JPH05346353A - Optical fiber temperature distribution sensor - Google Patents

Abstract

PURPOSE:To accurately calculate temperature by compensating the attenuation rate difference between anti-stokes light and stokes light and a temperature error by obtaining the temperature when an ideal intensity ratio is calculated using the relationship between the ideal intensity ratio and temperature. CONSTITUTION:Pulse light from a pulse light generator 1 passes through a directional property coupler 2 and then enters an optical fiber probe 3 for measuring temperature. Pulse light passing through the probe 3 is partially subjected to Raman scattering and then returns as rear scattering light. The rear scattering light passes through the coupler 3 again and then anti-stokes light and stokes light are taken out by an optical filter 4. The intensities of both lights are detected by a signal detector 4 and are converted into a time-series digital signal train. Then, temperature is calculated from the digital signal train by a signal processor 6. The temperatures of all measurement points are calculated by successively calculating the temperatures of near-by points whose temperatures were not calculated yet according to a point whose temperature was obtained using a known reference temperature and a specific function. The signal processor 6 is provided with the function.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention measures the time-series data by injecting pulsed light into an optical fiber and sampling the intensity of anti-Stokes light and Stokes light of backscattered light generated by the Raman effect in the optical fiber. , The order of the time-series data corresponds to the distance along the optical fiber, and relates to the optical fiber temperature distribution sensor that obtains the temperature distribution along the optical fiber from these intensity data, especially with anti-Stokes light. By compensating for the temperature error caused by the Stokes' light attenuation difference and its temperature dependence,
The present invention relates to an extremely high-precision optical fiber temperature distribution sensor capable of calculating accurate temperature.

[0002]

2. Description of the Related Art Recently, in the field of optical sensing, there has been known an optical fiber temperature distribution sensor for obtaining a temperature distribution along an optical fiber by utilizing Raman effect in the optical fiber.

That is, an optical fiber temperature distribution sensor of this type uses the pulsed light emitted from the pulsed light generator,
Through a directional coupler, it leads to an optical fiber probe for temperature measurement. Part of the pulsed light passing through the optical fiber probe is Raman-scattered and returns as backscattered light.
The backscattered light then passes through the directional coupler again,
The optical filter extracts the anti-Stokes light and the Stokes light. The intensities of the anti-Stokes light and the Stokes light are respectively detected by the signal detection device and converted into a time-series digital signal sequence. Finally, the signal processing device calculates the temperature from the digital signal sequence. In this case, the conventional optical fiber temperature distribution sensor uses the following method when converting the intensity signal into temperature.

That is, first, the intensity ratio of anti-Stokes light and Stokes light is taken. Long (DALong, "Raman Spe
ctroscopy ", McGraw-Hi11, London, 1977), the intensity ratio of anti-Stokes light and Stokes light is expressed by the following equation (1): R (T) = (λ _s / λ _a ) ⁴ exp (-hcν / kT) ・ ・ ・ ・ ・ ・ (1)

In the equation (1), T is temperature and R is
(T) is the intensity ratio of anti-Stokes light and Stokes light,
λ _s is the wavelength of Stokes light, λ _a is the wavelength of anti-Stokes light, h is Planck's constant, c is the speed of light, ν is Raman shift, and k is Boltzmann's constant. Ideally, the temperature can be calculated from the relation of this equation (1).

However, in the actual measurement, it is difficult to measure the absolute intensity ratio because the intensity ratio varies in the performance of the photodetector, and the amount measured by the actual device is relatively small. The intensity ratio is

Therefore, a method is well known in which a temperature reference point is set and the temperature thereat is obtained by another means to obtain the entire temperature distribution. Usually, the above (1)
Dakin (JPDakin, DJPratt, GWBibby, an derived from the formula
d JNRoss, "Temperature Distribution Measurement U
sing Raman Ratio Thermometry ", SPIE Vo1.566 Fiber O
pticand Laser Sensors III 249, 1985) to calculate the temperature using the following relational expression.

[0008]

[Equation 4] However, in the equation (2), θ is the strength at the temperature reference point,
R ′ (θ) is the observed intensity ratio at the temperature reference point, and R (T) is the observed intensity ratio at the temperature calculation point.

In the method using the equation (2), it is assumed that the attenuation rates of the anti-Stokes light and the Stokes light are equal. Therefore, the contributions of the respective attenuation rates of the anti-Stokes light and the Stokes light are not included in the equation (2).

