Disaster warning device and method for drainage pipeline based on distributed optical fiber

Abstract

The present application belongs to the technical field of drainage pipeline structural health monitoring, and discloses a drainage pipeline disaster warning device and method based on distributed optical fiber, the device comprising: an optical fiber sensing module, including an optical frequency domain reflection demodulator and a plurality of sensing optical fibers; various sensing optical fibers are attached to the inner wall of the drainage pipeline according to a preset scheme, and the optical frequency domain reflection demodulator is used to obtain the sensing data of various sensing optical fibers; a data analysis module, which is used to analyze the sensing data by statistical methods, and when abnormal data is obtained by analysis, the disaster risk level corresponding to the abnormal data is determined by a machine learning algorithm; a disaster warning module, which generates warning information according to abnormal data and disaster risk level; an operation and maintenance decision module, which is used to make drainage pipe operation and maintenance decisions based on a decision-making system and warning information based on risk-informed theory. The present application can timely and accurately identify abnormal conditions in drainage pipelines at an early stage, extend the service performance of drainage pipelines, and reduce operation and maintenance costs.

Description

Technical Field

[0001] The present application relates to the technical field of drainage pipe structure health monitoring, and in particular to a drainage pipe disaster warning device and method based on distributed optical fiber. Background Art

[0002] Drainage pipes play an important role in the collection, purification and discharge of urban domestic sewage, rainfall runoff and other waste water. However, the construction quality of buried drainage pipe networks is poor, the management system is backward, and the aging of drainage pipe networks has led to ineffective flood discharge (rainwater and sewage confluence), ground collapse, black and smelly water bodies and other disasters that are very common in various cities.

[0003] The current assessment of the safety and service status of buried pipe networks is usually carried out in the form of regular or fixed-area inspections. Inspectors go to designated pipelines to carry out work, use on-site operating equipment to shoot and record videos of the inside of the pipelines, and then bring the videos back indoors for manual playback by professionals to identify the internal conditions of the drainage pipes. Deformation, cracking and leakage of the pipe walls caused by aging of buried drainage pipes are often difficult to detect in time by the human eye, which can easily lead to a secondary disaster chain, resulting in a high risk of major accidents and casualties. Summary of the invention

[0004] The present application provides a drainage pipe disaster warning device and method based on distributed optical fiber, which can timely and accurately identify abnormal conditions in the drainage pipe at an early stage, extend the service performance of the drainage pipe, and reduce operation and maintenance costs.

[0005] In a first aspect, an embodiment of the present application provides a drainage pipeline disaster warning device based on distributed optical fiber, comprising:

[0006] The optical fiber sensing module includes an optical frequency domain reflection demodulator and a plurality of sensing optical fibers; the various sensing optical fibers are attached to the inner wall of the drainage pipe according to a preset scheme, and the optical frequency domain reflection demodulator is used to obtain sensing data of the various sensing optical fibers;

[0007] A data analysis module is used to analyze the sensing data using statistical methods, and when abnormal data is obtained through the analysis, determine the disaster risk level corresponding to the abnormal data using a machine learning algorithm;

[0008] Disaster warning module, which generates warning information based on abnormal data and disaster risk levels;

[0009] The operation and maintenance decision module is used to make drainage pipe operation and maintenance decisions based on the risk-informed decision-making system and early warning information.

[0010] Further, the sensing optical fiber includes a temperature sensing optical fiber and a strain sensing optical fiber, and the sensing data includes temperature data and strain data.

[0011] Furthermore, the temperature sensing optical fiber is arranged at the top axis position of the inner wall of the drainage pipe; the strain sensing optical fiber is arranged at the top axis position, the bottom axis position and the side axis position of the inner wall of the drainage pipe respectively.

[0012] Furthermore, the temperature sensing optical fiber is encapsulated in an alumina tube.

[0013] Furthermore, the strain sensing optical fiber is also spirally arranged on the inner wall of the drainage pipe with a preset spiral pitch.

[0014] Furthermore, the optical frequency domain reflection demodulator is used to obtain temperature data of the temperature sensing optical fiber and strain data of the strain sensing optical fiber based on the optical frequency domain reflection technology of back Rayleigh scattering, and compensate the strain data according to the temperature data.

[0015] Furthermore, the optical frequency domain reflectometry demodulator is also used to upload temperature data and strain data to the cloud.

[0016] Further, the data analysis module includes a data transmission unit, a disaster identification unit, and a risk assessment unit;

[0017] The data transmission unit is used to obtain the sensing data of the optical frequency domain reflection demodulator through the cloud;

[0018] The disaster identification unit is used to analyze the strain data using the sliding window outlier analysis method;

[0019] The risk assessment unit is used to determine the disaster risk level of abnormal data using machine learning algorithms.

