JP2723661B2 - Optical fiber break prediction method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a method for predicting breakage of an optical fiber constituting a wired transmission line or the like.

(Prior Art) In recent years, optical fibers have been widely used as transmission media for wired transmission lines because of their characteristics of large capacity and low loss. However, since the optical fiber is made of a brittle material such as quartz glass, the optical fiber breaks due to a static fatigue phenomenon when strain is applied for a long period of time, so that an optical signal is not transmitted.

The breakage due to static fatigue occurs suddenly without a sign such as a change in transmission characteristics, and since the optical fiber has a large transmission capacity, such an unexpected breakage seriously hinders the operation of a communication network.

Here, the findings that have been clarified in the past regarding the static fatigue fracture phenomenon of optical fibers are described below.

The static fatigue fracture phenomenon of an optical fiber is explained by the theory of brittle fracture of glass.

According to the theory of brittle fracture, microscopic flaws on the glass surface grow gradually in the presence of stress and eventually break. The time t until breakage is represented by s as the applied strain and x as the size of the scratch initially present on the glass surface.
Where A and n are constants, it is obtained by the following equation (1) (Reference: IEICE Transactions J66-B, 7, P829-8
36 “Optical fiber strength guarantee method by screening test”).

t = Ax− ^{(n−2) / 2s−n} (1) Therefore, in order to predict the breaking time t of the optical fiber,
It is necessary to determine the size of the flaw x and the size of the applied strain s (or the size of the stress proportional to the stress s).

However, in reality, the size x of the flaw is usually 1 μm or less, and it is almost impossible to grasp the size of such a flaw over the entire length of the optical fiber constituting the transmission line in the length direction. is there.

Therefore, conventionally, a method has been proposed in which a test sample is extracted from an optical fiber manufactured under the same manufacturing conditions, a strength test is performed, and the size of the flaw is statistically grasped. If the size of the flaw can be grasped in this way, the applied strain s
If it becomes clear, even if the break time t of the optical fiber itself cannot be predicted, it is possible to calculate the probability p (hereinafter, referred to as a break rate) that the optical fiber breaks during the fixed time t instead. .

The breaking rate p is given by the following equation (2), using n, m, and k as parameters representing the distribution of scratches and the growth rate of the scratches obtained from the strength test of the test sample.

ln (p) = (m / (n-2)) t + (mn / (n-2)) ln (s) +
k (2) In addition, in order to lower the breaking rate p of the optical fiber, a relatively large portion of the scratches existing on the surface of the optical fiber is removed in advance by applying a short time immediately after the production of the optical fiber. You may need to use it later. Breakage rate p in this case
The parameters n, m, N _p , strain s, time t, which represent the distribution of wounds and the growth rate of the wounds determined from the strength test of the test sample
Strain amount s _p added immediately after preparation, is given by the following equation (3) as a function of strain application time t _p.

ln (p) = ln (m / (n−2)) + ln (N _p ) + n · ln (s / s _p ) + ln (t / t _p ) (3) (Means for solving the problem) Conventionally, methods proposed to predict breakage of a wired transmission line using an optical fiber have been proposed only for determining or predicting an optical fiber failure only by measuring the optical transmission loss of the optical fiber. (Reference: Japanese Patent Publication No. Hei 2-20049, “Optical Fiber Cable Breakage Prediction Device”, 1990 IEICE Spring Conference. B
−888 “Method of configuring optical line test and management system”).

Such a break prediction method is effective to some extent in post-judgment and prediction of a failure due to an increase in optical transmission loss of an optical fiber, and post-judgment of a breakage failure of an optical fiber.
As described above, the breakage of the optical fiber is caused by the strain applied to the optical fiber, and the magnitude of this strain cannot be directly obtained by measuring the optical transmission characteristics of the optical fiber. It is not useful for predicting the fracture failure of steel.

More specifically, there are two types of strain applied to an optical fiber, elongational strain and bending strain. Among them, no matter how much the elongational strain is applied, the optical transmission loss does not increase at all. , It is not possible to measure this kind of strain at all,
Subsequent breakage of the optical fiber cannot be predicted at all.