However, in reality, since the wavelengths of anti-Stokes light and Stokes light are different, the attenuation factors are generally different. Further, the difference in attenuation rate between anti-Stokes light and Stokes light actually has temperature dependence, and the difference in attenuation rate generally differs depending on temperature. Therefore, when the temperature is calculated by such a method, a temperature error cannot be accurately measured due to a temperature error due to the difference in the attenuation rate between the anti-Stokes light and the Stokes light and the temperature dependence of the difference in the attenuation rate.

[0011]

As described above, in the conventional optical fiber temperature distribution sensor, the temperature error due to the difference in the attenuation rate between the anti-Stokes light and the Stokes light and the temperature dependence of the difference in the attenuation rate occurs, and There was a problem that it was not possible to calculate the proper temperature.

An object of the present invention is to provide an extremely high-precision optical fiber capable of calculating an accurate temperature by compensating for a temperature error caused by a difference in attenuation rate between anti-Stokes light and Stokes light and its temperature dependence. It is to provide a temperature distribution sensor.

[0013]

In order to achieve the above object, pulsed light is incident on an optical fiber and the intensity of the backscattered anti-Stokes light and Stokes light generated by the Raman effect in the optical fiber is sampled. The temperature of the optical fiber is measured as time series data, and the temperature distribution along the optical fiber is calculated from these intensity data by utilizing the fact that the order of the time series data corresponds to the distance along the optical fiber. In the distribution sensor,

In the first aspect of the present invention, the intensity ratio of anti-Stokes light and Stokes light is measured as time-series data by sampling, and at least 1
The temperature of the reference point is obtained by another means, and the ideal intensity ratio at the temperature reference point is calculated by using the relationship between the ideal intensity ratio of anti-Stokes light and Stokes light at the back-scattered point obtained in advance and the temperature. The intensity ratio data of each of a plurality of measurement points in the vicinity of the measurement point where the temperature is known, and the anti-Stokes light in the optical fiber obtained in advance that holds between the ideal intensity ratio and the relationship between temperature and Using the relational expression including the relationship between the Stokes' light attenuation difference and temperature, the temperature-ideal intensity ratio has already been calculated or is calculated by another means. Calculate the ideal intensity ratio at the measurement point where the temperature is not calculated, and calculate the temperature at the point where the ideal intensity ratio is calculated from the calculated ideal intensity ratio using the relationship between the ideal intensity ratio and temperature. It is configured to include a temperature calculation section that.

Here, in particular, as the temperature calculating means,
The number of measurement points is N, and the intensity ratio data at each measurement point is {R '
_i ; i = 1,2, ..., N}, and the set of ideal intensity ratios is {R _i ; i =
1,2, ..., N}, the temperature at each measurement point is {T _i ; i = 1,2, ...,
N}, R is the ideal intensity ratio, R (T) is the function that represents the relationship between the ideal intensity ratio and temperature when T is the temperature, T is the attenuation rate difference, and α is the attenuation rate difference when the temperature is T. Where α (T) is a function representing the relationship between temperature and temperature, and Δx is the distance between adjacent intensity ratio data, the intensity ratio data of each of a plurality of measurement points near the measurement point whose temperature is known. And as a relational expression including the relationship between the attenuation rate difference and the temperature, which holds between the ideal intensity ratio and the temperature,

[0016]

[Equation 5] At least one of them is used to determine the temperature.

As the temperature calculation means, the number of measurement points is N, and the intensity ratio data at each measurement point is {R '_i; i.
= 1,2, ..., N}, the set of ideal intensity ratios is {R _i ; i = 1,2,
…, N}, the temperature at each measurement point is {T _i ; i = 1,2, ..., N},
R (T) is a function that represents the relationship between the ideal intensity ratio and temperature when R is the ideal intensity ratio and T is the temperature, and α is the attenuation rate difference and α is the attenuation rate difference when the temperature is T. When the function representing the relationship is α (T) and the distance between adjacent intensity ratio data is Δx, the intensity ratio data of each of a plurality of measurement points near the measurement point where the temperature is known and the ideal As a relational expression including the relationship between the attenuation rate difference and the temperature, which holds between the relative intensity ratio and the temperature,

[0018]

[Equation 6] At least one of them is used to determine the temperature.

On the other hand, in the invention described in claim 4, the intensity ratio of anti-Stokes light and Stokes light is measured as time-series data by sampling, and at least 1
Temperature of the point The temperature of the reference point is obtained by another means, and the intensity ratio data of each of the multiple measurement points near the measurement point where the temperature is known, the temperature, and the anti-Stokes at the previously determined backscattered point The relationship between the ideal intensity ratio of light and Stokes light and the relationship between temperature and the relationship between temperature and the relationship between temperature and the anti-Stokes light and Stokes light in the optical fiber obtained in advance is used to calculate the temperature. Is calculated or is obtained by another means from the temperature at the measurement point, the temperature calculation means for obtaining the temperature at the measurement point in the vicinity where the temperature has not been calculated yet is provided.