[0020] Further, the risk assessment unit is used to calculate the pipeline deformation corresponding to the abnormal data using a machine learning algorithm;

[0021] And, obtaining the pipeline cross-section location corresponding to the abnormal data in the optical frequency domain reflection demodulator and multiple strain data of the pipeline cross-section location, and calculating the pipeline deformation corresponding to the multiple strain data;

[0022] Furthermore, the deformation of each pipeline is compared to obtain the maximum deformation of the pipeline cross-section positioning, and the disaster risk level corresponding to the abnormal data is determined based on the maximum deformation and the pipeline diameter.

[0023] In a second aspect, an embodiment of the present application provides a drainage pipeline disaster warning method based on distributed optical fiber, comprising:

[0024] Acquire sensing data by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme;

[0025] Analyze the sensory data through statistical methods;

[0026] If abnormal data is obtained through analysis, the disaster risk level corresponding to the abnormal data is determined through machine learning algorithms;

[0027] Generate early warning information based on abnormal data and disaster risk levels;

[0028] Make drainage pipe operation and maintenance decisions based on a risk-informed decision-making system and early warning information.

[0029] Furthermore, the sensing optical fiber includes a temperature sensing optical fiber and a strain sensing optical fiber; the sensing data includes temperature data and strain data; the sensing data is obtained by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme, including:

[0030] The temperature data of each measuring point on the temperature sensing optical fiber and the strain data of each measuring point on the strain sensing optical fiber are obtained based on the optical frequency domain reflection technology of Rayleigh backscattering;

[0031] The temperature data is compensated to the strain data based on the location of each measuring point.

[0032] Furthermore, the above-mentioned acquisition of sensing data by various sensing optical fibers pasted on the inner wall of the drainage pipe according to a preset scheme also includes: uploading the acquired temperature data and strain data to the cloud for storage.

[0033] Furthermore, the statistical method is a sliding window outlier analysis method;

[0034] The above-mentioned analysis of the sensing data by statistical methods includes:

[0035] Applying the sliding window to the strain data of each measuring point at the first moment based on a preset sliding window length;

[0036] Calculate the inconsistency index corresponding to each strain data in the sliding window;

[0037] Calculate the threshold value corresponding to the sliding window based on each inconsistency indicator;

[0038] Slide the sliding window on each strain data at the first moment to obtain the inconsistency index corresponding to each strain data and the threshold value corresponding to each window;

[0039] According to the threshold value, it is judged whether there are outliers in each inconsistency indicator in the corresponding window;

[0040] If it exists, the outlier is detected based on the abnormal state diagnosis window to obtain abnormal data;

[0041] If no abnormal data exists or is not detected, the sliding window is applied to the strain data at the second moment.

[0042] Furthermore, the inconsistency index corresponding to each strain data in the above-mentioned calculation sliding window includes:

[0043] The mean and standard deviation were calculated based on each strain data;

[0044] The corresponding inconsistency index is calculated based on the mean, standard deviation and strain data.

[0045] Furthermore, the above-mentioned calculation of the threshold value corresponding to the sliding window based on each inconsistency index includes:

[0046] Sort each inconsistency indicator in descending order to obtain an indicator queue;

[0047] The threshold value is obtained according to the preset sliding window length, the preset guarantee rate and the indicator queue.

[0048] Furthermore, the above-mentioned outlier detection based on the abnormal state diagnosis window to obtain abnormal data includes:

[0049] Determine the target threshold value and each target measurement point of the window corresponding to the outlier;

[0050] Obtaining the preset diagnostic window length of the abnormal state diagnostic window;

[0051] Obtain strain data of the target measuring point at multiple times based on the preset diagnostic window length;

[0052] Calculate the inconsistency index corresponding to the target measuring point at each moment according to the strain data at each moment;

[0053] Compare the inconsistency index corresponding to each moment with the target threshold value to obtain the abnormal moment ratio;

[0054] If the abnormal time ratio exceeds the preset ratio, the target measurement point will be regarded as abnormal data.

[0055] Furthermore, the above-mentioned determination of the disaster risk level corresponding to the abnormal data by the machine learning algorithm includes:

[0056] Use machine learning algorithms to calculate the pipe deformation corresponding to abnormal data;

[0057] Obtain the pipeline cross-section location corresponding to the abnormal data and multiple strain data of the pipeline cross-section location;

[0058] Calculate the pipe deformation corresponding to multiple strain data;

[0059] The deformation of each pipeline is compared to obtain the maximum deformation of the pipeline cross-section positioning;

[0060] The disaster risk level corresponding to the abnormal data is determined based on the maximum deformation and pipeline diameter.

[0061] Furthermore, the warning information includes the disaster risk level and the location of the pipeline section corresponding to the abnormal data.

[0062] In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the drainage pipe disaster warning method based on distributed optical fiber as described in any of the above embodiments are performed.

