A flexible composite pipe leak self-aware self-adaptive repair system and method

Abstract

The application relates to a flexible composite pipe leakage self-sensing self-adaptive repair system and method, and belongs to the technical field of pipeline leakage detection and repair, in particular to a flexible composite pipe leakage self-sensing self-adaptive repair system and method, which comprises the following modules: a self-sensing module, a data processing and decision module, a self-adaptive repair module for executing the hierarchical repair instruction; an energy function module; the self-sensing module, the data processing and decision module and the self-adaptive repair module are sequentially connected, and the energy function module is connected with the edge calculation nodes of the pipeline field in the self-sensing module, the data processing and decision module and the self-adaptive repair module. The application constructs a full-closed-loop autonomous system integrating real-time sensing, intelligent decision, active execution and effect evaluation, and changes the traditional passive emergency response mode into active preventive maintenance and autonomous emergency disposal.

Description

Technical Field

[0001] This invention belongs to the field of pipeline leakage detection and repair technology, specifically relating to a real-time sensing and adaptive repair system for small cracks in flexible composite pipes based on intelligent sensing and self-healing materials. It is applicable to crude oil transportation pipelines in high-risk environments such as offshore oil fields and permafrost areas, and particularly relates to a flexible composite pipe leakage self-sensing adaptive repair system and method. Background Technology

[0002] Flexible composite pipes face severe challenges in safe operation and maintenance, with existing technologies exhibiting fundamental deficiencies in both monitoring and repair. First, traditional pipeline leak detection relies on manual inspections or single sensors (such as pressure sensors), which are severely inadequate in detecting minute, slow-moving leaks. Response delays are typically long (>10 minutes), and location accuracy is low (±10 meters or more), making early warning and precise intervention difficult. Second, existing repair technologies for micro-cracks (≤2mm) easily generated in flexible composite pipes due to their material properties, such as heat fusion or external patching, are passive interventions. These technologies not only require pipeline shutdown but also necessitate external professionals and equipment, failing to achieve rapid in-situ repair. Finally, in complex environments such as high-pressure subsea conditions and permafrost deformation, single sensor signals are highly susceptible to environmental interference, leading to frequent false alarms or missed alarms in the monitoring system, resulting in severely compromised overall system reliability. Summary of the Invention

[0003] This invention aims to address the problems of slow response, low positioning accuracy, and susceptibility to environmental interference in existing pipeline repair technologies for monitoring microcracks in flexible composite pipes; as well as the fundamental defects of repair methods that rely on manual intervention, cannot perform rapid on-site repairs, have high operation and maintenance costs, and lack autonomous decision-making and closed-loop control capabilities.

[0004] To address the aforementioned technical problems, this invention provides a flexible composite pipe leakage self-sensing and adaptive repair system and method.

[0005] This invention discloses a flexible composite pipe leakage self-sensing and adaptive repair system, comprising the following modules: The self-sensing module is used to monitor the multi-dimensional status data of the flexible composite pipe; The data processing and decision-making module is used to analyze the multidimensional state data, diagnose leaks, and generate graded repair instructions. An adaptive repair module is used to execute the graded repair instructions; Energy function module, used for energy supply; The self-sensing module, data processing and decision-making module, and adaptive repair module are connected in sequence, and the energy function module is connected to the self-sensing module, data processing and decision-making module, and adaptive repair module respectively.

[0006] Preferably, the self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, distributed acoustic sensor network, or PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

[0007] Preferably, the self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network and / or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic wave signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensing network, and the PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

[0008] Preferably, the flexible composite tube includes an outer protective layer and a reinforcing layer, wherein the reinforcing layer is located inside and adjacent to the outer protective layer, and the fiber Bragg grating strain / temperature sensor array is embedded in the reinforcing layer of the flexible composite tube; The distributed acoustic sensing network is embedded in the reinforcing layer of the flexible composite pipe; The PZT piezoelectric thin film sensor array is disposed on the outer protective layer of the flexible composite tube.

