Pipeline non-excavation repair process

Abstract

The invention discloses a pipeline trenchless repair process, and relates to the technical field of pipeline repair, and the pipeline trenchless repair process comprises the steps of S01, trenchless detection and evaluation, S02, intelligent excavation of a construction well, S03, pipeline pretreatment, S04, intelligent CCTV detection, S05, automatic repair, S06, pipe-in-pipe insertion, S07, sealing and pressure testing, S08, pipeline connection and S09, intelligent monitoring and maintenance. The position, the length and the damage condition of the pipeline can be more accurately determined through detection such as intelligent radar detection, so that accurate data support is provided for repairing work, human errors can be reduced through an automatic repairing robot and an intelligent CCTV detection system, the precision and the consistency of the repairing work are improved, and the repairing efficiency is improved. According to the improved process, the pipe-in-pipe with intelligent materials and structural design is adopted, the complex pipeline environment can be better adapted, and the adaptability and durability of repairing are improved.

Description

Technical Field

[0001] The invention relates to the technical field of pipeline repair, in particular to a pipeline trenchless repair process. Background Art

[0002] After searching, the Chinese patent publication number CN109707949A discloses a trenchless pipeline repair process. Although this process has the advantages of existing technical solutions, namely, simple and fast construction and small environmental impact, and solves the defects of existing solutions, enhances the service life of repaired pipelines, and reduces the repeated waste of social resources, the existing CCTV detection robots have limited functions and may not be able to accurately identify all pipeline problems, such as tiny leaks or hidden misalignments. The existing pipeline repair process lacks an effective intelligent monitoring system and cannot monitor the operating status of the pipeline in real time, resulting in potential problems that cannot be discovered and handled in a timely manner. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a trenchless pipeline repair process, which solves the problems mentioned in the background technology that the existing CCTV detection robots are limited in identifying tiny leaks and hidden dislocations, and the pipeline repair process lacks intelligent monitoring, making it difficult to detect and handle potential problems in real time.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a pipeline trenchless repair process, the steps of which are as follows:

[0005] S01. Trenchless detection and assessment: Use high-precision radar detection technology to conduct comprehensive inspections of pipelines and assess the extent of damage and repair needs of pipelines;

[0006] S02. Intelligent excavation of construction wells: Use intelligent robots to accurately excavate construction wells to reduce damage to the surrounding environment;

[0007] S03. Pipeline pretreatment: Use high-pressure water jet technology to remove silt and debris in the pipeline to ensure the smooth passage of the CCTV detection robot;

[0008] S04. Intelligent CCTV detection: Upgrade the CCTV detection robot and equip it with an intelligent analysis system to automatically detect and record the leakage points, dislocation, blockage and bends of the pipeline;

[0009] S05. Automated repair: Develop automated repair equipment to automatically perform internal repairs on pipelines based on CCTV inspection results, including sealing of leaks and correction of misalignment;

[0010] S06. Pipe-in-pipe insertion: The pipe-in-pipe adopts intelligent materials, which can automatically deform into the best repair shape according to the internal environment of the pipeline, and can be precisely inserted through the intelligently controlled guide rope system;

[0011] S07. Sealing and pressure testing: Intelligent sealing technology is used at both ends of the pipe-in-pipe to ensure the sealing effect; the intelligent pressure testing system is used to perform pressure testing on the repaired pipeline to ensure no leakage;

[0012] S08. Pipeline connection: Develop intelligent joints that can automatically adapt to the size and shape of different pipes to ensure the stability and tightness of pipeline connections;

[0013] S09. Intelligent monitoring and maintenance: After the repair is completed, an intelligent monitoring system is installed to monitor the operation status of the pipeline in real time, and to detect and warn of potential problems in a timely manner;

[0014] Preferably, the intelligent analysis system includes a high-resolution imaging module, an image enhancement module, an intelligent module, a sensor integration module, a navigation and positioning module, a real-time data processing module, a user interface module and an automatic report generation module;

[0015] The high-resolution imaging module is equipped with a high-resolution camera to obtain a clear image of the interior of the pipeline;

[0016] The image enhancement module is used to enhance the captured image using an image processing algorithm to improve the contrast and clarity of the image, making tiny leakage points and misalignment more obvious;

[0017] The intelligent module integrates machine learning algorithms to identify different types of pipeline defects through training; the algorithm learns from historical data to continuously improve recognition accuracy, and uses convolutional neural networks (CNN) to improve the ability to identify small leaks and hidden dislocations;

