Composite detection optical fiber and valve leakage detection device and method using same

Abstract

The invention relates to a composite detection optical fiber and a valve leakage detection device, the composite detection optical fiber comprises an optical fiber unit and a grating unit, the optical fiber unit comprises a temperature sensing optical fiber and a vibration sensing optical fiber, and the grating unit comprises a pressure optical fiber grating and a sound wave optical fiber grating; the valve leakage detection device comprises a composite detection optical fiber, an optical fiber data acquisition module, a data processing module, a neural network processing module and an output display module. The one-cable multi-core composite detection optical fiber is simple to install and small in laying amount; four types of sensors are adopted to analyze data of different areas of the valve, so that the inner leakage of the valve can be detected, and the outer leakage of the valve can also be detected; the neural network processing module not only can accurately judge the leakage of the valve, but also can accurately position the leakage area of the valve, predict the leakage range of the valve, establish a multi-parameter monitoring model of the valve, and provide visual and accurate display for the safe operation of the valve.

Description

Technical Field

[0001] The present invention relates to the technical field of leakage monitoring, and particularly to a composite detection optical fiber and a valve leakage detection device and method using the optical fiber. Background Art

[0002] In pipeline systems such as petrochemical and natural gas industries, valves are an indispensable part of the equipment, and valves are also widely used in liquefied hydrocarbon pipelines and equipment. Valves on liquefied hydrocarbon pipelines and equipment are often in contact with liquefied hydrocarbon chemicals, and these liquefied hydrocarbon chemicals often have high pressure and flow rate. After long-term use, valve leakage accidents often occur. Valve leakage can cause the loss of transmission medium and production failures. Therefore, in the actual production process, timely and accurate detection of valve leakage can avoid major accident disasters and significant economic losses, which is of great significance.

[0003] Valve leakage can be divided into two types: external leakage and internal leakage, depending on the flow direction of the medium. Valve external leakage refers to the leakage of the transmission medium to the outside of the valve body, mainly caused by failures at the sealing packing of the valve stem and the valve body connection. The main reasons for valve external leakage are improper installation of flanges or packings, etc. The leaked medium is directly exposed to the on-site environment and is extremely likely to cause a fire when encountering high-energy equipment or high-temperature areas. Valve internal leakage means that when the fluid passes through the valve, the fluid cannot be blocked by the valve plate, valve flap, etc. and continues to flow out in the outlet direction. Valve internal leakage occurs inside the valve cavity and is not easily observable, so it is often easily overlooked. However, the internal leakage failure of the valve has a very significant impact on the safe operation of production.

[0004] There are mainly two existing conventional valve leakage detection methods: one is direct observation detection, that is, technicians directly observe with the naked eye and listen with the ear. However, when the leaked substance is seen and the sound of steam leakage is heard, the leakage amount may already be too large; the other is to use a handheld valve leakage device to detect whether there is leakage in the valve. For valve internal leakage, means such as acoustic sensors, ultrasonic detection, and infrared thermal imaging detection are currently used.

[0005] A Chinese invention patent with an application number of 201010590565.1 (publication number CN102095083A) discloses a detection system for gas valve leakage, mainly using temperature transmitters, ultrasonic detectors, and pressure detectors installed before and after the valve for leakage detection. This invention sets up multiple independent detection instruments, with complex equipment installation and configuration, and it is an active detection device that is not intrinsically safe. Moreover, the three sensors are independently set and cannot determine multiple parameters for the same area, so the leakage positioning accuracy is insufficient. At the same time, it does not analyze different working conditions. There are many working conditions during the operation of the valve, and there are different interferences, which bring interference to the detection of temperature, pressure, and ultrasonic waves. Secondly, it does not propose a design for data fusion determination of several sensors, and simply determines leakage from the detection value of a single sensor, with limited accuracy and precision and a high false alarm rate.