[0009] Preferably, the data processing and decision-making module includes an edge computing node deployed at the pipeline site and a central control unit located in a remote control unit that uses a CNN-LSTM hybrid deep learning algorithm to fuse and analyze multi-dimensional state data in order to diagnose leakage events, assess damage levels, and generate corresponding repair instructions. All edge computing nodes are connected to the central control unit; Each edge computing node is connected to the fiber Bragg grating strain / temperature sensor in the adjacent fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensor in the distributed acoustic sensor network, and the PZT piezoelectric thin film sensor in the PZT piezoelectric thin film sensor array. The edge computing node is connected to the energy function module.

[0010] Preferably, the adaptive repair module includes microcapsules that are passively ruptured due to the propagation of the crack itself, and the microcapsules are filled with a repair agent for sealing the crack; The flexible composite tube also includes an inner liner for contact with fluid, the inner liner being located inside the reinforcing layer, and the microcapsules being embedded in the inner liner of the flexible composite tube.

[0011] Preferably, the adaptive repair module further includes patches that are activated by heating or electric current and deform to apply a compressive force to the crack area, thereby achieving active repair. All patches are connected to the data processing and decision-making module. The patch is a shape memory polymer patch based on an epoxy resin matrix; The patches are all pre-embedded in the patch layer in the flexible composite pipe. The patch layer is wrapped around the inner lining layer and is adjacent to the inner lining layer.

[0012] Preferably, the adaptive repair module further includes a repair robot that moves to a designated location and injects a rapid-curing material after receiving a signal, the repair robot being connected to the data processing and decision-making module.

[0013] Preferably, the energy function module includes a photovoltaic module, a piezoelectric energy harvesting unit based on pipeline fluid vibration, and an energy storage and management unit; The flexible composite pipe includes an outer protective layer, a reinforcing layer, and an inner lining layer, with the reinforcing layer located between the outer protective layer and the inner lining layer. The photovoltaic module is laid on the outer protective layer surface of the flexible composite pipe; The piezoelectric energy harvesting unit is embedded between the reinforcing layer and the inner lining layer of the flexible composite tube; The energy storage and management unit is connected to the edge computing node; Both the photovoltaic module and the piezoelectric energy harvesting unit are connected to the energy storage and management unit.

[0014] A self-sensing and adaptive repair method for leaks in flexible composite pipes, comprising the following steps: A self-sensing and adaptive repair system for leaks in flexible composite pipes is used to repair cracks generated on the flexible composite pipes. (a) The self-sensing module acquires multi-dimensional state data of the flexible composite pipe in real time; (b) The data processing and decision-making module performs fusion analysis on the multidimensional state data obtained in step (a) to diagnose leakage events, assess damage levels, generate corresponding repair instructions, and send the repair instructions to the adaptive repair module. (c) The adaptive repair module receives and executes repair instructions; (d) After the repair operation is completed, the multi-dimensional status data of the repair area is continuously monitored through the self-sensing module to evaluate the repair effect; if the evaluation result is that the repair fails, the process returns to step (b). (e) If the evaluation result is still a repair failure after repeating steps (b) and (c) a maximum number of times, the central control unit of the data processing and decision-making module will be automatically triggered to issue an alarm.

[0015] This invention transforms the traditional passive emergency response model into proactive preventative maintenance and autonomous emergency handling by constructing a fully closed-loop autonomous system integrating "real-time perception, intelligent decision-making, proactive execution, and effect evaluation." This invention also establishes a novel self-correcting methodology of "diagnosis-repair-evaluation-re-decision" to address the industry pain points of high false alarm rates and the lack of an autonomous repair closed loop.

[0016] The combination and coupling principle of multiple sensors in this invention lies in "multimodal information fusion diagnosis." This principle is first based on "structural integration": all sensors are tightly integrated with the pipeline's reinforcement layer, outer protective layer, and inner lining layer. This ensures that any single physical event (such as crack propagation or media leakage) will simultaneously generate monitorable signals in multiple physical dimensions (strain, temperature, and sound waves). Based on this, the three sensors form a "macro-meso-micro" multi-scale monitoring network through "multi-scale collaboration": a fully distributed acoustic sensor network provides macroscopic positioning; a fiber Bragg grating strain / temperature sensor array provides high-precision strain and temperature data for that point; and a PZT piezoelectric thin film sensor array captures the microscopic acoustic emission signals at that point.