[0063] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a drainage pipe disaster warning method based on distributed optical fiber as in any of the above embodiments.

[0064] In summary, compared with the prior art, the technical solution provided in the embodiment of the present application has at least the following beneficial effects:

[0065] The embodiment of the present application provides a drainage pipe disaster warning device based on distributed optical fiber, which obtains sensing data through distributed sensing optical fibers pasted on the inner wall of the drainage pipe, analyzes the sensing data through statistical methods, and when abnormal data is obtained through analysis, judges the disaster risk level corresponding to the abnormal data through machine learning algorithms, and generates warning information based on the abnormal data and the disaster risk level; the decision-making system based on risk-informed theory and the warning information make drainage pipe operation and maintenance decisions. By laying distributed optical fibers, the present application can sense the state data of the inner wall of the pipeline with high spatial resolution and high monitoring accuracy over a long range without destroying the original structure of the drainage pipe, and analyzes the sensing data and judges the disaster risk level in combination with statistical methods and machine learning algorithms, so that the abnormal situation of the drainage pipe can be identified in a timely and accurate manner at an early stage, and warnings and drainage pipe operation and maintenance decisions can be made when abnormal data is detected, thereby extending the service performance of the drainage pipe and reducing operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A structural diagram of a drainage pipe disaster warning device based on distributed optical fiber provided in one embodiment of the present application.

[0067] Figure 2 A schematic diagram of a distributed arrangement of sensing optical fibers in a drainage pipe provided in one embodiment of the present application.

[0068] Figure 3 A flowchart of a drainage pipe disaster warning method based on distributed optical fiber is provided as an embodiment of the present application.

[0069] Figure 4 A flowchart of a sliding window outlier analysis method provided for one embodiment of the present application. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0071] Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of this application.

[0072] See also Figure 1 In view of the problem that the existing buried pipe network status relies on human identification and investigation, the detection accuracy is low, and abnormal conditions such as aging and deformation of drainage pipes cannot be detected in time at an early stage, the embodiment of the present application provides a drainage pipe disaster warning device based on distributed optical fiber, which specifically includes the following modules:

[0073] The optical fiber sensing module 101 includes an optical frequency domain reflection demodulator and a plurality of sensing optical fibers; the various sensing optical fibers are pasted on the inner wall of the drainage pipe according to a preset scheme, and the optical frequency domain reflection demodulator is used to obtain sensing data of the various sensing optical fibers.

[0074] The sensing optical fiber includes temperature sensing optical fiber and strain sensing optical fiber, and the sensing data includes temperature data and strain data. The optical frequency domain reflectometry demodulator is a device with control function, data processing function and communication function, which is powered by a civilian power supply. The control function can set the sampling frequency and spatial resolution of the distributed optical fiber monitoring system; the communication function can be WiFi communication, Bluetooth communication or USB interface communication.

[0075] In some embodiments, the optical frequency domain reflectometry demodulator is used to obtain the temperature data of the temperature sensing optical fiber and the strain data of the strain sensing optical fiber based on the optical frequency domain reflectometry technology of backscattering Rayleigh, and compensate the strain data according to the temperature data. The temperature data and strain data obtained by the optical frequency domain reflectometry demodulator include the pipe cross-section location of the corresponding measuring point of the data.

[0076] The above-mentioned optical frequency domain reflection technology based on back-Rayleigh scattering can sense the deformation and temperature changes of the inner wall of the drainage pipe in real time. It has extremely high spatial resolution, which can reach the millimeter level at most. At the same time, it can also provide high measurement accuracy of 0.1με and 0.1℃, and can accurately measure the temperature data and strain data of the inner wall of the pipe.

[0077] In some embodiments, the optical frequency domain reflectometry demodulator can also upload the sensed temperature data and the temperature-compensated strain data to the cloud in real time for storage through the WiFi communication function.

[0078] Specifically, see Figure 2 The temperature sensing optical fiber can be arranged at the top axis position of the inner wall of the drainage pipe; the strain sensing optical fiber can be arranged at the top axis position, the bottom axis position and the side axis position of the inner wall of the drainage pipe respectively; the temperature sensing optical fiber and the strain sensing optical fiber are both glued with epoxy resin glue, and are connected to the optical frequency domain reflection demodulator through an optical fiber jumper.

[0079] The temperature sensing optical fiber in the above embodiment is arranged on the axis of the top of the pipe, which makes full use of the characteristic that the hot steam in the drainage pipe flows upward to ensure that the temperature data is not falsely low; and the strain sensing optical fiber is arranged on four axes of the pipe, such that the sensed strain data is more complete and comprehensive, avoiding omissions and blind spots.

