A pipeline shape change monitoring system and method

Abstract

The application relates to the technical field of pipeline deformation monitoring, and discloses a pipeline deformation monitoring system and method, which comprises a multi-frequency channel acquisition device, a data processing module in a sensor system, an environment compensation module in the sensor system and a multi-level early warning and risk assessment module in the sensor system. The multi-frequency channel acquisition device is used for collecting monitoring data of a pipeline wall based on optical fiber grating sensors and micro-electro-mechanical system sensors which are arranged at equal intervals on the pipeline wall. The data processing module in the sensor system is used for alternately processing fusion and compensation of the monitoring data to obtain first fusion data. The environment compensation module in the sensor system is used for performing external environment compensation on the first fusion data to obtain second fusion data. The multi-level early warning and risk assessment module in the sensor system is used for performing multi-level early warning according to the second fusion data and determining a pipeline deformation monitoring result. The application can solve the problem that current technologies cannot meet the deformation monitoring demand for safe and efficient operation of pipelines.

Description

Technical Field

[0001] This application relates to the field of pipeline deformation monitoring technology, and more specifically, to a pipeline deformation monitoring system and method. Background Technology

[0002] Buried pipelines, as a crucial infrastructure, are widely used in oil, natural gas, water supply, and heating. However, due to their long-term exposure to complex geological environments, pipeline deformation is becoming an increasingly prominent issue. Pipeline deformation can be caused by various factors, including ground movement, soil subsidence, external loads, corrosion, and the pipeline's own aging. Failure to monitor and implement effective measures in a timely manner may lead to pipeline leaks, ruptures, and other accidents, resulting in severe economic losses and environmental hazards.

[0003] Traditional methods for monitoring the deformation of buried pipelines mainly include manual inspection and simple sensor deployment. These methods suffer from insufficient monitoring accuracy, poor real-time performance, and limited monitoring range, making it difficult to meet the requirements of modern engineering for the safe and efficient operation of pipelines. Summary of the Invention

[0004] In view of this, this application proposes a pipeline deformation monitoring system and method, aiming to solve the problem that current technology is unable to meet the deformation monitoring needs for safe and efficient pipeline operation.

[0005] In a first aspect, this application proposes a pipeline deformation monitoring system, comprising: a multi-frequency channel acquisition device for acquiring monitoring data of the pipeline wall based on fiber optic grating sensors and microelectromechanical system sensors equally spaced on the pipeline wall, and transmitting the monitoring data to a sensor system; the monitoring data includes at least one of: pipeline wall deformation data, pipeline wall temperature data, soil moisture data, and formation pressure data; a data processing module in the sensor system for performing alternating fusion and compensation processing on the monitoring data to obtain first fused data, and sending the first fused data to an environmental compensation module in the sensor system; the environmental compensation module for performing external environmental compensation on the first fused data to obtain second fused data, and sending the second fused data to a multi-level early warning and risk assessment module in the sensor system; the multi-level early warning and risk assessment module for performing multi-level early warning based on the second fused data and determining the pipeline deformation monitoring results.

[0006] Optionally, the data processing module is also used to send the first fused data into the prediction module of the sensor system; the prediction module is used to perform deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction result, and send the prediction result into the early warning module of the sensor system; the early warning module is used to determine whether to trigger an early warning based on the prediction result, and transmit the first fused data and the early warning data to the data storage module of the sensor system; the multi-level early warning and risk assessment module is also used to transmit the second fused data, multi-level early warning data and pipeline deformation monitoring results to the data storage module.

[0007] Optionally, when the monitoring data includes: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data, the data processing module performs alternating fusion and compensation processing on the monitoring data to obtain the first fused data. This includes: performing spatiotemporal fusion of the pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data to obtain fused data; performing environmental compensation and soil settlement compensation on the fused data based on the pipe wall temperature data, soil moisture data, and formation pressure data using an environmental compensation algorithm to obtain compensated fused data; and performing preprocessing and feature fusion on the compensated fused data to obtain the first fused data. The preprocessing includes noise reduction and cleaning.

[0008] Optionally, when the data processing module performs spatiotemporal fusion of pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data to obtain fused data, it includes:

[0009] F(t,x,y,z)=w1·D s (t,x,y,z)+w2·T(t,x,y,z)+w3·H(t,x,y,z)+w4·

[0010] P(t,x,y,z)+b;

[0011] Where w1 is the first weighting coefficient, w2 is the second weighting coefficient, w3 is the third weighting coefficient, w4 is the fourth weighting coefficient, and D s (t,x,y,z) represents the deformation data of the pipe wall as a function of time t and spatial location (x,y,z), T(t,x,y,z) represents the temperature data of the pipe wall as a function of time t and spatial location (x,y,z), H(t,x,y,z) represents the soil moisture data as a function of time t and spatial location (x,y,z), P(t,x,y,z) represents the formation pressure data as a function of time t and spatial location (x,y,z), b is the bias term, and F(t,x,y,z) is the fused data.

[0012] Optionally, the data processing module performs environmental compensation and soil settlement compensation on the fused data based on pipe wall temperature data, soil moisture data, and formation pressure data using an environmental compensation algorithm. The resulting compensated fused data includes:

[0013] F c (t,x,y,z)=F(t,x,y,z)-C e (T(t,x,y,z),H(t,x,y,z),P(t,x,y,z));

[0014] Among them, F c (t,x,y,z) represents the fused data after environmental compensation, C e For environmental compensation factor functions;

[0015] F cs (t,x,y,z)=F c (t,x,y,z)-C s (t,x,y,z);

[0016] Among them, F cs (t,x,y,z) represents the compensated fused data, C s Soil settlement compensation factor function.

[0017] Optionally, when the data processing module preprocesses and fuses the compensated fused data to obtain the first fused data, it includes:

[0018] F1(t,x,y,z)=Ψ(F cs (t,x,y,z));

[0019] Where F1(t,x,y,z) is the first fused data, and Ψ() is the feature fusion function.

[0020] Optionally, when the prediction module performs deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction result, it includes: using a time series prediction algorithm to predict the deformation trend of the first fused data to obtain predicted deformation data, and using a machine learning classification model to detect abnormal patterns in the first fused data; and determining the prediction result based on the predicted deformation data and abnormal patterns.

[0021] Optionally, when determining the prediction result based on the predicted deformation data and anomaly patterns, the prediction module includes:

[0022]

[0023] in, To predict deformation data, f trend This is a function for a trend prediction model;

[0024] Determining whether the predicted deformation data exceeds a preset threshold includes:

[0025]

[0026] Among them, Trigger trend To predict deformation results, Trigger trend =1 means the predicted deformation data exceeds the preset threshold, trigger. trend =0 means that the predicted deformation data has not exceeded the preset threshold, D threshold For preset thresholds;

[0027] A(t,x,y,z)=f anomaly (F1(t,x,y,z));

[0028] Where A(t,x,y,z) is the anomaly indicator for the abnormal mode; A(t,x,y,z) = 1 indicates an anomaly, and A(t,x,y,z) = 0 indicates normal operation. anomaly For machine learning model functions;

[0029] The early warning module determines whether to trigger an early warning based on the prediction results, including:

[0030]

[0031] Where Alarm is the prediction result, Alarm=1 will trigger an alarm signal, and Alarm=0 will not trigger an alarm signal.

[0032] Optionally, when the environmental compensation module performs external environmental compensation on the first fused data to obtain the second fused data, it includes: dynamically compensating the first fused data based on the external air pressure, external humidity, and external temperature using an adaptive compensation model to obtain the second fused data.