[0006] Generally speaking, the following problems still exist in the existing technical solutions: (1) When using multi-sensor detection, it is often impossible to conduct centralized detection of multiple parameters for a certain area, cannot centrally detect various leakage states of a certain area, and fails to detect the characteristics of the valve and classify and detect the areas prone to leakage of the valve by region; (2) When setting multiple detectors to detect leakage, there are many types of sensors, a large amount of sensor equipment installation, high installation requirements, a large amount of fiber optic sensor laying, and it also fails to detect internal and external leakage of the valve simultaneously; (3) The existing technology does not solve the interference problem, and the detection of the sensor is affected by different interference factors such as valve operation, pipeline operation, and external environment; (4) It fails to analyze according to different valve specifications, parameters, valve states, pipeline states, material pressure, material flow rate, etc., and does not have good applicability and pertinence, and the detection accuracy is limited. Summary of the Invention

[0007] The first technical problem to be solved by the present invention is to provide a composite detection optical fiber in view of the current situation of the prior art.

[0008] The second technical problem to be solved by the present invention is to provide a valve leakage detection device applying the above composite detection optical fiber in view of the current situation of the prior art.

[0009] The third technical problem to be solved by the present invention is to provide a method for detecting valve leakage used in conjunction with the above valve leakage detection device in view of the current situation of the prior art.

[0010] The technical solution adopted by the present invention to solve the above first technical problem is as follows: The composite detection optical fiber is characterized in that: the composite detection optical fiber includes an optical fiber unit and a grating unit, the optical fiber unit includes a temperature sensing optical fiber and a vibration sensing optical fiber, and the grating unit includes a pressure fiber grating and an acoustic fiber grating.

[0011] In order to prevent the detection light of both the optical fiber unit and the grating unit from interfering with each other, an isolation layer is provided inside the composite detection optical fiber. The isolation layer separates the optical fiber unit and the grating unit. The temperature sensing optical fiber and the vibration sensing optical fiber are located on one side of the isolation layer, and the pressure optical fiber grating and the acoustic optical fiber grating are located on the other side of the isolation layer.

[0012] Further, a grating unit is provided at the same interval along the length direction of the composite detection optical fiber, and the pressure optical fiber grating and the acoustic optical fiber grating in each grating unit are encapsulated at the same position.

[0013] Further, the interval distance of each grating unit is within 1 meter. The interval distance is selected according to the requirements of the positioning accuracy.

[0014] In order to facilitate the fixing and installation of the composite detection optical fiber, welding points and / or magnetic attraction points are provided at the encapsulation positions of the pressure optical fiber grating and the acoustic optical fiber grating. The welding points and the magnetic attraction points enable the composite detection optical fiber to be fixed more accurately at specific positions near the valve and the valve, providing a basis for accurately detecting the leakage point.

[0015] Further, an outer sheath is provided outside the composite detection optical fiber, and the cross section of the outer sheath is flat. The flat outer sheath is convenient for attaching and fixing at positions near the valve and the valve.

[0016] The technical solution adopted by the present invention to solve the above second technical problem is as follows: The valve leakage detection device applying the above composite detection optical fiber is characterized in that: the composite detection optical fiber is laid on the outer wall of the valve body of the valve, or the inlet flange of the valve, or the outlet flange of the valve, or the pipeline at the upstream and downstream of the valve. The composite detection optical fiber is laid at the above positions for detecting external leakage.

[0017] Further, the valve stem and the valve body are connected by a connecting flange, and the composite detection optical fiber is laid upstream and downstream of the connecting flange and in the connecting surface area of the connecting flange, or the composite detection optical fiber is laid at the valve packing. The composite detection optical fiber is laid at the above positions for detecting internal leakage.

[0018] In order to accurately locate the position where the internal leakage occurs, the pressure optical fiber grating and the acoustic optical fiber grating are arranged in the connecting surface area of the connecting flange or at the valve packing.

[0019] Further, the valve leakage detection device further includes an optical fiber data acquisition module, a data processing module, a neural network processing module, and an output display module; a laser is used as the incident light source, and the incident light source enters the composite detection optical fiber and reaches the temperature sensing optical fiber, the vibration sensing optical fiber, the pressure fiber grating, and the acoustic fiber grating respectively. The composite detection optical fiber is connected to the optical fiber data acquisition module, and the optical fiber data acquisition module, the data processing module, the neural network processing module, and the output display module are connected in sequence.