In addition, when bending strain is applied, the optical transmission loss of the optical fiber increases.However, this loss increase is not only due to the bending strain, but also the bending length and the structure of the bent optical fiber, such as core diameter and refraction. It also depends heavily on the rate distribution.

For this reason, special measures are required to calculate the magnitude of bending strain from the measurement results of optical transmission loss.However, inspection systems based on the conventional method do not take any measures for that, and the bending strain And the subsequent prediction of the breakage of the optical fiber cannot be performed at all.

As described above, one of the reasons that the fracture prediction is not performed by the conventional method is that, among the two types of strains described above, only a method of measuring elongation strain has been developed so far (see References). : O
ptical Fiber Communication conference'90.PD-15 “F
irst measurement of strain along field-installed
optical fiber using brillouin spectroscopy "), a method for measuring bending strain has not been developed yet because it is not possible to calculate the total strain applied to the optical fiber and to measure the optical fiber when the strain cannot be measured. This is because a method for predicting the fracture of the steel has not been developed yet.

The present invention has been made in view of such circumstances,
An object of the present invention is to provide a method for predicting breakage of an optical fiber that can accurately predict breakage of an optical fiber.

(Means for Solving the Problems) In order to achieve the above object, in claim (1), a strength test is performed on an optical fiber of the same type as the optical fiber to be measured, and a standard for the breakage rate and strain of the optical fiber is set. The relationship is stipulated, an optical fiber of the same type as the optical fiber to be measured is installed under the same conditions as the optical fiber to be measured, the breakage rate of the same type of optical fiber from the installation time to the current time is measured, and this measurement is performed. The amount of strain of the same kind of optical fiber was calculated from the breaking rate and the reference relationship, and the future breaking rate of the optical fiber to be measured was calculated from the calculated strain amount and the reference relationship.

Further, in claim (2), a strength test is performed on an optical fiber of the same type as the optical fiber to be measured, and a reference relationship between the breakage rate and the elongation strain of the optical fiber is defined. Propagates light of a predetermined wavelength, detects the propagated light, measures the elongation strain of the optical fiber to be measured, and installs an optical fiber of the same type as the optical fiber to be measured under the same conditions as the optical fiber to be measured. Then, the breaking rate of the same kind of optical fiber from the time of installation to the present time is measured, and the elongation strain amount of the same kind of optical fiber is calculated from the measured breaking rate and the reference relation, and the calculated elongation strain amount and the An estimated amount of elongation strain was calculated from the measured amount of elongation strain, and a future breakage rate of the optical fiber to be measured was calculated from the estimated amount of elongation strain and the reference relationship.

(Operation) According to claim (1), a strength test is performed on an optical fiber of the same type as the optical fiber to be measured, and a reference relationship between the breakage rate and the strain of the optical fiber is defined.

Next, an optical fiber of the same type as the optical fiber to be measured is installed under the same conditions as the optical fiber to be measured, and the breakage rate of the same type of optical fiber from the installation time to the current time is measured.

Next, the strain amount of the same kind of optical fiber is calculated from the measured breaking rate and a reference relation obtained in advance.

Subsequently, a future breakage rate of the optical fiber to be measured is calculated from the calculated strain amount and the reference relation, and a breakage prediction of the optical fiber is performed based on the calculation result.

According to claim (2), a strength test is performed on an optical fiber of the same type as the optical fiber to be measured, and a reference relationship between the breaking rate and the elongation strain of the optical fiber is defined.

Here, light of a predetermined wavelength is incident on the optical fiber to be measured and propagated. The propagating light is detected by the light receiver, and the elongation strain of the optical fiber to be measured is measured based on the detection result.

Further, an optical fiber of the same type as the optical fiber to be measured is installed under the same conditions as the optical fiber to be measured, and the breakage rate of the same type of optical fiber from the installation time to the current time is measured.

Next, the amount of elongation strain of the same kind of optical fiber is calculated from the measured breaking rate and a reference relation obtained in advance.

Subsequently, an estimated amount of elongation strain is calculated from the calculated amount of elongation strain and the measured amount of elongation strain, and a future breakage rate of the optical fiber to be measured is calculated from the estimated amount of elongation strain and the reference relationship. The breakage of the optical fiber is predicted based on the calculation result.