[0080] In some embodiments, the strain sensing optical fiber can also be spirally arranged on the inner wall of the drainage pipe with a preset spiral pitch. The spiral optical fiber is mainly arranged in key monitoring pipe sections, such as the socket pipe interface, the joint between the pipeline and the inspection well, etc., and the spiral pitch is selected according to the importance of the monitored pipe section.

[0081] Furthermore, the temperature sensing optical fiber is encapsulated in an alumina tube, which can make the temperature sensing optical fiber not affected by the deformation of the drainage pipe. The measured data is only related to the change of ambient temperature in the pipe. It is used for temperature compensation of the strain data measured by the strain sensing optical fiber, thereby obtaining real-time and true strain data of the inner wall of the drainage pipe, ensuring the accuracy of drainage pipe identification and detection.

[0082] The data analysis module 102 is used to analyze the sensing data by using a statistical method, and when abnormal data is obtained through the analysis, determine the disaster risk level corresponding to the abnormal data by using a machine learning algorithm.

[0083] In some embodiments, the data analysis module 102 includes a data transmission unit, a disaster identification unit, and a risk assessment unit.

[0084] The data transmission unit is used to obtain the sensing data of the optical frequency domain reflectometry demodulator through the cloud.

[0085] The disaster identification unit is used to analyze the strain data using a sliding window outlier analysis method.

[0086] The risk assessment unit is used to determine the disaster risk level of abnormal data using machine learning algorithms.

[0087] Specifically, the risk assessment unit is used to calculate the pipeline deformation corresponding to the abnormal data using a machine learning algorithm; and, obtain the pipeline cross-section location corresponding to the abnormal data in the optical frequency domain reflection demodulator, and multiple strain data of the pipeline cross-section location, and calculate the pipeline deformation corresponding to the multiple strain data; and, compare the deformation amounts of each pipeline to obtain the maximum deformation amount of the pipeline cross-section location, and determine the disaster risk level corresponding to the abnormal data based on the maximum deformation amount and the pipeline diameter.

[0088] The disaster warning module 103 generates warning information according to abnormal data and disaster risk level.

[0089] Among them, the warning information includes the disaster risk level and the pipeline section location corresponding to the abnormal data.

[0090] Specifically, the output of the risk assessment unit is the disaster risk level and the abnormal data obtained by the disaster identification unit and the corresponding pipeline cross-section location. Therefore, the disaster warning module can determine the measurement point location based on the output result of the risk assessment unit, and use wireless transmission methods such as 5G to package the disaster warning time, pipeline cross-section location and disaster risk level and send them to the operation and maintenance decision module 104.

[0091] The operation and maintenance decision module 104 is used to make operation and maintenance decisions for the drainage pipe based on the decision-making system of risk-informed theory and early warning information.

[0092] Specifically, the operation and maintenance decision module 104 obtains a method for processing abnormal data of drainage pipes based on the decision system of risk-informed theory and the early warning information, that is, the operation and maintenance decision module 104 decides what measures to take to deal with the potential disasters of drainage pipes corresponding to the early warning information. For example, if the disaster risk level in the early warning information indicates a relatively serious degree in the decision system, or the pipeline section location in the early warning information is a key pipeline stage, the operation and maintenance decision module 104 will order the staff to immediately go to the corresponding pipeline for repair. If the disaster risk level is low, the operation and maintenance decision module will generate a detection cycle for the corresponding pipeline section location based on the decision system of risk-informed theory, and regularly remind the staff to conduct safety inspections, etc.

[0093] The above embodiment provides a drainage pipe disaster warning device based on distributed optical fiber, which obtains sensing data through distributed sensing optical fibers attached to the inner wall of the drainage pipe, analyzes the sensing data through statistical methods, and when abnormal data is obtained through analysis, determines the disaster risk level corresponding to the abnormal data through a machine learning algorithm, and generates warning information based on the abnormal data and the disaster risk level; and makes drainage pipe operation and maintenance decisions based on the decision-making system and warning information of risk-informed theory. By laying distributed optical fibers, the present application can sense the state data of the inner wall of the pipeline with high spatial resolution and high monitoring accuracy over a long range without destroying the original structure of the drainage pipe, and analyzes the sensing data and determines the disaster risk level in combination with statistical methods and machine learning algorithms, so that abnormal conditions of the drainage pipe can be identified in a timely and accurate manner at an early stage, and warnings and drainage pipe operation and maintenance decisions can be made when abnormal data is detected, thereby extending the service performance of the drainage pipe and reducing operation and maintenance costs.

[0094] See also Figure 3 The embodiment of the present application also provides a drainage pipeline disaster warning method based on distributed optical fiber, including:

[0095] Step S1, acquiring sensing data by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme.

[0096] The sensing optical fiber includes a temperature sensing optical fiber and a strain sensing optical fiber; and the sensing data includes temperature data and strain data.