[0017] Based on the aforementioned fusion diagnostic principle, the collected data is analyzed by the data acquisition and processing unit. The core of this unit is a central data processing and decision-making module deployed in a remote control unit, which utilizes a CNN-LSTM hybrid deep learning model to analyze sensor data. In this model, a convolutional neural network (CNN) is used to extract spatial features from sensors such as PZT arrays, while a long short-term memory network (LSTM) is used to analyze the dynamic evolution trends of time-series signals such as FBG and DAS.

[0018] This invention can accurately determine the location and severity of leaks, and as mentioned above, effectively distinguish between normal pipeline operation fluctuations (such as pump start-up and shutdown) and real leak signals (i.e., a "real leak" event must meet the requirement of simultaneously detecting a complete set of characteristics at the same time and place, including continuous leaking sound from DAS, sudden strain change and / or temperature drop from FBG, and acoustic emission signal from PZT, while "false alarm interference" lacks key characteristics), thereby greatly reducing the false alarm rate.

[0019] This invention triggers a corresponding graded repair mechanism based on a high-confidence judgment result. This mechanism is executed by a self-healing material integration module that integrates multiple self-healing materials: when the damage is judged to be a microcrack (Level 1 damage), without active command, the microcapsules pre-embedded in the corrosion-resistant lining will passively rupture due to the crack's own expansion, and the repair agent inside will flow out and rapidly polymerize to seal the crack; when the damage is judged to be more severe (Level 2 or Level 3 damage), the data processing and decision module will issue an active repair command, which will activate the shape memory polymer (SMP) patch based on epoxy resin matrix located in the SMP patch layer by heating or electric current, causing it to deform, thereby applying a huge compressive force to the crack area to achieve active repair.

[0020] After repair, this invention continuously monitors and evaluates the repair effect. If the assessment indicates repair failure (e.g., FBG strain data not returning to normal), the system will automatically send the current status data back to the decision-making unit to upgrade the repair strategy and initiate a new round of repair attempts (e.g., upgrading from Level 1 microcapsule repair failure to Level 2 SMP patch repair). To ensure system safety under long-term monitoring, if repair fails within a preset maximum number of attempts (e.g., 3 times), the system will terminate autonomous repair and issue the highest-level alarm, thus forming a self-correcting closed loop that combines high autonomy with "failure-safety" characteristics. Attached Figure Description

[0021] Figure 1 This is a logic block diagram of the repair process of the present invention.

[0022] Figure 2 This is a system logic block diagram according to an embodiment of the present invention.

[0023] Figure 3 This is a diagram showing the design of the flexible composite pipe structure and the arrangement of self-healing materials according to the present invention. Detailed Implementation

[0024] This invention discloses a flexible composite pipe leakage self-sensing and adaptive repair system, comprising the following modules: The self-sensing module is used to monitor the multi-dimensional status data of the flexible composite pipe; The data processing and decision-making module is used to analyze the multidimensional state data, diagnose leaks, and generate graded repair instructions. An adaptive repair module is used to execute the graded repair instructions; Energy function module, used for energy supply; The self-sensing module, data processing and decision-making module, and adaptive repair module are connected in sequence, and the energy function module is connected to the edge computing node of the pipeline site and the adaptive repair module in the self-sensing module, data processing and decision-making module, respectively.

[0025] In one embodiment, the self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, distributed acoustic sensor network, or PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

[0026] In one embodiment, the self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network and / or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic wave signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensing network, and the PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

[0027] In one embodiment, the flexible composite tube includes an outer protective layer and a reinforcing layer, the reinforcing layer being located within and adjacent to the outer protective layer, and the fiber Bragg grating strain / temperature sensor array being embedded within the reinforcing layer of the flexible composite tube; The distributed acoustic sensing network is embedded in the reinforcing layer of the flexible composite pipe; The PZT piezoelectric thin film sensor array is disposed on the outer protective layer of the flexible composite tube.

