Flexible composite pipe safety state monitoring method, device and system

Abstract

The invention discloses a flexible composite pipe safety state monitoring method, device and system, and the method comprises the steps: predicting the current condition of a to-be-detected pipeline through obtaining historical monitoring data, obtaining a prediction state vector, then building a coupling observation function to determine a Kalman gain, determining whether a predicted value is believed or not based on the Kalman gain, and after the believed predicted value is determined, carrying out the monitoring of the safety state of the to-be-detected pipeline. Correcting the predicted value based on Kalman gain to obtain a corrected target state vector, then determining a soil state mark based on the target state vector so as to determine the current maximum internal stress of the to-be-detected pipeline, and determining a pipeline state risk level in combination with a function that the allowable stress of the material changes along with the temperature. Finally, the leakage identification result, the leakage position and the leakage severity are determined through the signal leakage identification model, automatic monitoring of the safety state of the buried flexible composite pipe is achieved, and meanwhile the safety monitoring accuracy of the buried flexible composite pipe is improved.

Description

Technical Field

[0001] This application relates to the field of safety monitoring technology for oil and gas gathering and transportation pipelines, and in particular to a method, device and system for monitoring the safety status of flexible composite pipes. Background Technology

[0002] Flexible composite pipes, due to their excellent corrosion resistance, flexibility, and ability to be laid in integral rolls, are gradually replacing traditional steel pipes in oil and gas gathering and transportation and municipal pipeline networks in cold regions. Flexible composite pipes typically consist of a multi-layered composite structure including an inner lining, a reinforcing layer, and an outer protective layer, maintaining good service performance under high internal pressure and complex geological conditions. However, in extremely cold environments (where ambient temperatures can drop to -30℃ or even -40℃) and under frequent freeze-thaw cycles, buried flexible composite pipes may face the following problems: 1. Freeze-thaw cycles can easily induce adverse conditions such as pipe bending, stretching, and local buckling; 2. At low temperatures, material stiffness increases and fracture toughness decreases, making them prone to microcracks, interlayer delamination, and other damage during long-term service, significantly reducing the pipeline's safety margin; 3. In extremely cold environments, the increased temperature difference between the inside and outside of the pipe generates additional thermal stress in the pipeline structure. Combined with soil frost heave loads, this creates a complex temperature-stress coupling effect, significantly increasing the risk of interface leakage and pipe cracking.

[0003] To monitor the condition of buried flexible composite pipes and ensure their safety, safety monitoring of buried flexible composite pipes is necessary. Existing buried pipeline safety monitoring mainly has the following shortcomings: slow response to early damage such as minor leaks and crack initiation, often only being discovered when the accident has developed to a more serious stage; the single vibration and single-point pressure sensing methods used are easily affected by environmental noise interference such as pump start-up and shutdown, ground machinery, and soil disturbance, making it difficult to distinguish between actual damage and normal operating condition fluctuations, lacking accurate measurement, unable to quantitatively assess structural safety margin, and unable to predict the long-term performance degradation trend of buried flexible composite pipes. Summary of the Invention

[0004] This application provides a method, device, and system for monitoring the safety status of flexible composite pipes. It addresses the problem in existing technologies where it's difficult to accurately distinguish between actual damage and normal operating condition fluctuations in buried flexible composite pipes, leading to an inability to quantitatively assess the structural safety margin of the pipeline. It also solves the problem of existing technologies failing to predict the long-term performance degradation trend of buried flexible composite pipes, thus failing to guarantee their safety status. By correcting the predicted state vector using Kalman gain, the cross-sensitivity of temperature to mechanical strain in the monitoring data is automatically decoupled, and random noise is suppressed, improving data confidence. Then, through a dual-modal fusion mechanism of a signal leakage identification model and a pipeline stress estimation model, false alarms caused by a single signal source are avoided, effectively improving the accuracy of monitoring the safety status of buried flexible composite pipes. It also solves the problem of traditional monitoring systems failing in winter due to changes in soil acoustic properties.

[0005] In a first aspect, embodiments of this application provide a method for monitoring the safety status of flexible composite pipes, including: Historical monitoring data of the pipeline to be tested is obtained, and the predicted state vector is determined by calculating the historical monitoring data using the state vector prediction formula. Acquire the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function; When the Kalman gain meets the preset judgment range, the predicted state vector is corrected based on the Kalman gain and the current monitoring data to obtain the corrected target state vector; The maximum internal stress of the pipeline under test is determined based on the target state vector, and the safety margin is determined by combining the allowable stress of the material with the function of temperature. The risk level of the pipeline condition is determined based on the safety margin. Acquire acoustic signals from the pipeline under test and determine the leak location based on the acoustic signals; Calculate the signal energy intensity of the acoustic signal, determine the pressure drop rate based on the current monitoring data, and assess the severity of the leak using the signal energy intensity and the pressure drop rate; The safety status of the pipeline under test is monitored based on the pipeline's risk level, the location of the leak, and the severity of the leak.

[0006] In conjunction with the first aspect, in one possible implementation, the step of calculating the predicted state vector from the historical monitoring data using a state vector prediction formula includes: The historical monitoring data is converted into a historical state vector, and the predicted state vector is obtained by calculating the historical state vector, the state transition matrix, and the process noise using the state vector prediction formula. The state vector prediction formula is as follows:

[0007] In the formula, express The predicted state vector of the pipeline under test at time t, where A represents the state transition matrix. express The historical state vector of the pipeline under test at any given time. express The process noise of the pipeline under constant monitoring.

[0008] In conjunction with the first aspect, in one possible implementation, determining the Kalman gain based on the coupled observation function includes: Establish a coupled observation function and determine the Jacobian matrix by taking the partial derivative of the coupled observation function; The prior error covariance matrix is ​​determined using the covariance prediction formula. Obtain the preset measurement noise covariance, and calculate the Kalman gain by applying the Kalman gain formula to the Jacobian matrix, the prior error covariance matrix, and the preset measurement noise covariance.

[0009] In conjunction with the first aspect, in one possible implementation, The coupled observation function is:

[0010] In the formula, Represents the coupled observation function, This represents the strain coefficient of the optical fiber in the pipe under test. express The real mechanical strain of the pipeline to be measured at all times This represents the temperature coefficient of the optical fiber in the pipe under test. express The actual temperature of the pipeline and surrounding soil is to be measured at all times. express The actual internal pressure of the medium in the pipeline to be measured at all times. This indicates the reference temperature for calibrating the optical fiber in the pipe under test; The covariance prediction formula is as follows:

[0011] In the formula, Let A represent the prior error covariance matrix at time k calculated based on historical monitoring data at time k-1, and let A represent the state transition matrix. This represents the posterior error covariance matrix at time k-1, calculated based on historical monitoring data at time k-1. This represents the transpose of the state transition matrix. This represents the preset process noise covariance; The Kalman gain calculation formula is as follows:

[0012] In the formula, This represents the Kalman gain at time k. Let represent the Jacobian matrix obtained after linearizing the coupled observation function at time k. denoted as the preset measurement noise covariance, and T represents the matrix transpose.

[0013] In conjunction with the first aspect, in one possible implementation, the step of correcting the predicted state vector based on the Kalman gain and the current monitoring data to obtain the corrected target state vector includes: The sensor's own measurement noise is acquired and combined with the coupled observation function. The current monitoring data is then corrected using a fusion decoupling formula to obtain the current state vector. Perform spatiotemporal alignment processing on the current state vector to construct the current alignment vector; Based on the Kalman gain and the current alignment vector, the predicted state vector is corrected using a vector correction formula to obtain the target state vector.