[0020] Further, the data processing module also receives the valve status, material flow rate, and material pressure data sent by the DCS distributed control system.

[0021] Further, the training of the neural network model of the neural network processing module includes the following steps:

[0022] A1: Collect the parameter signals of each parameter during leakage under multiple working conditions of the pipeline and valve, as well as the parameter signals of common non-leakage.

[0023] A2: Extract the sample data sets of the temperature, pressure, vibration, and sound waves of each valve in the leakage and non-leakage states; among them, the leakage state includes the parameter data of leakage in different regions, different working conditions, different valves, and different sizes, and the non-leakage state includes the parameter data of different regions, different working conditions, and different valves during normal operation.

[0024] A3: Divide the sample data set into a training set, a validation set, and a test set.

[0025] A4: Process the training set samples to optimize the training model.

[0026] A5: Optimize the neural network model parameters based on the validation set.

[0027] Further, the input values of the neural network model include at least one of the following: the detection value of the temperature sensing optical fiber, the detection value of the vibration sensing optical fiber, the detection value of the pressure fiber grating sensor, the detection value of the acoustic fiber grating sensor, the valve type, the valve status, the valve size, the material pressure, the material flow rate, and the detection area.

[0028] Further, the output values of the neural network model include at least one of the following: the leakage type, the leakage degree, and the leakage area.

[0029] The technical solution adopted by the present invention to solve the above-mentioned third technical problem is as follows: The valve leakage detection method, used in cooperation with the above-mentioned valve leakage detection device, is characterized by including the following steps:

[0030] S1: Lay a composite detection optical fiber near the valve to be detected.

[0031] S2: The laser serves as an incident light source and enters the composite detection optical fiber. The composite detection optical fiber detects the temperature, vibration, stress, and acoustic wave parameters in the detected area and transmits the parameters to the optical fiber data acquisition module.

[0032] S3: The data processing module receives the temperature, vibration, stress, and acoustic wave parameter data collected by the optical fiber data acquisition module and the data from the DCS distributed control system, and processes the data.

[0033] S4: The neural network processing module determines the leakage model based on the data processing results.

[0034] S5: The output display module outputs the determination result.

[0035] Compared with the prior art, the advantages of the present invention are as follows:

[0036] (1) A composite detection optical fiber is proposed, in which a temperature sensing optical fiber, a vibration sensing optical fiber, a pressure fiber grating, and an acoustic wave fiber grating are arranged in one cable with multiple cores. The composite detection optical fiber is simple to install and has a small laying amount. Moreover, four types of sensors are used to analyze the data in different areas of the valve, which can not only detect the internal leakage of the valve but also detect the external leakage of the valve, and the detection range is complete.

[0037] (2) On the premise that the positioning accuracy of the temperature sensing optical fiber and the vibration sensing optical fiber is 0.5 - 1 m, the encapsulation parts of the pressure fiber grating and the acoustic wave fiber grating are accurately laid in the areas prone to leakage, which is convenient for the pressure fiber grating and the acoustic wave fiber grating to transmit the position signals to the data processing system, and the position signals of the optical fiber unit and the grating unit are combined to more accurately locate the position where the leakage occurs.

[0038] (3) The state of the valve, the working state of the pipeline, the pipeline pressure, the pipeline flow rate, etc. are comprehensively considered, and the determination is carried out through neural network learning to eliminate the interference signals in the working process of the pipeline valve, greatly improving the accuracy of the determination, and having pertinence and applicability.