[0033] Optionally, when the multi-level early warning and risk assessment module performs multi-level early warning based on the second fused data, it includes: analyzing the second fused data based on deformation characteristics and dynamic trends to obtain deformation characteristic data and dynamic trend data; if the deformation characteristic data exceeds the normal threshold but does not exceed the safety threshold, the multi-level early warning is determined to be a minor risk warning; if the deformation characteristic data exceeds the normal threshold and the absolute value of the difference between the dynamic trend data and the critical threshold is less than a preset value, the multi-level early warning is determined to be a moderate risk warning; if the deformation characteristic data exceeds the safety threshold and the dynamic trend data exceeds the critical threshold, the multi-level early warning is determined to be a high-risk warning.

[0034] Secondly, a pipeline deformation monitoring method is provided, comprising: collecting monitoring data of the pipeline wall based on fiber optic grating sensors and microelectromechanical system sensors evenly spaced on the pipeline wall; the monitoring data includes at least one of the following: pipeline wall deformation data, pipeline wall temperature data, soil moisture data, and formation pressure data; performing alternating fusion and compensation processing on the monitoring data to obtain first fused data; performing external environmental compensation on the first fused data to obtain second fused data; and performing multi-level early warning and determining the pipeline deformation monitoring results based on the second fused data.

[0035] Thirdly, a pipeline deformation monitoring device is provided, including a memory and a processor; the memory is used to store computer execution instructions, and the processor is connected to the memory via a bus; when the pipeline deformation monitoring device is running, the processor executes the computer execution instructions stored in the memory, so that the pipeline deformation monitoring device performs the pipeline deformation monitoring method described in the second aspect.

[0036] The pipeline deformation monitoring device can be a network device or a component of a network device, such as a chip system within the network device. This chip system supports the network device in implementing the functions involved in the first aspect and any of its possible implementations, such as acquiring, determining, and transmitting the data and / or information involved in the aforementioned pipeline deformation monitoring method. The chip system includes a chip, but may also include other discrete devices or circuit structures.

[0037] Fourthly, a computer-readable storage medium is provided, comprising computer-executable instructions that, when executed on a computer, cause the computer to perform the pipeline deformation monitoring method described in the second aspect.

[0038] Fifthly, a computer program product is also provided, which includes computer instructions that, when executed on a pipeline deformation monitoring device, cause the pipeline deformation monitoring device to perform the pipeline deformation monitoring method as described in the second aspect above.

[0039] It should be noted that the aforementioned computer instructions may be stored, in whole or in part, on a computer-readable storage medium. This computer-readable storage medium may be packaged together with the processor of the pipeline deformation monitoring device, or it may be packaged separately from the processor of the pipeline deformation monitoring device; this application does not limit this.

[0040] The descriptions of the second, third, fourth, and fifth aspects of this application can be referenced to the detailed description of the first aspect.

[0041] In the embodiments of this application, the names of the aforementioned pipeline deformation monitoring devices do not limit the devices or functional modules themselves. In actual implementation, these devices or functional modules may appear under other names. For example, the receiving unit may also be called a receiving module, receiver, etc. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of the claims of this application and their equivalents.

[0042] As can be seen from the above, this application can bring the following beneficial effects:

[0043] (1) This application can acquire multi-dimensional data such as pipeline deformation, temperature, soil moisture and formation pressure in real time by jointly deploying fiber optic grating sensors and microelectromechanical systems (MEMS) sensors. Through spatiotemporal fusion of data, the blind spots and insufficient accuracy that may occur with single sensor monitoring are eliminated, and the integrity and accuracy of the data are greatly improved.

[0044] (2) This application utilizes an adaptive compensation model to dynamically compensate deformation monitoring data based on real-time monitored external air pressure, humidity and temperature data, thereby completely eliminating the interference of environmental variables on the monitoring results and ensuring the long-term stability and reliability of the data.

[0045] (3) This application designs a multi-level early warning and risk assessment mechanism. Based on the analysis of deformation characteristics and dynamic trends, the risk level can be divided into slight risk, medium risk and high risk. This avoids unnecessary early warning interference and can issue accurate alarm signals in a timely manner when the risk is high, thus ensuring the safety of pipeline operation.

[0046] (4) This application analyzes deformation trends through time series prediction algorithms and identifies abnormal patterns in the data by combining machine learning models. It can detect potential risks in advance and avoid major safety accidents such as pipeline leaks and ruptures caused by deformation.

[0047] (5) This application is designed for buried pipeline environments under different geological conditions. It can dynamically adjust the compensation factor to adapt to complex environmental changes (such as ground subsidence, seasonal humidity changes, etc.) and ensure the applicability of the system in diverse scenarios.

[0048] (6) This application adopts a high-efficiency data processing module and an automated compensation algorithm, which supports real-time monitoring, compensation, analysis and risk assessment, greatly shortens the response time from monitoring to decision-making, and provides efficient technical support for pipeline operation and management.

[0049] (7) This application uses a data storage module to store fused data, compensation data and early warning information for a long time, supporting the traceability and analysis of historical data, and providing rich data support for subsequent pipeline health management and life prediction.

[0050] (8) This application achieves a high degree of automation in data collection, analysis and early warning, reducing the need for traditional manual detection, which not only reduces labor costs, but also improves the efficiency and accuracy of monitoring.

[0051] (9) The design of this application adopts a modular structure, which can flexibly adjust the sensor deployment scheme and algorithm model according to the pipeline type and application scenario, adapt to the pipeline monitoring needs of different scales, and has good scalability and economic applicability.

[0052] (10) This application comprehensively applies multi-source data fusion, environmental compensation and risk assessment technologies, which effectively improves the safety monitoring and fault prediction capabilities of pipelines, significantly reduces the accident rate caused by pipeline deformation problems, and provides a solid technical guarantee for ensuring public safety and reducing environmental pollution. Attached Figure Description

[0053] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0054] Figure 1 A structural block diagram of a pipeline deformation monitoring system provided in this application embodiment;

[0055] Figure 2 A flowchart of a pipeline deformation monitoring method provided in this application embodiment;

[0056] Figure 3 This is a structural block diagram of a pipeline deformation monitoring device provided in an embodiment of this application. Detailed Implementation

[0057] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present application will now be described in detail with reference to the accompanying drawings and embodiments.

[0058] In some embodiments of this application, see Figure 1As shown, a pipeline deformation monitoring system includes a multi-frequency channel acquisition device 101 and a sensor system 102. The sensor system 102 includes a data processing module 1021, an environmental compensation module 1022, a multi-level early warning and risk assessment module 1023, a prediction module 1024, an early warning module 1025, and a data storage module 1026.

[0059] The multi-frequency channel acquisition device is used to acquire monitoring data of the pipe wall based on fiber optic grating sensors and microelectromechanical system sensors that are equally spaced on the pipe wall, and transmit the monitoring data to the sensor system; the monitoring data includes at least one of the following: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data.

[0060] In other words, fiber optic grating sensors and microelectromechanical system (MEMS) sensors are arranged at equal intervals on the pipe wall, and multi-frequency channel acquisition equipment is set up.