[0014] In conjunction with the first aspect, in one possible implementation, the fusion decoupling formula is:

[0015] In the formula, express The monitoring data collected continuously from the pipeline under test Represents the coupled observation function, express The measurement noise of the time sensor itself. ,in, This represents the raw frequency shift of the pipe under test as measured by the distributed fiber optic sensor, including temperature and strain information. This indicates the temperature of the pipe under test as measured by the DTS distributed temperature sensor, including the temperature of the pipe wall and the surrounding soil. This indicates the pressure of the medium inside the pipe being tested, as measured by the pressure sensor. The vector correction formula is:

[0016] In the formula, express The target state vector at time t. Indicates according to Predicted from historical monitoring data at any given time The predicted state vector at time t. express Kalman gain at time step express The current aligned vector is obtained by performing spatiotemporal alignment processing on the current state vector at time t. This represents the value obtained by calculating the predicted state vector using the coupled observation function. This represents the observation residual, which is the deviation between the current alignment vector and the predicted state vector.

[0017] In conjunction with the first aspect, in one possible implementation, determining the current maximum internal stress of the pipe under test based on the target state vector includes: The target state vector is subjected to permafrost environment feature identification to obtain soil state labels; The current mechanical strain, current temperature, current internal medium pressure, and soil state label in the target state vector are input into the pipeline stress estimation model trained by XGBoost regression tree for identification, so as to obtain the current maximum internal stress of the pipeline under test.

[0018] In conjunction with the first aspect, in one possible implementation, the step of acquiring the acoustic signal of the pipe under test and determining the leak location based on the acoustic signal includes: Acoustic signals from the pipe under test are collected as the original signals, and wavelet transform is used to perform time-frequency decomposition on the original signals to obtain detail components and approximate components in different frequency bands. The detailed components of different frequency bands are calculated to obtain the energy proportion and Shannon entropy of each frequency band; Feature vectors are constructed based on the energy proportions of all frequency bands and Shannon entropy. The feature vector is input into the signal leakage identification model for identification, and the identification result is obtained. When the identification result is a leak, it is determined whether a negative pressure wave exists within the corresponding time window based on the current time; if the existence of the negative pressure wave is confirmed, the leak identification result is that a leak exists. Once a leak is confirmed, the leak location is determined using the TDOA positioning algorithm based on the original signal.

[0019] Secondly, embodiments of this application provide a flexible composite pipe safety status monitoring device, comprising: The historical monitoring data processing module is used to acquire historical monitoring data of the pipeline under test, and to calculate the predicted state vector by using the state vector prediction formula. The current monitoring data processing module is used to acquire the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function. The predicted state vector correction module is used to correct the predicted state vector based on the Kalman gain and the current monitoring data when the Kalman gain meets the preset judgment range, so as to obtain the corrected target state vector. The risk level determination module is used to determine the current maximum internal stress of the pipeline under test based on the target state vector, and to determine the safety margin by combining the allowable stress of the material with the function of temperature change, and to determine the risk level of the pipeline state based on the safety margin. The leak location determination module is used to collect the acoustic signal of the pipeline under test and determine the leak location based on the acoustic signal; The leakage severity assessment module is used to calculate the signal energy intensity of the acoustic signal and determine the pressure drop rate based on the current monitoring data, and assess the leakage severity by using the signal energy intensity and the pressure drop rate. The pipeline safety status monitoring module is used to monitor the safety status of the pipeline under test based on the pipeline status risk level, the leak location, and the leak severity.

[0020] Thirdly, this application provides a flexible composite pipe safety status monitoring system, including a cloud monitoring platform, a monitoring station, a multi-parameter sensing and detection unit, and an intelligent analysis and early warning unit; The cloud monitoring platform obtains the data retrieval command, retrieves historical monitoring data according to the data retrieval command, and sends it to the intelligent analysis and early warning unit; The multi-parameter sensing and detection unit acquires the current monitoring data through sensors arranged in the buried flexible composite pipe and transmits it to the monitoring station. After receiving the current monitoring data, the monitoring station preprocesses the current monitoring data through the data acquisition and processing unit and converts the current monitoring data into digital signals for transmission to the intelligent analysis and early warning unit. After receiving historical and current monitoring data, the intelligent analysis and early warning unit executes the aforementioned flexible composite pipe safety status monitoring method to obtain the pipeline status risk level, leakage location, and leakage severity, and transmits this information to the cloud monitoring platform. After receiving data on the pipeline status risk level, leak location, and leak severity, the cloud monitoring platform activates the emergency command and dispatch module and sends early warning information to the corresponding client via a mobile app.

[0021] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: This application embodiment predicts the current condition of the pipeline under test by acquiring historical monitoring data, obtaining a predicted state vector. Then, a coupled observation function is established to determine the Kalman gain. Based on the Kalman gain, it is determined whether to believe the predicted value (i.e., the predicted state vector). When the Kalman gain meets the preset judgment range (i.e., the Kalman gain infinitely approaches 0 but is not 0), the predicted value is chosen to be believed. To improve the accuracy of the data, after determining to believe the predicted value, it is necessary to correct the predicted value based on the Kalman gain to obtain the corrected target state vector. Then, based on the target state vector, the soil state marker is determined, thereby determining the current maximum internal stress of the pipeline under test. Combined with the function of allowable stress of the material changing with temperature, the risk level of the pipeline condition is determined. Finally, the leakage identification result is determined through a signal leakage identification model, and the leakage location is determined by combining the TDOA positioning algorithm. Based on the signal energy intensity and pressure drop rate in the current monitoring data, the severity of the leakage is determined. This effectively solves the problem of inaccurate monitoring caused by the inability to predict the state of buried flexible composite pipes in the prior art, realizes automatic monitoring of the safety status of buried flexible composite pipes, and improves the accuracy of safety monitoring of buried flexible composite pipes. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a method for monitoring the safety status of a flexible composite pipe, as provided in this application embodiment; Figure 2 A flowchart for calculating the current maximum internal stress of a pipe under test is provided as an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a flexible composite pipe safety status monitoring device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a flexible composite pipe safety status monitoring system provided in an embodiment of this application; Figure 5 A logic block diagram of a flexible composite pipe safety status monitoring system provided in this application embodiment; Figure 6 This is a schematic diagram of a distributed optical fiber laid along a flexible composite tube, as provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] The following description of some technologies involved in the embodiments of this application is provided to aid understanding and should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, some descriptions of well-known functions and structures are omitted in the following description.

[0026] Figure 1 This is a flowchart of a method for monitoring the safety status of a flexible composite pipe according to an embodiment of this application, including steps S10 to S70. Figure 1 This is merely one execution order shown in the embodiments of this application and does not represent the only execution order for a flexible composite pipe safety status monitoring method. Where the final result can be achieved, Figure 1 The steps shown can be performed in parallel or in reverse order.

[0027] S10: Obtain historical monitoring data of the pipeline to be tested, and calculate the predicted state vector by using the state vector prediction formula.

[0028] In this embodiment, the monitoring data refers to data acquired by sensors deployed in the buried flexible composite pipe. The sensors in this embodiment include, but are not limited to, distributed fiber optic strain gauges (BOTDR), distributed temperature sensors (DTS), and pressure sensors. Accelerometers and leak detection sensors. Among them, the distributed fiber optic strain sensor BOTDR is used to acquire the actual mechanical strain along the pipeline under test. Distributed temperature sensor (DTS) is used to monitor the temperature of the pipeline under test and the surrounding soil. Pressure sensor Used to monitor the pressure of the medium inside the pipe under test. An accelerometer is used to monitor the vibration response of the pipe under test; a leak detection sensor is used to monitor acoustic signals or pressure disturbance signals related to leaks in the pipe under test. In this embodiment, the leak detection sensor is a distributed acoustic sensor (DAS).

