KR102197696B1 - Structure health monitoring system using optic fiber-based hybrid nerve network sensor, and method for the same - Google Patents

Abstract

The present invention relates to a hybrid neural network sensor measuring system which is able to use a single optic fiber but to select an FBG measuring method and a BOCDA measuring method for measuring. The present invention is able to efficiently assess the condition of a facility based on a dynamic response in accordance with the measuring by the FBG method and a static response in accordance with the BOCDA method, to primarily measure the dynamic response by using the FBG from a hybrid neural network sensor installed on the facility by using a single optic fiber, and to secondarily measure the static response of the facility by using a high resolution of the BOCDA measuring method when an abnormal signal is detected.

Description

Facility health monitoring system using optical fiber-based hybrid neural network sensor and its method {STRUCTURE HEALTH MONITORING SYSTEM USING OPTIC FIBER-BASED HYBRID NERVE NETWORK SENSOR, AND METHOD FOR THE SAME}

The present invention relates to a facility health monitoring system, and more specifically, a fiber Bragg grating (FBG) measurement method and a Brillouin optical correlation domain analysis (BOCDA) measurement using one optical fiber As a hybrid neural network sensor measurement system that can measure facilities according to the method, it monitors to evaluate the condition of facilities based on dynamic response according to FBG measurement and static response according to BOCDA measurement, and facility health using optical fiber-based hybrid neural network sensor It relates to a road monitoring system and method thereof.

Typically, large structures and facilities built in the process of developing into an industrial society are damaged due to defects in the design and construction process or various factors that were not considered at the time of design, and the use of these facilities. As the period elapses, it gradually deteriorates, and its safety is greatly threatened. For example, in the case of facilities with severe structural damage, there are frequent cases where the service life is shortened to a degree that is significantly less than the design service life planned at the time of design.

Accordingly, efforts to ensure long-term safety and operability of facilities are urgently required. In particular, large structures such as buildings, bridges, and dams are constantly exposed to various operating loads, impacts from external objects, earthquakes, wind loads, wave loads, and corrosion, so the problem of securing the safety of facilities from them is economical and social. It has become an issue of great interest. In order to accurately diagnose the safety of such large structures, it is required to monitor the behavior of facilities through appropriate experimental measurements, a technology to analyze facility damage mechanically, and a diagnosis technology through an analysis technology modeling facility damage.

In addition, although periodic safety inspections are carried out on major infrastructure facilities, they are mainly at the level of visual inspection at points accessible by the inspector. Also, there is a lack of manpower and resources necessary for inspection, and facilities that are inaccessible. The reality is that the inspection period is limited due to the difficulty of inspection.

On the other hand, the start of maintenance of the infrastructure starts with the measurement of the response of the facility, for example, acceleration, strain, and displacement. At this time, the reliability of the measured response is high, and as the measurement information is varied, the condition of the facility is accurately evaluated, and it becomes an index that can determine whether, when, and how to repair and reinforce.

Through the response of the facility (or structure), many studies and practical applications have been conducted to evaluate the overall condition of the facility, but the local condition evaluation of the structure, such as damage, is performed through methods through non-destructive inspection and visual inspection. And manpower required.

Meanwhile, as a conventional technology for measuring the response of current facilities, 1) a response measurement method using an electric resistance point sensor, 2) a measurement method using an optical fiber Bragg Grating (FBG), and 3) an optical fiber itself. There are distributed sensor measurement methods used as sensors.

Specifically, in the case of a response measurement method using an electric resistance point sensor, as the number of sensors increases, problems such as cable wiring or power supply occur, and also, the measurement environment due to limitation of measurement location, durability, and electromagnetic interference. There is a problem that it is limited to

In addition, in the case of a measurement method using an optical fiber Bragg grating (FBG), multiple measurements can be performed with one optical fiber while having the same function as the electrical resistance strain gauge described above. At this time, the FBG sensor has a sampling rate of several kHz or more and a high A level of precision measurement is possible. However, there is a disadvantage in that the number of FBG sensors that can be measured through one FBG interrogator is limited to within tens.