[0028] In one embodiment, the data processing and decision-making module includes an edge computing node deployed at the pipeline site and a central control unit located in a remote control unit that uses a CNN-LSTM hybrid deep learning algorithm to fuse and analyze multi-dimensional state data in order to diagnose leakage events, assess damage levels, and generate corresponding repair instructions. All edge computing nodes are connected to the central control unit; Each edge computing node is connected to the fiber Bragg grating strain / temperature sensor in the adjacent fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensor in the distributed acoustic sensor network, and the PZT piezoelectric thin film sensor in the PZT piezoelectric thin film sensor array.

[0029] In one embodiment, the adaptive repair module includes microcapsules that are passively ruptured due to the propagation of the crack itself, the microcapsules being filled with a repair agent for sealing the crack; The flexible composite tube also includes an inner liner for contact with fluid, the inner liner being located inside the reinforcing layer, and the microcapsules being embedded in the inner liner of the flexible composite tube.

[0030] In one embodiment, the adaptive repair module further includes a patch that is activated by heating or electric current and deforms to apply a compressive force to the crack area to achieve active repair. The patches are all connected to the data processing and decision module. The patch is a shape memory polymer patch based on an epoxy resin matrix; The patches are all pre-embedded in the patch layer in the flexible composite pipe. The patch layer is wrapped around the inner lining layer and is adjacent to the inner lining layer.

[0031] In one embodiment, the adaptive repair module further includes a repair robot that moves to a designated location and injects a rapid-curing material after receiving a signal, the repair robot being connected to the data processing and decision-making module.

[0032] Preferably, the energy function module includes a photovoltaic module and a piezoelectric energy harvesting unit based on pipeline fluid vibration, wherein the energy function module includes a photovoltaic module, a piezoelectric energy harvesting unit based on pipeline fluid vibration, and an energy storage and management unit; Photovoltaic modules are laid on the outer protective layer surface of the flexible composite pipe; The piezoelectric energy harvesting unit is embedded between the reinforcing layer and the inner lining layer of the flexible composite tube. The energy storage and management unit is integrated into the hardware box of the edge computing node; The energy function module is connected to the edge computing node of the pipeline site in the self-sensing module, the data processing and decision-making module, and the adaptive repair module, respectively.

[0033] A self-sensing and adaptive repair method for leaks in flexible composite pipes, comprising the following steps: A self-sensing and adaptive repair system for leaks in flexible composite pipes is used to repair cracks generated on the flexible composite pipes. (a) The self-sensing module acquires multi-dimensional state data of the flexible composite pipe in real time; (b) The data processing and decision-making module performs fusion analysis on the multidimensional state data obtained in step (a) to diagnose leakage events, assess damage levels, generate corresponding repair instructions, and send the repair instructions to the adaptive repair module. (c) The adaptive repair module receives and executes repair instructions; (d) After the repair operation is completed, the multi-dimensional status data of the repair area is continuously monitored through the self-sensing module to evaluate the repair effect; if the evaluation result is that the repair fails, the process returns to step (b). (e) If the evaluation result is still a repair failure after repeating steps (b) and (c) a maximum number of times, the central control unit of the data processing and decision-making module will be automatically triggered to issue an alarm.

[0034] The core idea of ​​this invention lies in transforming the traditional passive emergency response mode into proactive preventative maintenance and autonomous emergency handling by constructing a fully closed-loop autonomous system integrating "real-time perception, intelligent decision-making, proactive execution, and effect evaluation." The core innovation of this invention is not simply the integration of existing technologies, but rather the construction of a novel self-correcting working method of "diagnosis-repair-evaluation-re-decision-making" to address the industry pain points of high false alarm rates and the lack of an autonomous repair closed loop.