[0018] The sensor integration module includes an acoustic sensor, an infrared sensor, and a gas sensor to assist image recognition and provide a more comprehensive assessment of pipeline conditions;

[0019] The navigation and positioning module is used to enable the robot to move autonomously in the pipeline through the developed automatic navigation system, reducing the dependence on manual operation, and to ensure that the position of the robot in the pipeline can be accurately tracked by integrating a laser rangefinder and GPS;

[0020] The real-time data processing module is used to process data, process the data collected by the robot in real time, and convert it into useful information, and use wireless communication technology, Wi-Fi or 4G / 5G, to transmit the data to the ground control center in real time;

[0021] The user interface module is used to enable an operator to easily monitor the status of the robot and the collected data;

[0022] The automated report generation module is used to present the inspection results to the user in an easy-to-understand format, including the type of defect, location, and recommended repair measures.

[0023] Preferably, in the artificial intelligence module, in order to improve the accuracy of pipeline defect recognition, for small leaks and hidden dislocations, the intelligent module adopts the following logical algorithm process:

[0024] Data preprocessing: applying denoising and image normalization techniques to prepare data for training and testing;

[0025] Feature extraction: Use edge detection and texture analysis to extract features from images that help identify defects;

[0026] Machine learning algorithms: Use algorithms such as support vector machines (SVM) and random forests to classify different types of pipeline defects;

[0027] Deep learning algorithm: Feature learning and classification through convolutional neural network (CNN), including convolutional layers, pooling layers and fully connected layers to optimize network weights and minimize loss functions;

[0028] Model evaluation and optimization: Use cross-validation and hyperparameter tuning techniques to evaluate model performance and optimize;

[0029] Real-time reasoning: Deploy the trained model on the robot to perform online reasoning, process the images captured by the camera in real time, and make threshold judgments based on the probability values ​​output by the model;

[0030] Feedback mechanism: Through incremental learning and user feedback, the model is continuously optimized to adapt to new pipeline defect types and improve recognition capabilities.

[0031] Preferably, the convolutional neural network (CNN) is used to process image data. The CNN learns the features of the image through a series of convolutional layers, pooling layers (or downsampling layers), and fully connected layers. The convolutional layers extract the features of the image by using filters (; each filter slides on the input image and calculates the dot product of the filter and the local area of ​​the image to generate a feature map.

[0032] Preferably, the feature map is represented by assuming that the input image is X and the convolution kernel is K. The convolution operation can be expressed as:

[0033]

[0034] Where X is the input image or input feature map. In the first convolution layer, X is usually the original image data. K is the convolution kernel or filter, which is a small window used to extract features in the convolution layer. Each convolution kernel corresponds to a feature detector, which is used to detect specific features (such as edges, textures, etc.) in the input image. Y i,j : The value of the output feature map (FeatureMap) at position (i, j); the feature map is the result of the convolution operation, which represents the characteristic response of the input image at a certain position after being processed by the convolution kernel; m, n: the index of the convolution kernel, indicating the position of the convolution kernel in space; b: bias term (Bias), used to adjust the result of the convolution operation and increase the flexibility of the model; Y: the output of the convolution layer, usually before applying the activation function; Y activated : Output after applying the activation function; Activation functions (such as ReLU) are used to introduce nonlinearity, enabling the model to learn more complex features. In this way, the convolutional layer can extract useful features from the input image and map these features to the output feature map.

[0035] Preferably, the high-precision radar detection technology in S01 is as follows:

[0036] Using ground-penetrating radar (GPR): deploying high-frequency radar systems to transmit and receive electromagnetic waves that penetrate the ground, analyzing the reflected signals to identify the precise location and characteristics of underground pipelines;

[0037] High-resolution data acquisition: High-sensitivity sensors are used for data acquisition to ensure that the acquired radar data has high resolution, so that tiny anomalies can be detected;

[0038] Signal processing: Use signal processing techniques, such as filtering and denoising algorithms, to improve the quality of radar signals and reduce the impact of environmental noise on detection results;

[0039] Intelligent data analysis: Apply machine learning algorithms to analyze radar data to automatically identify the material, shape, and possible damage areas of the pipeline; use deep learning models, such as convolutional neural networks (CNNs), to extract complex patterns from radar images and improve the accuracy of identifying pipeline damage;

[0040] 3D modeling and GIS integration: Convert radar detection data into 3D models to intuitively display the spatial layout and damage of pipelines; integrate detection results with geographic information systems (GIS) to provide accurate geographic location information and environmental background to assist decision-making;

[0041] Real-time monitoring and intelligent assessment: Implement real-time data processing to quickly respond to possible problems in the pipeline, improve response speed and processing efficiency; develop an intelligent assessment system to automatically assess the damage degree and repair priority of the pipeline, and provide support for maintenance decisions;

[0042] Predictive maintenance: Build a predictive model to predict pipeline damage trends and potential risks based on historical and real-time data; provide data-driven maintenance decision support to help develop preventive maintenance plans and reduce unexpected failures and repair costs.