[0061] Specifically, fiber Bragg grating (FBG) sensors feature high sensitivity, strong electromagnetic interference resistance, and long-distance transmission capability. When evenly spaced and positioned at key locations on the pipe wall, they can accurately measure minute deformations, making them suitable for long-term monitoring in complex operating environments. FBG sensors reflect dynamic changes in stress and strain through minute wavelength variations, providing high-precision deformation information for subsequent data fusion. MEMS sensors are small sensors based on microfabrication technology; their sensing elements can measure various parameters, including pressure, temperature, and acceleration. Deploying them at equidistant points near the pipe wall allows for real-time acquisition of environmental data such as pipe wall temperature, soil moisture, and ground pressure, providing accurate data support for environmental compensation and trend prediction. Multi-channel acquisition equipment receives multi-source data from FBG and MEMS sensors, supporting time-division synchronous acquisition and transmission across multiple signal frequency bands. The equipment is equipped with a high-precision sampling module and a wideband decoding module, ensuring efficient integration and accurate analysis of data from different types of sensors. Furthermore, the equipment possesses anti-interference capabilities, adapting to the data acquisition needs of complex pipeline operating environments.

[0062] The multi-frequency channel acquisition device can collect data on pipe wall deformation, pipe wall temperature, soil moisture, and formation pressure, and transmit these data to the sensor system.

[0063] Specifically, fiber optic grating sensors are used to capture minute deformations in the pipe wall caused by changes in internal pressure, external loads, or geological activity, ensuring high accuracy and sensitivity in deformation monitoring.

[0064] By utilizing the temperature-sensitive properties of fiber optic grating sensors, temperature data of the pipe wall can be collected synchronously to monitor the impact of temperature changes on material properties and deformation.

[0065] By using microelectromechanical systems (MEMS) sensors to monitor the moisture content of the soil around the pipeline, the potential impact of environmental moisture on ground stability and external pipeline corrosion can be assessed.

[0066] Collecting ground pressure information reflects the stress distribution of the soil or rock strata around the pipeline, providing a basis for assessing the stress state of the pipeline.

[0067] By setting up multi-frequency channel acquisition equipment, synchronous acquisition and frequency division processing of multiple parameters can be achieved, avoiding data acquisition conflicts and improving acquisition efficiency and data quality.

[0068] The data acquisition frequency is dynamically adjusted according to the operating status (such as pipeline transport flow rate and temperature fluctuations) and environmental changes (such as rainfall and seismic activity) to adapt to the complex operating environment.

[0069] High-speed data transmission protocols (such as Ethernet and fiber optic communication) are used to transmit the collected deformation data, temperature data, humidity data and pressure data to the sensor system in real time to ensure the timeliness of the data.

[0070] By using timestamps to synchronize and calibrate different types of data, we can ensure the consistency of deformation data with environmental parameters in time and space, and provide accurate basic data for subsequent analysis.

[0071] Redundant acquisition mechanisms (such as multi-point deployment and two-way communication) can improve the reliability of data transmission and avoid data loss due to equipment failure or environmental interference.

[0072] The data processing module in the sensor system is used to perform alternating fusion and compensation processing on the monitoring data to obtain first fused data, and then send the first fused data to the environmental compensation module in the sensor system.

[0073] Optionally, the data processing module is further configured to send the first fused data into the prediction module of the sensor system.

[0074] The prediction module is used to predict deformation trends and identify abnormal patterns in the first fused data, obtain prediction results, and send the prediction results to the early warning module in the sensor system.

[0075] The early warning module is used to determine whether to trigger an early warning based on the prediction result, and transmits the first fused data and the early warning data to the data storage module in the sensor system.

[0076] In other words, the data processing module in the sensor system performs alternating fusion and compensation processing on the deformation data of the pipe wall to obtain the first fused data, which is then sent to the prediction module and the environmental compensation module.

[0077] The prediction module performs deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction result, and sends the prediction result to the early warning module. The early warning module determines whether to trigger an early warning based on the prediction result, and transmits the first fused data and the early warning data to the data storage module.

[0078] The environmental compensation module is used to perform external environmental compensation on the first fused data to obtain the second fused data, and then sends the second fused data to the multi-level early warning and risk assessment module in the sensor system; the multi-level early warning and risk assessment module is used to perform multi-level early warning and determine the pipeline deformation monitoring results based on the second fused data.

[0079] Optionally, the multi-level early warning and risk assessment module is also used to transmit the second fused data, the multi-level early warning data, and the pipeline deformation monitoring results to the data storage module.

[0080] In other words, the environmental compensation module performs external environmental compensation on the first fused data to obtain the second fused data, sends the second fused data to the multi-level early warning and risk assessment module for multi-level early warning and risk assessment report generation, and transmits the second fused data, multi-level early warning data and risk assessment report to the data storage module.

[0081] Optionally, when the monitoring data includes: the deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, the data processing module performs alternating fusion and compensation processing on the monitoring data to obtain the first fused data, including:

[0082] The deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data are spatiotemporally fused to obtain fused data. The fused data is then subjected to environmental compensation and soil settlement compensation based on the temperature data of the pipe wall, the soil moisture data, and the formation pressure data using an environmental compensation algorithm to obtain compensated fused data. The compensated fused data is then preprocessed and feature fusion is performed to obtain the first fused data. The preprocessing includes noise reduction and cleaning.

[0083] In other words, the data processing module in the sensor system performs alternating fusion and compensation processing on the deformation data of the pipe wall to obtain the first fused data. The first fused data is then sent to the prediction module and the environmental compensation module. Specifically, the data processing module performs spatiotemporal fusion on the deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data to obtain fused data. Based on the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, the fused data is subjected to environmental compensation using an environmental compensation algorithm, followed by soil settlement compensation to obtain the compensated fused data. The compensated fused data is then preprocessed and feature fusion is performed to obtain the first fused data.

[0084] Specifically, for pipeline wall deformation data and environmental data, short-term dynamic features and long-term trend features are extracted respectively, and multi-scale analysis algorithms (such as wavelet transform or empirical mode decomposition) are used to decompose the change patterns at different time scales.

[0085] By combining convolutional neural networks (CNNs) with temporal models (such as LSTM or GRU), spatiotemporal features of data can be jointly modeled, thereby enhancing the correlation of data in the spatiotemporal dimension.

[0086] By learning the model, dynamic weights are assigned to temperature, humidity, formation pressure, and deformation data, enabling the fusion results to reflect the comprehensive impact of environmental factors on deformation in real time. For example, during the rainy season, the weight of soil moisture is automatically increased, while in winter, the weight of temperature data is higher.

[0087] To address potential bias errors in multi-sensor data fusion, an iterative correction algorithm is used to reduce deviations caused by equipment errors or external interference, thereby improving the accuracy of the fused data.

[0088] To address the nonlinear effects of temperature, humidity, and formation pressure, a compensation network based on a multilayer perceptron (MLP) is constructed. This network can learn complex nonlinear mapping relationships, thereby better correcting deformation data.

[0089] By incorporating historical environmental monitoring data from multiple years as training samples for the compensation model, the generalization ability of the compensation algorithm to different environmental conditions is improved.

[0090] The compensation model dynamically adjusts its parameters during operation. For example, in extreme weather conditions (such as heavy rain or extreme cold), the system will automatically correct the sensitivity of the compensation function to ensure data accuracy.

[0091] By utilizing ground pressure, soil moisture, and deformation trends, a multivariate coupled model of soil settlement is constructed to achieve precise dynamic compensation for soil settlement.

[0092] Based on the settlement data from monitoring points, combined with geological characteristics and settlement patterns, the settlement impact in areas without sensor deployment is predicted using a regional interpolation algorithm, further improving the comprehensiveness of compensation data.

[0093] Correcting long-term cumulative settlement displacement separates the long-term settlement trend of the strata from short-term dynamic fluctuations, reducing the impact of accumulated errors on pipeline monitoring results.