In addition, in the case of a distributed sensor measurement method using an optical fiber itself as a sensor, unlike the measurement method using an optical fiber Bragg grating (FBG) described above, there is an advantage in that it is possible to measure a continuous distribution of physical quantities. For example, as a distributed sensor, in the Brillouin optical correlation domain analysis (BOCDA) method, the number of effective measurement points that can be measured is usually 1000 or more, and an arbitrary point can be selected and measured. However, the distributed sensor measurement method using the optical fiber itself as a sensor has a disadvantage that it is practically impossible to measure the dynamic response because the sampling rate is slow, and the measurement error is about 5 times larger than that of the FBG sensor.

As described above, there are several methods of measuring the response of other facilities in the conventional technology, but there is a problem in that it is difficult to efficiently measure the response of the facility according to the advantages and disadvantages of each.

Korean Patent Registration No. 10-1408954 (Registration Date: June 11, 2014), Title of Invention: "Strain Sensing Device of Stranded Stranded Wire and Sensing System Equipped with the Same" Korean Patent Registration No. 10-1326859 (Registration Date: November 1, 2013), Title of Invention: "Inductive Brillouin Scattering-Based Fiber Optic Sensor Device Using Optical Frequency Modulation" Republic of Korea Patent No. 10-1182650 (Registration date: September 7, 2012), title of invention: "Distribution type optical fiber sensor and sensing method using Brillouin scattering" Republic of Korea Patent Publication No. 2010-26145 (published date: March 10, 2010), title of the invention: "Method of measuring tension or strain using an optical fiber Bragg grating sensor"

The technical problem to be achieved by the present invention for solving the above-described problem is a hybrid neural network sensor measurement system capable of selecting and measuring an FBG measurement method and a BOCDA measurement method while using one optical fiber, and provides a dynamic response and a dynamic response according to FBG measurement. It is to provide a facility health monitoring system and method using an optical fiber-based hybrid neural network sensor that can efficiently evaluate the condition of a facility based on a static response according to BOCDA measurement.

Another technical problem to be achieved by the present invention is to primarily measure the dynamic response using an optical fiber Bragg grating from a hybrid neural network sensor installed in a facility using one optical fiber, and analyze the Brillouin scattering correlation region when detecting an abnormal signal (BOCDA ) It is to provide a facility health monitoring system and method using an optical fiber-based hybrid neural network sensor that can secondarily measure the static response of the facility using the high resolution of the measurement method.

As a means for achieving the above-described technical problem, the facility health monitoring system using an optical fiber-based hybrid neural network sensor according to the present invention uses one optical fiber installed at one side of the facility, and a response corresponding to the deformation rate of the facility A hybrid neural network sensor that measures a value; An optical fiber Bragg grating (FBG) interrogator connected to both ends of the hybrid neural network sensor to always measure a dynamic response of the facility corresponding to an FBG wavelength, which is an optical signal in a length direction of an optical fiber; An abnormal signal determination unit that analyzes the dynamic response measured by the FBG interrogator to check whether an abnormal signal exceeding a preset range is detected; Analysis of the Brillouin scattering correlation region that is connected to both ends of the hybrid neural network sensor and measures the static response of the entire optical fiber or the suspected damage area of the facility corresponding to the Brillouin frequency, which is an optical signal in the length of the optical fiber when the abnormal signal is detected (BOCDA) interrogator; And a local damage check unit that analyzes the static response measured by the BOCDA interrogator to check whether the facility is locally damaged, wherein the hybrid neural network sensor uses an optical fiber Bragg grating (FBG) to determine the dynamic response of the facility. It is characterized by primary measurement, and secondary measurement of the static response of the facility using a high resolution of Brillouin scattering correlation domain analysis (BOCDA).

Here, the FBG interrogator measures the initial dynamic response of the facility and then firstly measures the facility at all times, and the BOCDA interrogator measures the initial static response of the facility and then detects the abnormal signal. The facility can be measured secondary.

The facility health monitoring system using an optical fiber-based hybrid neural network sensor according to the present invention further includes a facility health evaluation unit that evaluates the health of the facility in response to the degree of local damage of the facility identified by the local damage check unit. Can include.

Here, the abnormal signal determination unit, the local damage check unit, and the facility health evaluation unit may be implemented on a manager terminal installed near the facility.