Here, particularly as the temperature calculating means,
The number of measurement points is N, and the intensity ratio data at each measurement point is {R '
_i ; i = 1,2, ..., N}, and the set of ideal intensity ratios is {R _i ; i =
1, 2, ..., N}, the temperature at each measurement point is {T _i ; i =
1,2, ..., N}, where R is the ideal intensity ratio and R is the temperature, and T is the function that represents the relationship between the ideal intensity ratio and temperature, the attenuation rate difference is α, and the temperature is T. When α (T) is the function that expresses the relationship between the temperature difference and the attenuation rate difference, and Δx is the distance between adjacent intensity ratio data, a plurality of measurement points near the temperature measurement point are known. Each intensity ratio data, temperature,
As a relational expression including the relationship between the attenuation rate difference and the temperature that holds between the ideal intensity ratio and the temperature,

[0021]

[Equation 7] At least one of them is used to determine the temperature.

[0022]

Therefore, first, in the optical fiber temperature distribution sensor according to the first aspect of the present invention, the intensity ratio of the anti-Stokes light and the Stokes light is measured as time series data by sampling, and at least one temperature reference is made. The temperature of the point is obtained by another means, and the ideal intensity ratio at the temperature reference point is obtained by using the relationship between the ideal intensity ratio of the anti-Stokes light and the Stokes light at the previously backscattered point and the temperature. Then, the intensity ratio data of each of a plurality of measurement points in the vicinity of the measurement point where the temperature is known, and the anti-Stokes light and the Stokes light in the optical fiber obtained in advance that holds between the relationship between the ideal intensity ratio and the temperature. Using the relational expression including the relationship between the attenuation rate difference and the temperature, the temperature and the ideal intensity ratio have already been calculated or the intensity ratio data at the measurement point obtained by another means can be used to calculate The ideal intensity ratio at the measurement points that have not been calculated is calculated, and the temperature at the point where the ideal intensity ratio is calculated using the relationship between the ideal intensity ratio and temperature is calculated from this calculated ideal intensity ratio. With this, it is possible to compensate the temperature error due to the difference between the attenuation rates of the anti-Stokes light and the Stokes light and the temperature dependence thereof, and calculate the accurate temperature.

On the other hand, in the optical fiber temperature distribution sensor of the fourth aspect of the present invention, the intensity ratio of anti-Stokes light and Stokes light is measured as time-series data by sampling, and at least one temperature reference point The temperature is obtained by another means, and the intensity ratio data for each of the multiple measurement points near the measurement point where the temperature is known, the temperature, and the ideal anti-Stokes light and Stokes light at the backscattered point obtained in advance The temperature has already been calculated or calculated using a relational expression that includes the relationship between the temperature difference between the anti-Stokes light and the Stokes light in the optical fiber, which is obtained in advance, and that holds between the dynamic intensity ratio and the temperature. The temperature at the measurement point that has not yet been calculated can be calculated from the temperature at the measurement point obtained by the method. The attenuation index difference of the anti-Stokes light and the Stokes light, and to compensate for temperature errors due to the temperature dependence, it is possible to calculate an accurate temperature.

[0024]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an example of the overall configuration of an optical fiber temperature distribution sensor according to the embodiment of the first invention.

That is, as shown in FIG. 1, the optical fiber temperature distribution sensor according to the first embodiment of the present invention provides a pulsed light generator 1 for generating pulsed light and a pulsed light from the pulsed light generator 1. Directional coupler 2 for receiving light, optical fiber probe 3 for temperature measurement for receiving pulsed light from directional coupler 2, and a part of pulsed light returned by Raman scattering by optical fiber probe 3 The optical filter 4 that receives the Stokes light and the anti-Stokes light through the rectifier 2, and the intensities of the anti-Stokes light and the Stokes light from the optical filter 4 are respectively detected and converted into a time series digital signal sequence. The signal detection device 5 and the signal processing device 6 that calculates the temperature from the digital signal train from the signal detection device 5 are included.