[0039] (4) The neural network processing module can not only accurately determine the leakage of the valve, but also accurately locate the leakage area of the valve, predict the leakage range of the valve, and establish a multi-parameter monitoring model for the valve, providing an intuitive and accurate display for the safe operation of the valve. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic structural diagram of the composite detection optical fiber in the embodiment of the present invention;

[0041] Figure 2 It is a schematic laying diagram of the composite detection optical fiber when detecting external leakage in the embodiment of the present invention;

[0042] Figure 3Schematic diagram of the laying of the composite detection optical fiber for detecting internal leakage in the embodiments of the present invention;

[0043] Figure 4 Schematic diagram of the composition of the valve leakage detection device in the embodiments of the present invention;

[0044] Figure 5 Flow chart of valve leakage detection in the embodiments of the present invention;

[0045] Figure 6 Schematic diagram of the neural network model of the neural network processing module in the embodiments of the present invention. Detailed implementation manners

[0046] The present invention will be further described in detail below in conjunction with the embodiments with reference to the drawings.

[0047] Embodiment 1

[0048] As Figure 1 shown, a preferred embodiment of the composite detection optical fiber of the present invention is provided. The composite detection optical fiber 1 has multiple cores in one cable, and includes an optical fiber unit, a grating unit, and an isolation layer 15 therein. An outer sheath 17 is provided outside the composite detection optical fiber. The isolation layer 15 separates the optical fiber unit and the grating unit, and the outer sheath 17 with a flat cross-section is sleeved outside the isolation layer 15, the optical fiber unit, and the grating unit.

[0049] Among them, the optical fiber unit includes a temperature sensing optical fiber 11 and a vibration sensing optical fiber 12, and the grating unit includes a pressure optical fiber grating 13 and an acoustic wave optical fiber grating 14. The temperature sensing optical fiber 11 and the vibration sensing optical fiber 12 are located on one side of the isolation layer 15, and the pressure optical fiber grating 13 and the acoustic wave optical fiber grating 14 are located on the other side of the isolation layer 15.

[0050] A grating unit is provided at a certain distance in the length direction of the composite detection optical fiber 1, and the pressure optical fiber grating 13 and the acoustic wave optical fiber grating 14 in each grating unit are encapsulated at the same position. The temperature sensing optical fiber 11 and the vibration sensing optical fiber 12 in the composite detection optical fiber can determine the location of the leakage by using the optical fiber transmission characteristics. Usually, the detection and positioning accuracy is 0.5 - 1 m. Then, how to improve the positioning accuracy within 1 m? The present invention proposes to set a grating unit at a certain distance (selected within 1 m according to the requirements of the positioning accuracy) in the composite detection optical fiber 1. Through the alarm positioning of the above grating unit at a fixed point, the changing position within 1 m can be accurately located. In addition, welding points 161 and / or magnetic attraction points 162 are provided at the encapsulation positions of the pressure optical fiber grating 13 and the acoustic wave optical fiber grating 14. The welding points 161 and the magnetic attraction points 162 are fixed at the positions of the valve and near the valve where leakage is likely to occur, providing support for accurately detecting the leakage point.

[0051] A variety of optical fiber sensors are used to detect the leakage state of the valve. By connecting optical fiber filters with different wavelengths, the optical fiber sensors do not interfere with each other, improving the detection accuracy. Ordinary optical fiber is used as the detection optical fiber, and a temperature measurement optical fiber core and a vibration optical fiber core are set. Pressure fiber gratings and acoustic fiber gratings are arranged at equal intervals inside. The quasi-distributed optical fiber sensing technology of acoustic waves can multiplex multiple stress sensors with the same optical fiber, so as to realize the accurate measurement of fixed-point temperature, pressure and acoustic waves at key parts of the pipeline to be measured, such as flange parts, packing parts, valve cover parts, etc.

[0052] Embodiment 2

[0053] As Figures 2 to 4 shown is a preferred embodiment of the valve leakage detection device of the present invention. The valve leakage detection device includes the above-mentioned composite detection optical fiber 1, and also includes an optical fiber data acquisition module, a data processing module, a neural network processing module and an output display module; a laser is used as the incident light source, and the incident light source enters the composite detection optical fiber 1 and reaches the temperature sensing optical fiber 11, the vibration sensing optical fiber 12, the pressure fiber grating 13 and the acoustic fiber grating 14 respectively. The composite detection optical fiber 1 is connected to the optical fiber data acquisition module, and the optical fiber data acquisition module, the data processing module, the neural network processing module and the output display module are connected in sequence. At the same time, the data processing module receives the data of the valve 2 state, the material flow rate and the material pressure sent by the DCS distributed control system, as Figure 4 shown. When the data processing module receives the data transmitted from the composite detection optical fiber 1 and the DCS distributed control system, the valve leakage detection device can judge and determine the leakage area of the valve 2, send out a leakage alarm message, and can also predict the leakage range. The data of the composite detection optical fiber 1 includes temperature changes, vibration signals, pressure changes and acoustic signals generated due to the leakage of liquefied hydrocarbon medium, as well as position data for positioning the leakage area by the composite detection optical fiber 1.