Here, the BOCDA interrogator is characterized in that it applies a wavelength of 1548 nm to the optical fiber, which is the hybrid neural network sensor.

Here, the FBG interrogator considers that the BOCDA interrogator applies an optical signal of a wavelength of 1548 nm, so that mutual interference does not occur, except for the wavelength of 1545 nm to 1551 nm. It is characterized by applying to.

Here, the facility may be a large structure including a building, a bridge, and a dam.

On the other hand, as another means for achieving the above-described technical problem, a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to the present invention includes: a) installing a hybrid neural network sensor using one optical fiber in the facility; b) measuring the initial response of the facility using an optical fiber Bragg grating (FBG) interrogator and a Brillouin scattering correlation domain analysis (BOCDA) interrogator, respectively; c) constantly measuring the dynamic response of the facility corresponding to the wavelength of the FBG, which is an optical signal in the longitudinal direction of the optical fiber, using the FBG interrogator; d) analyzing the measured dynamic response to evaluate the global condition of the facility and determining whether an abnormal signal is detected; e) when the abnormal signal is detected, measuring, by a BOCDA interrogator, a static response of the entire section or a partial section of the facility corresponding to the Brillouin frequency, which is an optical signal in the longitudinal direction of the optical fiber; And f) confirming local damage of the facility according to the static response measurement of the BOCDA interrogator, wherein the hybrid neural network sensor uses one optical fiber, and the hybrid neural network sensor is an optical fiber Bragg grating (FBG) The dynamic response of the facility is primarily measured by using, and the static response of the facility is secondarily measured using a high resolution of the Brillouin scattering correlation domain analysis (BOCDA) measurement method.

According to the present invention, a hybrid neural network sensor measurement system capable of selecting and measuring an FBG measurement method and a BOCDA measurement method using one optical fiber, based on a dynamic response according to FBG measurement and a static response according to BOCDA measurement. You can evaluate the condition efficiently.

According to the present invention, a dynamic response is primarily measured using an optical fiber Bragg grating from a hybrid neural network sensor installed in a facility using one optical fiber, and when an abnormal signal is detected, the Brillouin scattering correlation region analysis (BOCDA) measurement method is high. Using the resolution, the static response of the facility can be measured secondarily.

According to the present invention, it is possible to efficiently perform decision-making, such as a maintenance plan and maintenance/reinforcement timing and method, during the life cycle of a facility according to facility health monitoring using an optical fiber-based hybrid neural network sensor.

1 is a view for explaining an optical fiber-based hybrid neural network sensor in a facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
2 is a block diagram of a facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
3 is an operation flow diagram of a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
4 is a diagram illustrating installation of a hybrid neural network sensor in a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
5 is a diagram illustrating an initial response measurement of a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
6 is a diagram illustrating dynamic response measurement using an FBG interrogator in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
7 is a diagram illustrating detection of an abnormal signal according to dynamic response measurement in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
FIG. 8 is a diagram specifically illustrating detection of an abnormal signal according to the dynamic response measurement shown in FIG. 7.
9 is a diagram illustrating static response measurement using a BOCDA interrogator in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.
10 is a diagram illustrating local damage verification of a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are assigned to similar parts throughout the specification.

Throughout the specification, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless otherwise stated. In addition, terms such as "... unit" described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware or software or a combination of hardware and software.

Hereinafter, a facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention will be described with reference to FIGS. 1 to 2, and an optical fiber according to an embodiment of the present invention will be described with reference to FIGS. 3 to 10. A method for monitoring facility health using a hybrid neural network sensor is described.

[Facility health monitoring system 100 using an optical fiber-based hybrid neural network sensor]

First, FIG. 1 is a view for explaining an optical fiber-based hybrid neural network sensor in a facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, and FIG. 1A shows a hybrid neural network sensor, and FIG. 1b) shows a Point Optical Fiber Sensor using an optical fiber Bragg grating (FBG), and FIG.1C shows a Distributed Optical Fiber Sensor using Brillouin Scattering Correlation Area Analysis (BOCDA). Sensor) respectively.