[0043] Preferably, the system logic for automatically evaluating the pipeline damage degree and repair priority in S01 is:

[0044] Data preparation: Integrate historical and real-time data of pipelines, including age, material type, environmental factors, etc.; clean data, process missing values ​​and outliers, and ensure data quality;

[0045] Model development: Extract key features from the data, such as pipeline conditions and historical damage records, select appropriate machine learning algorithms, such as random forests or neural networks, train the model using labeled data, and optimize model performance through cross-validation;

[0046] Assessment and Prioritization: The model predicts the extent of pipeline damage based on input features, combines the extent of damage with pipeline key indicators, calculates the repair priority, and designs an intuitive user interface to display the assessment results and priorities, ensuring that the system can receive and process data in real time and update the assessment;

[0047] Continuous optimization: Monitor model performance, continuously optimize the model based on user feedback and new data, provide evaluation reports and maintenance recommendations, and assist in decision-making.

[0048] Preferably, the intelligent pipeline monitoring system in S09 mainly includes the following steps:

[0049] Sensor deployment: Install sensors such as pressure, flow and leakage sensors at key locations in pipelines to collect data in real time;

[0050] Data transmission: Transmit sensor data to the central monitoring system through wireless communication technology;

[0051] Data analytics: Using machine learning algorithms to analyze data and identify unusual patterns, such as leaks or pressure anomalies;

[0052] Real-time warning: Once an abnormality is detected, the system automatically triggers an early warning and notifies relevant personnel through multiple channels (SMS, application push, etc.);

[0053] GIS integration: Integration with geographic information system (GIS) to accurately locate problem areas and improve response speed;

[0054] System maintenance: Regularly update analytical models and calibrate sensors to ensure the long-term accuracy and reliability of the monitoring system.

[0055] Preferably, the method for identifying abnormal patterns is:

[0056] Scaling the data to a uniform scale, such as a range of 0 to 1, or to have zero mean and unit variance;

[0057]

[0058] Where X is the original data, μ is the mean, and σ is the standard deviation. Features that are helpful for anomaly detection are extracted from the original data. The principal component analysis (PCA) and recursive feature elimination (RFE) algorithms are used to select the most influential features. The optimal hyperplane is found in the feature space to distinguish normal and abnormal data. The accuracy, recall, F1 score and other indicators are used to evaluate the anomaly detection ability of the model. A threshold is set according to the output of the model, and data points exceeding the threshold are considered abnormal.

[0059] The present invention provides a pipeline trenchless repair process, which has the following beneficial effects:

[0060] (1) This trenchless pipeline repair process significantly improves the efficiency and quality of repair: by introducing high-precision detection technology, such as intelligent radar detection, the location, length and damage of the pipeline can be determined more accurately, thereby providing precise data support for the repair work. The use of automated repair robots and intelligent CCTV detection systems can reduce human errors and improve the accuracy and consistency of the repair work. The improved process uses pipe-in-pipe with intelligent materials and structural design, which can better adapt to complex pipeline environments and improve the adaptability and durability of the repair.