[0029] Specifically, to obtain After the monitoring data of the pipeline under test is used as historical monitoring data, the historical monitoring data is arranged according to a preset order to obtain the corresponding historical state vector. In this embodiment, the state vector is specifically as follows:

[0030] In the formula, express Mechanical strain is constantly measured by distributed fiber optic strain sensors. express The temperature of the pipe under test and the surrounding soil is constantly measured by distributed temperature sensors. express The pressure of the medium inside the pipe under test is constantly measured by a pressure sensor.

[0031] After obtaining the historical state vector Then, the historical state vector is predicted using the state vector prediction formula. State transition matrix A and process noise Perform calculations to determine the predicted state vector. .

[0032] Process noise refers to noise signals emitted through pipe vibration response and acoustic or pressure disturbance signals associated with pipe leakage. This process noise represents environmental uncertainties, such as sudden changes in frozen soil pressure.

[0033] In extremely cold environments, barring unforeseen events, changes in pipeline conditions are typically continuous.

[0034]

[0035] Assuming the rate of change is constant over a short period, a linear state vector prediction formula can be established, which is:

[0036] In the formula, express The predicted state vector of the pipeline under test at time t, where A represents the state transition matrix. express The historical state vector of the pipeline under test at any given time. express The process noise of the pipeline under constant monitoring.

[0037] The expanded form of the above state transition matrix is:

[0038] in, express Mechanical strain is constantly measured by distributed fiber optic strain sensors. express The temperature of the pipe under test and the surrounding soil is constantly measured by distributed temperature sensors. express The pressure of the medium inside the pipe under test is constantly measured by a pressure sensor. express The rate of change of strain at any given time (used to capture frost heave or deformation trends). express The rate of temperature change over time (used to capture the movement of freezing fronts). express Time and The time difference between moments The process noise (representing environmental uncertainty, such as sudden changes in permafrost pressure) follows a Gaussian distribution. .

[0039] S20: Obtain the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function.

[0040] Among them, the current monitoring data refers to The monitoring data of the pipeline under test is collected in real time by sensors deployed in the buried flexible composite pipe.

[0041] Specifically, due to environmental interference and inherent errors in the sensor itself, the data monitored by the sensor will inevitably differ from the actual data in reality. To improve the accuracy of the data, this embodiment establishes a coupled observation function to correct the current monitoring data monitored by the sensor, thereby obtaining corrected monitoring data.

[0042] The coupled observation function is:

[0043] In the formula, Represents the coupled observation function, This represents the strain coefficient of the optical fiber in the pipe under test. express The real mechanical strain of the pipeline to be measured at all times This represents the temperature coefficient of the optical fiber in the pipe under test, which is the source of temperature interference. express The actual temperature of the pipeline and surrounding soil is to be measured at all times. express The actual internal pressure of the medium in the pipeline to be measured at all times. This indicates the reference temperature used for calibrating the optical fiber in the pipe under test.

[0044] By establishing a coupled observation function, distributed temperature sensors will be automatically utilized during subsequent Extended Kalman Filter (EKF) processing. The provided temperature information is used to correct the distributed fiber optic sensor. By obtaining strain readings, temperature interference can be eliminated, and thus, the following can be obtained. The actual mechanical strain of the pipeline is to be measured at all times.

[0045] Establishing the coupled observation function After that, the Jacobian matrix is ​​determined by taking the partial derivative of the coupled observation function. Then, the covariance prediction formula is used to determine... Prior error covariance matrix at time 1 .

[0046] The formula for predicting covariance is:

[0047] In the formula, Let A represent the prior error covariance matrix at time k calculated based on historical monitoring data at time k-1, and let A represent the state transition matrix. This represents the posterior error covariance matrix at time k-1, calculated based on historical monitoring data at time k-1. This represents the transpose of the state transition matrix. This represents the preset process noise covariance.

[0048] After obtaining the prior error covariance matrix Then, the preset measurement noise covariance is obtained. Furthermore, the Jacobian matrix and the prior error covariance matrix were calculated using the Kalman gain formula. and preset measurement noise covariance The Kalman gain was calculated. .

[0049] The Kalman gain calculation formula is:

[0050] In the formula, This represents the Kalman gain at time k. Let the Jacobian matrix at time k be denoted as . This represents the coupled observation function after linearization approximation at time k. denoted as the preset measurement noise covariance, and T represents the matrix transpose.

[0051] Is the value predicted by the state vector prediction formula more accurate by calculating the Kalman gain, or is the value predicted by the actual current monitoring data measured by the sensor more accurate? If the sensor noise... The larger the gain If it is small, it represents the value predicted by the state vector prediction formula, that is, the predicted state vector. More accurate.

[0052] S30: When the Kalman gain meets the preset judgment range, the predicted state vector is corrected based on the Kalman gain and the current monitoring data to obtain the corrected target state vector.

[0053] Among them, the preset judgment range refers to the range set for judging the Kalman gain. Whether it is a range of values ​​that infinitely approaches 0 but is not equal to 0.

[0054] Specifically, after determining the Kalman gain, the sensor's own measurement noise is obtained. And combined with the coupled observation function The current monitoring data is corrected by a fusion decoupling formula to obtain the current state vector. .

[0055] The fusion decoupling formula is used to describe the sensor readings. ) and the actual state of the pipeline under test The relationship between the sensor readings and the state vector is determined by the fusion decoupling formula used to correct the current monitoring data. This step is crucial for fusion and requires mapping sensor readings onto the state vector.

[0056] The fusion decoupling formula in this embodiment is:

[0057] In the formula, express The monitoring data collected from the pipeline under test at all times (i.e., the current monitoring data). Represents the coupled observation function, express The measurement noise of the sensor itself (i.e., the sensor's own error, such as the white noise of a fiber optic demodulator). ,in, This represents the raw frequency shift of the pipe under test as measured by the distributed fiber optic sensor BOTDR, including temperature and strain information. This represents the temperature of the pipe under test as measured by the distributed temperature sensor (DTS), including the temperature of the pipe wall and the surrounding soil. Indicates pressure sensor The measured pressure of the medium inside the pipe to be tested.

[0058] Furthermore, since the data measured by distributed fiber optic sensors and distributed temperature sensors are distributed data, while the data measured by pressure sensors are point data, pressure sensors typically have a high sampling rate. Fiber optic demodulation is typically slow (on the order of minutes). Therefore, after obtaining the current state vector, it is necessary to perform spatiotemporal alignment processing on the current state vector to construct the current aligned vector. .

[0059] The spatiotemporal alignment process includes the following steps: 1. Establishing a pipeline mileage mapping table, gridding the original frequency shift in the current state vector by meter, and broadcasting the pipe medium pressure data in the current state vector to the pipe segment grid where it is located, or interpolating it to the entire line through a fluid dynamics model to complete the spatial alignment of the current state vector; 2. Using the fiber demodulation period as a reference, downsampling the pipe medium pressure data or interpolating the original frequency shift of the pipeline under test at time k-1 to complete the temporal alignment of the current state vector; 3. Based on the spatial alignment and the spatially aligned current state vector, constructing a current alignment vector at a unified time. .

[0060] After performing spatiotemporal alignment on the current state vector, the vector correction formula is applied based on the Kalman gain. and the current alignment vector For the predicted state vector After correction, the target state vector is obtained. .

[0061] The vector correction formula is: K≥1 In the formula, express The target state vector at time t. Indicates according to Predicted from historical monitoring data at any given time The predicted state vector at time t. express Kalman gain at time step express The current alignment vector after aligning the current monitoring data at any given time. This represents the value obtained by calculating the predicted state vector using the coupled observation function. This represents the observation residual, which is the deviation between the current alignment vector and the predicted state vector.