[0094] Statistical features (mean, standard deviation, extreme values), frequency domain features (dominant frequency, spectral energy), and time series features (trend slope, rate of change) of environmental and deformation data are extracted to enrich the feature dimensions of the first fused data.

[0095] By using kernel function methods (such as Gaussian kernel or polynomial kernel), multidimensional features are mapped to a high-dimensional space for fusion, capturing more complex interaction relationships and forming high-quality first fused data.

[0096] It should be noted that by employing spatiotemporal fusion and multivariate compensation, the system comprehensively considers the impact of various factors such as temperature, humidity, and formation pressure on deformation data, significantly improving data accuracy and consistency. The introduction of multi-layer denoising mechanisms and bias correction algorithms effectively suppresses random noise and systematic errors in the data. Adaptive weighting and real-time parameter update mechanisms enable the system to maintain accurate perception and response to deformation even when the environment undergoes rapid changes (such as drastic temperature differences or sudden increases in soil moisture). The dynamic feature analysis model, driven by historical data, allows the system to address both short-term sudden risks and provide long-term trend predictions and assessments. Data fusion and feature enhancement enable the system to more accurately identify risk points, and combined with regional subsidence prediction capabilities, it can assess potential risks in unmonitored areas. The system not only identifies the current risk status but also analyzes the possible causes and influencing factors of risks through fused data, providing strong support for decision-making. The generation of high-dimensional fused data optimizes the data storage structure and provides high-quality input for the prediction, assessment, and early warning modules. The fused data carries rich multidimensional information, making subsequent risk assessment reports more scientific and persuasive. Faced with complex external disturbances (such as earthquakes and equipment errors), the data fusion and compensation mechanism makes the system's deformation sensing capability more stable and reliable. When new sensors or compensation factors are added, the system can be quickly integrated and operate efficiently, facilitating subsequent upgrades.

[0097] Optionally, the compensated fused data may be preprocessed, specifically by denoising and cleaning the compensated fused data.

[0098] Specifically, the data curves are smoothed using a moving average method and a low-pass filter to remove the influence of high-frequency noise. A Fast Fourier Transform (FFT) is used to transform the data to the frequency domain, filtering out high-frequency or low-frequency noise components in the spectrum. Then, the smoothed signal is reconstructed through an inverse transform and introduced into a Kalman filter for real-time denoising of dynamic data. A prediction-correction mechanism adaptively adjusts the noise model to obtain the optimal estimate. Outliers are identified using the mean and standard deviation of the data, and extreme values ​​deviating from the mean by more than three times the standard deviation are removed. Data distribution is analyzed based on quartiles, and outliers outside the upper and lower limits are marked and corrected. Wavelet decomposition is performed on the data to separate signal and noise components, retaining coefficients that contribute significantly to the signal for reconstruction, thereby removing noise. The data is decomposed into multiple intrinsic mode functions (IMFs), retaining only the main components to filter out noise. Data timestamps are verified, and time-disrupted data caused by sensor synchronization errors are removed or repaired. For multi-dimensional data such as pipe wall temperature, soil moisture, and formation pressure, their physical correlations are detected, and logically contradictory abnormal records are removed. Short-term missing data are imputed using linear or Lagrange interpolation. Missing values ​​are predicted and imputed using multivariate regression models to ensure data integrity. Data segments with missing rates exceeding a certain threshold are directly removed. For redundant data collected repeatedly from multiple sensors, approximate values ​​are merged using similarity analysis (such as Euclidean distance or cosine similarity) to retain the most representative sample points. Data is scaled to the range [0,1] to eliminate the impact of dimensional differences on subsequent analysis. Z-score standardization is used to transform the data to zero mean and unit variance to improve the consistency of data distribution.

[0099] It should be noted that multi-level filtering and nonlinear denoising methods significantly reduce the impact of environmental noise and sensor errors on the data, making the final processed data clearer and more accurate. Statistical anomaly detection and cleaning algorithms eliminate physically unreasonable and logically contradictory data, improving the data's authenticity and reliability. Interpolation and predictive completion techniques ensure the continuity of data over long periods, providing complete input for subsequent analysis. Data consistency verification eliminates potential inconsistencies between dimensions, further enhancing data integrity and usability. The adoption of automated cleaning rules significantly improves the processing efficiency of large-scale, multi-source data, reducing manual intervention. Merging duplicate data reduces the waste of storage and computing resources, improving system operating efficiency. Data standardization makes data from different dimensions comparable, providing optimized input for subsequent machine learning models and trend analysis algorithms. Adaptive filters and other technologies make the processing methods highly adaptable to complex environmental changes, helping to improve the robustness of the pipeline monitoring system. High-quality data provides a reliable foundation for risk warning, trend analysis, and decision optimization, significantly improving the effectiveness of the monitoring system.

[0100] Optionally, when the data processing module performs spatiotemporal fusion of the pipe wall deformation data, the pipe wall temperature data, the soil moisture data, and the formation pressure data to obtain fused data, it includes:

[0101] F(t,x,y,z)=w1·D s (t,x,y,z)+w2·T(t,x,y,z)+w3·H(t,x,y,z)+w4·

[0102] P(t,x,y,z)+b;

[0103] Where w1 is the first weighting coefficient, w2 is the second weighting coefficient, w3 is the third weighting coefficient, w4 is the fourth weighting coefficient, and D s (t,x,y,z) represents the deformation data of the pipe wall as a function of time t and spatial location (x,y,z), T(t,x,y,z) represents the temperature data of the pipe wall as a function of time t and spatial location (x,y,z), H(t,x,y,z) represents the soil moisture data as a function of time t and spatial location (x,y,z), P(t,x,y,z) represents the formation pressure data as a function of time t and spatial location (x,y,z), b is the bias term, and F(t,x,y,z) is the fused data.

[0104] Optionally, when the data processing module performs environmental compensation and soil settlement compensation on the fused data based on the pipe wall temperature data, soil moisture data, and formation pressure data using an environmental compensation algorithm to obtain the compensated fused data, it includes:

[0105] F c (t,x,y,z)=F(t,x,y,z)-C e (T(t,x,y,z),H(t,x,y,z),P(t,x,y,z));

[0106] Among them, F c (t,x,y,z) represents the fused data after environmental compensation, C e For environmental compensation factor functions;

[0107] F cs (t,x,y,z)=F c (t,x,y,z)-C s (t,x,y,z);

[0108] Among them, F cs (t,x,y,z) represents the compensated fused data, C s Soil settlement compensation factor function.

[0109] Optionally, when the data processing module preprocesses and fuses the compensated fused data to obtain the first fused data, it includes:

[0110] F1(t,x,y,z)=Ψ(F cs (t,x,y,z));

[0111] Where F1(t,x,y,z) is the first fused data, and Ψ() is the feature fusion function.

[0112] Optionally, when performing deformation trend prediction and abnormal pattern recognition on the first fused data to obtain a prediction result, the method includes: using a time series prediction algorithm to predict the deformation trend of the first fused data to obtain predicted deformation data, and using a machine learning classification model to detect abnormal patterns in the first fused data; and determining the prediction result based on the predicted deformation data and the abnormal patterns.

[0113] In other words, the prediction module performs deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction result, and sends the prediction result to the early warning module. The early warning module determines whether to trigger an early warning based on the prediction result. Specifically, the prediction module uses a time series prediction algorithm to predict the deformation trend of the first fused data, uses a machine learning classification model to detect abnormal patterns in the first fused data, compares the deformation trend with a preset threshold, and then combines it with the abnormal patterns to form a prediction result. The early warning module determines whether to issue an early warning based on the comparison between the prediction result and the preset threshold.