[0097] The sensing data obtained by various sensing optical fibers attached to the inner wall of the drainage pipe according to a preset scheme include:

[0098] The optical frequency domain reflection technology based on back Rayleigh scattering obtains the temperature data of each measuring point on the temperature sensing optical fiber and the strain data of each measuring point on the strain sensing optical fiber; the temperature data is compensated to the strain data based on the position of each measuring point.

[0099] Specifically, the temperature data of each measuring point is only used to compensate the strain data of the same pipe section.

[0100] In some embodiments, the above-mentioned acquisition of sensing data by various sensing optical fibers attached to the inner wall of the drainage pipe according to a preset scheme also includes: uploading the acquired temperature data and strain data to the cloud for storage.

[0101] Among them, the strain data uploaded to the cloud is the real strain data after temperature compensation.

[0102] Step S2: analyzing the sensing data by using statistical methods.

[0103] Step S3: If abnormal data is obtained through analysis, the disaster risk level corresponding to the abnormal data is determined through a machine learning algorithm.

[0104] Step S4, generating warning information based on abnormal data and disaster risk level.

[0105] Step S5, making drainage pipe operation and maintenance decisions based on the risk-informed decision-making system and early warning information.

[0106] The above embodiment provides a drainage pipe disaster warning method based on distributed optical fiber, which obtains sensing data through distributed sensing optical fibers attached to the inner wall of the drainage pipe, analyzes the sensing data through statistical methods, and when abnormal data is obtained through analysis, determines the disaster risk level corresponding to the abnormal data through a machine learning algorithm, and generates warning information based on the abnormal data and the disaster risk level; and makes drainage pipe operation and maintenance decisions based on a decision-making system and warning information based on risk-informed theory. This application deploys distributed optical fibers to sense the state data of the inner wall of the drainage pipe, and combines statistical methods and machine learning algorithms to analyze the sensing data and determine the disaster risk level, so that abnormal conditions of the drainage pipe can be identified in a timely and accurate manner at an early stage, and warnings and drainage pipe operation and maintenance decisions can be made when abnormal data is detected, thereby extending the service performance of the drainage pipe and reducing operation and maintenance costs.

[0107] See also Figure 4 In some embodiments, the statistical method is a sliding window outlier analysis method.

[0108] The above-mentioned analysis of the sensing data by statistical methods includes:

[0109] Step S21: applying the sliding window to the strain data of each measuring point at the first moment based on a preset sliding window length.

[0110] Among them, the first moment is a selected sampling moment, and the preset sliding window length is set to S. The sliding window is applied to the strain data of each measuring point at the first moment. That is, at the first moment, the strain data of S measuring points on the inner wall of the drainage pipe are selected by the sliding window for outlier analysis. After sliding, there are N sliding windows in total, and N*S measuring points are analyzed.

[0111] Step S22, calculating the inconsistency index corresponding to each strain data in the sliding window.

[0112] The inconsistency index corresponding to each strain data in the above-mentioned calculation sliding window includes:

[0113] The mean and standard deviation were calculated from the data for each strain.

[0114] The corresponding inconsistency index is calculated based on the mean, standard deviation and strain data.

[0115] Specifically, for the strain data X of S measuring points in the nth sliding window n (x n1 ,x n2 ,x n3 …x nS ), calculate its mean mean(X n ) and standard deviation std(X n ), n = 1, 2...N; then use the mean and standard deviation to calculate the value of X in the window n The strain data x of each measuring point ni Inconsistency Index The calculation formula is:

[0116]

[0117] Step S23, calculating a threshold value corresponding to the sliding window based on each inconsistency index.

[0118] The above-mentioned calculation of the threshold value corresponding to the sliding window based on each inconsistency index includes:

[0119] Sort each inconsistency indicator in descending order to obtain an indicator queue.

[0120] The threshold value is obtained according to the preset sliding window length, the preset guarantee rate and the indicator queue.

[0121] Specifically, for the S strain data corresponding to the nth (n=1, 2...N) sliding window, Sort in descending order and take the S×(1-a)th As the threshold value of the nth sliding window, a is the preset guarantee rate of the normal state of the drainage pipe structure. For example, the index queues after descending sorting are [5 4 3 2 1], the preset guarantee rate a is 80%, and the threshold value is 4.

[0122] Step S24, sliding the sliding window on each strain data at the first moment to obtain the inconsistency index corresponding to each strain data and the threshold value corresponding to each window.

[0123] Specifically, after traversing N sliding windows, the inconsistency indicators of N*S measurement points at the first moment are calculated.

[0124] Step S25, judging whether there are outliers in each inconsistency indicator in the corresponding window according to the threshold value.

[0125] Specifically, each inconsistency indicator is compared with the threshold value of the window where it is located. If it does not exceed the threshold value, the corresponding strain data is normal. If it exceeds the threshold value, the measurement point corresponding to the inconsistency indicator exceeding the threshold value is regarded as an outlier.