[0054] The leakage caused by the valve 2 during installation and use includes internal leakage and external leakage. Therefore, in a specific implementation manner, the composite detection optical fiber 1 of the valve leakage detection device is laid on the outer wall of the valve body of the valve 2, or the inlet flange 41 of the valve 2, or the outlet flange 42 of the valve 2, or the pipeline 3 upstream and downstream of the valve 2 for detecting the external leakage of the valve, as Figure 2 shown; the composite detection optical fiber 1 of the valve leakage detection device is also laid and connected to the upstream and downstream of the connecting flange 21 connecting the valve stem and the valve body and the connecting surface area of the connecting flange 21. The composite detection optical fiber 1 is also laid at the valve packing 22 for detecting the internal leakage of the valve, as Figure 3 shown.

[0055] Specifically, regarding internal leakage of the valve: The valve stem and the valve body of the valve 2 are connected by a connecting flange 21. The composite detection optical fiber 1 is wound around the upstream and downstream sides of the connecting flange 21 and the connecting surface area of the connecting flange 21, and at least the pressure fiber Bragg grating 13 and the acoustic fiber Bragg grating 14 are arranged on the connecting surface of the connecting flange 21; the composite detection optical fiber 1 is wound inside the valve packing 22, and at the same time the pressure fiber Bragg grating 13 and the acoustic fiber Bragg grating 14 are arranged inside the valve packing 22.

[0056] When there is internal leakage in the valve, the transport medium inside the valve will form a multiphase turbulent jet, which impacts the valve wall and radiates energy outward, generating a high-speed contraction flow noise. The stress wave of the generated noise signal will propagate along components such as the valve body, that is, acoustic emission signals such as jet noise, vortex noise, and turbulent noise; when there is a leakage hole on the valve sealing surface, the liquefied hydrocarbon medium jets through the valve damage point due to the pressure difference and collides with the inner wall of the valve and the pipe wall to form turbulence, generating transient acoustic signals. The acoustic fiber Bragg grating 14 of the composite detection optical fiber 1 can detect the acoustic signal during leakage. And when there is internal leakage in the valve, it will cause weak vibration of the pipeline. The vibration sensing optical fiber 12 attached to the outer wall of the valve and the pipeline near the valve can collect the vibration signal of the leakage, and the internal leakage fault of the valve can be detected by analyzing the vibration signal of the pipe wall when there is internal leakage in the valve. In addition, when there is internal leakage in the valve, it will cause a temperature difference between the upstream and downstream of the valve. By using the temperature difference between the liquefied hydrocarbon temperature transmitted by the valve and the ambient temperature, a temperature sensing optical fiber 11 is arranged near the valve body to measure the temperature of the valve body. When there is leakage in the valve, the heat loss caused will lead to an obvious change in the temperature of the outside of the valve or the pipeline downstream of the valve. By using the temperature sensing optical fibers 11 arranged upstream and downstream of the valve, the leakage fault of the valve can be detected by detecting the temperature change.