[0062] By dynamically adjusting the weights of model predictions and sensor measurements using Kalman gain, the cross-sensitivity of temperature to strain measurements is automatically decoupled, random noise is suppressed, and the target state vector is improved. Based on the confidence level of the data, key features needed for subsequent steps are extracted, making the extracted key features more consistent with the real situation.

[0063] S40: Determine the current maximum internal stress of the pipeline under test based on the target state vector, and determine the safety margin by combining the allowable stress of the material with the function of temperature change, and determine the pipeline state risk level based on the safety margin.

[0064] Furthermore, such as Figure 2 As shown, after obtaining the target state vector, the current maximum internal stress of the pipe under test is determined based on the target state vector, specifically including the following steps: S41: Identify the permafrost environment features of the target state vector to obtain soil state labels.

[0065] Specifically, after obtaining the target state vector, the current soil temperature in the target state vector is obtained, and then the current soil temperature is input into the frozen soil environment feature recognition program for recognition. When the current soil temperature is greater than or equal to the preset temperature threshold, and it is determined based on the current soil temperature that the soil temperature is continuously rising, the thawing period analysis process is entered to determine whether the thawing depth and strain abrupt change characteristics both conform to the thawing settlement characteristics. When both the thawing depth and strain abrupt change characteristics conform to the thawing settlement characteristics, a thawing settlement early warning label is output as the first soil state label. When the thawing depth or strain abrupt change characteristics do not conform to the thawing settlement characteristics, a normal thawing label is output as the second soil state label.

[0066] If the current soil temperature is less than or equal to the preset temperature threshold, and the soil temperature is determined to be continuously decreasing based on the current soil temperature, the freezing period analysis process is initiated to determine whether the temperature change rate and strain rate both meet the frost heave characteristics. If both the temperature change rate and strain rate meet the frost heave characteristics, a frost heave warning label is output as the third soil state label. If the temperature change rate or strain rate does not meet the frost heave characteristics, a normal freezing label is output as the fourth soil state label.

[0067] Furthermore, in the melting period region, as the ice lens melts, the soil volume shrinks, and the pipeline loses support and sinks (becomes suspended or bends downward). The melting period analysis process in this embodiment specifically includes: determining the strain change time based on the current mechanical strain in the target state vector, and determining the temperature change time based on the current temperature in the target state vector; when the mechanical strain change time lags behind the temperature change time, and there is abrupt change characteristics in the mechanical strain. Proportional to the depth of melting ,Right now If the current mechanical strain changes abruptly and is located at the melting front, then it is determined whether the current mechanical strain changes abruptly and whether it is located at the melting front. If the current mechanical strain changes abruptly and is located at the melting front, then it is determined that it meets the melting and sinking characteristics.

[0068] in, This indicates the abrupt change in mechanical strain, specifically the abrupt change in strain caused by melt deposition. This indicates the depth of melting, specifically for temperatures above the freezing threshold. The portion is integrated over time, representing the total heat input into the soil. If the current soil temperature of a section of the pipeline under test... If the temperature continues to drop and the 0℃ isotherm moves deeper into the soil, then it is determined to be the freezing period.

[0069] Specifically, the current abrupt change in mechanical strain refers to the period during melting ( When temperatures exceed 0°C, the ice in the soil melts and is drained away, causing the soil volume to shrink. The bottom of the pipe loses soil support (forming a cavity), and the pipe bends downwards under its own weight or the pressure of the overlying soil (settlement). Abrupt strain changes refer to a sudden increase in compressive strain at the top of the pipe and tensile strain at the bottom at the melting front, or a localized concentration of tensile stress in the axial direction. This change is typically step-like or rapidly nonlinear, and occurs slightly later than the temperature rise.

[0070] This embodiment uses three dimensions to quantify strain mutation characteristics: A. Amplitude quantization: Residuals exceeding the "reference thermal strain" First, calculate a "theoretical strain reference value" (strain under pure temperature influence):

[0071] In the formula, This represents the coefficient of linear expansion of the pipeline. This represents the current temperature in the target state vector. This indicates the reference base value.

[0072] The quantitative criterion for "mutation" is:

[0073] In the formula, This represents the actual mechanical strain in the target state vector. This indicates the preset safety tolerance.

[0074] B. Rate Quantization: Anomaly in Time-Domain Rate of Change When melting and settling occur, it is often accompanied by the sudden collapse of the soil structure.

[0075] Quantitative criteria:

[0076] If, under conditions of slow temperature change, the rate of change of strain over time (slope) suddenly increases significantly, and the direction is either stretching (positive) or bending, this is considered an abrupt change. This represents the preset theoretical coefficients.

[0077] C. Spatial morphology quantification: Curvature abrupt change (bending characteristics) Melting and settling are usually localized, which can cause “peaks” or “troughs” in the strain distribution along the axial direction of the pipe.

[0078] The quantitative criterion is:

[0079] The system will calculate along the length of the pipe. The second derivative of strain (curvature). Under normal, uniform thermal expansion and contraction, the second derivative is close to 0; however, during melting and sinking (bending), this value increases significantly, forming a stress concentration point. Among these, This represents the set curvature threshold, used to determine whether there is abnormal local bending.

[0080] Melting depth is typically estimated using the Stefan Formula or a simplified form, which utilizes the accumulated temperature.

[0081] The specific calculation steps include: 1. Calculate the melting index ( .

[0082] 2. For those detected exceeding the freezing threshold Integrate the temperature over time.

[0083]

[0084] In the formula, This indicates the melting index (accumulated temperature, unit: °C·h). Indicates the first The actual measured temperature at any given time. This indicates the soil freezing temperature threshold (e.g., 0℃).

[0085] 3. Calculate the melting depth ( .

[0086] 4. Use empirical coefficients to convert accumulated temperature into depth.

[0087]

[0088] In the formula, This indicates the currently calculated melting depth. This represents the soil thermal conductivity correction factor, set based on local geological data. This indicates that the depth of melting is usually proportional to the square root of the accumulated temperature.

[0089] During the freezing period, the pipeline will experience localized bending strain (typically upward arching deformation). The freezing period analysis process in this embodiment specifically includes: obtaining the current strain rate based on the target state vector. and current temperature change rate If the current strain rate is greater than 0 and the current temperature change rate is less than 0, then determine whether the difference between the current mechanical strain and the reference thermal stress is greater than a preset threshold; if the difference between the current mechanical strain and the reference thermal stress is greater than the preset threshold, then determine that it meets the characteristics of frost heave.

[0090] S42: Input the current mechanical strain, current temperature, current internal medium pressure and soil state label in the target state vector into the pipeline stress estimation model trained by XGBoost regression tree for identification, and obtain the current maximum internal stress of the pipeline to be tested.

[0091] Specifically, after obtaining the soil condition marker, the current mechanical strain, current temperature, current pipe medium pressure, and soil condition marker in the target state vector are input into the pipeline stress estimation model. The pipeline stress estimation model identifies the current mechanical strain, current temperature, current pipe medium pressure, and soil condition marker in the target state vector to obtain the current maximum internal stress of the pipeline under test.

[0092] The pipeline stress estimation model in this embodiment adopts... Regression trees are highly capable of handling nonlinear mappings. During the regression tree training process, the training data used includes a simulation database obtained from finite element simulation analysis and measured data. The simulation database contains the mapping relationship between temperature, pressure, and frost heave and the external surface strain and the maximum internal von Mises stress.