[0061] (2) This is a trenchless pipeline repair process. The improved repair process uses high-quality materials and advanced repair technology to improve the structural strength and sealing performance of the repaired pipeline, thereby extending the service life of the pipeline. The introduction of intelligent joints and intelligent monitoring systems ensures the stability of pipeline connections and the reliability of long-term operation, reducing the frequency of future repairs and replacements. Through real-time monitoring and early warning systems, problems in pipeline operation can be discovered and handled in a timely manner, avoiding greater losses caused by pipeline failures, thereby reducing long-term operating and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] See also Figure 1 , the embodiment of the present invention provides a technical solution: a pipeline trenchless repair process, the steps of which are as follows: S01. Trenchless detection and assessment: use high-precision radar detection technology to conduct a comprehensive inspection of the pipeline to assess the damage degree and repair needs of the pipeline; S02. Intelligent excavation of construction wells: use intelligent robots to accurately excavate construction wells to reduce damage to the surrounding environment; S03. Pipeline pretreatment: use high-pressure water jet technology to remove silt and debris in the pipeline to ensure the smooth passage of CCTV detection robots; S04. Intelligent CCTV detection: upgrade the CCTV detection robot, equip it with an intelligent analysis system, automatically detect and record the leakage points, dislocations, blockages and bends of the pipeline; S05. Automated repair: develop automated repair equipment, according to CCT Based on the detection results of V, the internal repair of the pipeline is automatically carried out, including sealing of the leakage point and correction of the misalignment; S06. Insertion of pipe in pipe: the pipe in pipe adopts intelligent materials, which can automatically deform into the best repair shape according to the internal environment of the pipeline, and accurately insert through the intelligent control guide rope system; S07. Sealing and pressure testing: intelligent sealing technology is used at both ends of the pipe in pipe to ensure the sealing effect; the repaired pipeline is pressure tested using the intelligent pressure testing system to ensure no leakage; S08. Pipeline connection: develop intelligent joints that can automatically adapt to the sizes and shapes of different pipelines to ensure the stability and sealing of pipeline connections; S09. Intelligent monitoring and maintenance: after the repair is completed, install an intelligent monitoring system to monitor the operation status of the pipeline in real time, and promptly discover and warn of potential problems;

[0065] The intelligent analysis system includes a high-resolution imaging module, an image enhancement module, an intelligent module, a sensor integration module, a navigation and positioning module, a real-time data processing module, a user interface module, and an automatic report generation module; the high-resolution imaging module is equipped with a high-resolution camera to obtain clear images of the inside of the pipeline; the image enhancement module is used to enhance the captured images using image processing algorithms to improve the contrast and clarity of the images, making tiny leakage points and dislocations more obvious; the intelligent module integrates machine learning algorithms to identify different types of pipeline defects through training; and the algorithm learns from historical data to continuously improve recognition accuracy, and uses convolutional neural networks (CNN) to improve the ability to identify tiny leaks and hidden dislocations; the sensor integration module includes acoustic sensors, infrared sensors and Gas sensors are used to assist image recognition and provide a more comprehensive assessment of pipeline conditions. The navigation and positioning module is used to enable the robot to move autonomously in the pipeline through the developed automatic navigation system, reducing dependence on manual operation, and integrating laser rangefinders and GPS to ensure that the robot's position in the pipeline can be accurately tracked. The real-time data processing module is used to process data, process the data collected by the robot in real time, and convert it into useful information, and use wireless communication technology, Wi-Fi or 4G / 5G, to transmit the data to the ground control center in real time. The user interface module is used to enable operators to easily monitor the status of the robot and the collected data. The automated report generation module is used to present the inspection results to the user in an easy-to-understand format, including the type and location of the defects and the recommended repair measures.

[0066] In the artificial intelligence module, in order to improve the accuracy of pipeline defect identification, the intelligent module adopts the following logical algorithm flow for small leaks and hidden dislocations: Data preprocessing: Apply denoising and image normalization techniques to prepare data for training and testing; Feature extraction: Use edge detection and texture analysis and other methods to extract features from images that are helpful for identifying defects; Machine learning algorithm: Use algorithms such as support vector machine (SVM) and random forest to classify different types of pipeline defects; Deep learning algorithm: Perform feature learning and classification through convolutional neural network (CNN), including convolutional layer, pooling layer and fully connected layer to optimize network weights and minimize loss function; Model evaluation and optimization: Use cross-validation and hyperparameter tuning techniques to evaluate model performance and optimize; Real-time reasoning: Deploy the trained model on the robot for online reasoning, process the images captured by the camera in real time, and make threshold judgments based on the probability value output by the model; Feedback mechanism: Continuously optimize the model through incremental learning and user feedback to adapt to new types of pipeline defects and improve recognition capabilities.

[0067] Convolutional neural networks (CNNs) are used to process image data. CNNs learn image features through a series of convolutional layers, pooling layers (or downsampling layers), and fully connected layers. The convolutional layers extract image features by using filters. Each filter slides on the input image and calculates the dot product between the filter and the local area of ​​the image to generate a feature map. The feature map is represented by setting the input image as X and the convolution kernel as K. The convolution operation can be expressed as:

[0068]

[0069] Where X is the input image or input feature map. In the first convolution layer, X is usually the original image data. K is the convolution kernel or filter, which is a small window used to extract features in the convolution layer. Each convolution kernel corresponds to a feature detector, which is used to detect specific features (such as edges, textures, etc.) in the input image. Y i,j : The value of the output feature map (FeatureMap) at position (i, j); the feature map is the result of the convolution operation, which represents the characteristic response of the input image at a certain position after being processed by the convolution kernel; m, n: the index of the convolution kernel, indicating the position of the convolution kernel in space; b: bias term (Bias), used to adjust the result of the convolution operation and increase the flexibility of the model; Y: the output of the convolution layer, usually before applying the activation function; Y activated : Output after applying the activation function; Activation functions (such as ReLU) are used to introduce nonlinearity, enabling the model to learn more complex features. In this way, the convolutional layer can extract useful features from the input image and map these features to the output feature map.