[0114] It should be noted that combining traditional time series forecasting with modern machine learning methods allows for a comprehensive consideration of the temporal trends and potential anomalies in pipeline deformation data, thus providing more accurate deformation predictions. Machine learning classification models can promptly identify early anomaly patterns in the data, helping the prediction system to detect potential pipeline risks in advance and prevent accidents. By comprehensively judging deformation trends and anomaly patterns, the early warning module can accurately determine the severity of pipeline risks and issue timely warning signals, allowing for necessary preventative measures to be taken in advance. The system's automated early warning mechanism reduces the need for manual monitoring, ensuring immediate response in emergencies and improving monitoring efficiency and safety. By integrating data such as pipeline wall deformation, temperature, humidity, and pressure, the early warning module can make a more comprehensive and accurate risk assessment, helping decision-makers develop more reasonable emergency response plans. Real-time early warning information helps maintenance personnel quickly locate potential risk areas, make timely adjustments, and optimize resource allocation. The system can issue timely warnings when pipeline deformation or abnormal changes in environmental conditions occur, allowing for early measures such as pipeline maintenance and reinforcement to prevent major accidents such as pipeline ruptures or leaks. By continuously monitoring and providing early warnings of pipeline operating status, we can ensure the safe operation of pipelines, extend their service life, and reduce maintenance costs.

[0115] Optionally, when the prediction module determines the prediction result based on the predicted deformation data and the anomaly pattern, it includes:

[0116] The prediction module uses a time series prediction algorithm to predict the deformation trend of the first fused data, and a machine learning classification model to detect abnormal patterns in the first fused data. The deformation trend is compared with a preset threshold, and then combined with the abnormal patterns to form the prediction result. The early warning module determines whether to issue an early warning based on the comparison between the prediction result and the preset threshold. The specific details are as follows:

[0117]

[0118] in, To predict deformation data, f trend This is a function for a trend prediction model;

[0119] Determining whether the predicted deformation data exceeds a preset threshold includes:

[0120]

[0121] Among them, Trigger trend To predict deformation results, Trigger trend =1 means that the predicted deformation data exceeds the preset threshold, Trigger trend =0 means that the predicted deformation data does not exceed the preset threshold, D thresholdFor preset thresholds;

[0122] A(t,x,y,z)=f anomaly (F1(t,x,y,z));

[0123] Where A(t,x,y,z) is the anomaly identifier of the abnormal pattern, A(t,x,y,z) = 1 indicates an anomaly, and A(t,x,y,z) = 0 indicates normality, f anomaly For machine learning model functions;

[0124] The early warning module determines whether to trigger an early warning based on the prediction result, including:

[0125]

[0126] Where Alarm is the prediction result, Alarm=1 will trigger an alarm signal, and Alarm=0 will not trigger an alarm signal.

[0127] Optionally, the environmental compensation module performs external environmental compensation on the first fused data to obtain the second fused data, which includes: dynamically compensating the first fused data based on external air pressure, external humidity and external temperature using an adaptive compensation model to obtain the second fused data.

[0128] Specifically, to improve the accuracy and reliability of the first fused data, especially considering the impact of environmental factors (such as air pressure, humidity, and temperature) on pipeline monitoring data, an adaptive compensation model is used to dynamically compensate the first fused data, resulting in more accurate second fused data. The adaptive compensation model is based on external environmental factors (air pressure, humidity, and temperature) and historical monitoring data, and dynamically corrects the first fused data by continuously adjusting compensation parameters. The compensation process includes the following steps:

[0129] The system collects real-time data on external air pressure, humidity, and temperature. This data is provided by environmental sensors working in conjunction with the sensor system and uploaded to the data processing platform in real time. Compensation coefficients are calculated using a model of the relationship between air pressure and pipe wall deformation, typically obtained through a mechanical model of the pipe or regression calculations based on historical data. Humidity changes affect the soil and the environment surrounding the pipe, thus requiring the establishment of a model to assess the impact of pipe deformation. Statistical analysis is used to obtain the relationship between humidity changes and pipe deformation, thereby calculating compensation coefficients. Temperature changes are closely related to the expansion and contraction characteristics of pipe materials. Using the material's coefficient of thermal expansion and the rate of temperature change, the impact of temperature on pipe deformation data is calculated, and compensation coefficients are generated. An adaptive algorithm automatically adjusts the compensation coefficients based on real-time environmental data. This process, based on a feedback mechanism, dynamically optimizes the model according to the deviation between the actual data after each compensation and the true state, adapting to environmental changes. The environmental compensation coefficients are combined with the first fused data, and linear regression is used to adjust errors in the data to obtain the second fused data. A regression analysis model, based on historical data, is used to derive regression coefficients for the impact of factors such as temperature, humidity, and air pressure on pipe data, thereby correcting the first fused data.

[0130] With dynamic compensation, the second fused data can more accurately reflect the true state of the pipeline, eliminate interference from environmental factors, and make the monitoring data more stable and reliable. The adaptive compensation model can adjust the compensation strategy in real time according to changes in the external environment, avoiding parameter failure problems in traditional static compensation models and ensuring high accuracy of long-term monitoring data. Compared with the uncompensated first fused data, the second fused data can effectively correct errors caused by environmental factors, improving the accuracy and consistency of the data.

[0131] It should be noted that the adaptive compensation model can dynamically correct sensor data based on real-time changes in external air pressure, humidity, and temperature, eliminating interference from environmental factors on pipeline monitoring data and thus improving data accuracy. It can automatically adapt to different environmental conditions and adjust compensation strategies in real time, avoiding the error accumulation caused by environmental changes in static compensation methods. Through dynamic compensation, the system can cope with complex and ever-changing external environments, avoiding the impact of external climate factors on long-term monitoring. The automatic adjustment and real-time compensation mechanism ensures stable operation under different weather and climate conditions, guaranteeing data continuity and consistency. The adaptive compensation model can automatically adjust compensation parameters according to different climatic environments, making the system more adaptable to various external changes and enabling more intelligent real-time monitoring. It can not only address short-term sudden climate changes but also respond to long-term climate trend changes. More accurate second-fusion data provides reliable input for the risk assessment model, helping to more accurately predict pipeline deformation and potential failure points. The system can provide more accurate pipeline health assessments through corrected data, helping decision-makers formulate more scientific pipeline maintenance and management strategies. Dynamic compensation avoids misdiagnosis caused by environmental influences, promptly identifies potential problems, and thus prevents over-maintenance and incorrect repairs, reducing pipeline maintenance costs. The system can accurately predict faults without interference from external environmental factors, thereby extending the service life of pipeline equipment.

[0132] Optionally, when the multi-level early warning and risk assessment module performs multi-level early warning based on the second fused data, it includes:

[0133] The second fused data is analyzed based on deformation characteristics and dynamic trends to obtain deformation characteristic data and dynamic trend data. If the deformation characteristic data exceeds the normal threshold but does not exceed the safety threshold, the multi-level warning is determined to be a minor risk warning. If the deformation characteristic data exceeds the normal threshold and the absolute value of the difference between the dynamic trend data and the critical threshold is less than a preset value, the multi-level warning is determined to be a moderate risk warning. If the deformation characteristic data exceeds the safety threshold and the dynamic trend data exceeds the critical threshold, the multi-level warning is determined to be a high-risk warning.