[0093] This embodiment uses simulation software (such as ABAQUS or ANSYS) to build a physical model and generate training data. The regression tree dataset. A three-dimensional solid model of the flexible composite pipe is established to accurately delineate the inner lining, reinforcing winding layer (pressure-bearing core), and outer sheath. Additionally, during the simulation, an extreme cold environment coupling field needs to be set up, which includes: Temperature field: Set the material constitutive relationship from -50℃ to +80℃ (considering the hardening and brittleness changes of polymer materials at low temperatures).

[0094] Frozen soil interaction: Simulating frost heave (applying upward non-uniform displacement load) and thaw settlement (applying gravity load when the bottom loses support) conditions.

[0095] Massive operating condition simulation : Perform a parametric scan to calculate the response under different combinations: Combination: (temperature) ×Internal pressure × Freezing heave .

[0096] Record: Corresponding external surface strain (The maximum von Mises stress that can be measured by optical fiber) and its internal structure. The mapping relationship between (the most dangerous but undetectable by optical fibers).

[0097] Furthermore, after obtaining the current maximum internal stress, the safety margin is determined by combining the allowable stress of the material with the function of temperature, and the pipeline condition risk level is determined based on the safety margin.

[0098] Specifically, after obtaining the current maximum internal stress, the safety margin is determined by combining the allowable stress of the material with the function of temperature, and then the pipeline condition risk level is determined based on the safety margin calculation formula.

[0099] The formula for calculating the safety margin is:

[0100] In the formula, Indicates safety margin, This represents a function that expresses the allowable stress of a material as a function of temperature. This indicates the current maximum internal stress of the pipe under test. , This represents the model function corresponding to the pipeline stress estimation model. This represents the current mechanical strain in the target state vector. This represents the current temperature in the target state vector. This represents the current pressure of the medium inside the pipe in the target state vector.

[0101] The risk level of the pipeline status is determined based on the safety margin. In this embodiment, the risk levels include, but are not limited to, safe, concern, early warning, and alarm.

[0102] To facilitate understanding, this embodiment illustrates the relationship between the aforementioned safety margin and state risk level through examples. The following values ​​are merely illustrative examples in this embodiment and do not represent a limitation on the range of values ​​for the safety margin.

[0103] : Safe (green); Attention (yellow); Warning (orange); Alarm (red, indicating risk of failure).

[0104] S50: Collect the acoustic signal of the pipeline under test and determine the location of the leak based on the acoustic signal.

[0105] Specifically, the acoustic signal of the pipeline under test is collected by a distributed acoustic sensor (DAS) as the raw signal, and the raw signal is decomposed into time and frequency components by wavelet transform to obtain detail components and approximate components of different frequency bands. The energy proportion and Shannon entropy of each frequency band are calculated for the detail components of different frequency bands. A feature vector F is constructed based on the energy proportion and Shannon entropy of all frequency bands. The feature vector F is input into the signal leakage identification model for identification to obtain the identification result. When the identification result is a leakage, it is determined whether there is a negative pressure wave in the corresponding time window based on the current time. If the existence of a negative pressure wave is confirmed, the leakage identification result is that there is a leakage. After the existence of a leakage is confirmed, the leakage location is determined by the TDOA positioning algorithm based on the raw signal.

[0106] In this embodiment, the wavelet transform is selected. or Wavelet bases decompose the original signal into detail components in different frequency bands. and approximate components .

[0107] Leakage characteristics: typically concentrated in the high-frequency band (e.g., (The hissing sound of jetting).

[0108] Environmental noise is usually concentrated in the low-frequency range (such as vehicle vibration and geological subsidence).

[0109] Energy characteristic calculation: Calculate the energy percentage and Shannon entropy of each frequency band. Construct feature vectors :

[0110] Constructing feature vectors It can effectively highlight the "continuous broadband" characteristic of the leaked signal, distinguishing it from transient interference.

[0111] By performing multi-scale decomposition of the original signal using wavelet transform, high-frequency components that match the leakage characteristic frequency of the pipeline under test are further extracted, effectively suppressing background noise interference.

[0112] The feature vector is input into the signal leakage detection model for identification, and the probability of the current signal belonging to the following categories is output. The category with the highest probability is taken as the identification result. The categories in this embodiment include: Normal fluid noise Leakage incident

[0113] Mechanical interference

[0114] Noise from frozen soil activity

[0115] Furthermore, in this embodiment, after determining that there is a "leak" based on the signal leakage identification model, the data collected by the pressure sensor in the pipeline under test will be checked. If a pressure transient drop (i.e. a negative pressure wave) is detected within the same time window, it is confirmed that there is a leak and an alarm is triggered. This "sound wave + pressure" dual confirmation mechanism can greatly reduce the false alarm rate.

[0116] Furthermore, the signal leakage identification model in this embodiment uses a BP neural network, and the training process of the BP neural network includes: By employing computational fluid dynamics and acoustic finite element method, the propagation and attenuation characteristics of sound waves in the pipe-soil coupling medium under extremely cold conditions are simulated to generate an acoustic fingerprint of leakage. Pipeline acoustic signals are acquired, and positive samples are generated based on the pipeline acoustic signals, the acoustic fingerprint of leakage, and leakage labels. Interference noise signals are acquired, and negative samples are generated based on the interference noise signals and non-leakage labels. The positive and negative samples are used as a sample set, which is then divided into a training set and a test set. The training set is input into a BP neural network for training to obtain a signal leakage training model. The test set is input into the signal leakage training model for testing; if the test results are accurate, the signal leakage training model is used as the signal leakage identification model.

[0117] Due to the hardening of soil in extremely cold environments, acoustic properties change drastically, rendering summer leakage characteristic databases ineffective in winter. Therefore, it is necessary to use simulation to "reverse engineer" the unique acoustic fingerprints of winter leaks.

[0118] The specific implementation process includes three steps: sound source generation, propagation simulation, and fingerprint extraction. Step 1: Generate the "initial sound source" using CFD (Computational Fluid Dynamics). Modeling: Build a pipeline failure model in CFD software (such as Fluent) and set up high-pressure fluid jet conditions; Calculation: Simulate turbulent fluctuations of fluid at the leak point (Lighthill acoustic analogy theory); Output: Obtain near-field pressure pulsation data near the leak. This is the unattenuated, raw signal from the broadband noise source.

[0119] Step 2: Simulate the "environmental filtering effect" using acoustic finite element method. Multiphysics coupling: The "near-field pressure" obtained in step one is used as the sound source load and applied to the pipe model in acoustic finite element software (such as ACTRAN or ANSYS Acoustics).

[0120] Define the properties of frozen soil: Set the soil to a frozen state: significantly increase the elastic modulus (harden it), adjust the density and Poisson's ratio, and modify the sound velocity and sound absorption coefficient (frozen soil has better conductivity for high-frequency sound waves than soft soil). Propagation calculations were performed to simulate the entire process of sound waves traveling from the leak point through the pipe wall, into the frozen soil, and finally reaching the fiber optic sensor. During this process, propagation attenuation characteristics came into play—some frequencies were absorbed by the soil (attenuation), while others were preserved due to pipe-soil resonance.

[0121] Step 3: Extract the "acoustic fingerprint" of the leak. Virtual monitoring: Virtual sensor probes are set at different distances in the simulation model (e.g., 10m, 20m from the leak point).

[0122] Fingerprint generation: 1. Frequency domain analysis: Perform FFT (Fast Fourier Transform) on the time domain signal received by the virtual probe; 2. Feature locking: Observe the spectrum and find the specific frequency band with the most concentrated energy and the slowest attenuation in the permafrost environment (usually the high frequency band of 1kHz-5kHz). 3. Fingerprint definition: The combination of "center frequency + bandwidth + energy attenuation slope" is the leakage acoustic fingerprint under this operating condition.