[0070] The system logic for automatically assessing pipeline damage and repair priority in S01 is as follows:

[0071] Data preparation: Integrate historical and real-time data of pipelines, including age, material type, environmental factors, etc.; clean data, handle missing values ​​and outliers, and ensure data quality; model development: extract key features from the data, such as pipeline conditions and historical damage records, select suitable machine learning algorithms, such as random forests or neural networks, train models using labeled data, and optimize model performance through cross-validation; evaluation and priority: the model predicts the extent of pipeline damage based on input features, calculates repair priorities based on the extent of damage and key pipeline indicators, designs an intuitive user interface, displays evaluation results and priorities, ensures that the system can receive and process data in real time, and updates evaluations; continuous optimization: monitor model performance, continuously optimize models based on user feedback and new data, provide evaluation reports and maintenance recommendations, and assist in decision-making.

[0072] The intelligent pipeline monitoring system in S09 mainly includes the following steps: sensor deployment: installing sensors at key locations of the pipeline, such as pressure, flow and leakage sensors, to collect data in real time; data transmission: transmitting sensor data to the central monitoring system through wireless communication technology; data analysis: using machine learning algorithms to analyze data and identify abnormal patterns, such as leakage or pressure abnormalities; real-time warning: once an abnormality is detected, the system automatically triggers a warning and notifies relevant personnel through multiple channels (SMS, application push, etc.); GIS integration: integration with the geographic information system (GIS) to accurately locate problem areas and improve response speed; system maintenance: regularly update analysis models and calibrate sensors to ensure the long-term accuracy and reliability of the monitoring system.

[0073] The method of identifying abnormal patterns is to: scale the data to a uniform scale, such as a range of 0 to 1, or to have zero mean and unit variance;

[0074]

[0075] Where X is the original data, μ is the mean, and σ is the standard deviation. Features that are helpful for anomaly detection are extracted from the original data. The principal component analysis (PCA) and recursive feature elimination (RFE) algorithms are used to select the most influential features. The optimal hyperplane is found in the feature space to distinguish normal and abnormal data. The accuracy, recall, F1 score and other indicators are used to evaluate the anomaly detection ability of the model. A threshold is set according to the output of the model, and data points exceeding the threshold are considered abnormal.

[0076] Table 1 below is a pipeline repair efficiency and quality evaluation table:

[0077]

[0078] Table 1

[0079] Combined with the data in the table, it can be concluded that the trenchless pipeline repair process has achieved remarkable results in improving repair efficiency and quality, reducing costs, reducing environmental impact, and improving long-term operational reliability. By using high-precision radar detection technology and intelligent CCTV detection, the process can accurately identify pipeline damage and carry out targeted repairs. The application of automated repair technology and intelligent materials further improves the accuracy and consistency of repair work and reduces the possibility of human error. Cost savings have been achieved in each step of the process, especially in the automated repair link, which saves up to 20%. This shows that the use of advanced automation technology and intelligent equipment can effectively reduce repair costs while improving work efficiency. Steps such as intelligent excavation of construction wells and pipeline pretreatment reduce damage to the surrounding environment, with an environmental damage reduction rate of more than 80%. This not only helps to protect the environment, but also reduces the social costs caused by construction. Through the introduction of intelligent joints and intelligent monitoring systems, the process ensures the stability of pipeline connections and the reliability of long-term operation. The installation of real-time monitoring and early warning systems enables potential problems to be discovered and handled in a timely manner, avoiding greater losses caused by pipeline failures.