As shown in Fig. 1a), in the facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, the hybrid neural network sensor 110 uses one optical fiber installed on one side of the facility. , The response corresponding to the strain rate of the facility is measured. In this case, the hybrid neural network sensor 110 primarily measures the dynamic response of the facility using an optical fiber Bragg grating (FBG), as shown in b) of FIG. 1, and also, when detecting an abnormal signal, As shown in Fig. 1c), the static response of the facility is secondarily measured using the high resolution of the Brillouin scattering correlation region analysis (BOCDA) measurement method.

In other words, as shown in b) of FIG. 1, a point optical fiber sensor using an optical fiber Bragg grating (FBG) indicates that measurement can be performed at only a few points on one optical fiber, and is also shown in c) of FIG. As described above, the distributed optical fiber sensor using the Brillouin scattering correlation domain analysis (BOCDA) indicates that measurement is possible at a number of consecutive points on one optical fiber. At this time, as shown in c) of FIG. 1, since the optical fiber sensor based on the Brillouin scattered light can continuously measure the deformation and temperature distribution of the entire section to which the optical fiber sensor is attached, a large structure with a large size or It is known to be effectively used for monitoring of long distance lines. At this time, a method of continuously measuring the strain occurring in the longitudinal direction of the optical fiber by using the linear relationship between the strain applied to the optical fiber and the Brillouin frequency change is called Brillouin-based distributed optical fiber sensing. That is, distributed optical fiber sensing is a method of using the entire optical fiber as a strain sensor, and has the same effect as installing a large number of strain sensors in the longitudinal direction of the optical fiber sensor. The Brillouin scattering correlation region (BOCDA) analysis method is a distributed optical fiber sensing technique having a high spatial resolution of 5 cm, and a distributed optical fiber sensor interrogator can be used.

)를 취득하는 분포측정 센서를 말하며, 이러한 브릴루앙 주파수 변이 검출 원리를 간략히 설명하면 다음과 같다.More specifically, a phenomenon in which light interacts with sound waves generated in a material and is scattered at a frequency different from the frequency of incident light is called Brillouin scattering, and the difference between these frequencies is called Brillouin frequency variation. At this time, the optical fiber BOCDA (Brillouin Optical Correlation Domain Analysis) sensor refers to the Brillouin frequency shift at all positions along the length of the sensing optical fiber.

It refers to a distribution measuring sensor that acquires ), and a brief description of the Brillouin frequency shift detection principle is as follows.

)를 상기 피측정 광섬유의 브릴루앙 주파수 변이값(

)과 일치하도록 광주파수를 조정하면, 펌핑광은 유도 브릴루앙 산란에 의해 탐색광으로 광에너지 변환을 하고, 탐색광은 광섬유 내에서 브릴루앙 광증폭을 하게 된다. 이렇게 증폭된 탐색광의 광신호는 광검출기에 의해 전기신호로 변환됨으로써 측정이 가능하다.Pumping light and probe light are respectively incident at both ends of the optical fiber, but the optical frequency difference between the pumping light and the search light (

) To the Brillouin frequency shift value of the optical fiber to be measured (

If the optical frequency is adjusted to match ), the pumped light converts light energy into search light by guided Brillouin scattering, and the search light performs Brillouin light amplification in the optical fiber. The optical signal of the amplified search light can be measured by converting it into an electrical signal by a photodetector.

)은, 광이 진행되는 물질, 즉, 광섬유의 재료에 크게 영향을 받을 뿐만 아니라 광섬유에 인가되는 변형률에 따라서 변화한다. 외부에서 가해지는 응력에 의한 광섬유의 변형률을

라 하고, 온도 변화를

라 할 때, 브릴루앙 주파수 변이값의 변화량(

)은 다음의 수학식 1과 같이 나타낼 수 있다. These Brillouin frequency shift values (

) Is not only greatly affected by the material through which light proceeds, that is, the material of the optical fiber, but also changes according to the strain applied to the optical fiber. The strain of the optical fiber due to external stress

And the temperature change

When d, the amount of change in the Brillouin frequency shift value (

) Can be expressed as in Equation 1 below.