[0134] In other words, the second fused data is sent to the multi-level early warning and risk assessment module for multi-level early warning and risk assessment report generation. The specific content of the multi-level early warning and risk assessment module is as follows: The multi-level early warning and risk assessment module analyzes the second fused data based on deformation characteristics and dynamic trends. If the deformation characteristic data exceeds the normal threshold but does not exceed the safety threshold, it is judged as a slight risk. If the deformation characteristic data exceeds the normal threshold and the dynamic trend tends to the critical threshold, it is judged as a moderate risk. If the deformation characteristic data exceeds the safety threshold and the dynamic trend exceeds the critical threshold, it is judged as a high risk. The slight risk, moderate risk and high risk are displayed and an assessment report is generated.

[0135] Specifically, the normal threshold is: pipe deformation within this range is considered normal, the changes are within the pipe's operating range, and will not cause harm.

[0136] Safety threshold: When the deformation of the pipeline exceeds this threshold, there may be a certain degree of damage or failure risk, but it will not necessarily lead to pipeline rupture immediately. It is still possible to avoid greater problems through maintenance and repair.

[0137] Critical threshold: When the deformation of the pipeline approaches or reaches this threshold, the deformation trend shows signs of severe acceleration. If it continues to develop, it may lead to pipeline failure or even catastrophic failure.

[0138] The assessment report includes the following components: Risk Level: The report clearly indicates the current risk level of the pipeline, specifying whether it is minor, moderate, or high risk. Deformation Data Analysis: Provides trend analysis of deformation data, including current and historical deformation, and gives specific values ​​exceeding thresholds. Dynamic Trend Analysis: Includes deformation acceleration, rate of change, and possible future trend predictions, providing decision-makers with a clear understanding of future risk trends. Compensation and Correction Data: If a compensation model has been applied, the report will display the compensated data and correction results. Recommendations and Measures: Based on the assessment results, the report will provide recommended measures for different risk levels, such as whether increased monitoring is needed, whether immediate repair or suspension of use is required, etc.

[0139] It should be noted that the environmental compensation in the first fused data is a basic environmental compensation that corrects for the direct impact of known common environmental factors (such as temperature, humidity, soil pressure, etc.) on deformation data. Typically, compensation based on fixed algorithms or rules is used in the early stages of data acquisition to eliminate or reduce the static influence of the environment on the data, making the data more reliable.

[0140] The environmental compensation in the second fusion of data is more dynamic and adaptive. After the data has undergone basic compensation, it incorporates real-time changes in the external environment (such as seasonal changes, sudden drastic changes in soil moisture, changes in pipeline operating status, etc.) and further refines the data through an adaptive compensation model. This step, through more flexible model adjustment, adapts to dynamic environmental changes to ensure the long-term accuracy of the data.

[0141] Improving accuracy: Basic compensation is effective in static environments, but it cannot fully cover all influencing factors under complex and constantly changing conditions. Adaptive compensation can overcome this deficiency, ensuring that data remains accurate even in volatile environments.

[0142] Enhanced system robustness: Dual compensation makes the system more adaptable to the environment. Whether it is a short-term drastic change or a long-term small change, the data can be properly corrected to avoid the monitoring accuracy from being lost due to sudden environmental changes.

[0143] To meet the correction needs at different stages: In the early stages of data acquisition, basic compensation removes significant environmental interference, while in the later stages of data processing, adaptive compensation fine-tunes the data in response to dynamic changes in the environment, resulting in multi-level data refinement.

[0144] It should be noted that the pipeline deformation monitoring system and method provided in this application embodiment can be applied to the deformation monitoring of buried pipelines, as well as to the deformation monitoring of other types of pipelines. This application embodiment does not limit this.

[0145] In some embodiments, such as Figure 2 As shown in the embodiments of this application, a pipeline deformation monitoring method is also provided, applied to the above-mentioned... Figure 1 The pipeline deformation monitoring system includes:

[0146] S201. Monitoring data of the pipe wall is collected based on fiber optic grating sensors and microelectromechanical system sensors that are equally spaced on the pipe wall.

[0147] The monitoring data includes at least one of the following: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data;

[0148] S202. The monitoring data is subjected to alternating fusion and compensation processing to obtain the first fused data.

[0149] S203. Perform external environment compensation on the first fused data to obtain the second fused data.

[0150] S204. Perform multi-level early warning and determine pipeline deformation monitoring results based on the second fused data.

[0151] Optionally, the above-mentioned pipeline deformation monitoring methods also include:

[0152] S205. Perform deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction results.

[0153] S206. Determine whether to trigger an early warning based on the prediction result, and store the first fused data and the early warning data.

[0154] S207. Store the second fused data, multi-level early warning data, and the pipeline deformation monitoring results.

[0155] Optionally, when the monitoring data includes: the deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, the above-mentioned fusion and compensation alternating processing of the monitoring data yields first fused data, including:

[0156] The deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data are spatiotemporally fused to obtain fused data;

[0157] The fused data is subjected to environmental compensation and soil settlement compensation based on the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, using an environmental compensation algorithm to obtain the compensated fused data;

[0158] The compensated fused data is preprocessed and feature fused to obtain the first fused data; the preprocessing includes noise reduction and cleaning.

[0159] Optionally, the above-mentioned spatiotemporal fusion of the pipe wall deformation data, the pipe wall temperature data, the soil moisture data, and the formation pressure data to obtain fused data includes:

[0160] F(t,x,y,z)=w1·D s (t,x,y,z)+w2·T(t,x,y,z)+w3·H(t,x,y,z)+w4·

[0161] P(t,x,y,z)+b;

[0162] Where w1 is the first weighting coefficient, w2 is the second weighting coefficient, w3 is the third weighting coefficient, w4 is the fourth weighting coefficient, and D d(t,x,y,z) represents the deformation data of the pipe wall as a function of time t and spatial location (x,y,z), T(t,x,y,z) represents the temperature data of the pipe wall as a function of time t and spatial location (x,y,z), H(t,x,y,z) represents the soil moisture data as a function of time t and spatial location (x,y,z), P(t,x,y,z) represents the formation pressure data as a function of time t and spatial location (x,y,z), b is the bias term, and F(t,x,y,z) is the fused data.

[0163] Optionally, the aforementioned fused data, based on the pipe wall temperature data, soil moisture data, and formation pressure data, undergoes environmental compensation and soil settlement compensation using an environmental compensation algorithm to obtain compensated fused data, including:

[0164] F c (t,x,y,z)=F(t,x,y,z)-C e (T(t,x,y,z),H(t,x,y,z),P(t,x,y,z));

[0165] Among them, F c (t,x,y,z) represents the fused data after environmental compensation, C e For environmental compensation factor functions;

[0166] F cs (t,x,y,z)=F c (t,x,y,z)-C s (t,x,y,z);

[0167] Among them, F cs (t,x,y,z) represents the compensated fused data, C s Soil settlement compensation factor function.

[0168] Optionally, the above-mentioned preprocessing and feature fusion of the compensated fused data to obtain the first fused data includes:

[0169] F1(t,x,y,z)=Ψ(F cs (t,x,y,z));

[0170] Where F1(t,x,y,z) is the first fused data, and Ψ() is the feature fusion function.

[0171] Optionally, the above-mentioned deformation trend prediction and abnormal pattern recognition of the first fused data to obtain the prediction result includes:

[0172] The deformation trend is predicted by using a time series prediction algorithm on the first fused data to obtain predicted deformation data, and an abnormal pattern is detected by using a machine learning classification model on the first fused data.

[0173] The prediction result is determined based on the predicted deformation data and the anomaly pattern.