[0057] During the actual operation process, since the externally leaked medium of the valve will escape into the external environment, the harm caused by external leakage will be more serious than internal leakage. Specifically, regarding external leakage of the valve: The main external leakage parts of the valve are summarized as follows: (1) Valve body leakage, that is, leakage occurs when there are cracks or other conditions in the valve body itself; (2) Leakage at the valve packing, including loose packing gland, imperfect sealing packing, aging or wear of the packing by the valve stem. The contact part between the valve stem and the packing is the most prone to leakage; (3) Leakage at the valve body connection part: The seal at the valve body connection part refers to the seal between the valve body and the valve cover, which is usually flange connection seal. Due to reasons such as improper tightening of the connection bolts and poor processing quality of the flange sealing surface, leakage can occur at the valve body connection part; (4) Leakage at the valve pipeline connection: Usually also flange seal, and external leakage can occur at the seal.

[0058] When external leakage of the valve occurs, the outflow of the medium outside the valve body will cause temperature changes. For example, the gasification of liquefied hydrocarbon medium brings about temperature changes. After leakage, there will be a local temperature mutation. The liquefied hydrocarbon leaks and vaporizes by absorbing heat, causing the surrounding temperature to decrease. Therefore, the temperature change near the valve is monitored in real time through the temperature sensing optical fiber 11 wound around the valve; after the external leakage of the valve, a pressurized gas-liquid two-phase will be released at the leakage point. Pressure fiber Bragg gratings are arranged at the flange sealing surface, packing sealing surface, packing gland, valve cover sealing surface, etc. where leakage is likely to occur to monitor the pressure value in real time. The pressurized liquefied hydrocarbon leakage will impact the pressure fiber Bragg grating 13; when the valve has external leakage, friction will occur between the gas-liquid and the leakage hole wall, and the fluid medium leaking from the valve will impact the sealing surface and generate elastic waves. The valve leakage information is obtained by detecting the high-frequency vibration signal generated during leakage. The high-frequency vibration signal generated during the leakage of the valve is detected by the vibration sensing optical fiber 12 to achieve the purpose of determining leakage.

[0059] Embodiment 3

[0060] As Figure 6 shown, the present invention also introduces a neural network processing module to improve the accuracy of detection and determination. By collecting and simulating the valve in normal state, internal leakage state, and external leakage state, and combining different parameter values in cases such as valve type, valve size, valve opening, closing, and opening degree, the architecture and hierarchy of the neural network are established. The neural network of the present invention is a deep neural network, which is an algorithm for modeling high-complexity data through multiple non-linear transformations. The deep neural network is divided into three layers, namely the input layer, the hidden layer, and the output layer. The input values of the input layer of the neural network model: classify and grade the parameters of each sensor, and at the same time normalize each value. The input variables include:

[0061] The detection value T of the temperature sensing optical fiber;

[0062] The detection value Z of the vibration sensing optical fiber;

[0063] The detection value P1 of the pressure fiber Bragg grating sensor;

[0064] The detection value V of the acoustic fiber Bragg grating sensor;

[0065] Valve type: butterfly valve, gate valve, ball valve, etc.;

[0066] Valve state: open, closed, opening degree;

[0067] Valve size: S;

[0068] Medium pressure: P2, output from the DCS to the data processing module;

[0069] Medium flow rate: V, output from the DCS to the data processing module;

[0070] Valve inspection areas: inlet flange, outlet flange, valve stem, packing, valve cover, valve body.

[0071] Output value of the output layer of the neural network model: Valve leakage determination output of the present invention:

[0072] Leakage type (non - leakage, internal leakage, external leakage);

[0073] Leakage degree (micro - leakage, medium leakage, severe leakage);

[0074] Leakage area (valve body, valve cover, packing, interface flange, etc.).

[0075] The neural network uses the detection values of the temperature - sensing optical fiber 11, vibration - sensing optical fiber 12, pressure fiber grating 13, acoustic fiber grating 14, sensor detection area, valve type, valve size, valve status, material pressure, etc. as monitoring parameters, thereby constructing a neural network leakage determination model. The parameter values of the sensors are respectively normalized into N variables as the input of the neural network, and at the same time, the output values are selected as leakage determination, leakage degree, and leakage area. In the design of the neural network, the most important thing is to determine the optimal number of neurons in the hidden layer. The correct selection of the number of hidden - layer nodes is very important and requires comprehensive consideration of various conditions (such as comprehensive consideration of learning and training time and error, etc.) to determine the number of hidden - layer nodes. Through learning samples, the learning and training of the network are realized. The selected neural network algorithm can better describe the correlation between various parameters by selecting an appropriate transfer function.