[0035] Reference Figure 2 The system provided by the present invention includes a distributed sensor network (self-sensing module), a data acquisition and processing unit (data processing and decision-making module), a self-healing material integration module (adaptive repair module), and a photovoltaic power supply module (energy supply function module).

[0036] Reference Figure 3 (a) shows the pipe cross-section structure. In this invention, a fiber Bragg grating (FBG) strain / temperature sensor array is pre-embedded in key stress points such as the pipe reinforcement layer. Combined with a fully distributed acoustic sensing (DAS) network (also located in the reinforcement layer), the pipe's temperature, strain, and acoustic signals are continuously monitored along the entire length of the pipe. Furthermore, a high-sensitivity PZT (lead zirconate titanate) piezoelectric thin film array (arranged in the outer protective layer) is integrated to capture high-frequency acoustic emission signals, thereby enabling multi-dimensional and multi-scale monitoring of the pipe's temperature, strain, and acoustic signals.

[0037] The aforementioned "multi-dimensional, multi-scale monitoring" refers to achieving high-confidence diagnosis through the complementary functions and data synergy of three types of sensors. (1) Multi-dimensional monitoring (physical quantities): The system simultaneously acquires key point strain and temperature data provided by the FBG array and acoustic wave data provided by the DAS network and PZT array.

[0038] (2) Multi-scale monitoring (range and type): The system achieves cross-validation through the synergy of three sensors on the spatial and signal scales: (a) The FBG array, in "sentinel" mode, performs high-precision "quasi-distributed" strain / temperature monitoring of key stress points such as welds and elbows; (b) The DAS network, in "full coverage" mode, uses the entire optical cable as a continuous acoustic sensor to perform "fully distributed" macroscopic monitoring of the entire pipeline length, and is good at locating continuous leakage acoustic signals; (c) The PZT array, in "high sensitivity" mode, performs local monitoring of specific high-risk areas, and is good at capturing extremely short-lived, high-frequency "acoustic emission" microscopic signals emitted when tiny cracks propagate inside the material.

[0039] To achieve the above monitoring, the preferred arrangement of the "self-sensing module" of this invention is as follows: (1) Fiber Bragg grating (FBG) sensor array: During the manufacturing process of flexible composite pipe, the encapsulated FBG sensing fiber is pre-embedded and integrated into the reinforcement layer structure of the pipe to ensure optimal strain coupling. Its arrangement adopts the "critical point sentinel" mode, that is, it is densely arranged in the high stress concentration areas of the pipe (such as elbows, T-joints, flange connections and support points) (for example, 3-4 FBGs are arranged in a ring on the cross section), and a sensing point is set in a quasi-distributed manner (for example, every 5 to 10 meters) on straight sections.

[0040] (2) Distributed Acoustic Sensing (DAS) Network: A single sensing optical cable is used, which is also integrated into the reinforcement layer of the pipe during manufacturing and is spirally wound along the entire length of the pipe. This optical cable is functionally equivalent to thousands of continuous virtual microphones, covering the entire length of the pipe, and realizing fully distributed macroscopic monitoring with a spatial resolution of 1-5 meters.

[0041] (3) PZT piezoelectric film sensor array: attached or integrated into the outer protective layer of the pipeline and arranged in coordination with the "key points" of the FBG sensor. That is, in each high-risk area (such as elbows and joints) where the FBG is densely arranged, a sensor array consisting of 3-4 PZT patches is arranged to capture micro signals.

[0042] (4) Reference Figure 3 (c) shows the sensor integration arrangement scheme, in which the PZT piezoelectric film sensor array and FBG sensor are arranged together at the key points of the pipeline, while the distributed acoustic sensing (DAS) network and the sensors in the fiber Bragg grating (FBG) sensor array together cover the straight section of the pipeline.

[0043] Reference Figure 1 , Figure 2 and Figure 3 The working method of this invention is as follows: First, the self-sensing modules deployed at the pipeline site perform long-term, uninterrupted monitoring of the pipeline. When an abnormal signal is detected, the data is sent to edge computing nodes for preprocessing, and then the "preliminary abnormal data" is uploaded to the remote central data processing and decision-making module.