[0123] This process fills the gap in the measured data regarding the lack of data on extreme freezing conditions.

[0124] Furthermore, this embodiment processes the original signal using the TDOA positioning algorithm to determine the leak location, specifically including the following steps: Step 1: In the distributed acoustic wave sensor monitoring array, once a leak is confirmed based on the leak identification results, the sensor points with the highest and second-highest signal-to-noise ratio (SNR) along the distributed acoustic wave sensor line are automatically selected as monitoring points. The fiber optic cable length between the two monitoring points is [length missing]. The sensor point with the highest signal-to-noise ratio (SNR) is selected as the first monitoring point. The sensor point with the second highest signal-to-noise ratio (SNR) was selected as the second monitoring point. The raw signals collected from the two monitoring points were obtained respectively.

[0125] Step 2: After obtaining the original signal, denoise the original signal to obtain the time-domain signal of the leakage sound wave. and The time delay between the two leaking acoustic wave time-domain signals was calculated using the generalized cross-correlation function (GCC). .

[0126] The generalized cross-correlation function in this embodiment is specifically as follows:

[0127] In the formula, Represents the mathematical expectation. Represents the time-domain signal of the leaking acoustic wave and The time lag variable between them.

[0128] Since the leakage sound waves originate from the same source (the leak point), the time-domain signals of the leakage sound waves received by the two distributed sound wave sensors are similar, with only a time difference in arrival. .

[0129] On the spectrum of the generalized cross-correlation function, the time corresponding to the maximum correlation value (peak value) is the arrival time difference between the two leaking acoustic wave time-domain signals. :

[0130] Step 3: Determine the effective propagation velocity of the leakage acoustic wave time-domain signal in the flexible composite pipe medium. .

[0131] Considering the variation of the elastic modulus of the flexible composite pipe under different temperatures and pressures, the effective propagation velocity of the leakage sound wave time-domain signal in the flexible composite pipe medium is... It is not a fixed value, but rather based on the currently monitored temperature. and current monitoring pressure Data, determined by querying a pre-set sound velocity-operating condition mapping table, helps improve the positioning accuracy of the positioning algorithm.

[0132] Step 4: Based on the fiber optic cable length between the two monitoring points Arrival time difference and effective speed of sound propagation Calculate the distance from the leak point to the first monitoring point. distance Based on the location of the first monitoring point and the distance between the leak point and the first monitoring point To determine the location of the leak.

[0133] Specifically, when the leak point is far from the first monitoring point The distance is The distance between the leak point and the second monitoring point is... The distance is .

[0134] The time-domain signal of the leaking acoustic wave from the leak point is transmitted to the first monitoring point. The time is:

[0135] The time-domain signal of the leaking acoustic wave from the leak point is transmitted to the second monitoring point. The time is:

[0136] The time difference of arrival of the two leaking acoustic wave time-domain signals:

[0137] The distance between the leak point and the first monitoring point was finally determined. The distance is:

[0138] In the formula, This represents the known length of the optical fiber between the two monitoring points. This represents the propagation speed of the leakage acoustic wave in the medium or wall inside the flexible composite pipe, and is a calibration value. This represents the time difference calculated using the generalized cross-correlation function (if it is). First, the signal was received. (Negative; conversely, positive).

[0139] S60: Calculate the signal energy intensity of the acoustic signal and determine the pressure drop rate based on the current monitoring data. Assess the severity of the leak by using the signal energy intensity and pressure drop rate.

[0140] 1. The calculation process for signal energy intensity specifically includes: After acquiring the acoustic signal using a distributed acoustic sensor, the acoustic signal is used as the raw signal and denoised. Then, the root mean square energy within a time window (e.g., 1 second) before and after the leak occurs is calculated based on the denoised time-domain signal of the leaking acoustic wave.

[0141] In the formula, This represents the signal energy intensity of the leaked acoustic wave in the time domain. This represents the amplitude of the sound wave signal after noise reduction processing. This indicates the number of sampling points, which reflects the flow velocity and turbulence intensity at the leak point.

[0142] 2. The calculation process for the pressure drop rate specifically includes: Using the pressure of the medium inside the pipe collected by the pressure sensor, the rate of change (derivative) of the pressure of the medium inside the pipe with time is calculated, i.e., the rate of pressure drop:

[0143] In the formula, Indicates the rate of pressure decrease. This indicates the pressure sensor data collected. The pressure of the medium inside the pipe at any given time, i.e., the pressure value at the current moment.

[0144] The rate of pressure drop can be calculated to reflect the speed at which fluid leaks out of the flexible composite pipe. When a flexible composite pipe bursts, it causes a sudden and rapid drop in the pressure of the medium inside the pipe. (Extremely high), while when a flexible composite pipe experiences a minor leak, the pressure drop of the medium inside the pipe is extremely slow or even masked by the pressure replenishment from the pump station, making... .

[0145] To improve the accuracy of leakage severity assessment, this embodiment employs a "sound-pressure joint judgment logic" to classify leakage severity into three levels, specifically including: Level I: Minor Leak (See page) When the signal energy intensity of the leaked acoustic wave time-domain signal exceeds the background noise threshold (high-frequency hissing), that is... And the rate of pressure drop Extremely low or no significant decrease was detected, i.e. If so, it may be due to pinholes in the inner lining of the flexible composite pipe, or slight leakage at the joint of the flexible composite pipe, where the medium enters the annulus but does not spray out in large quantities, which is judged as a minor leakage.

[0146] Level II: Moderate Leak When the signal energy intensity of the leaked acoustic wave time-domain signal increases significantly, that is And this is accompanied by an observable slow decrease in pressure, i.e. If the leak is due to cracks or holes in the flexible composite pipe, the fluid will leak continuously, but the pressure will be maintained within a certain range under the action of the pump, which is judged as a moderate leak.

[0147] Level III: Severe Burst / Rupture When the signal energy intensity of the leakage acoustic wave time-domain signal instantaneously reaches saturation (or a huge impact signal occurs), and the pressure drop rate... Extremely high (negative pressure wave appears) If the flexible composite pipe has completely broken or the pressure-bearing layer has failed, resulting in pressure loss, it is considered a severe pipe burst. (In this case, pressure is the primary factor in the determination, with acoustic waves as a secondary confirmation).

[0148] In this embodiment .in, This represents the first background noise threshold, used to determine whether a leak is minor. This represents the second background noise threshold, used to determine whether the noise level indicates a moderate leak. This indicates the first pressure threshold, used to distinguish between minor leaks and moderate leaks; This indicates the second pressure threshold, used to distinguish between moderate leaks and severe pipe ruptures.

[0149] S70: Monitor the safety status of the pipeline under test based on the pipeline condition risk level, as well as the location and severity of the leak.

[0150] This application embodiment predicts the current condition of the pipeline under test by acquiring historical monitoring data, obtaining a predicted state vector. Then, it determines the Kalman gain by acquiring the current monitoring data of the pipeline under test, and determines whether to believe the predicted value (i.e., the predicted state vector) based on the Kalman gain. When the Kalman gain meets the preset judgment range (i.e., the Kalman gain infinitely approaches 0 but is not 0), the predicted value is chosen to be believed. To improve the accuracy of the data, after determining to believe the predicted value, it is necessary to correct the predicted value based on the Kalman gain to obtain the corrected target state vector. Then, based on the target state vector, the soil state marker is determined, thereby determining the current maximum internal stress of the pipeline under test. Combined with the function of allowable stress of the material changing with temperature, the pipeline state risk level is determined. Finally, the leakage identification result is determined by the signal leakage identification model, and the leakage location is determined by the TDOA positioning algorithm. Based on the signal energy intensity and pressure drop rate in the current monitoring data, the severity of the leakage is determined. This effectively solves the problem of inaccurate monitoring caused by the inability to predict the state of buried flexible composite pipes in the prior art, realizes the state prediction of buried flexible composite pipes, and improves the accuracy of safety monitoring of buried flexible composite pipes.