[0076] Among them, the training of the neural network model of the neural network processing module includes the following steps:

[0077] A1: Collect the parameter signals of each parameter under leakage conditions in multiple working conditions of the pipeline 3 and the valve 2 and the parameter signals of common non - leakage.

[0078] A2: Extract the sample data sets of the temperature, pressure, vibration, and sound waves of each valve in the leakage and non - leakage states; among them, the leakage state includes the parameter data of different regions, different working conditions, different valves, and different sizes of leakage, and the non - leakage state includes the parameter data of different regions, different working conditions, and different valves during normal operation.

[0079] A3: Divide the sample data set into a training set, a validation set, and a test set;

[0080] A4: Process the training - set samples to optimize the training model;

[0081] A5: Optimize the neural network model parameters based on the validation set.

[0082] During the detection of the sensor, there will be different working conditions such as valves and pipelines. Different interferences will be generated under various working conditions, such as environmental background noise, interference from valve opening and closing actions, working states of different pipelines, interference from pump startup and shutdown, etc. The neural network has strong environmental adaptability, learning ability, fault tolerance and parallel processing ability. When the input-output relationship cannot be described by a specific function expression, the neural network intelligent algorithm can simulate the internal connection between the input and output through learning and training to achieve the effect of artificial intelligence judgment.

[0083] Embodiment 4

[0084] As Figure 5 shown is the flowchart of the valve leakage detection method of the present invention. The method includes the following steps:

[0085] S1: Lay the composite detection optical fiber 1 near the valve 2 to be detected;

[0086] S2: The laser serves as an incident light source and enters the composite detection optical fiber 1. The composite detection optical fiber 1 detects the temperature, vibration, stress, and acoustic wave parameters of the detected area and transmits the parameters to the optical fiber data acquisition module;

[0087] S3: The data processing module receives the temperature, vibration, stress, acoustic wave parameter data collected by the optical fiber data acquisition module and the data of the DCS distributed control system and performs data processing;

[0088] S4: The neural network processing module determines the leakage model for the data processing result;

[0089] S5: The output display module outputs the determination result.

[0090] This method uses the temperature sensing optical fiber 11, vibration sensing optical fiber 12, pressure fiber grating 13, and acoustic wave fiber grating 14 to collect the leakage signal of the valve. When the valve leaks, the leakage position of the valve can be located by analyzing the changes of various parameters in different detection areas on the composite detection optical fiber 1, and the leakage size and trend can be predicted; due to the correlation of temperature, vibration, stress, and acoustic wave parameters, a detection and determination model is established through the neural network algorithm to perform real-time online intelligent detection, identification and classification of valve leakage events, and solve the problem of accurate online leakage detection and positioning of valves in complex environments.

Claims

1. A composite detection optical fiber, characterized in that: The composite detection optical fiber (1) includes an optical fiber unit and a grating unit. The optical fiber unit includes a temperature sensing optical fiber (11) and a vibration sensing optical fiber (12). The grating unit includes a pressure optical fiber grating (13) and an acoustic wave optical fiber grating (14).

2. The composite detection optical fiber according to claim 1, characterized in that: An isolation layer (15) is provided inside the composite detection optical fiber (1). The isolation layer (15) separates the optical fiber unit and the grating unit. The temperature sensing optical fiber (11) and the vibration sensing optical fiber (12) are located on one side of the isolation layer (15), and the pressure optical fiber grating (13) and the acoustic wave optical fiber grating (14) are located on the other side of the isolation layer (15).

3. The composite detection optical fiber according to claim 2, wherein: A grating unit is provided at the same interval along the length direction of the composite detection optical fiber. The pressure optical fiber grating (13) and the acoustic wave optical fiber grating (14) in each grating unit are encapsulated at the same position.

4. The composite detection optical fiber according to claim 3, wherein: The interval distance of each grating unit is within 1 meter.