(Example) FIG. 1 is a configuration diagram showing an example of a measurement system used in the present invention. In FIG. 1, 1 is a main signal transmitting device, 2
Is a main signal receiving device, 3 is a plurality of optical fibers to be measured, 4a, 4b, 4c are optical branching devices, 5a, 5b, 5c are optical switches, 6,
Reference numeral 7 denotes a known optical pulse tester whose measurement wavelengths are different from each other. 8 is a pulse light generator, 9 is a light receiving device, 10
Is a wavelength sweep type light source, 11 is a data analyzer, and 12 is a database including a magnetic disk storage device and the like. In addition,
In the wiring shown in FIG. 1, a solid line indicates an optical wiring, and a broken line indicates an electric wiring.

Since a transmitting device 1 and a receiving device 2 for a main signal are usually connected to each optical fiber 3, the measuring system used in the present embodiment is, as shown in FIG. Each optical fiber 3 is connected via 4b. In addition, for selection of each optical fiber 3 to be actually measured and selection of a measuring device,
Optical switches 5a, 5b, 5c are used.

Next, the procedure and principle of the method of the present invention using the measurement system having the above configuration will be described step by step.

 First, a method for measuring bending strain will be described.

In order to measure the bending strain, pulse lights having different wavelengths emitted from the optical pulse testers 6 and 7 are appropriately selected by the optical switch 5c, and are incident on the optical fiber 3 to be measured via the optical switch 5a. . Further, the Rayleigh scattered light from the optical fiber 3 is again appropriately switched by the optical switch 5c via the optical switch 5a and received by the optical pulse tester 6 or the optical pulse tester 7. The optical transmission loss of the fiber 3 is measured remotely.

These measurement results are sent from the optical pulse testers 6 and 7 to the data analyzer 11, where the data analyzer 11 performs predetermined data processing and obtains the bending diameter and bending strength of the bending strain applied to the optical fiber 3. Detected separately in length.

Here, the number of optical pulse testers 6 and 7 is not limited to two, and can be increased according to the number of measurement wavelengths of pulse light required for data processing in the data analyzer 11. In addition, a single optical pulse tester that mounts a plurality of pulse light sources and shares a display unit, an operation unit, and the like may be used. Further, the measurement of the optical transmission loss amount may be performed without using the optical pulse testers 6 and 7, for example, by installing a semiconductor laser and an optical power meter at both ends of the optical fiber 3 to be measured. However, even when there are two or more bends, if the measurement is performed using the optical pulse testers 6 and 7, there is an advantage that the amount of optical transmission loss at each bend can be separated and measured.

As described above, bending strain (bending diameter, bending length) is calculated from a plurality of optical transmission losses measured using pulsed lights of a plurality of wavelengths.
In order to detect the optical transmission loss due to bending strain and the bending diameter, an optical fiber of the same type as the optical fiber 3 to be measured and an optical pulse tester of the same type as the optical pulse testers 6 and 7 are used in advance. It is necessary to grasp the relationship and store it in the data analyzer 11 as reference data.

That is, when bending is applied to the optical fiber, the amount of optical transmission loss of this optical fiber increases. Therefore, bending distortion can be detected by measuring the amount of optical transmission loss of the optical fiber to be measured. However, the amount of optical transmission loss is large even when the bending strain applied to the optical fiber is large, that is, even when the bending diameter is small, or when the bending strain is applied (hereinafter referred to as bending length). Even increase. For this reason,
Even if the amount of optical transmission loss is measured using pulsed light of one type of wavelength, it is impossible to separate and detect the bending diameter and the bending length.

In addition, by using an optical pulse tester to measure the amount of optical transmission loss, it is possible to measure the place where bending strain occurs, but secure the dynamic range necessary to measure the actual transmission path. For this purpose, the distance resolution of the optical pulse tester must be generally about 10 m to 100 m, whereas the bending diameter of the bending strain that leads to the breakage of the optical fiber is several mm to several cm. Therefore, with the distance resolution of the optical pulse tester, it is impossible at all to detect the bending diameter associated with the breakage of the optical fiber separately from the bending length.