[Equation 1]

) 및 온도 변환계수(

)는 미리 알려진 값이지만, 정확성을 높이기 위해 실제 응용조건에 따라 정확하게 조사하여 사용하는 것이 더 바람직하다. 예를 들면, 표준 단일모드 광섬유를 감지 광섬유로 사용할 경우, 이러한 변형률 변환계수(

)는 약 0.05㎒/μ_ε와 온도 변환계수(

)는 약 1㎒/℃값을 갖지만 정확한 값은 실제 응용조건에 따라 조사하여 사용해야 한다.Here, the strain conversion factor (

) And temperature conversion factor (

) Is a known value in advance, but it is more preferable to accurately investigate and use it according to actual application conditions in order to increase accuracy. For example, when a standard single-mode fiber is used as a sensing fiber, this strain conversion factor (

) Is about 0.05MHz/μ _ε and the temperature conversion factor (

) Has a value of about 1㎒/℃, but the exact value should be investigated and used according to the actual application conditions.

Accordingly, when the Brillouin frequency shift is obtained, the strain or temperature distributed in the detected optical fiber can be measured using Equation 1, and a sensor made by this principle is called an optical fiber BOCDA sensor. In order to obtain the Brillouin frequency shift distributed in such a sensing optical fiber, the Brillouin gain spectrum must be obtained from all positions of the sensing optical fiber.

Meanwhile, FIG. 2 is a configuration diagram of a facility health monitoring system using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.

Referring to FIG. 2, a facility health monitoring system 100 using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention includes a hybrid neural network sensor 110, an optical fiber Bragg grating (FBG) interrogator 120, An abnormal signal determination unit 130, a Brillouin scattering correlation area measurement method (BOCDA) interrogator 140, a local damage check unit 150, and a facility health evaluation unit 160, and the facility 200 is a building , It may be a large structure including a bridge and a dam, but is not limited thereto.

The hybrid neural network sensor 110 uses one optical fiber installed on one side of the facility 200 and measures a response corresponding to the strain rate of the facility 200. At this time, the hybrid neural network sensor 110 primarily measures the dynamic response of the facility 200 using an optical fiber Bragg grating (FBG), and uses the high resolution of the Brillouin scattering correlation region analysis (BOCDA) measurement method. Thus, the static response of the facility 200 is secondarily measured.

The FBG interrogator 120 is connected to both ends of the hybrid neural network sensor 110 and primarily always measures the dynamic response of the facility 200 corresponding to the FBG wavelength, which is an optical signal in the length direction of the optical fiber. That is, the FBG interrogator 120 always measures the facility 200 after measuring the initial dynamic response of the facility 200.

The abnormal signal determination unit 130 analyzes the dynamic response measured by the FBG interrogator 120 and checks whether an abnormal signal exceeding a preset range is detected.

The BOCDA interrogator 140 is connected to both ends of the hybrid neural network sensor 110, and when the abnormal signal is detected, a damage suspicious area of the facility 200 corresponding to the Brillouin frequency, which is an optical signal in the length of the optical fiber. Or, measure the static response of the entire optical fiber. That is, the BOCDA interrogator 140 measures the initial static response of the facility 200, and then measures the facility 200 secondarily when detecting the abnormal signal.

The local damage check unit 150 analyzes the static response measured by the BOCDA interrogator 140 to check whether the facility 200 is locally damaged.

The facility health evaluation unit 160 evaluates the health of the facility 200 in response to the degree of local damage to the facility 200 identified by the local damage check unit 150. Here, the abnormal signal determination unit 130, the local damage check unit 150, and the facility health evaluation unit 160 may be implemented on a manager terminal installed near the facility 200.

In addition, in the case of the facility health monitoring system 100 using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, the wavelength of an optical fiber Bragg grating (FBG) to be engraved on an optical fiber for measurement when measuring through the hybrid neural network sensor 110 Be careful with your choices. Specifically, since it is 1548 nm, which is the wavelength used to measure the Brillouin signal, when considering the situation in which an optical fiber Bragg grating (FBG) is installed at a wavelength interval of 3 nm, an optical fiber Bragg grating (FBG) in the wavelength range of 1545 to 1551 nm is measured. When present in the optical fiber, the pump light and the probe light may be reflected back from the fiber Bragg grating (FBG) depending on the incident case.At this time, the reflection of the probe light causes signal reduction, and the reflection of the pump light causes a photodetector. Excessive input may be generated, and thus saturation and noise increase may be caused, so that Brillouin scattering correlation domain analysis (BOCDA) measurement may become impossible.