5. The composite detection optical fiber according to claim 4, wherein: Welding points (161) and / or magnetic attraction points (162) are provided at the encapsulation positions of the pressure optical fiber grating (13) and the acoustic wave optical fiber grating (14).

6. The composite detection optical fiber according to claim 5, wherein: An outer sheath (17) is provided outside the composite detection optical fiber (1). The cross section of the outer sheath (17) is flat.

7. A valve leakage detection device using the composite detection optical fiber as described in any one of claims 1 to 6, characterized in that: The composite detection optical fiber (1) is laid on the outer wall of the valve body of the valve (2), or the inlet flange (41) of the valve (2), or the outlet flange (42) of the valve (2), or the pipeline (3) upstream and downstream of the valve (2).

8. The valve leakage detection device according to claim 7, wherein: The valve stem and the valve body are connected by a connecting flange (21). The composite detection optical fiber (1) is laid upstream and downstream of the connecting flange (21) and in the connecting surface area of the connecting flange (21), or the composite detection optical fiber (1) is laid at the valve packing (22).

9. The valve leakage detection device according to claim 8, characterized in that: The pressure optical fiber grating (13) and the acoustic wave optical fiber grating (14) are arranged in the connecting surface area of the connecting flange (21) or at the valve packing (22).

10. The valve leakage detection device according to claim 9, characterized in that: It also includes an optical fiber data acquisition module, a data processing module, a neural network processing module, and an output display module. A laser is used as the incident light source. The incident light source enters the composite detection optical fiber (1) and reaches the temperature sensing optical fiber (11), the vibration sensing optical fiber (12), the pressure optical fiber grating (13), and the acoustic wave optical fiber grating (14) respectively. The composite detection optical fiber (1) is connected to the optical fiber data acquisition module, and the optical fiber data acquisition module, the data processing module, the neural network processing module, and the output display module are connected in sequence.

11. The valve leakage detection device according to claim 10, wherein: The data processing module also receives the valve (2) status, material flow rate, and material pressure data sent by the DCS distributed control system.

12. The valve leakage detection device according to claim 11, characterized in that: Training of the neural network model of the neural network processing module includes the following steps: A1: Collect the parameter signals of each parameter under leakage in multiple working conditions of the pipeline (3) and the valve (2) and the parameter signals of common non-leakage. A2: Extract the sample data sets of the temperature, pressure, vibration, and acoustic wave of each valve in the leakage and non-leakage states. Among them, the leakage states include the parameter data of leakage in different regions, different working conditions, different valves, and different sizes, and the non-leakage states include the parameter data of different regions, different working conditions, and different valves during normal operation. A3: Divide the sample data set into a training set, a validation set, and a test set; A4: Process the training set samples to optimize the training model; A5: Optimize the neural network model parameters based on the validation set.

13. The valve leakage detection device according to claim 12, characterized in that: The input values of the neural network model include at least one of the following: the detection value of the temperature sensing optical fiber (11), the detection value of the vibration sensing optical fiber (12), the detection value of the pressure fiber grating (13) sensor, the detection value of the acoustic fiber grating (14) sensor, the type of the valve (2), the state of the valve (2), the size of the valve (2), the material pressure, the material flow rate, and the detection area.

14. The valve leakage detection device according to claim 12, wherein: The output values of the neural network model include at least one of the following: the leakage type, the leakage degree, and the leakage area.

15. A valve leakage detection method, used in conjunction with the valve leakage detection device according to any one of claims 7 to 14, characterized in that, It includes the following steps: S1: Lay the composite detection optical fiber (1) near the valve (2) to be detected; S2: The laser serves as the incident light source and enters the composite detection optical fiber (1). The composite detection optical fiber (1) detects the temperature, vibration, stress, and acoustic wave parameters of the detected area and transmits the parameters to the optical fiber data acquisition module; S3: The data processing module receives the temperature, vibration, stress, and acoustic wave parameter data collected by the optical fiber data acquisition module and the data of the DCS distributed control system and performs data processing; S4: The neural network processing module determines the leakage model for the data processing result; S5: The output display module outputs the determination result.

Images