Here, the signal processing device 6 measures the intensity ratio of the anti-Stokes light and the Stokes light as time-series data by sampling, and at the same time obtains the temperature of at least one temperature reference point by another means and obtains it in advance. Using the relationship between the ideal intensity ratio of anti-Stokes light and Stokes light at the backscattered point and the temperature, the ideal intensity ratio at the temperature reference point is obtained, and a plurality of measurements near the measurement point where the temperature is known are obtained. Intensity ratio data for each point,
Using a relational expression that includes the relationship between the temperature difference between the anti-Stokes light and the Stokes light in the optical fiber, which is obtained in advance and holds between the ideal intensity ratio and the temperature,
Calculate the ideal intensity ratio at the measurement point where the temperature and the ideal intensity ratio have already been calculated or the intensity ratio data at the measurement point that has been obtained by another means have not yet calculated the temperature in the vicinity, It has a temperature calculation function for obtaining the temperature at the point where the ideal intensity ratio is calculated from the calculated ideal intensity ratio using the relationship between the ideal intensity ratio and the temperature. Next, the operation of the optical fiber temperature distribution sensor of the present embodiment configured as above will be described.

In FIG. 1, the pulsed light emitted from the pulsed light generator 1 is guided to an optical fiber probe 3 for temperature measurement through a directional coupler 2. Part of the pulsed light passing through the optical fiber probe 3 is Raman-scattered and returns as backscattered light. This backscattered light is
The anti-Stokes light and the Stokes light are extracted by the optical filter 4 through the directional coupler 2 again. The intensities of the anti-Stokes light and the Stokes light are respectively detected by the signal detection device 5 and converted into a time-series digital signal sequence. Then, the signal processing device 6 calculates the temperature from the digital signal sequence. In this case, the signal processing device 6 calculates the temperature by the following method. (A) First calculation method

The number of measurement points is N, the intensity ratio data at each measurement point is {R '_i; i = 1,2, ..., N}, and the set of ideal intensity ratios is {R _i ; i = 1, 2, ..., N}, the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R (T),
If the attenuation rate difference is α and the temperature is T, the function expressing the relationship between the attenuation rate difference and temperature is α (T), and the distance between adjacent intensity ratio data is Δx, the temperature is known. As a relational expression including the relationship between the attenuation rate difference and the temperature that holds between the intensity ratio data of a plurality of measurement points near a certain measurement point and the ideal intensity ratio and the temperature,

[0029]

[Equation 8] The temperature is determined using at least one of Here, the expressions (3) and (4) are derived by the following method.

The intensities of the anti-Stokes light and the Stokes light back-scattered at the distance x and the temperature T (x) at the scattering points are _Ia (T (x)) and _Is (T (x)), respectively.
The intensity ratio R (T (x)) at the scattering point is calculated by the following equation.

[0031]

[Equation 9]

The actual observed quantities are the intensity of anti-Stokes light and Stokes light affected by the attenuation, and their intensity ratio. These are I _a '(x), I _s ' (x
), R ′ (x). Also, the attenuation factor of anti-Stokes light and Stokes light is μ as a function of temperature, respectively.
_{Letting a} (T) and μ _s (T) and the attenuation rate difference be α (T), the following relationships are established.

[0033]

[Equation 10] By converting this equation (5), the following equation is obtained.

[0034]

[Equation 11] Differentiating this expression with respect to x gives the following expression.

[0035]

[Equation 12]

Where R '(x) is the observed quantity, and R'
Assuming that functional forms of (T) and α (T) are given, the temperature distribution can be obtained by solving equation (6) as an equation for T (x). However, since equation (6) includes T (x) in a complicated form, it is difficult to solve analytically. Therefore, the following method is used as a more realistic solution method.

That is, the actually measured intensity ratio R '
(X) is not continuous, but discrete values _R'i ; i = 1,2, ..., N}
Given as. So replace the derivative with the difference,
Equation (6) is converted into a discrete value type equation. In fact, (6)
The following substitutions are made to the expression:

[0038]

[Equation 13] The above replacement is carried out, and by modifying this, the above equation (3) is obtained. The equation (4) can also be obtained by the same method as above. The specific method of calculating the temperature using the above equations (3) and (4) is as follows.

First, any one or more of the measurement points is set as a temperature reference point, and the temperature at that point is obtained by another means.
Also, a function R representing the relationship between the ideal intensity ratio and temperature
(T), a function α (T) representing the relationship between the attenuation rate difference and temperature is given. Here, R (T) does not have to be an absolute intensity ratio but may be a relative intensity ratio, but it is necessary that the inverse function R ⁻¹ (R) be uniquely obtained.

It is now assumed that the r-th measurement point is the temperature reference point and the temperature is T _r . Then, the ideal intensity ratio R _{r there} is determined by R _r = R (T _r ). Next, the ideal intensity ratio R _{r + 1} at the r + 1-th measurement point is calculated by (3)
Calculate using an equation. Then, the temperature at the r + 1-th measurement point is obtained by T _{r + 1} = R ⁻¹ (R _{r + 1} ) using the inverse equation. Similarly, the r-1 th temperature T _r-1 can be obtained by using the equation (4).