[0174] Optionally, determining the prediction result based on the predicted deformation data and the anomaly pattern includes:

[0175]

[0176] in, To predict deformation data, f trend This is a function for a trend prediction model;

[0177] The above determination of whether the predicted deformation data exceeds a preset threshold includes:

[0178]

[0179] Among them, Trigger trend To predict deformation results, Trigger trend =1 means that the predicted deformation data exceeds the preset threshold, Trigger trend =0 means that the predicted deformation data does not exceed the preset threshold, D threshold The preset threshold;

[0180] A(t,x,y,z)=f anomaly (F1(t,x,y,z));

[0181] Where A(t,x,y,z) is the anomaly identifier of the abnormal pattern, A(t,x,y,z) = 1 indicates an anomaly, and A(t,x,y,z) = 0 indicates normality, f anomaly For machine learning model functions;

[0182] The above-mentioned determination of whether to trigger an early warning based on the prediction results includes:

[0183]

[0184] Where Alarm is the prediction result, Alarm=1 will trigger an alarm signal, and Alarm=0 will not trigger an alarm signal.

[0185] Optionally, the above-mentioned external environment compensation of the first fused data to obtain the second fused data includes:

[0186] The first fused data is dynamically compensated based on external air pressure, external humidity, and external temperature using an adaptive compensation model to obtain the second fused data.

[0187] Optionally, the above-mentioned multi-level early warning based on the second fused data includes:

[0188] The second fused data is analyzed based on deformation characteristics and dynamic trends to obtain deformation characteristic data and dynamic trend data.

[0189] If the deformation feature data exceeds the normal threshold but does not exceed the safety threshold, the multi-level warning is determined to be a minor risk warning.

[0190] If the deformation feature data exceeds the normal threshold and the absolute value of the difference between the dynamic trend data and the critical threshold is less than a preset value, then the multi-level warning is determined to be a medium-risk warning.

[0191] If the deformation feature data exceeds the safety threshold and the dynamic trend data exceeds the critical threshold, then it is determined to be a high-risk warning under the multi-level early warning system.

[0192] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0193] This application embodiment can divide the pipeline deformation monitoring device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0194] like Figure 3 The diagram shown is a structural schematic of a pipeline deformation monitoring device provided in an embodiment of this application. Figure 3 The pipeline deformation monitoring device shown includes: a communication unit 301 and a processing unit 302;

[0195] The communication unit 301 is used to collect monitoring data of the pipe wall based on fiber optic grating sensors and microelectromechanical system sensors that are equally spaced on the pipe wall.

[0196] The monitoring data includes at least one of the following: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data;

[0197] The processing unit 302 is used to perform alternating fusion and compensation processing on the monitoring data to obtain the first fused data.

[0198] The processing unit 302 is also used to perform external environment compensation on the first fused data to obtain the second fused data.

[0199] The processing unit 302 is also used to perform multi-level early warning and determine pipeline deformation monitoring results based on the second fused data.

[0200] Optionally, the processing unit 302 is also used to perform deformation trend prediction and abnormal pattern recognition on the first fused data to obtain prediction results.

[0201] The processing unit 302 is also used to determine whether to trigger an early warning based on the prediction result, and to store the first fused data and the early warning data.

[0202] Processing unit 302 is also used to store the second fused data, multi-level early warning data, and the pipeline deformation monitoring results.

[0203] Optionally, when the monitoring data includes: the deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, the processing unit 302 is specifically used for:

[0204] The deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data are spatiotemporally fused to obtain fused data;

[0205] The fused data is subjected to environmental compensation and soil settlement compensation based on the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, using an environmental compensation algorithm to obtain the compensated fused data;

[0206] The compensated fused data is preprocessed and feature fused to obtain the first fused data; the preprocessing includes noise reduction and cleaning.

[0207] Optionally, the processing unit 302 is specifically used for:

[0208] F(t,x,y,z)=w1·D s(t,x,y,z)+w2·T(t,x,y,z)+w3·H(t,x,y,z)+w4·

[0209] P(t,x,y,z)+b;

[0210] Where w1 is the first weighting coefficient, w2 is the second weighting coefficient, w3 is the third weighting coefficient, w4 is the fourth weighting coefficient, and D s (t,x,y,z) represents the deformation data of the pipe wall as a function of time t and spatial location (x,y,z), T(t,x,y,z) represents the temperature data of the pipe wall as a function of time t and spatial location (x,y,z), H(t,x,y,z) represents the soil moisture data as a function of time t and spatial location (x,y,z), P(t,x,y,z) represents the formation pressure data as a function of time t and spatial location (x,y,z), b is the bias term, and F(t,x,y,z) is the fused data.

[0211] Optionally, the processing unit 302 is specifically used for:

[0212] F c (t,x,y,z)=F(t,x,y,z)-C e (T(t,x,y,z),H(t,x,y,z),P(t,x,y,z));

[0213] Among them, F c (t,x,y,z) represents the fused data after environmental compensation, C e For environmental compensation factor functions;

[0214] F cs (t,x,y,z)=F c (t,x,y,z)-C s (t,x,y,z);

[0215] Among them, F cs (t,x,y,z) represents the compensated fused data, C s Soil settlement compensation factor function.

[0216] Optionally, the processing unit 302 is specifically used for:

[0217] F1(t,x,y,z)=Ψ(F cs (t,x,y,z));

[0218] Where F1(t,x,y,z) is the first fused data, and Ψ() is the feature fusion function.

[0219] Optionally, the processing unit 302 is specifically used for:

[0220] The deformation trend is predicted by using a time series prediction algorithm on the first fused data to obtain predicted deformation data, and an abnormal pattern is detected by using a machine learning classification model on the first fused data.

[0221] The prediction result is determined based on the predicted deformation data and the anomaly pattern.

[0222] Optionally, the processing unit 302 is specifically used for:

[0223]

[0224] in, To predict deformation data, f trend This is a function for a trend prediction model;

[0225]

[0226] Among them, Trigger trend To predict deformation results, Trigger trend =1 means that the predicted deformation data exceeds the preset threshold, Trigger trend =0 means that the predicted deformation data does not exceed the preset threshold, D threshold The preset threshold;

[0227] A(t,x,y,z)=f anomaly (F1(t,x,y,z));

[0228] Where A(t,x,y,z) is the anomaly identifier of the abnormal pattern, A(t,x,y,z) = 1 indicates an anomaly, and A(t,x,y,z) = 0 indicates normality, f anomaly For machine learning model functions;

[0229] The above-mentioned determination of whether to trigger an early warning based on the prediction results includes:

[0230]

[0231] Where Alarm is the prediction result, Alarm=1 will trigger an alarm signal, and Alarm=0 will not trigger an alarm signal.

[0232] Optionally, the processing unit 302 is specifically used for:

[0233] The first fused data is dynamically compensated based on external air pressure, external humidity, and external temperature using an adaptive compensation model to obtain the second fused data.

[0234] Optionally, the processing unit 302 is specifically used for:

[0235] The second fused data is analyzed based on deformation characteristics and dynamic trends to obtain deformation characteristic data and dynamic trend data.

[0236] If the deformation feature data exceeds the normal threshold but does not exceed the safety threshold, the multi-level warning is determined to be a minor risk warning.

[0237] If the deformation feature data exceeds the normal threshold and the absolute value of the difference between the dynamic trend data and the critical threshold is less than a preset value, then the multi-level warning is determined to be a medium-risk warning.