Therefore, in the present embodiment, the effect that the bending diameter or bending length due to bending strain has on the increase in the amount of optical transmission loss depends on the wavelength of light propagating through the optical fiber,
The bending diameter and the bending length are separately detected by measuring the amount of increase in the optical transmission loss using a plurality of lights of different wavelengths.

In this case, the relationship between the amount of increase in the optical transmission loss due to the bending strain of the optical fiber and the bending shape, that is, the relationship between the bending diameter and the bending length differs depending on the type of the optical fiber to be measured (for example, a single mode optical fiber and a multimode optical fiber). (Optical fibers are completely different, and even the same single-mode optical fiber depends on how the refractive index distribution is given.) Therefore, using an optical fiber of the same type as the optical fiber to be measured, the increase in optical transmission loss due to bending It is necessary to grasp the relationship between the shape and the bending shape in advance.

Next, the present invention will be described in detail by taking, as an example, the case of a single mode optical fiber having a step-like refractive index distribution (hereinafter, referred to as an SI SM optical fiber).

It is well known that the relationship between the optical transmission loss S due to bending strain and the bending shape of the SI SM optical fiber is given by equation (4) (Reference: D. Marcuse, “Cuvature loss formula f
or optical fibers "Applied Optics, 19,9, pp1493-150
0).

Here, R is the bending diameter, L is the bending length, b is the propagation constant in the axial direction of the optical fiber, k is the propagation constant in the core perpendicular to the axis, r is the propagation constant in the cladding perpendicular to the axis, K ₊₁ , K _-1
Is a modified Bessel function, V is a normalized frequency, a is a core radius, and π is a pi.

As shown in the above equation (4), the optical transmission loss S
Are parameters related to the bending shape such as bending radius R and bending length L (hereinafter referred to as bending parameters) and b, k, r, V,
It is calculated from a parameter such as a that defines the transmission characteristics of the optical fiber (hereinafter referred to as a transmission parameter). These transmission parameters can be grasped by inspection after manufacturing the optical fiber. Therefore, the two unknown parameters in the equation (4) are the bending radius R and the bending length L of the bending parameters.

What is important here is that the transmission parameter changes according to the wavelength of the light propagating in the optical fiber, so that the above (4)
As can be inferred from the equation, the optical transmission loss due to bending has the property of being different for each wavelength (hereinafter referred to as the wavelength dependence of the loss), and this wavelength dependence itself varies depending on the bending diameter R. . Therefore, by examining the wavelength dependence of the loss, the bending diameter R can be calculated, and as a result,
The bending length L can also be calculated. The above-described wavelength dependence of the loss can be examined by measuring the amount of optical transmission loss using, for example, a plurality of pulsed lights having different wavelengths.

In the example described above, the wavelength dependence of the loss corresponds to the bending radius R
In the case where the wavelength dependence of the loss also depends on the bending length L in an optical fiber having a different shape of the refractive index distribution, the manner of the dependence is usually the same as the bending diameter R. Since the bending length L is different from the bending length L, the bending diameter R and the bending length L can be calculated from the measurement result of the increase amount of the optical transmission loss using a plurality of lights of different wavelengths.

Anyway, if you know the transmission parameters,
As described above, since the two unknown parameters in the equation (4) are the bending diameter R and the bending length L, if the optical transmission loss is measured using at least two different wavelengths of light, the bending diameter R And the bending length L can be calculated. Of course, in order to reduce the influence of the measurement error accompanying the measurement, the amount of optical transmission loss is measured using three or more types of light having different wavelengths, and the bending diameter R and the bending length are obtained from the measurement result using the least square method. L may be calculated.

Next, as a more specific example, an example of a method of calculating the bending diameter R and the bending length L of the SI SM optical fiber will be described.