Thereafter, the above method is repeatedly applied to
The temperatures at all the measurement points can be calculated by sequentially calculating the temperature at the point where the temperature has not been calculated yet one point before or after the point where the temperature was obtained. (B) Second calculation method

[0042] The number of measurement points N, the intensity ratio data at each measurement point _{{R 'i; i = 1,2} , ..., N}, a set of ideal intensity ratio {R _{i; i} = 1, 2, ..., N}, the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R (T),
If the attenuation rate difference is α and the temperature is T, the function expressing the relationship between the attenuation rate difference and temperature is α (T), and the distance between adjacent intensity ratio data is Δx, the temperature is known. As a relational expression including the relationship between the attenuation rate difference and the temperature that holds between the intensity ratio data of a plurality of measurement points near a certain measurement point and the ideal intensity ratio and the temperature,

[0043]

[Equation 14] The temperature is determined using at least one of Here, the equations (7) and (8) are derived by the following method.

First, the expression (5) is established at an arbitrary position in the optical fiber. Therefore, the i-th measured intensity ratio R '
_The position where the light giving _i is scattered is x _i , and the position where the light giving the i + 1-th measurement intensity ratio R ′ _{i + 1} is scattered is x
_{If i + 1} , the following relations are established from the equation (5).

[0045]

[Equation 15] Taking the ratio of both sides of these two equations, the following equation is obtained.

[0046]

[Equation 16]

Attention is now paid to the integral in the exponential function.
The point where the intensity ratio is actually observed in the integration range of the above equation is x _i
And x _{i + 1} only, and the intensity ratio in the region between them is not measured. Therefore, the integral is replaced by the sum of trapezoidal formulas.

[0048]

[Equation 17] By substituting this equation into the equation (9), the above equation (7) is obtained. The equation (8) can also be obtained by the same method as above. The specific method of calculating the temperature using the above equations (7) and (8) is as follows.

First, any one or more of the measurement points is set as a temperature reference point, and the temperature at that point is obtained by another means.
Also, a function R representing the relationship between the ideal intensity ratio and temperature
(T), a function α (T) representing the relationship between the attenuation rate difference and the temperature is given. Here, R (T) does not have to be an absolute intensity ratio but may be a relative intensity ratio, but it is necessary that the inverse function R ⁻¹ (R) be uniquely obtained.

The method for obtaining the temperature is as described above in equation (3),
It is similar to the specific method of deriving the temperature from the equation (4). That is, the formal difference between the expressions (3) and (7) is that T _{i + 1} is included on the right side of the expression (7).

Therefore, as a first approximation, when T _{i + 1} on the right side is replaced with T _i , the temperature is calculated by the same method as the concrete method for deriving the temperature from the above-mentioned equations (3) and (4). You can Usually, this is enough accuracy.
When higher accuracy is required, the following method is used.

That is, it is assumed that the i-th temperature T _i has already been calculated, and the i + 1-th temperature T _{i + 1} that has not yet been calculated is calculated. First, the first approximate temperature T ⁽¹⁾ _{i + 1} is obtained by the method described above. And
The second approximate temperature T ⁽²⁾ _{i + 1} is calculated by replacing T _{i + 1} on the right side of the equation (7) with T ⁽¹⁾ _{i + 1} . Hereinafter, the above method is repeatedly applied to obtain the nth approximate temperature T ⁽ⁿ⁾ _{i + 1} . This repetition may be performed any number of times, but normally sufficiently high accuracy can be obtained by calculating up to the second approximate temperature. Further, a means for determining the repeated convergence may be provided. Further, the exponent part of the exponential function of the equations (3) and (4) usually has an absolute value smaller than 1,
Formula of exponential function

[0053]

[Equation 18] Thus, the calculation can be speeded up by expanding to an arbitrary order.

In the above example, from the i-th temperature to i +
Although the method of calculating the first temperature has been described, the i-1th temperature can be calculated from the i-th temperature by the completely same method. After that, the temperatures at all measurement points can be obtained in the same manner as in the specific method of deriving the temperature from the above-described equations (3) and (4). Next, an embodiment of the second invention will be described. The optical fiber temperature distribution sensor of the second embodiment of the present invention has exactly the same configuration as that shown in FIG. 1 and has only the temperature calculation function of the signal processing device 6.

That is, the signal processing device 6 of the second embodiment of the present invention measures the intensity ratio of the anti-Stokes light and the Stokes light as time-series data by sampling, and at least the temperature of at least one temperature reference point. Intensity ratio data for multiple measurement points in the vicinity of the measurement point where temperature is known by different means, temperature, and ideal intensity of anti-Stokes light and Stokes light at the backscattered point obtained in advance The temperature has already been calculated by using a relational expression including the relationship between the temperature and the attenuation factor difference between the anti-Stokes light and the Stokes light in the optical fiber, which is obtained in advance and holds between the ratio and the temperature. It has a temperature calculation function that calculates the temperature at the measurement point that has not yet been calculated from the temperature at the measurement point calculated by It is intended.