[0238] If the deformation feature data exceeds the safety threshold and the dynamic trend data exceeds the critical threshold, then it is determined to be a high-risk warning under the multi-level early warning system.

[0239] This application also provides a computer-readable storage medium, which includes computer-executable instructions. When the computer-executable instructions are run on a computer, the computer performs the pipeline deformation monitoring method provided in the above embodiments.

[0240] This application also provides a computer program product that can be directly loaded into a memory and contains software code. After being loaded and executed by a computer, the computer program product can implement the pipeline deformation monitoring method provided in the above embodiments. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

[0241] The system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0242] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. The programs corresponding to the software modules and method steps can be placed in random access memory (RAM), main memory, read-only memory (ROM), electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the invention.

Claims

1. A pipeline deformation monitoring system, characterized in that, include: A multi-frequency channel acquisition device is used to acquire monitoring data of the pipe wall based on fiber optic grating sensors and microelectromechanical system sensors that are equally spaced on the pipe wall, and to transmit the monitoring data to the sensor system. The monitoring data includes at least one of the following: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data; The data processing module in the sensor system is used to perform alternating fusion and compensation processing on the monitoring data to obtain first fused data, and then send the first fused data to the environmental compensation module in the sensor system. The environmental compensation module is used to perform external environmental compensation on the first fused data to obtain the second fused data, and then send the second fused data to the multi-level early warning and risk assessment module in the sensor system. The multi-level early warning and risk assessment module is used to perform multi-level early warning and determine pipeline deformation monitoring results based on the second fused data.

2. The system according to claim 1, characterized in that, The data processing module is also used to send the first fused data into the prediction module of the sensor system; The prediction module is used to predict deformation trends and identify abnormal patterns in the first fused data, obtain prediction results, and send the prediction results to the early warning module in the sensor system. The early warning module is used to determine whether to trigger an early warning based on the prediction result, and transmit the first fused data and the early warning data to the data storage module of the sensor system; The multi-level early warning and risk assessment module is also used to transmit the second fused data, the multi-level early warning data, and the pipeline deformation monitoring results to the data storage module.

3. The system according to claim 1, characterized in that, When the monitoring data includes: the deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, the data processing module performs alternating fusion and compensation processing on the monitoring data to obtain the first fused data, including: The deformation data of the pipe wall, the temperature data of the pipe wall, the soil moisture data, and the formation pressure data are spatiotemporally fused to obtain fused data; The fused data is subjected to environmental compensation and soil settlement compensation based on the temperature data of the pipe wall, the soil moisture data, and the formation pressure data, using an environmental compensation algorithm to obtain the compensated fused data; The compensated fused data is preprocessed and feature fused to obtain the first fused data; the preprocessing includes noise reduction and cleaning.

4. The system according to claim 3, characterized in that, When the data processing module performs spatiotemporal fusion of the pipe wall deformation data, the pipe wall temperature data, the soil moisture data, and the formation pressure data to obtain fused data, it includes: F(t,x,y,z)=w1·D s (t,x,y,z)+w2·T(t,x,y,z)+w3·H(t,x,y,z)+w4· P(t,x,y,z)+b; Where w1 is the first weighting coefficient, w2 is the second weighting coefficient, w3 is the third weighting coefficient, w4 is the fourth weighting coefficient, and D s (t,x,y,z) represents the deformation data of the pipe wall as a function of time t and spatial location (x,y,z), T(t,x,y,z) represents the temperature data of the pipe wall as a function of time t and spatial location (x,y,z), H(t,x,y,z) represents the soil moisture data as a function of time t and spatial location (x,y,z), P(t,x,y,z) represents the formation pressure data as a function of time t and spatial location (x,y,z), b is the bias term, and F(t,x,y,z) is the fused data.

5. The system according to claim 4, characterized in that, The data processing module performs environmental compensation and soil settlement compensation on the fused data based on the pipe wall temperature data, soil moisture data, and formation pressure data using an environmental compensation algorithm to obtain the compensated fused data, including: F c (t,x,y,z)=F(t,x,y,z)-C e (T(t,x,y,z),H(t,x,y,z),P(t,x,y,z)); Among them, F c (t,x,y,z) represents the fused data after environmental compensation, C e For environmental compensation factor functions; F cs (t,x,y,z)=F c (t,x,y,z)-C s (t,x,y,z); Among them, F cs (t,x,y,z) represents the compensated fused data, C s Soil settlement compensation factor function.

6. The system according to claim 5, characterized in that, When the data processing module preprocesses and fuses features of the compensated fused data to obtain the first fused data, it includes: F1(t,x,y,z)=Ψ(F cs (t,x,y,z)); Where F1(t,x,y,z) is the first fused data, and Ψ is the feature fusion function.

7. The system according to claim 2, characterized in that, The prediction module performs deformation trend prediction and abnormal pattern recognition on the first fused data to obtain the prediction result, including: The deformation trend is predicted by using a time series prediction algorithm on the first fused data to obtain predicted deformation data, and an abnormal pattern is detected by using a machine learning classification model on the first fused data. The prediction result is determined based on the predicted deformation data and the anomaly pattern.

8. The system according to claim 7, characterized in that, When the prediction module determines the prediction result based on the predicted deformation data and the anomaly pattern, it includes: in, To predict deformation data, f trend This is a function for a trend prediction model; Determining whether the predicted deformation data exceeds a preset threshold includes: Among them, Trigger trend To predict deformation results, Trigger trend =1 means that the predicted deformation data exceeds the preset threshold, Trigger trend =0 means that the predicted deformation data does not exceed the preset threshold, D threshold The preset threshold; A(t,x,y,z)=f anomaly (F1(t,x,y,z)); Where A(t,x,y,z) is the anomaly identifier of the abnormal pattern, A(t,x,y,z) = 1 indicates an anomaly, and A(t,x,y,z) = 0 indicates normality, f anomaly For machine learning model functions; The early warning module determines whether to trigger an early warning based on the prediction result, including: Where Alarm is the prediction result, Alarm=1 will trigger an alarm signal, and Alarm=0 will not trigger an alarm signal.

9. The system according to claim 1, characterized in that, When the environmental compensation module performs external environmental compensation on the first fused data to obtain the second fused data, it includes: The first fused data is dynamically compensated based on external air pressure, external humidity, and external temperature using an adaptive compensation model to obtain the second fused data.

10. The system according to claim 9, characterized in that, When the multi-level early warning and risk assessment module performs multi-level early warning based on the second fused data, it includes: The second fused data is analyzed based on deformation characteristics and dynamic trends to obtain deformation characteristic data and dynamic trend data. If the deformation feature data exceeds the normal threshold but does not exceed the safety threshold, the multi-level warning is determined to be a minor risk warning. If the deformation feature data exceeds the normal threshold and the absolute value of the difference between the dynamic trend data and the critical threshold is less than a preset value, then the multi-level warning is determined to be a medium-risk warning. If the deformation feature data exceeds the safety threshold and the dynamic trend data exceeds the critical threshold, then the multi-level warning is determined to be a high-risk warning.

11. A method for monitoring pipeline deformation, characterized in that, include: Monitoring data of the pipe wall is collected by fiber optic grating sensors and microelectromechanical system sensors that are evenly spaced on the pipe wall. The monitoring data includes at least one of the following: pipe wall deformation data, pipe wall temperature data, soil moisture data, and formation pressure data; The monitoring data is subjected to alternating fusion and compensation processing to obtain the first fused data; External environment compensation is applied to the first fused data to obtain the second fused data; Multi-level early warnings and pipeline deformation monitoring results are determined based on the second fused data.

Images