In other words, the BOCDA interrogator 140 applies an optical signal having a wavelength of 1548 nm to the optical fiber, which is the hybrid neural network sensor 110, as shown in c) of FIG. 1, and further, the FBG interrogator The logger 120 considers that the BOCDA interrogator 140 applies an optical signal having a wavelength of 1548 nm, as shown in b) of FIG. 1, so that interference does not occur between 1545 nm and 1551. It is preferable to apply an optical signal having a wavelength excluding a wavelength of nm to the optical fiber, which is the hybrid neural network sensor 110.

After all, in the case of the facility health monitoring system 100 using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, a measurement method using an optical fiber Bragg grating (FBG) and a Brillouin scattering correlation area analysis (BOCDA) measurement method As a measurement method using an optical fiber-based hybrid neural network sensor that combines only the advantages, the dynamic response is primarily measured using an optical fiber Bragg grating (FBG) from the hybrid neural network sensor 110 installed in the facility 200 using only one optical fiber. Then, when an abnormal signal is detected, the static response of the facility 200 may be secondarily measured using the high resolution of the Brillouin scattering correlation region analysis (BOCDA) measurement method. Accordingly, in the case of the facility health monitoring system 100 using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, the facility is more efficiently monitored than conventional methods through measurement of the hybrid neural network sensor 110 can do.

[Facility health monitoring method using optical fiber-based hybrid neural network sensor]

3 is an operation flow diagram of a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention.

Referring to FIG. 3, in the facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, first, a hybrid neural network sensor 110 using one optical fiber is installed in the facility 200 ( S110).

Specifically, FIG. 4 is a diagram illustrating installation of a hybrid neural network sensor in a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention. For example, the facility 200 is In the case of a suspension bridge structure, as shown in FIG. 4, a pylon 210 of a steel or reinforced concrete structure, a cable 220 suspended from the pylon 210, a top plate or girder 230 of a long bridge, and the girder ( It consists of a pier 240 supporting the 230, a hanger 250 hanging from the cable 220 to guide the tension of the cable in the ground direction, and an anchorage for fixing one side of the cable 220 to the ground, It will be described as installing an optical fiber-based hybrid neural network sensor 110 using one optical fiber for the suspension bridge structure 200 on the side of the upper plate or the girder 230. In this case, the facility 200 may be a large structure such as a building, a bridge, or a dam, but is not limited thereto.

Next, the initial response of the facility 200 is measured using an optical fiber Bragg grating (FBG) interrogator 120 and a Brillouin scattering correlation region analysis (BOCDA) interrogator 140, respectively (S120).

Specifically, FIG. 5 is a diagram illustrating an initial response measurement of a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention. The FBG interrogator 120 is The initial dynamic response of the facility 200 corresponding to the FBG wavelength as an optical signal is measured, and the BOCDA interrogator 140 is the initial of the facility 200 corresponding to the Brillouin frequency, which is an optical signal in the length direction of the optical fiber. Measure the static response.

Next, the dynamic response of the facility 200 is always measured using the FBG interrogator 120 (S130).

Specifically, FIG. 6 is a diagram illustrating dynamic response measurement using an FBG interrogator in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, wherein the FBG interrogator 120 is After measuring the initial dynamic response of the facility 200, the facility 200 is primarily always measured, and the BOCDA interrogator 140 measures the initial static response of the facility 200 and then the abnormality When a signal is detected, the facility 200 is secondarily measured. At this time, the BOCDA interrogator 140 applies a wavelength of 1548 nm to the optical fiber, which is the hybrid neural network sensor 110, and the FBG interrogator 120 is the BOCDA interrogator 140 In consideration of applying an optical signal having a wavelength of nm, a wavelength excluding a wavelength of 1545 nm to 1551 nm is applied to the optical fiber of the hybrid neural network sensor 110 so as not to cause mutual interference.

Next, the measured dynamic response is analyzed to evaluate the global state of the facility 200, and it is determined whether an abnormal signal is detected (S140).