Next, the operation of the optical fiber temperature distribution sensor of the present embodiment constructed as described above will be explained. The description of the same operation as that of the above embodiment will be omitted, and only different parts will be described here. That is, in the signal processing device 6, the temperature is calculated by the following method.

The number of measurement points is N, the intensity ratio data at each measurement point is {R ′ _i ; i = 1,2, ..., N}, and the set of ideal intensity ratios is {R _i ; i = 1, 2, ..., N}, the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R (T),
If the attenuation rate difference is α and the temperature is T, the function expressing the relationship between the attenuation rate difference and temperature is α (T), and the distance between adjacent intensity ratio data is Δx, the temperature is known. Intensity ratio data of each of a plurality of measurement points in the vicinity of a certain measurement point, temperature, as a relational expression including the relationship between the attenuation rate difference and the temperature that holds between the ideal intensity ratio and the temperature,

[0058]

[Formula 19] The temperature is determined using at least one of Here, the equations (11) and (12) are derived by the following method. First, the part of the equation (6) that differentiates R with respect to x can be decomposed as follows.

[0059]

[Equation 20] Using this relationship, equation (6) can be modified as follows.

[0060]

[Equation 21]

Then, from the above equation (6), the above (3)
Equation (4) is replaced with a discrete value system by the same method as when the equation (4) is derived.
Expression (12) is obtained. The specific method of calculating the temperature using the above equations (11) and (12) is as follows.

First, any one or more of the measurement points is set as a temperature reference point, and the temperature at that point is obtained by another means.
In addition, the function expressing the relationship between the ideal intensity ratio and temperature is R
(T), α (T) is a function that represents the relationship between the attenuation rate difference and temperature.
Is given. Here, R (T) does not have to be an absolute intensity ratio, and may be a relative intensity ratio, but an expression including the following derivative

[0063]

[Equation 22] Must be calculable.

It is now assumed that the r-th measurement point is the temperature reference point and the temperature is T _r . Then, the temperature at the r + 1-th measurement point can be calculated using the equation (11). Similarly, the equation (12) can be used to obtain the r-1th temperature T _r-1 .

Thereafter, the above method is repeatedly applied to
The temperatures at all measurement points can be calculated by sequentially calculating the temperature at the point before or after the point at which the temperature is obtained and at the point at which the temperature has not yet been calculated.

FIG. 2 is a diagram showing an example of a numerical simulation assuming a situation where the temperature continuously changes depending on the position. FIG. 2A shows the temperature distribution used for the calculation, and FIG.
(B) is a temperature calculation result by the conventional method using the equation (2), and FIG. 2 (c) is a temperature calculation result by the equation (3) and the equation (4) according to the first calculating method of the first invention. , Fig. 2 (d)
Are equations (7) and (8) according to the second calculation method of the first invention.
FIG. 2E shows the calculation result using the formula, and the calculation result using the formula (11) and the formula (12) according to the calculation method of the second invention.

FIG. 3 is a diagram showing an example of a numerical simulation assuming a situation where the temperature changes rapidly depending on the position. FIG. 3A shows the temperature distribution used for the calculation, and FIG.
(B) is the temperature calculation result by the conventional method using the equation (2). FIG. 3 (c) is a temperature calculation result using the equations (3) and (4) according to the first calculation method of the first invention, and FIG.
Are equations (7) and (8) according to the second calculation method of the first invention.
Calculation results using the equations, FIG. 3E shows the calculation results using the equations (11) and (12) according to the calculation method of the second invention.

Note that these are numerical simulations in which the measurement intensity ratio similar to the actual measurement environment is obtained by numerical calculation, and the temperature is calculated by both the method of the present invention and the conventional method based on the result. .. Next, the calculation results shown in FIGS. 2 and 3 will be described.

First, as shown in FIGS. 2B and 3B, in the conventional temperature calculation method, the original temperature distribution shown in FIG.
Compared to (a) and FIG. 3 (a), there is a large deviation, and it can be seen that the temperature is not correctly obtained. In this case, the maximum error between the original temperature and the calculated temperature is
34.4 ° C in the example of (b), 22.0 ° C in the example of FIG. 3 (b)
Is.