[0126] Step S26: If it exists, detect the outlier based on the abnormal state diagnosis window to obtain abnormal data.

[0127] Step S27: If no abnormal data exists or is not detected, the sliding window is applied to the strain data at the second moment.

[0128] Specifically, if the inconsistency index of each strain data at the first moment does not exceed the threshold value of the window where it is located, it means that the strain data at the first moment is normal, and the strain data at the next sampling moment, that is, the second moment, is detected until the sliding window outlier analysis of the strain data at all moments is completed.

[0129] The sliding window outlier analysis method in the above embodiment can quickly detect abnormal data in the strain data, thereby achieving the purpose of identifying abnormal conditions of the drainage pipe in a timely manner at an early stage.

[0130] In some embodiments, the above-mentioned outlier detection based on the abnormal state diagnosis window to obtain abnormal data includes:

[0131] Step S261, determining the target threshold value and each target measuring point of the window corresponding to the outlier.

[0132] Specifically, after comparing the inconsistency index and the threshold value, the outlier is obtained, and the threshold value of the sliding window where the outlier is located is the target threshold value. The S measurement points x in the window n1 ~x nS The target measuring point.

[0133] Step S262, obtaining a preset diagnosis window length of the abnormal state diagnosis window.

[0134] Step S263: acquiring strain data of the target measuring point at multiple moments based on a preset diagnosis window length.

[0135] Specifically, the number of moments is the preset diagnostic window length. Let k be the preset diagnostic window length of the abnormal state diagnostic window. Then, the time interval of each target measurement point at t j ―t j+k Strain data at the moment.

[0136] Among them, t j For the first moment, the value of k is usually much smaller than S.

[0137] Step S264: Calculate the inconsistency index corresponding to the target measuring point at each moment according to the strain data at each moment.

[0138] Specifically, the calculation method of the inconsistency index here is the same as that of the sliding window analysis method, that is, the target measurement point is calculated at t j ―tj+k The mean and standard deviation of the strain data at time t are calculated based on the mean and standard deviation. j ―t j+k The non-consistency index corresponding to the strain data at time t is understandable that a target measuring point has j ―t j+k There are k inconsistency indicators at the moment.

[0139] Step S265, comparing the inconsistency index corresponding to each moment with the target threshold value to obtain the abnormal moment ratio.

[0140] The inconsistency index of the target measuring point at each moment is compared with the target threshold value. If the inconsistency index of consecutive moments exceeds the target threshold value, the number of consecutive abnormal moments is compared with all moments, that is, the abnormal moment ratio.

[0141] Step S266: If the abnormal time ratio exceeds the preset ratio, the target measurement point is regarded as abnormal data.

[0142] The preset ratio may be k / 2, k / 3, etc. Taking k / 2 as an example, if the abnormal ratio is greater than k / 2, the target measuring point is regarded as abnormal data, and the pipeline section location where the target measuring point is located is determined at the same time.

[0143] The above embodiment re-detects the strain data of the target measuring point at multiple times and confirms the abnormal data through the abnormal state diagnosis window, thereby achieving early abnormality diagnosis of the drainage pipe and making the diagnosis more accurate.

[0144] In some embodiments, the above-mentioned determination of the disaster risk level corresponding to the abnormal data by a machine learning algorithm includes:

[0145] Step S31, using a machine learning algorithm to calculate the pipeline deformation corresponding to the abnormal data.

[0146] Step S32, obtaining the pipeline cross-section location corresponding to the abnormal data and a plurality of strain data of the pipeline cross-section location.

[0147] Specifically, considering that the strain sensing optical fiber is arranged on the top, side and bottom of the inner wall of the drainage pipe, the strain data of the pipe section positioning here refers to the strain data of the other three strain sensing optical fibers at the same section; for example, if the abnormal data is the measuring point at the top of the pipe, the strain data of the two sides and bottom of the pipe in the cross section where the measuring point is located are obtained to calculate the deformation.

[0148] Step S33, calculating the pipe deformation corresponding to the plurality of strain data.

[0149] Among them, the machine learning algorithm is also used to calculate the pipeline deformation corresponding to multiple strain data.

[0150] Step S34, comparing the deformation amounts of each pipeline to obtain the maximum deformation amount of the pipeline cross section positioning.

[0151] Step S35, determining the disaster risk level corresponding to the abnormal data according to the maximum deformation and the pipe diameter.

[0152] Specifically, the disaster risk level is divided into 5 levels from low to high, 0-4. Level 0 means no deformation of the pipeline, level 1 means the maximum deformation is less than 5% of the pipeline diameter, level 2 means the maximum deformation is 5%-15% of the pipeline diameter, level 3 means the maximum deformation is 15-25% of the pipeline diameter, and level 4 means the maximum deformation is greater than 25% of the pipeline diameter.