Specifically, FIG. 7 is a diagram illustrating detection of an abnormal signal according to a dynamic response measurement in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, and FIG. 8 is a dynamic response shown in FIG. As a diagram specifically showing the detection of an abnormal signal according to measurement, FIG. 8a) shows normal signal detection, and FIG. 8b) shows abnormal signal detection. That is, in the case of the measurement point indicated by reference numeral A in FIG. 7, as shown in a) of FIG. 8, the normal signal is sensed, but in the case of the measurement points indicated by reference numerals B and C in FIG. As shown in b), it indicates detecting an abnormal signal.

Next, when the abnormal signal is detected, the BOCDA interrogator 140 measures the static response of the entire section or some section of the facility 200 (S150). Specifically, FIG. 9 is a diagram illustrating static response measurement using a BOCDA interrogator in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, wherein the BOCDA interrogator 140 The static response of the entire optical fiber or the area of the facility 200 suspected of being damaged is measured.

Next, local damage of the facility 200 is checked according to the static response measurement of the BOCDA interrogator 140 (S160). Specifically, FIG. 10 is a diagram illustrating local damage confirmation of a facility in a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, and is used to measure the static response of the BOCDA interrogator 140. Accordingly, local damage to the facility 200 can be confirmed.

Subsequently, the health of the facility 200 may be evaluated in response to the degree of local damage to the facility 200 in which the local damage is confirmed. Here, since it is obvious to those skilled in the art to evaluate the health of the facility 200 in response to the degree of the local damage, a detailed description will be omitted.

In the case of a facility health monitoring method using an optical fiber-based hybrid neural network sensor according to an embodiment of the present invention, the hybrid neural network sensor 110 uses one optical fiber, and the hybrid neural network sensor 110 is an optical fiber Bragg grating (FBG). ) To first measure the dynamic response of the facility 200, and secondly measure the static response of the facility 200 using the high resolution of the Brillouin scattering correlation region analysis (BOCDA) measurement method. I can.

In conclusion, according to an embodiment of the present invention, as a hybrid neural network sensor measurement system that can be measured by selecting an optical fiber Bragg grating (FBG) measurement method and a Brillouin correlation region analysis (BOCDA) measurement method while using one optical fiber, FBG Based on the dynamic response according to the measurement and the static response according to the BOCDA measurement, the condition of the facility can be efficiently evaluated. In addition, the dynamic response is performed using an optical fiber Bragg grating from a hybrid neural network sensor installed in the facility using one optical fiber. Secondly, the static response of the facility can be measured secondarily when an abnormal signal is detected by using the high resolution of the Brillouin scattering correlation domain analysis (BOCDA) measurement method.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

100: Fiber-based hybrid neural network measurement system
200: facility
110: hybrid neural network sensor 120: FBG interrogator
130: abnormal signal determination unit 140: BOCDA interrogator
150: local damage check unit 160: facility health evaluation unit

Claims (12)