On the other hand, FIG. 2 shows the calculation results using the equations (3) and (4) according to the first calculation method of the first invention.
2 (c), FIG. 3 (c), and FIG. 2 which is a calculation result using the formulas (7) and (8) according to the second calculation method of the first invention.
In both (d) and FIG. 3 (d), the original temperature distribution is almost reproduced, and the effect of the present invention is well represented. In this case, the maximum error between the original temperature and the calculated temperature is
0.033 ° C. in the example of (c), 0.18 in the example of FIG. 3 (b).
5 ° C., 0.004 ° C. in the example of FIG. 2 (d), and 0.086 ° C. in the example of FIG. 3 (d). The calculation time is shown in FIG.
2 (c) and FIG. 3 (c) are almost the same as the conventional example.
In the example of (d) and FIG. 3 (d), it is almost double that of the conventional example.

Further, according to the calculation method of the second invention, (1
FIG. 2 is a calculation result using the equations (1) and (12).
2 (e) and FIG. 3 (e) in which there is no abrupt temperature change
In (e), the original temperature distribution is almost reproduced.
In the example of (e), a large error occurs in that the temperature changes abruptly. In this case, the maximum error between the original temperature and the calculated temperature is 0.67 ° C. in the example of FIG. 2 (e) and 27.4 ° C. in the example of FIG. 3 (e). Further, the calculation time is a little shorter than that of the conventional example.

Summarizing the above results, the temperature accuracy is highest in the results using the expressions (7) and (8), and
The result using the equation (4) is the second highest. Considering that the temperature accuracy of an actual optical fiber temperature distribution sensor is limited by noise due to the light receiving element and the amplifier, and at present, the error due to noise is about 0.5 ° C. in terms of temperature, the above two methods are used. Both of the calculation errors are less than that, and the attenuation rate difference can be corrected almost completely.

Further, from the results obtained by using the equations (11) and (12), it is difficult to calculate the correct temperature when there is a rapid temperature change as shown in FIG.
When the temperature change is gradual as shown in (e), a nearly correct temperature can be calculated. Furthermore, equation (11), (1
When the equation (2) is used, the temperature calculation time may be greatly shortened as shown below.

That is, in the above calculation, since a high-performance computer for performing high-speed real number calculation is used, no significant difference appears in the calculation time. However, in a processing system in which a real number calculation takes time, a large difference may occur in calculation time. In the simulation this time, the equation (1) is used as the R (T), but when the temperature is calculated using the equations (11) and (12), the equation (13) necessary for the calculation is expressed as follows. Can be expressed as elementary.

[0075]

[Equation 23]

Therefore, it is not necessary to calculate the exponential function or logarithmic function. The exponential function and the logarithmic function are numerically calculated by using an expansion equation such as the equation (10), which requires many real number calculations. Therefore, when the equations (11) and (12) are used, in a processing system that takes a long time to calculate the real number, the calculation time can be significantly shortened as compared with the other methods described above.

As the expression of R (T), the expression (1) was used in the above embodiment, but as described above, (3)
The conditions required when using the equations, (4), (7), and (8) are that the inverse function R ⁻¹ (R) is uniquely obtained, equations (11), (12) The condition necessary to use is that equation (13) can be calculated, and mathematically, the temperature can be calculated using any function as long as the above is satisfied.

When the measured temperature range is wide, the intensity ratio obtained experimentally may not be applied to the theoretical formula (1). In this case, the formula (1) may be corrected, or an approximate formula such as an experimentally obtained polynomial formula or one obtained by interpolating a plurality of coordinate points may be used. The same applies to α (T).

[0079]

As described above, according to the present invention, an accurate temperature can be calculated by compensating for the temperature error caused by the difference in attenuation rate between anti-Stokes light and Stokes light and its temperature dependence. A highly accurate optical fiber temperature distribution sensor can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of the overall configuration of an optical fiber temperature distribution sensor according to the present invention.

FIG. 2 is a diagram showing an example of a result of a numerical calculation simulation assuming a situation where the temperature continuously changes depending on the position.

FIG. 3 is a diagram showing an example of a result of a numerical calculation simulation assuming a situation where the temperature continuously changes depending on the position.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Pulse light generation device, 2 ... Directional coupler, 3 ... Optical fiber probe, 4 ... Optical filter, 5 ... Signal detection device, 6
... Signal processing device.