[0044] The central module processes and locates the data, makes decisions, and generates hierarchical policy instructions. For example, if the damage is determined to be minor, a Level 1 repair instruction is issued; if it is a larger crack, a Level 2 repair instruction is issued; and if it is severe damage, a Level 3 repair instruction is issued.

[0045] Upon receiving the instruction, the on-site adaptive repair module executes the corresponding repair operations. After repair is complete, the self-sensing module continuously monitors the repaired area and feeds back the "post-repair status data" to the central module for effectiveness evaluation. Figure 1 The illustrated process demonstrates the core innovative closed loop of this invention: if the "repair successful" condition is "no", the process will proceed to the "maximum number of attempts reached" check. If the maximum limit (e.g., 3 attempts) is not reached, the process will return to the "decision (tiered strategy)" stage, initiating a strategy adjustment and upgrade repair attempt based on new data. If the maximum limit is reached and the process still fails, the system will trigger an "alarm".

[0046] The following will be illustrated with a specific embodiment and referenced. Figure 1 , Figure 2 and Figure 3 This invention provides a detailed description of the system and method. This embodiment simulates the entire process from the formation of a micro-crack to its successful repair on a flexible composite pipe in an unattended, high-risk environment.

[0047] Step 1: Long-term monitoring and event triggering. (Refer to...) Figure 2 and Figure 3 (c) Self-sensing modules deployed at the pipeline site perform long-term, uninterrupted monitoring of the pipeline. Suppose that at a certain moment, a tiny crack appears due to pipeline fatigue or external damage.

[0048] Step 2: Multimodal perception and edge preprocessing. Based on the principle of "multimodal information fusion diagnosis" described in this invention: (1) First, the brief high-frequency acoustic emission signal generated by the crack propagation is captured by the PZT piezoelectric thin film sensor array (located on the outer protective layer) arranged at the key point; (2) Almost simultaneously, the strain at the crack changes abruptly, and this abrupt change is accurately captured by the sensor in the fiber Bragg grating (FBG) sensor array located in the reinforcement layer as a "sentinel" signal; (3) If the crack causes the medium to seep out, the distributed acoustic sensing (DAS) network, which is also located in the reinforcement layer, also captures the continuous acoustic feature signal generated by the leakage point, and the sensor in the fiber Bragg grating (FBG) sensor array simultaneously detects the temperature anomaly caused by the medium leakage. The edge computing node deployed nearby receives the synchronization anomaly signals from these three sensors of different scales. The node uses a built-in lightweight CNN model to analyze the time difference (Δt) of the signals from the PZT piezoelectric thin film sensor array (space) arriving at different sensors, quickly performs spatial localization preprocessing of the sound source, and confirms it as a high-confidence micro-leakage event in a very short time (e.g., within 1 second) through the aforementioned feature fusion verification, and immediately reports the "preliminary anomaly data" containing information such as location and signal characteristics.

[0049] Step 3: Central Decision-Making and Command Issuance. The central data processing and decision-making module within the remote control unit receives the "preliminary abnormal data." Its core CNN-LSTM hybrid deep learning model performs in-depth analysis of the data: the CNN module analyzes the spatial stress distribution characteristics reported by the PZT array and FBG array; the LSTM module analyzes the temporal evolution characteristics of the FBG temperature sensors in the Distributed Acoustic Sensing (DAS) network and fiber Bragg grating (FBG) sensor array. By fusing spatiotemporal features, the model can not only calculate the damage size with high accuracy but also compare it with the "normal operating condition model" (such as pump and valve start-up and shutdown) in the database to achieve accurate confirmation of leakage events. Subsequently, the damage level is determined according to the preset rule base. Scenario 1 (Level 1 Damage): If the damage is determined to be a microcrack (e.g., ≤2mm), it is classified as Level 1 damage. In this scenario, the system defaults to the microcapsules pre-embedded in the pipeline lining autonomously completing the repair (e.g., Figure 3 (b) shows the Level 1 repair mechanism. Therefore, the central unit decides to "not issue an active repair instruction for the time being, and directly enter the continuous monitoring and evaluation phase." Scenario 2 (Level 2 damage): If it is determined to be a large crack (e.g., between 2mm and 5cm), it is classified as Level 2 damage. The central unit will generate and issue a Level 2 repair instruction, the goal of which is to activate the nearest epoxy resin matrix-based shape memory polymer patch (e.g., ...) to the crack location. Figure 3 (b) shows the level 2 repair mechanism.

