optical fiber are fabricated by continuously placing two mirrors at a distance (L).

The FP interferometer used in the present invention uses the light reflected by the interferometer as an interferometer whose reflectance of the two mirrors constituting the interferometer is sufficiently smaller than one. The reflectivity and transmittance of the two mirrors are R ₁ , R ₂ and T ₁ . When T ₂ the reflectance R of the interferometer is

Is given by

In the optical fiber temperature sensor using the mirror-type FP interferometer, each FP interferometer, laser diode, photodetector, and directional coupler are used as optical components, and are composed of short-mode optical fibers connecting these optical components. In addition, a laser driving circuit and a signal processing circuit for driving the light source and signal processing are added, and FIG. 5 is a schematic block diagram thereof.

The block diagram shown in FIG. 5 includes a laser drive circuit 100, a pulse width modulation circuit 110, a crystal oscillator 120, laser diodes 130 and 140, an optical separator 150, and a mirror 160. The optical fiber 170, the signal processor 180, the sampling pulse generator 190, the analog / digital conversion circuit 200, and the counter 210 are configured.

In FIG. 5, the laser is pulse-modulated in the pulse width modulation circuit 110 by applying a pulse along with a direct current bias current as an injection current, and the modulated light output reaches the interferometer through the inductive fiber and then It is reflected by the indicated reflectance R and returned to the photodetector 140. The frequency is also modulated by frequency chirping at the same time as the output light intensity is modulated while the current pulse is applied, and the change of the optical frequency due to frequency chirping during the laser pulse modulation period is shown in FIG. It is given in the form of an exponential function of time.

Equation (2) can be expressed as follows to clarify the influence of the measured signal M of the interferometer and R on the laser frequency ν.

In formula (3), ψ ₀ is a phase difference when M = M ₀ and ν = ν ₀ ,

Where ΔM = MM ₀ Δν-ν ₀ and M ₀ and ν ₀ are the reference values (or initial values) of the measured signal and optical frequency, and ν ₀ at _a time delayed by _a given time t _a after the current pulse is applied. Is the optical frequency of.

In general, the second term of Eq. (5) is given much smaller than 1 in an optical system using a laser as a light source, so the dispersion effect of the optical fiber is negligible to zero. Equation (5) is

Fig. 7 shows the optical output of an FP interferometer with a pore length of 1 centimeter (cm) without any change in the measurement signal, when the maximum change in optical frequency due to chipping is 12 GHz, and the time constant of chipping is 470 ns. It is shown.

In order to measure the amount of optical phase change using the intensity information of the light from the interferometer, the present invention uses the frequency chirping during the modulation current pulse period of the laser diode. That is, the signal processor 180 in performing the sampling at three points after t _a, two points of t _b or t _a, t _b, t _c applied to the laser modulation pulse, and the sensor output using the one of the two data signal Determination of phase increase and decrease, intensity intensity referencing and half fringe counting are performed. Therefore, more than two samplings are performed per laser pulse, and when the laser modulation frequency is referred to as f _mod , an oscillator having a higher frequency is required.

In formula (1), R ₁ = R ₂ = R _m ≪1, T ₁ Assuming C1,

Is given by The sensor signal from the interferometer passes through the optical connection line and reaches the photodetector 140, where a change in the environmental environment applied to the optical connection line causes a change in the light phase, a change in polarization state, and a change in light loss. In the case of the FP interferometer sensor, the change of the optical phase and the polarization state is applied to both interference waves and can be ignored. Therefore, in the FP interferometer type sensor, the change in the environment around the optical line may be considered only as the light loss factor α of the section between the interferometer and the photodetector. In addition, the effect by the finite coherence length of a light source is represented by (beta). Optical frequency at the sampling time t _b and t _c (t _b, t _c is ₀ + Δν _b ν ₀ + ν Δν each

It is to be set, and, assuming ψ ₀ = 0, the sampling data at _a t R _a, R _b sampling data at t _b, t _c and the sampling data at _c is R

Is given by Using 2αR _m from formulas (9a)-(9c)

To be induced, and expression (10a) of the Δø _M - shows the relationship (10c) to the seventh FIG.

From Fig. 8, a method of counting the number of half fringes due to the change of the phase difference between the interference waves Δø _M when the measured amount is changed can be found. Table 1 shows the result. Therefore, in order to measure the magnitude of the change in the measured amount, the amount of Δø _M , that is, the amount of change in the measured amount can be measured by increasing or decreasing the coefficient according to each case summarized in the table below.

In order to detect a sign change of R 'while R' is + (or-), the amount of phase change Δø of two laser modulation pulse periods must be smaller than π / 2. Therefore, when the laser modulation frequency is f, the optical phase change rate is

Should be limited to Therefore, the maximum measurement speed (dΔM / dt) _max is obtained from equations (4) and (11).