Claims (5)

[Claims] 1. Intensity of anti-Stokes light and Stokes light of back-scattered light generated by Raman effect in the optical fiber is sampled and measured as time series data. In the optical fiber temperature distribution sensor configured to obtain the temperature distribution along the optical fiber from these intensity data by utilizing the fact that the order of corresponds to the distance along the optical fiber, the anti-Stokes light and Stokes The light intensity ratio is measured as time series data by sampling, and the temperature of at least one temperature reference point is obtained by another means.
The ideal intensity ratio at the temperature reference point is obtained by using the relationship between the ideal intensity ratio of anti-Stokes light and Stokes light and the temperature at the backscattered point obtained in advance, and the vicinity of the measurement point where the temperature is known Intensity ratio data of each of a plurality of measurement points of, and the difference between the attenuation rate of the anti-Stokes light and the Stokes light in the optical fiber and the temperature which are obtained in advance between the ideal intensity ratio and the temperature Using a relational expression including a relation, the temperature and the ideal intensity ratio have already been calculated, or a measurement in which the temperature in the vicinity has not yet been calculated from the intensity ratio data at the measurement point obtained by another means. The temperature at which the ideal intensity ratio at a point is calculated, and the temperature at the point at which the ideal intensity ratio is calculated using the relationship between the ideal intensity ratio and temperature from the calculated ideal intensity ratio Optical fiber temperature distribution sensor, characterized in that it comprises an output means. 2. The temperature calculation means includes the number of the measurement points N, the intensity ratio data at each measurement point {R ′ _i ; i = 1, 2, ..., N}, the ideal intensity. The set of ratios is {R _i ; i = 1,2, ..., N}, and the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R
(T), α is the attenuation rate difference, α (T) is the function representing the relationship between the attenuation rate difference and temperature when T is the temperature, and Δx is the distance between adjacent intensity ratio data. In this case, the intensity ratio data of each of a plurality of measurement points in the vicinity of the measurement point where the temperature is known, and the relationship including the relationship between the attenuation rate difference and the temperature that holds between the relationship between the ideal intensity ratio and the temperature. As an expression, The optical fiber temperature distribution sensor according to claim 1, wherein the temperature is obtained using at least one of the above. 3. The temperature calculating means includes the number of the measurement points as N, the intensity ratio data at each measurement point as {R ′ _i ; i = 1, 2, ..., N}, and the ideal intensity. The set of ratios is {R _i ; i = 1,2, ..., N}, and the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R
(T), α is the attenuation rate difference, α (T) is the function representing the relationship between the attenuation rate difference and temperature when T is the temperature, and Δx is the distance between adjacent intensity ratio data. In this case, the intensity ratio data of each of a plurality of measurement points in the vicinity of the measurement point where the temperature is known, and the relationship including the relationship between the attenuation rate difference and the temperature that holds between the relationship between the ideal intensity ratio and the temperature. As an expression, The optical fiber temperature distribution sensor according to claim 1, wherein the temperature is obtained using at least one of the above. 4. The pulsed light is incident on the optical fiber, and the intensities of the anti-Stokes light and the Stokes light of the backscattered light generated by the Raman effect in the optical fiber are sampled and measured as time-series data, and the time-series data is obtained. In the optical fiber temperature distribution sensor configured to obtain the temperature distribution along the optical fiber from these intensity data by utilizing the fact that the order of corresponds to the distance along the optical fiber, the anti-Stokes light and Stokes The light intensity ratio is measured as time series data by sampling, and the temperature of at least one temperature reference point is obtained by another means.
Each intensity ratio data of a plurality of measurement points in the vicinity of the measurement point temperature is known, the temperature, the ideal intensity ratio of anti-Stokes light and Stokes light and the temperature of the backscattered point obtained in advance Using a relational expression containing the relationship between the temperature and the attenuation factor difference between the anti-Stokes light and the Stokes light in the optical fiber, which has been obtained in advance, the temperature has already been calculated or is obtained by another means. An optical fiber temperature distribution sensor and a temperature calculation method therefor, comprising temperature calculation means for calculating the temperature at a measurement point in the vicinity of which the temperature has not been calculated yet from the temperature at the measurement point. 5. The temperature calculation means comprises the number of the measurement points as N, the intensity ratio data at each measurement point as {R ′ _i ; i = 1, 2, ..., N}, and the ideal intensity. The set of ratios is {R _i ; i = 1,2, ..., N}, and the temperature at each measurement point is {T _i ;
i = 1,2, ..., N}, where R is the ideal intensity ratio and T is the temperature, a function representing the relationship between the ideal intensity ratio and temperature is R
(T), α is the attenuation rate difference, α (T) is the function representing the relationship between the attenuation rate difference and temperature when T is the temperature, and Δx is the distance between adjacent intensity ratio data. In this case, the intensity ratio data of each of a plurality of measurement points in the vicinity of the measurement point where the temperature is known, the temperature, and the relationship between the attenuation rate difference and the temperature that holds between the ideal intensity ratio and the temperature relationship. As a relational expression including The optical fiber temperature distribution sensor according to claim 4, wherein the temperature is obtained using at least one of the above.