[0151] While this application provides method operation steps as shown in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive labor. The order of steps listed in this embodiment is merely one possible execution order among many and does not represent the only execution order. In actual device or client product execution, the method can be executed sequentially according to this embodiment or the accompanying drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0152] like Figure 3 As shown in the figure, this application embodiment also provides a flexible composite pipe safety status monitoring device, the device comprising: The historical monitoring data processing module 10 is used to acquire historical monitoring data of the pipeline to be tested, and to calculate the predicted state vector by using the state vector prediction formula.

[0153] The current monitoring data processing module 20 is used to acquire the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function.

[0154] The predicted state vector correction module 30 is used to correct the predicted state vector based on the Kalman gain and the current monitoring data when the Kalman gain meets the preset judgment range, so as to obtain the corrected target state vector.

[0155] The risk level determination module 40 is used to determine the current maximum internal stress of the pipeline under test based on the target state vector, and to determine the safety margin by combining the allowable stress of the material with the function of temperature change, and to determine the risk level of the pipeline state based on the safety margin.

[0156] The leak location determination module 50 is used to collect the acoustic signal of the pipeline under test and determine the leak location based on the acoustic signal.

[0157] The leakage severity assessment module 60 is used to calculate the signal energy intensity of the acoustic signal and determine the pressure drop rate based on the current monitoring data, and assess the leakage severity by combining the signal energy intensity and the pressure drop rate.

[0158] The pipeline safety status monitoring module 70 is used to monitor the safety status of the pipeline under test based on the pipeline status risk level, leakage location, and leakage severity.

[0159] like Figure 4 As shown in the embodiments of this application, a safety status monitoring system for flexible composite pipes is also provided. The system includes a cloud monitoring platform, a monitoring station, a multi-parameter sensing and detection unit, and an intelligent analysis and early warning unit.

[0160] The cloud-based monitoring platform receives data retrieval commands, retrieves historical monitoring data based on these commands, and sends it to the intelligent analysis and early warning unit.

[0161] The multi-parameter sensing and detection unit acquires the current monitoring data through sensors arranged in the buried flexible composite pipe and transmits it to the monitoring station.

[0162] After receiving the current monitoring data, the monitoring station preprocesses the data through the data acquisition and processing unit and converts it into a digital signal for transmission to the intelligent analysis and early warning unit.

[0163] Specifically, the data acquisition and processing unit in this embodiment includes a signal conditioning circuit, an AD conversion module, an embedded processor, and a local storage unit.

[0164] The signal conditioning circuit is used to amplify, filter, isolate, and linearize the analog signals output by the pressure sensor and the acceleration sensor.

[0165] An AD conversion module is used to convert analog signals processed by the signal conditioning circuit into digital signals. Furthermore, in this embodiment, the AD conversion module employs an analog-to-digital converter with a resolution of at least 24 bits.

[0166] An embedded processor is used to perform feature extraction and edge-side data analysis on the digital signal output by the AD conversion module. Local storage units are used to cyclically store monitoring data, processing results, and event logs.

[0167] After receiving historical and current monitoring data, the intelligent analysis and early warning unit executes the aforementioned flexible composite pipe safety status monitoring method to obtain the pipeline status risk level, leakage location, and leakage severity, and then transmits this information to the cloud monitoring platform.

[0168] After receiving data on the pipeline's risk level, leak location, and leak severity, the cloud-based monitoring platform activates the emergency command and dispatch module and sends warning information to the corresponding clients via a mobile app.

[0169] Furthermore, users can send safety monitoring commands to the multi-parameter sensing and detection unit via the client. After receiving the safety monitoring command, the multi-parameter sensing and detection unit executes the steps of acquiring the current monitoring data through the sensors arranged in the buried flexible composite pipe and transmitting it to the monitoring station.

[0170] The aforementioned flexible composite pipe safety status monitoring system can achieve closed-loop monitoring of the safety status of buried flexible composite pipes in extremely cold environments, specifically as follows: Figure 5 As shown.

[0171] Furthermore, the cloud monitoring platform in this embodiment also includes a real-time monitoring interface and a data receiving and storage server. The real-time monitoring interface is used for centralized monitoring of the buried flexible composite pipe; the data receiving and storage server is used to receive and store data sent by the data acquisition and processing unit.

[0172] The monitoring station also includes an energy management unit, a communication and remote control unit, and a thermal insulation unit; among which, The energy management unit (EMU) is used to maintain the system's long-term independent operation under conditions of insufficient solar energy or continuous cloudy / snowy weather. This EMU includes solar panels, a battery storage array, a battery heating and insulation device, an intelligent power management chip, and a supercapacitor auxiliary power supply module. In extremely cold environments, after the monitoring station is powered on, the EMU first performs status checks on the solar panels, battery storage array, and supercapacitor, and raises the battery temperature to a set range using the battery heating and insulation device. When solar energy is insufficient or under conditions of continuous cloudy / snowy weather, the battery storage array maintains a suitable operating temperature through the battery heating and insulation device. The supercapacitor auxiliary power supply module provides peak power when the system experiences instantaneous demands such as high-current startup or communication transmission, ensuring the stability of the monitoring station's power supply. The intelligent power management chip enables maximum power point tracking and charging management for the solar panels.

[0173] The communication and remote control unit is used to enable two-way data transmission and remote control between the monitoring station and the cloud monitoring platform.

[0174] The thermal insulation protection unit includes a thermally insulated outer shell covering the monitoring station cabinet, a waterproof and moisture-proof sealing structure, an internal heating system, and a frost-resistant foundation, which is used to improve the reliability and structural stability of the monitoring station in extremely cold environments.

[0175] Furthermore, the sensors in this embodiment include a distributed fiber optic strain sensor (BOTDR), a distributed temperature sensor (DTS), and a pressure sensor. Accelerometers and leak detection sensors, such as... Figure 6 As shown, the distributed optical fiber sensor and the distributed optical fiber strain sensor are laid out along the flexible composite pipeline, which includes multiple bends, tees and valve chambers.

[0176] Some modules in the apparatus described in this application can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0177] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. When implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0178] The methods, apparatus, or modules described in this application can be implemented in a computer-readable program code manner. The controller can be implemented in any suitable manner, such as a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of a memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code manner, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included within it for implementing various functions can also be considered as structures within the hardware component. Alternatively, the device used to implement various functions can be viewed as either a software module that implements the method or a structure within a hardware component.

[0179] Furthermore, in the various embodiments of the present invention, each functional module can be integrated into one processing module, or each module can exist independently, or two or more modules can be integrated into one module.

[0180] The aforementioned storage media include, but are not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions.

[0181] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary hardware. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, or it can be embodied in the process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0182] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this application can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0183] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A method of monitoring the safety state of a flexible composite pipe, characterized by, include: Historical monitoring data of the pipeline to be tested is obtained, and the predicted state vector is determined by calculating the historical monitoring data using the state vector prediction formula. Acquire the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function; When the Kalman gain meets the preset judgment range, the predicted state vector is corrected based on the Kalman gain and the current monitoring data to obtain the corrected target state vector; The maximum internal stress of the pipeline under test is determined based on the target state vector, and the safety margin is determined by combining the allowable stress of the material with the function of temperature. The risk level of the pipeline condition is determined based on the safety margin. Acquire acoustic signals from the pipeline under test and determine the leak location based on the acoustic signals; Calculate the signal energy intensity of the acoustic signal, determine the pressure drop rate based on the current monitoring data, and assess the severity of the leak using the signal energy intensity and the pressure drop rate; The safety status of the pipeline under test is monitored based on the pipeline's risk level, the location of the leak, and the severity of the leak.