Is given by The other interferometer structure used as the sensor element in the coherent optical fiber sensor, that is, R in MZ type, Michelson, Sagnac, etc., is given by Eq. (7) and Eq. (8) can be obtained by applying length unbalance or by phase modulation. . In addition, the condition of equation (8) may also be achieved by switching the modulation current of the laser diode, the modulation current signals processed in the similar manner as the sensor output R _b '-R _c', respectively, when a pulse is applied is It is possible.

If the variation of α is small within 10%, it is possible to take a little loss at the maximum sensing speed and to process α by treating α as a constant. However, signal processing can be performed by treating α as a constant. However, the following two methods can be considered to obtain the equation (10) from the equation (9) in the general case where the variation of α is large, and an appropriate method can be selected and used depending on the situation.

First, the two-point sampling method, which requires two samplings per given laser current pulse, uses the value (4αβR _m ) by tracking the maximum and minimum of the data sampled at two points of t _a and t _b in FIG. Equation (10) is a method that can be used when the change of the sensor signal with time is large and the change of the surrounding environment on the optical connection line is a low speed. In contrast, the three-point sampling method, which requires three samplings per modulation current pulse, is sampled at three points of t _a , t _b , and t _c in FIGS. 6 and 7 to obtain the outputs of Equations 9a through 9c. Intensity reference signal using (9a) and (9c)

After obtaining, use R _ref to derive equation (10). From Eq. (10), fringe counting along with the direction of measurement change is possible using R _a 'and R _b ' or R _b 'and R _c ' data and the table above. In addition, the characteristic curves of R _a ′ and R _b ′, R _b ′ and R _c ′ in FIG. 8 have a phase difference of about 90 degrees, and the slope of the other curve becomes maximum when the slope of one curve is near zero.

Therefore, by alternately using the two characteristic curves of FIG. 8, it is possible to measure the amount of change in the phase difference according to the change of the signal, that is, the change in the measured amount with higher sensitivity in the same fringe.

In order to demonstrate the usefulness of the new signal processing method developed for the coherent fiber optic sensor, the built-in mirror FP interferometer fiber sensor uses one fiber optic FP interferometer, fiber optic laser diode and photodetector, fiber coupler, and short mode fiber. The configuration of the optical system is shown in FIG.

Mirror built-FP interferometer was fabricated by placing the reflectivity of the optical fiber coated with TiO ₂ in the wavelength 1.3㎛ made to optical fibers and melting the non-coated TiO ₂ 2.5% up mirror with two 1㎝ interval. The laser diode with a wavelength of 1.3 µm and a threshold current of 10 mA was driven with a 15 mA current pulse with a width of 1.56 Hz with a duty cycle of 15% and a bias current of 5 mA. The operating temperature of the laser diode was a thermoelectric cooler (TEC). Fixed at 22 ° C.

Light pulses from the laser diode are coupled to the optical fiber and then reflected by the optical fiber FP interferometer via the optical fiber coupler, the intensity of the reflected light being dependent on the temperature of the FP interferometer. The reflected light including the temperature information is converted into an electrical signal through a fiber coupler and then processed by the signal processor.

After amplifying the input signal converted from the PIN receiver module into an electrical signal, the signal processor generated and sampled a sampling pulse at a time interval as shown in FIG. 9 for half fringe counting using a two- or three-point sampling method. The crystal oscillator 120 having an oscillation frequency of 20 MHz was used as the oscillator of FIG. 5 to obtain accurate time intervals t _a -t _b, t _{b, and} -t _c satisfying the relationship of 8). FIG. 9 is an optical signal at an input terminal of the analog / digital conversion circuit 200 shown in FIG. 5 to reflect one signal line reflected from two interferometers, a sampling pulse string, and a data line of the analog / digital conversion circuit 200. Graph showing sample data of.

10 shows the results obtained by reading the temperature of the thermocouple used as a reference whenever the data of the counter 210 increases while putting the optical fiber FP interferometer in the oven to gradually check the characteristics of the optical fiber temperature sensor. It was.

The sensor sensitivity S _sen for the temperature of the FP interferometer with a pore length of 1 cm was measured at 1.14 rad / K (mean value in the measuring temperature range of 20 ℃ -360 ℃), where the maximum measuring speed was

Was calculated. Therefore, for a practical temperature sensor, it is necessary to increase the resolution by increasing the sensitivity of the sensor. In order to make the resolution of temperature measurement only 0.1 ℃ by half fringe counting, the hole length of FP interferometer should be about 30cm, and the maximum measurement speed is calculated as 1/30 of 1cm from Eq. (12). In the present invention, to prove the usefulness of the proposed method, because it was limited, the experiment for the implementation of a temperature sensor having a resolution of 0.1 ℃ was not performed.