A hybrid neural network sensor 110 that uses one optical fiber installed on one side of the facility 200 and measures a response corresponding to the strain of the facility 200;
An optical fiber Bragg grating (FBG) interrogator 120 connected to both ends of the hybrid neural network sensor 110 to always measure a dynamic response of the facility 200 corresponding to an FBG wavelength, which is an optical signal in the length of the optical fiber;
An abnormal signal determination unit 130 which analyzes the dynamic response measured by the FBG interrogator 120 to evaluate the global state of the facility, and checks whether an abnormal signal exceeding a preset range is detected;
Brillouin scattering correlation region analysis that is connected to both ends of the hybrid neural network sensor 110 and measures the static response of the facility 200 corresponding to the Brillouin frequency, which is an optical signal in the length direction of the optical fiber when the abnormal signal is detected (BOCDA) interrogator 140; And
Including a local damage check unit 150 to determine whether the local damage to the facility 200 by analyzing the static response measured by the BOCDA interrogator 140,
The hybrid neural network sensor 110 primarily measures the dynamic response of the facility 200 using an optical fiber Bragg grating (FBG), and uses the high resolution of the Brillouin scattering correlation region analysis (BOCDA) to use the facility ( 200), a facility health monitoring system using an optical fiber-based hybrid neural network sensor, characterized in that the static response is secondarily measured. The method of claim 1,
The FBG interrogator 120 primarily measures the facility 200 after measuring the initial dynamic response of the facility 200, and the BOCDA interrogator 140 measures the initial dynamic response of the facility 200. After measuring an initial static response, a facility health monitoring system using an optical fiber-based hybrid neural network sensor, characterized in that, when the abnormal signal is detected, the facility 200 is secondarily measured. The method of claim 1,
An optical fiber-based hybrid that further includes a facility health evaluation unit 160 that evaluates the health of the facility 200 in response to the local damage degree of the facility 200 identified by the local damage check unit 150 Facility health monitoring system using neural network sensor. The method of claim 3,
An optical fiber-based hybrid neural network sensor, characterized in that the abnormal signal determination unit 130, the local damage check unit 150 and the facility health evaluation unit 160 are implemented on a manager terminal installed near the facility 200 Facility health monitoring system using. The method of claim 1,
The BOCDA interrogator 140 applies a wavelength of 1548 nm to the optical fiber that is the hybrid neural network sensor 110. Facility health monitoring system using an optical fiber-based hybrid neural network sensor. The method of claim 5,
The FBG interrogator 120 considers that the BOCDA interrogator 140 applies an optical signal having a wavelength of 1548 nm, so that mutual interference does not occur, so that the hybrid wavelength is excluded from the wavelength of 1545 nm to 1551 nm. A facility health monitoring system using an optical fiber-based hybrid neural network sensor, which is applied to an optical fiber, which is a neural network sensor 110. The method of claim 1,
The facility 200 is a facility health monitoring system using an optical fiber-based hybrid neural network sensor, wherein the facility 200 is a large structure including a building, a bridge, and a dam. a) installing a hybrid neural network sensor 110 using one optical fiber in the facility 200;
b) measuring the initial response of the facility 200 using an optical fiber Bragg grating (FBG) interrogator 120 and a Brillouin scattering correlation region analysis (BOCDA) interrogator 140, respectively;
c) constantly measuring the dynamic response of the facility 200 corresponding to the FBG wavelength, which is an optical signal in the length direction of an optical fiber, using the FBG interrogator 120;
d) analyzing the measured dynamic response to evaluate the global state of the facility 200 and determining whether an abnormal signal is detected;
e) when the abnormal signal is detected, measuring, by the BOCDA interrogator 140, a static response of the entire section or a partial section of the facility 200 corresponding to the Brillouin frequency, which is an optical signal in the longitudinal direction of the optical fiber; And
f) including the step of confirming local damage to the facility 200 according to the static response measurement of the BOCDA interrogator 140,
The hybrid neural network sensor 110 uses one optical fiber, and the hybrid neural network sensor 110 primarily measures the dynamic response of the facility 200 using an optical fiber Bragg grating (FBG), and Brillouin scattering A facility health monitoring method using an optical fiber-based hybrid neural network sensor, characterized in that the static response of the facility 200 is secondarily measured using a high resolution of a correlation domain analysis (BOCDA) measurement method. The method of claim 8,
The FBG interrogator 120 primarily measures the facility 200 after measuring the initial dynamic response of the facility 200, and the BOCDA interrogator 140 measures the initial dynamic response of the facility 200. After measuring the initial static response, the facility health monitoring method using an optical fiber-based hybrid neural network sensor, characterized in that when the abnormal signal is detected, the facility 200 is secondarily measured. The method of claim 8,
g) A facility health monitoring method using an optical fiber-based hybrid neural network sensor, further comprising the step of evaluating the health of the facility 200 in response to the local damage degree of the facility 200 in which the local damage is confirmed. The method of claim 8,
The BOCDA interrogator 140 applies a wavelength of 1548 nm to the optical fiber, which is the hybrid neural network sensor 110. Facility health monitoring method using an optical fiber-based hybrid neural network sensor. The method of claim 11,
The FBG interrogator 120 considers that the BOCDA interrogator 140 applies an optical signal having a wavelength of 1548 nm, so that mutual interference does not occur, so that the hybrid wavelength is excluded from the wavelength of 1545 nm to 1551 nm. Facility health monitoring method using an optical fiber-based hybrid neural network sensor, characterized in that applying to an optical fiber, which is a neural network sensor 110.

Images