2. The method for monitoring the safety status of a flexible composite pipe according to claim 1, characterized in that, The step of calculating the predicted state vector from the historical monitoring data using the state vector prediction formula includes: The historical monitoring data is converted into a historical state vector, and the predicted state vector is obtained by calculating the historical state vector, the state transition matrix, and the process noise using the state vector prediction formula. The state vector prediction formula is as follows: In the formula, express The predicted state vector of the pipeline under test at time t, where A represents the state transition matrix. express The historical state vector of the pipeline under test at any given time. express The process noise of the pipeline under constant monitoring.

3. The method for monitoring the safety status of a flexible composite pipe according to claim 1, characterized in that, The determination of the Kalman gain based on the coupled observation function includes: Establish a coupled observation function and determine the Jacobian matrix by taking the partial derivative of the coupled observation function; The prior error covariance matrix is ​​determined using the covariance prediction formula. Obtain the preset measurement noise covariance, and calculate the Kalman gain by applying the Kalman gain formula to the Jacobian matrix, the prior error covariance matrix, and the preset measurement noise covariance.

4. The method for monitoring the safety status of a flexible composite pipe according to claim 3, characterized in that, The coupled observation function is: In the formula, Represents the coupled observation function. This represents the strain coefficient of the optical fiber in the pipe under test. express The real mechanical strain of the pipeline to be measured at all times This represents the temperature coefficient of the optical fiber in the pipe under test. express The actual temperature of the pipeline and surrounding soil is to be measured at all times. express The actual internal pressure of the medium in the pipeline to be measured at all times. This indicates the reference temperature for calibrating the optical fiber in the pipe under test; The covariance prediction formula is as follows: In the formula, Let A represent the prior error covariance matrix at time k calculated based on historical monitoring data at time k-1, and let A represent the state transition matrix. This represents the posterior error covariance matrix at time k-1, calculated based on historical monitoring data at time k-1. This represents the transpose of the state transition matrix. This represents the preset process noise covariance; The Kalman gain calculation formula is as follows: In the formula, This represents the Kalman gain at time k. Let represent the Jacobian matrix obtained after linearizing the coupled observation function at time k. denoted as the preset measurement noise covariance, and T represents the matrix transpose.

5. The method for monitoring the safety status of a flexible composite pipe according to claim 4, characterized in that, The step of correcting the predicted state vector based on the Kalman gain and the current monitoring data to obtain the corrected target state vector includes: The sensor's own measurement noise is acquired and combined with the coupled observation function. The current monitoring data is then corrected using a fusion decoupling formula to obtain the current state vector. Perform spatiotemporal alignment processing on the current state vector to construct the current alignment vector; Based on the Kalman gain and the current alignment vector, the predicted state vector is corrected using a vector correction formula to obtain the target state vector.

6. The method for monitoring the safety status of a flexible composite pipe according to claim 5, characterized in that, The fusion decoupling formula is as follows: In the formula, express The monitoring data collected continuously from the pipeline under test Represents the coupled observation function. express The measurement noise of the time sensor itself. ,in, This represents the raw frequency shift of the pipe under test as measured by the distributed fiber optic sensor, including temperature and strain information. This indicates the temperature of the pipe under test as measured by the DTS distributed temperature sensor, including the temperature of the pipe wall and the surrounding soil. This indicates the pressure of the medium inside the pipe being tested, as measured by the pressure sensor. The vector correction formula is: In the formula, express The target state vector at time t. Indicates according to Predicted from historical monitoring data at any given time The predicted state vector at time t. express Kalman gain at time step express The current aligned vector is obtained by performing spatiotemporal alignment processing on the current state vector at time t. This represents the value of the predicted state vector calculated using the coupled observation function. This represents the observation residual, which is the deviation between the current alignment vector and the predicted state vector.

7. The method for monitoring the safety status of a flexible composite pipe according to claim 1, characterized in that, Determining the current maximum internal stress of the pipe under test based on the target state vector includes: The target state vector is subjected to permafrost environment feature identification to obtain soil state labels; The current mechanical strain, current temperature, current internal medium pressure, and soil state label in the target state vector are input into the pipeline stress estimation model trained by XGBoost regression tree for identification, so as to obtain the current maximum internal stress of the pipeline under test.

8. The method for monitoring the safety status of a flexible composite pipe according to claim 1, characterized in that, The process of acquiring acoustic signals from the pipeline under test and determining the leak location based on the acoustic signals includes: Acoustic signals from the pipe under test are collected as the original signals, and wavelet transform is used to perform time-frequency decomposition on the original signals to obtain detail components and approximate components in different frequency bands. The detailed components of different frequency bands are calculated to obtain the energy proportion and Shannon entropy of each frequency band; Feature vectors are constructed based on the energy proportions of all frequency bands and Shannon entropy. The feature vector is input into the signal leakage identification model for identification, and the identification result is obtained. When the identification result is a leak, it is determined whether a negative pressure wave exists within the corresponding time window based on the current time; if the existence of the negative pressure wave is confirmed, the leak identification result is that a leak exists. Once a leak is confirmed, the leak location is determined using the TDOA positioning algorithm based on the original signal.

9. A flexible composite pipe safety status monitoring device, characterized in that, include: The historical monitoring data processing module is used to acquire historical monitoring data of the pipeline under test, and to calculate the predicted state vector by using the state vector prediction formula. The current monitoring data processing module is used to acquire the current monitoring data of the pipeline under test, establish a coupled observation function based on the current monitoring data, and determine the Kalman gain based on the coupled observation function. The predicted state vector correction module is used to correct the predicted state vector based on the Kalman gain and the current monitoring data when the Kalman gain meets the preset judgment range, so as to obtain the corrected target state vector. The risk level determination module is used to determine the current maximum internal stress of the pipeline under test based on the target state vector, and to determine the safety margin by combining the allowable stress of the material with the function of temperature change, and to determine the risk level of the pipeline state based on the safety margin. The leak location determination module is used to collect the acoustic signal of the pipeline under test and determine the leak location based on the acoustic signal; The leakage severity assessment module is used to calculate the signal energy intensity of the acoustic signal and determine the pressure drop rate based on the current monitoring data, and assess the leakage severity by using the signal energy intensity and the pressure drop rate. The pipeline safety status monitoring module is used to monitor the safety status of the pipeline under test based on the pipeline status risk level, the leak location, and the leak severity.

10. A flexible composite pipe safety status monitoring system, characterized in that, This includes a cloud-based monitoring platform, monitoring stations, multi-parameter sensing and detection units, and intelligent analysis and early warning units; The cloud monitoring platform obtains the data retrieval command, retrieves historical monitoring data according to the data retrieval command, and sends it to the intelligent analysis and early warning unit; The multi-parameter sensing and detection unit acquires the current monitoring data through sensors arranged in the buried flexible composite pipe and transmits it to the monitoring station. After receiving the current monitoring data, the monitoring station preprocesses the current monitoring data through the data acquisition and processing unit and converts the current monitoring data into digital signals for transmission to the intelligent analysis and early warning unit. After receiving historical and current monitoring data, the intelligent analysis and early warning unit executes the flexible composite pipe safety status monitoring method as described in any one of claims 1-8 to obtain the pipeline status risk level, leakage location, and leakage severity, and transmits it to the cloud monitoring platform. After receiving data on the pipeline status risk level, leak location, and leak severity, the cloud monitoring platform activates the emergency command and dispatch module and sends early warning information to the corresponding client via a mobile app.

Images