As a method for increasing the resolution of the temperature sensor, in addition to the method of increasing the pore length in the case of the FP interferometer mentioned above, the FP interferometer is planted in a material having a large output phase change for a given temperature change such as aluminum, The unbalanced MZ-type interferometer itself or a method of attaching signal optical fibers to aluminum can be used in which the length error between the optical fiber and the signal optical fiber is shorter than the coherence length (L _c ) of the light source. In addition, an FP type or MZ type sensor is also possible as the hole length is larger than L _c of the light source and a compensation interferometer is provided. However, in the case of a general FP type laser diode, the fringe can be observed even when the hole length is several centimeters, and thus the configuration of the optical fiber thermometer having a resolution of 0.1 ° C can be achieved only by fringe counting. In addition, as described above, by using the two characteristic curves of FIG. 8 to calculate the amount of change in the fringe, the measurement resolution can be further improved.

The signal processing method proposed in the present invention is also useful for signal processing of a multiplexed optical fiber sensor, and an example of signal processing performed with an additional interferometer in series with the FP interferometer of FIG. 2 is shown in FIG.

We proposed a new homodyne digital signal processing technique for signal processing of an optical fiber sensor using an unbalanced interferometer, and proved its usefulness by fabricating an optical fiber thermometer using a mirror-embedded FP interferometer and then performing a new signal processing. In this signal processing method, the frequency chirp generated during the modulation pulse time of the light source is changed to the change of the light intensity by using an unbalanced interferometer, and then the number of half fringes is counted using the data sampled at a predetermined time interval. It is proved useful by measuring 2.7 ℃ resolution from room temperature to 360 ℃ by applying to FP interferometer sensor element and laser light with wavelength 1.3㎛. Mentioned that it can be easily converted to a temperature sensor with the desired resolution, such as by increasing the hole length. In addition, the resolution of the sensor was inversely related to the maximum measurement speed, and was calculated to be 1.378 × 10 ⁵ ° C / sec with 2.7 ° C resolution.

The signal processing method of the present invention can be used for all kinds of interference type optical sensors using an unbalanced interferometer, and the error due to signal weakening due to the polarization change and the phase change caused by the change of the environment around the optical connection line between the sensor element and the sensor system. In addition, it is possible to deal with the influence of the light intensity change by the stress applied to the optical connection line. In addition, due to the absence of a proper signal processing method, many advantages of the structure, small size, simplicity, plantability, distinction of sensing area, connection of sensor element and system by a single fiber, use of the minimum number of optical parts, In spite of the realization of multiplexing which excludes the use of additional optical components in addition to the sensor element, the possibility of the practical use of the optical sensor using the FP interferometer, which has been avoided in use, has been increased.

Claims (2)

When sampling the sensor output obtained by transmitting the laser modulation pulses to the interferometer through the optical fiber connection line at two or three time points, the magnitude of the phase difference in the interferometer caused by the frequency change of the laser light source is 90 at each arbitrarily determined reference point. Determine the point of time difference by degrees, calculate the change in light loss due to the change of the optical environment connected from them, and measure the magnitude of the phase change due to the change of the measured amount along with the direction as the magnitude of the phase change amount. Performing a coefficient of the half fringe and subdividing the characteristic curve in the same fringe as necessary to measure the magnitude of the phase change amount so that the precision is greater than or equal to a predetermined magnitude; Using the frequency change of the laser light source by the temperature change during the switching of the laser driving current or the laser driving period to induce a frequency change of the laser light source in the sensor; Using the sampling data, the light intensity reference value is determined to consider the change of sensor output due to the change of the surrounding environment of light source, the increase / decrease direction of phase change by the measured amount, the coefficient of the number of half fringes, and the accuracy of the measurement. Subdividing the fringe such that is greater than or equal to a predetermined size; And a phase change amount caused by a change in the measured amount, and the number of half fringes to the number of half fringes, and further subdividing the measured value from the value of the phase change amount and the sensitivity of the sensor which are precisely measured to a predetermined precision or more. Signal processing method of an interference type optical sensor, characterized in that consisting of the step of converting to the size of. A signal processing circuit of an interference type optical fiber sensor, comprising: a laser diode and a single mode optical fiber for transmitting light to the optical fiber; Laser diode driving means for driving the laser diode to induce a change in optical frequency of a laser output; A crystal oscillator for generating a clock pulse signal of a predetermined frequency; Pulse width modulation means for supplying a pulse width modulated output pulse signal of the crystal oscillator to a laser diode driving means; Sampling pulse generating means for sampling the output pulse signal of the crystal oscillator; Light detecting means for detecting light reflected by the mirror; And sampling the sensor output from the light detecting means at two or three time points, calculating the change in light loss due to the change in the surrounding environment of the optical line, and the magnitude of the amount of phase change due to the change in the measured amount. In measuring the direction, the half fringe coefficient is used as the magnitude of the phase change amount, and if necessary, the magnitude of the phase change amount is precisely measured by subdividing the characteristic curve in the same fringe. And a signal processing means for obtaining a signal under measurement using the sensitivity of the sensor.