FIG. 2 is a diagram showing the relationship between the ratio of the amount of optical transmission loss and the bending radius R in the SI SM optical fiber, and shows the SI type to be measured.
A wavelength of 1.65 is applied to an optical fiber (core diameter 9 μm, non-refractive index difference 0.32%) of the same kind as SM optical fiber (hereinafter referred to as optical fiber).
Optical transmission loss S due to bending when light of μm is propagated
Curve C shows the dependence of the ratio of the ratio of the increase A on the bending diameter R to the increase B on the optical transmission loss S due to bending when light having a wavelength of 1.55 μm is propagated.

This relationship is obtained as follows. That is,
Optical fiber of the same type as the optical fiber to be measured and wavelength 1.65
Optical transmission loss at each wavelength in an optical fiber without bending strain using optical pulse testers 6 and 7 that emit pulsed light of μm and 1.55 μm wavelength
Measure S1a and S1b.

Next, in a state where a bending strain having a predetermined bending radius R is applied to the optical fiber, the optical transmission loss amounts S2a and S2b corresponding to the respective wavelengths are similarly obtained.
Is measured. After that, the difference between these measurement results is calculated,
Increase A (= S2) in optical transmission loss due to bending strain for each wavelength
a−S1a) and B (= S2b−S1b). Next, the ratio (A / B) of these increased amounts A and B is calculated, and the ratio (A / B) is associated with the bending diameter R.

It is clear from the above equation (4) that the relationship between the ratio (A / B) of the optical transmission loss amount irrespective of the bending length L and the bending diameter R can be obtained.

The bending diameter R in the optical fiber 3 to be measured can be easily detected using the relationship between the ratio (A / B) of the optical transmission loss amount thus obtained and the bending diameter R as a reference relation.

That is, pulse light having a wavelength of 1.65 μm and a wavelength of 1.55 μm are propagated through the optical fiber 3 to be measured, and the optical transmission loss amounts S3a and S3b are measured for each wavelength. Next, the optical transmission loss amounts S1a and S1b when the bending strain is not applied are subtracted from the measurement result, and the increase amount A (= S3a−S1a), B of the optical transmission loss amount for each wavelength is subtracted. (= S3b-S1b) is calculated.

Further, the ratio (A / B) of the increase amounts A and B is calculated, and the second
The bending diameter R can be detected based on the relationship shown in the drawing. Further, if the bending diameter R can be obtained, the bending length L can be calculated using the above equation (4).

The specific values of the transmission parameters mentioned above are the same SI type SM
Even in the case of an optical fiber, the bending radius R and the bending length L are calculated according to the above-described method because the transmission parameters are determined for each optical fiber 3 to be measured. We need to keep that in mind.

However, even when the transmission parameters cannot be determined after manufacturing the optical fiber, the bending diameter R and the bending length L can be calculated. That is, the transmission parameters of the SI SM optical fiber can be calculated using two parameters representing the refractive index distribution, for example, the core diameter and the relative refractive index difference (hereinafter, referred to as structural parameters). Therefore, if the transmission parameters are not known in advance, the unknown parameters in the above equation (4) are a total of four of the two structural parameters, the bending diameter R and the bending length L, and the optical fiber 3 to be measured is
If you know that it is SI type SM optical fiber,
By measuring the optical transmission loss S in the optical fiber 3 to be measured for each wavelength using pulsed light of two wavelengths, the unknown parameters, that is, the two structural parameters, the bending diameter R, and the bending length L are measured. Can be calculated.

On the other hand, when there are a plurality of optical fibers 3 to be measured, and the relationship between the increase in the amount of optical transmission loss due to bending of the plurality of optical fibers and the bent shape can only be statistically grasped (for example, the optical fiber This method can also be applied to the case where only the average value and the standard deviation of the parameters that define the relationship between the increase in the transmission loss and the bending shape are known. In this case, the detection result is also calculated as a statistical value.

That is, when there are a plurality of optical fibers to be measured and transmission parameters or structural parameters of these optical fibers can be statistically grasped, for example, when the average value and the standard deviation of the structural parameters are known, arbitrary The transmission loss amount S for each of a plurality of different wavelengths of light at the bending diameter R and the bending length L is also calculated as a statistical value (average value, standard deviation). The average value and the standard deviation of the bending diameter R and the bending length L of the optical fiber can be calculated.