[0153] The above embodiment determines the maximum deformation at the same cross section and determines the disaster risk level according to the maximum deformation, so that the most seriously deformed part of the pipeline can be screened in time, thus avoiding the omission of early diagnosis of deformation to a great extent.

[0154] In some embodiments, the warning information includes the disaster risk level and the pipeline cross-section location corresponding to the abnormal data.

[0155] Specifically, the warning information may also include the disaster warning time, the location of potential disaster points, etc. The location of the potential disaster point may be 2-3 measurement points before and after the location of the abnormal data, or it may be determined based on the pipeline direction and the surrounding environment.

[0156] The pipeline cross-section positioning in the warning information in the above embodiment is conducive to making more reasonable and accurate operation and maintenance decisions.

[0157] An embodiment of the present application provides a computer device, which may include a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the processor executes the steps of a drainage pipeline disaster warning method based on distributed optical fiber as in any of the above embodiments. The working process, working details and technical effects of the computer device provided in this embodiment can be found in the above embodiment of a drainage pipeline disaster warning method based on distributed optical fiber, which will not be repeated here.

[0158] The embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed, implements the steps of a drainage pipeline disaster warning method based on distributed optical fiber as in any of the above embodiments. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive and / or a memory stick, etc. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The working process, working details, and technical effects of the computer-readable storage medium provided in this embodiment can be found in the above embodiment of a drainage pipeline disaster warning method based on distributed optical fiber, and will not be repeated here.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0160] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. The above-mentioned embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several deformations and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of this application shall be based on the attached claims.

Claims

1. A drainage pipeline disaster warning device based on distributed optical fiber, characterized in that: include: Optical fiber sensing module, including an optical frequency domain reflectometry demodulator and a variety of sensing optical fibers; The various sensing optical fibers are attached to the inner wall of the drainage pipe according to a preset scheme, and the optical frequency domain reflection demodulator is used to obtain sensing data of the various sensing optical fibers; wherein the sensing optical fibers include temperature sensing optical fibers and strain sensing optical fibers, and the sensing data include temperature data and strain data; The optical frequency domain reflectometry demodulator is also used to upload the temperature data and the strain data to the cloud; The data analysis module includes a data transmission unit, a disaster identification unit and a risk assessment unit, which are used to analyze the sensing data by statistical methods, and when abnormal data is obtained through analysis, determine the disaster risk level corresponding to the abnormal data by machine learning algorithm; wherein the data transmission unit is used to obtain the sensing data of the optical frequency domain reflectometry demodulator through the cloud; the disaster identification unit is used to analyze the strain data by using a sliding window outlier analysis method; and the risk assessment unit is used to determine the disaster risk level of the abnormal data by using a machine learning algorithm; The analysis of the strain data by using the sliding window outlier analysis method is specifically as follows: based on a preset sliding window length, a sliding window is applied to the strain data of each measuring point at a first moment; the inconsistency index corresponding to each strain data in the sliding window is calculated; based on each inconsistency index, a threshold value corresponding to the sliding window is calculated; the sliding window is slid on each strain data at the first moment to obtain the inconsistency index corresponding to each strain data and the threshold value corresponding to each window; based on the threshold value, it is determined whether each inconsistency index in the corresponding window has an outlier; if so, the outlier is detected based on an abnormal state diagnosis window to obtain abnormal data; if the abnormal data does not exist or is not detected, the sliding window is applied to the strain data at a second moment; The risk assessment unit is used to calculate the pipeline deformation corresponding to the abnormal data by using the machine learning algorithm; and obtain the pipeline cross-section location corresponding to the abnormal data in the optical frequency domain reflection demodulator, and the multiple strain data of the pipeline cross-section location, and calculate the pipeline deformation corresponding to the multiple strain data; and compare the pipeline deformations to obtain the maximum deformation of the pipeline cross-section location, and determine the disaster risk level corresponding to the abnormal data according to the maximum deformation and the pipeline diameter; A disaster warning module generates warning information according to the abnormal data and the disaster risk level; The operation and maintenance decision module is used to make drainage pipe operation and maintenance decisions based on the decision-making system of risk-informed theory and the early warning information.

2. The drainage pipeline disaster warning device based on distributed optical fiber according to claim 1 is characterized in that: The temperature sensing optical fiber is arranged at the top axis position of the inner wall of the drainage pipe; the strain sensing optical fiber is arranged at the top axis position, the bottom axis position and the side axis position of the inner wall of the drainage pipe respectively.

3. The drainage pipeline disaster warning device based on distributed optical fiber according to claim 2 is characterized in that: The temperature sensing optical fiber is packaged in an alumina tube.