[0050] Step 4: Repair Execution and Effect Evaluation. The on-site adaptive repair module executes the corresponding repair operation based on whether an instruction has been received: In Scenario 1 above, the microcapsules torn by the crack have autonomously completed polymerization repair within minutes. In Scenario 2 above, the shape memory polymer (SMP) patch based on an epoxy resin matrix located in the SMP patch layer is electrothermally activated after receiving a Level 2 instruction. For example, in an embodiment applied in permafrost regions, the glass transition temperature (Tg) of the SMP patch can be set to 70°C, and its shrinkage rate after activation can reach over 40%, thereby applying a strong compressive force to the cracked area and achieving active sealing. After the repair action occurs, the self-sensing module continuously monitors the repaired area and feeds back the "post-repair status data" to the central module for effect evaluation.

[0051] Step 5: Closed-loop feedback and strategy upgrade. (Refer to...) Figure 1The logical flow involves the central module determining "Repair Successful?" based on feedback FBG strain and DAS acoustic data. The criteria for "Repair Successful" are: data showing key indicators such as pressure and strain returning to normal, and the disappearance of leakage acoustic signals. If the repair is deemed successful, the process returns to the "Start / Continuous Monitoring" state. Repair logs can be uploaded to the monitoring center via LoRaWAN or NB-IoT wireless communication. If the data remains abnormal (e.g., microcapsule repair failure in Scenario 1), "Repair Failed" is determined. At this point, since the maximum number of attempts (e.g., 3) has not been reached, the process returns to the "Decision" stage. The central unit will automatically upgrade the damage level from Level 1 to Level 2 based on the failure feedback, issuing a Level 2 repair command and activating the SMP patch for repair. If the evaluation still fails after SMP patch repair, the system will return to the decision stage again, upgrading the strategy to Level 3 and activating the in-pipe repair robot (e.g., Figure 3 (b) shows the 3-level repair mechanism. The robot will move to the designated location and inject a fast-curing material such as two-component polyurethane foam for emergency sealing. After implementing the upgraded repair strategy, the system will evaluate the effect again. If it still fails, the system will eventually trigger an "alarm" after reaching the maximum number of attempts, notifying manual intervention.

[0052] Through the above steps, this invention efficiently and autonomously completes the entire closed-loop operation of monitoring, diagnosing, and attempting multi-level repairs of pipeline cracks, to final confirmation or alarm, without any human intervention.

Claims

1. A flexible composite pipe leakage self-sensing and adaptive repair system, characterized in that, Includes the following modules: The self-sensing module is used to monitor the multi-dimensional status data of the flexible composite pipe; The data processing and decision-making module is used to analyze the multidimensional state data, diagnose leaks, and generate graded repair instructions. An adaptive repair module is used to execute the graded repair instructions; Energy function module, used for energy supply; The self-sensing module, data processing and decision-making module, and adaptive repair module are connected in sequence, and the energy function module is connected to the self-sensing module, data processing and decision-making module, and adaptive repair module respectively.

2. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 1, characterized in that, The self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic wave signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, distributed acoustic sensor network, or PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

3. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 1, characterized in that, The self-sensing module includes: An array of fiber Bragg grating strain / temperature sensors embedded in a flexible composite tube for monitoring the temperature and strain signals of the flexible composite tube. A distributed acoustic sensing network and / or PZT piezoelectric thin film sensor array embedded in a flexible composite tube for measuring acoustic wave signals in the flexible composite tube. The fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensing network, and the PZT piezoelectric thin film sensor array are all connected to the data processing and decision-making module.

4. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 3, characterized in that, The flexible composite tube includes an outer protective layer and a reinforcing layer. The reinforcing layer is located inside the outer protective layer and is adjacent to the outer protective layer. The fiber Bragg grating strain / temperature sensor array is embedded in the reinforcing layer of the flexible composite tube. The distributed acoustic sensing network is embedded in the reinforcing layer of the flexible composite pipe; The PZT piezoelectric thin film sensor array is disposed on the outer protective layer of the flexible composite tube.

5. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 3, characterized in that, The data processing and decision-making module includes edge computing nodes deployed at the pipeline site and a central control unit located in a remote control unit that uses a CNN-LSTM hybrid deep learning algorithm to fuse and analyze multi-dimensional state data in order to diagnose leakage events, assess damage levels, and generate corresponding repair instructions. All edge computing nodes are connected to the central control unit; Each edge computing node is connected to the fiber Bragg grating strain / temperature sensor in the adjacent fiber Bragg grating strain / temperature sensor array, the distributed acoustic sensor in the distributed acoustic sensor network, and the PZT piezoelectric thin film sensor in the PZT piezoelectric thin film sensor array. The edge computing node is connected to the energy function module.

6. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 1, characterized in that, The adaptive repair module includes microcapsules that are passively ruptured due to the expansion of the crack itself, and the microcapsules are filled with a repair agent for sealing the crack. The flexible composite tube also includes an inner liner for contact with fluid, the inner liner being located inside the reinforcing layer, and the microcapsules being embedded in the inner liner of the flexible composite tube.

7. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 6, characterized in that, The adaptive repair module also includes patches that are activated by heating or electric current and deform to apply a compressive force to the crack area, thereby achieving active repair. All patches are connected to the data processing and decision-making module. The patch is a shape memory polymer patch based on an epoxy resin matrix; The patches are all pre-embedded in the patch layer in the flexible composite pipe. The patch layer is wrapped around the inner lining layer and is adjacent to the inner lining layer.

8. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 7, characterized in that, The adaptive repair module also includes a repair robot that moves to a designated location and injects a fast-curing material after receiving a signal. The repair robot is connected to the data processing and decision-making module.

9. The flexible pipe leakage self-sensing and adaptive repair system as described in claim 5, characterized in that, The energy function module includes a photovoltaic module, a piezoelectric energy harvesting unit based on pipeline fluid vibration, and an energy storage and management unit. The flexible composite pipe includes an outer protective layer, a reinforcing layer, and an inner lining layer, with the reinforcing layer located between the outer protective layer and the inner lining layer. The photovoltaic module is laid on the outer protective layer surface of the flexible composite pipe; The piezoelectric energy harvesting unit is embedded between the reinforcing layer and the inner lining layer of the flexible composite tube; The energy storage and management unit is connected to the edge computing node; Both the photovoltaic module and the piezoelectric energy harvesting unit are connected to the energy storage and management unit.

10. A method for self-sensing and adaptive repair of leaks in flexible composite pipes, comprising using the self-sensing and adaptive repair system for leaks in flexible composite pipes as described in claim 1 to repair cracks generated on the flexible composite pipes, characterized in that... Includes the following steps: (a) The self-sensing module acquires multi-dimensional state data of the flexible composite pipe in real time; (b) The data processing and decision-making module performs fusion analysis on the multidimensional state data obtained in step (a) to diagnose leakage events, assess damage levels, generate corresponding repair instructions, and send the repair instructions to the adaptive repair module. (c) The adaptive repair module receives and executes repair instructions; (d) After the repair operation is completed, the multi-dimensional status data of the repair area is continuously monitored through the self-sensing module to evaluate the repair effect; if the evaluation result is that the repair fails, the process returns to step (b). (e) If the evaluation result is still a repair failure after repeating steps (b) and (c) a maximum number of times, the central control unit of the data processing and decision-making module will be automatically triggered to issue an alarm.

Images