Next, a method for measuring elongation strain will be described. Note that the principle of measuring the elongation strain has already been described in detail in the above-mentioned literature and the like, and will be briefly described here.

When elongation strain is measured in the measurement system shown in FIG. 1, the optical switches 5a and 5c and the optical branching devices 4a and 4c are transmitted from the pulse light generator 8.
While injecting an optical pulse into the optical fiber 3 through
An optical signal is injected from the wavelength-swept light source 10 into the same optical fiber 3 via the optical switch 5b and the optical branching device 4b.

Thereby, when the difference between the wavelength of the light injected from the pulse light generator 8 and the wavelength of the light injected from the wavelength-swept light source 10 coincides with the amount of Brillian wavelength shift caused by the elongation strain applied to the optical fiber 3. Then, the light injected from the wavelength sweep type light source 10 is Brillian amplified by the light injected from the pulse light generator 8. This wavelength shift amount is proportional to the magnitude of the distortion.

Next, the light receiving device 9 monitors the optical power from the wavelength sweep type light source 10 via the optical branching devices 4a and 4c and the optical switches 5a and 5c as a function of the wavelength of the injected light, and the result is sent to the data analyzer 11. Sent out. Thus, the data analyzer 11 calculates the Brillian wavelength shift amount and the elongation strain from this value.

The bending rate and the elongation strain thus measured are summed, and the breaking rate p is calculated as a value of the total strain applied to the optical fiber 3 to be measured for predicting the breaking of the optical fiber 3.

In order to obtain the breaking rate p, a strength test is performed on an optical fiber of the same type as the optical fiber 3 to be measured, and a reference relationship between the breaking rate p and the strain of the optical fiber is specified. Is calculated so that the breaking rate p can be calculated from the strain using the above-mentioned equation (2) or (3), so that the parameter n, m, k in equation (2) or the parameter n, m in equation (3) , must seek N _p, it is also be stored in the data analyzer 11 these parameters.

Therefore, in the data analyzer 11, following the above-described calculation of the total amount of strain, the calculation result is substituted into the equation (2) or the equation (3) to determine the breaking rate p.

As described above, the measurement of the other strain can be omitted in predicting the breaking rate p at a place where it is clear that only one of the two strains, namely, the bending strain and the elongation strain, is applied.

Next, a method of predicting a future break rate from a past break rate of the same type of optical fiber manufactured under the same conditions and installed under the same conditions will be described.

As described above, this method is used for estimating the rupture rate at a location where strain cannot be measured due to the dynamic range of the measuring device. Whether the optical fiber 3 has been broken can be easily detected by the optical pulse tester 6 or the optical pulse tester 7. Therefore, it is necessary to record the result in the database 12 together with information such as the rupture time and the rupture location. In addition, the database 12 also records in advance information indicating which optical fiber was manufactured under the same conditions and installed under the same conditions.

Stored in this database 12, manufactured under the same conditions as the optical fiber to be predicted, and based on the past breakage rate of the optical fiber installed under the same situation, the data analysis device 11, The distortion s is calculated from the equation (2) or (3), and based on the distortion s, (2)
The breaking rate at a future time t is calculated from the equation or the equation (3).

At this time, as described above, when a plurality of breaks have occurred in the past, the break time t _j (j = 1 to i) and the break probability p
_The distortion s is obtained from the i sets of data _j (j = 1 to j) by the least square method.

Next, a description will be given of a method of performing a highly accurate prediction of a future breakage rate p by combining the measurement results of both cases where the strain is also measured and the breakage rate p from installation to the present is measured.

In this method, the distortion value calculated by the data analyzer 11 based on the data from the optical pulse testers 6 and 7, the light receiving device 9, and the wavelength-swept light source 10 in the manner described above, and recorded in the database 12 The optical fiber is manufactured under the same conditions as the optical fiber to be predicted, and the strain value calculated by the data analyzer 11 based on the past breakage rate of the optical fiber installed under the same condition is also used. At the same time, the value of the distortion having a smaller error is calculated again by the least squares method in the data analyzer 11, and the calculated result is again expressed by the above equation (2) or (3).
Substituting into the equation, the breaking rate p is calculated by the data analysis device 11.