4. The drainage pipeline disaster warning device based on distributed optical fiber according to claim 2 is characterized in that: The strain sensing optical fiber is also spirally arranged on the inner wall of the drainage pipe with a preset spiral pitch.

5. The drainage pipeline disaster warning device based on distributed optical fiber according to claim 4 is characterized in that: The optical frequency domain reflection demodulator is used to obtain the temperature data of the temperature sensing optical fiber and the strain data of the strain sensing optical fiber based on the optical frequency domain reflection technology of back Rayleigh scattering, and compensate the strain data according to the temperature data.

6. A drainage pipeline disaster warning method based on distributed optical fiber, characterized in that: include: Acquire sensing data by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme; the sensing data includes temperature data and strain data; Analyzing the sensing data by a statistical method; the statistical method is a sliding window outlier analysis method; The analysis of the sensing data by statistical methods includes: applying a sliding window to the strain data of each measuring point at a first moment based on a preset sliding window length; calculating the inconsistency index corresponding to each strain data in the sliding window; calculating the threshold value corresponding to the sliding window based on each inconsistency index; sliding the sliding window on each strain data at the first moment to obtain the inconsistency index corresponding to each strain data and the threshold value corresponding to each window; judging whether there are outliers for each inconsistency index in the corresponding window according to the threshold value; if there are outliers, detecting the outliers based on an abnormal state diagnosis window to obtain abnormal data; if there are no outliers or no abnormal data is detected, applying the sliding window to the strain data at a second moment; If abnormal data is obtained through analysis, the disaster risk level corresponding to the abnormal data is determined by a machine learning algorithm; specifically, the pipeline deformation corresponding to the abnormal data is calculated by the machine learning algorithm; the pipeline cross-section location corresponding to the abnormal data and a plurality of strain data of the pipeline cross-section location are obtained; the pipeline deformation corresponding to the plurality of strain data are calculated; the pipeline deformations are compared to obtain the maximum deformation of the pipeline cross-section location; and the disaster risk level corresponding to the abnormal data is determined according to the maximum deformation and the pipeline diameter; generating early warning information according to the abnormal data and the disaster risk level; Drainage pipe operation and maintenance decisions are made based on a decision-making system based on risk-informed theory and the warning information.

7. The drainage pipeline disaster early warning method based on distributed optical fiber according to claim 6 is characterized in that: The sensing optical fiber includes a temperature sensing optical fiber and a strain sensing optical fiber; The sensing data is obtained by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme, including: Acquire the temperature data of each measuring point on the temperature sensing optical fiber and the strain data of each measuring point on the strain sensing optical fiber based on the optical frequency domain reflection technology of Rayleigh backscattering; The temperature data is compensated to the strain data based on the positions of the respective measuring points.

8. The drainage pipeline disaster early warning method based on distributed optical fiber according to claim 7 is characterized in that: The method of obtaining sensing data by attaching various sensing optical fibers to the inner wall of the drainage pipe according to a preset scheme also includes: The acquired temperature data and strain data are uploaded to the cloud for storage.

9. The drainage pipeline disaster early warning method based on distributed optical fiber according to claim 8, characterized in that: The calculating the inconsistency index corresponding to each strain data in the sliding window includes: Calculate the mean and standard deviation based on each of the strain data; The corresponding inconsistency index is calculated according to the mean, the standard deviation and the strain data.

10. The drainage pipeline disaster warning method based on distributed optical fiber according to claim 8, characterized in that: The calculating the threshold value corresponding to the sliding window based on each of the inconsistency indicators includes: Sorting each of the inconsistency indicators in descending order to obtain an indicator queue; The threshold value is obtained according to the preset sliding window length, the preset guarantee rate and the indicator queue.

11. The drainage pipeline disaster early warning method based on distributed optical fiber according to claim 8, characterized in that: The detecting the outlier based on the abnormal state diagnosis window to obtain abnormal data includes: Determine the target threshold value and each target measurement point of the window corresponding to the outlier; Obtaining a preset diagnostic window length of the abnormal state diagnostic window; Acquiring the strain data of the target measuring point at multiple times based on the preset diagnostic window length; Calculating the inconsistency index corresponding to the target measuring point at each moment according to the strain data at each moment; Compare the inconsistency index corresponding to each moment with the target threshold value to obtain the abnormal moment ratio; If the abnormal time ratio exceeds a preset ratio, the target measurement point is used as the abnormal data.

12. The drainage pipeline disaster early warning method based on distributed optical fiber according to claim 8, characterized in that: The warning information includes the disaster risk level and the pipeline cross-section location corresponding to the abnormal data.

13. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the drainage pipe disaster warning method based on distributed optical fiber as described in any one of claims 6 to 12 are implemented.

14. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the steps of the drainage pipe disaster warning method based on distributed optical fiber as described in any one of claims 6 to 12 are implemented.

Images