Next, a method of calculating a strain with a smaller error by the least square method will be described in detail, taking a case where the relationship between the breaking ratio p and the strain s follows the equation (2) as an example.

Here, the magnitude of the strain obtained by the measurement is s ₀ , and the measurement accuracy is δs. Also, the break time at the break occurrence recorded in the past is t _j (j = 1 to i), the break at time t _i is p _j , and the expected value of the error of ln (p _i ) based on the sampling error is δp _i. When a distortion with a small error calculated by the least square method is s _s , s _s is given as s that minimizes W in the equation (5).

Further, assuming that the number of optical fibers manufactured under the same conditions as the optical fiber to be predicted and installed under the same conditions is N, the breaking rate p _j (j = 1 to i) at time t _i
Is given by equation (6), and the expected value of sampling error δp _i is given by equation (7).

p _i = i / N (6) δp _i = (1 / ln (1-p _i )) ² · p _i (N−i + 1) / N ² (1-p _i ) (7) Measurement Parameters m, n, and k representing the distribution of flaws and the growth rate of flaws obtained from the strength test of the optical fiber of the same kind as the target optical fiber are not given as definite values, but are given with statistical variations (ie, average Values m ₀ , n ₀ , k ₀ and
Distributed .delta.m ^2, .DELTA.n ^2, and .delta.k ² is given), then, it may be estimated by the same method as the strain s in equation (5). That is, instead of equation (5), s, m, n, k that minimizes W in equation (8) are obtained.

(Effect of the Invention) As described above, according to claim (1) or (2), the breaking rate of the optical fiber can be predicted in advance. Therefore, based on this result, the use of the optical fiber predicted to have a high breakage rate is stopped, or the optical fiber at the place predicted to have a high breakage rate is replaced (for example, the strain is removed).
As a result, interruption of the transmission line due to static fatigue fracture of the optical fiber can be prevented.

Since the optical fiber has a large communication capacity, the prevention of communication disruption achieved here is extremely significant from the viewpoint of ensuring smooth activities throughout society. In addition, this method can be achieved by remotely measuring the characteristics of an optical fiber installed in a wide area outdoors, and can be said to be a very simple method.

In addition, there is an advantage that it is possible to predict a rupture rate in a place where strain measurement cannot be performed due to a dynamic range of the measuring device.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing an example of a measurement system used in the present invention,
FIG. 2 is a diagram showing the relationship between the ratio of the optical transmission loss and the bending diameter. In the figure, 1 ... a main signal transmitting device, 2 ... a main signal receiving device, 3 ... an optical fiber to be measured, 4a, 4b, 4c ... an optical branching device, 5a, 5b, 5c ... an optical switch. , 6,7 ... light pulse tester, 8 ... pulse light generator, 9 ... light receiving device, 10 ...
Wavelength-swept light source, 11 Data analyzer, 12 Database.

Claims (2)

(57) [Claims] An optical fiber of the same type as the optical fiber to be measured is subjected to a strength test to define a reference relationship between the breakage rate and the strain of the optical fiber. Installed under the same conditions as the optical fiber to be measured, measure the breakage rate of the same type of optical fiber from the time of installation to the current time, and calculate the strain amount of the same type of optical fiber from the measured breakage rate and the reference relationship. An optical fiber breakage prediction method, wherein a future breakage rate of the optical fiber to be measured is calculated from the calculated strain amount and the reference relation. 2. An optical fiber of the same kind as an optical fiber to be measured is subjected to a strength test to define a reference relationship between a breaking rate and an elongation strain of the optical fiber. And measure the amount of elongation strain of the optical fiber to be measured by detecting this propagated light, and install an optical fiber of the same type as the optical fiber to be measured under the same conditions as the optical fiber to be measured. From this to the present, measuring the breaking rate of the same kind of optical fiber, calculating the amount of elongation strain of the same kind of optical fiber from the measured breaking rate and the reference relationship, and calculating the calculated amount of elongation strain and the measured elongation strain An estimated elongation strain amount is calculated from the estimated elongation amount, and a future breakage rate of the optical fiber to be measured is calculated from the estimated elongation strain amount and the reference relation.