[0080] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0081] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A pipeline trenchless repair process, characterized in that: The steps are as follows: S01. Trenchless detection and assessment: Use high-precision radar detection technology to conduct comprehensive inspections of pipelines and assess the extent of damage and repair needs of pipelines; S02. Intelligent excavation of construction wells: Use intelligent robots to accurately excavate construction wells to reduce damage to the surrounding environment; S03. Pipeline pretreatment: Use high-pressure water jet technology to remove silt and debris in the pipeline to ensure the smooth passage of the CCTV detection robot; S04. Intelligent CCTV detection: Upgrade the CCTV detection robot and equip it with an intelligent analysis system to automatically detect and record the leakage points, dislocation, blockage and bends of the pipeline; S05. Automated repair: Develop automated repair equipment to automatically perform internal repairs on pipelines based on CCTV inspection results, including sealing of leaks and correction of misalignment; S06. Pipe-in-pipe insertion: The pipe-in-pipe adopts intelligent materials, which can automatically deform into the best repair shape according to the internal environment of the pipeline, and can be precisely inserted through the intelligently controlled guide rope system; S07. Sealing and pressure testing: Intelligent sealing technology is used at both ends of the pipe-in-pipe to ensure the sealing effect; the intelligent pressure testing system is used to perform pressure testing on the repaired pipeline to ensure no leakage; S08. Pipeline connection: Develop intelligent joints that can automatically adapt to the size and shape of different pipes to ensure the stability and tightness of pipeline connections; S09. Intelligent monitoring and maintenance: After the repair is completed, an intelligent monitoring system is installed to monitor the operating status of the pipeline in real time, and to detect and warn of potential problems in a timely manner.

2. A pipeline trenchless repair process according to claim 1, characterized in that: The intelligent analysis system includes a high-resolution imaging module, an image enhancement module, an intelligent module, a sensor integration module, a navigation and positioning module, a real-time data processing module, a user interface module and an automatic report generation module; The high-resolution imaging module is equipped with a high-resolution camera to obtain a clear image of the interior of the pipeline; The image enhancement module is used to enhance the captured image using an image processing algorithm to improve the contrast and clarity of the image, making tiny leakage points and misalignment more obvious; The intelligent module integrates machine learning algorithms to identify different types of pipeline defects through training; the algorithm learns from historical data to continuously improve recognition accuracy, and uses convolutional neural networks (CNN) to improve the ability to identify small leaks and hidden dislocations; The sensor integration module includes an acoustic sensor, an infrared sensor, and a gas sensor to assist image recognition and provide a more comprehensive assessment of pipeline conditions; The navigation and positioning module is used to enable the robot to move autonomously in the pipeline through the developed automatic navigation system, reducing the reliance on manual operation, and to ensure that the position of the robot in the pipeline can be accurately tracked by integrating a laser rangefinder and GPS; The real-time data processing module is used to process data, process the data collected by the robot in real time, and convert it into useful information, and use wireless communication technology, Wi-Fi or 4G / 5G, to transmit the data to the ground control center in real time; The user interface module is used to enable an operator to easily monitor the status of the robot and the collected data; The automated report generation module is used to present the inspection results to the user in an easy-to-understand format, including the type of defect, location, and recommended repair measures.

3. A pipeline trenchless repair process according to claim 2, characterized in that: In the artificial intelligence module, in order to improve the accuracy of pipeline defect recognition, the intelligent module adopts the following logical algorithm process for minor leaks and hidden dislocations: Data preprocessing: applying denoising and image normalization techniques to prepare data for training and testing; Feature extraction: Use edge detection and texture analysis to extract features from images that help identify defects; Machine learning algorithms: Use algorithms such as support vector machines (SVM) and random forests to classify different types of pipeline defects; Deep learning algorithm: Feature learning and classification through convolutional neural network (CNN), including convolutional layers, pooling layers and fully connected layers to optimize network weights and minimize loss functions; Model evaluation and optimization: Use cross-validation and hyperparameter tuning techniques to evaluate model performance and optimize; Real-time reasoning: Deploy the trained model on the robot to perform online reasoning, process the images captured by the camera in real time, and make threshold judgments based on the probability values ​​output by the model; Feedback mechanism: Through incremental learning and user feedback, the model is continuously optimized to adapt to new pipeline defect types and improve recognition capabilities.

4. A pipeline trenchless repair process according to claim 3, characterized in that: The convolutional neural network (CNN) is used to process image data. CNN learns the features of an image through a series of convolutional layers, pooling layers (or downsampling layers), and fully connected layers. The convolutional layers extract the features of an image by using filters. Each filter slides on the input image and calculates the dot product between the filter and the local area of ​​the image to generate a feature map.

5. A pipeline trenchless repair process according to claim 4, characterized in that: The feature map is represented by setting the input image as X and the convolution kernel as K. The convolution operation can be expressed as: Where X is the input image or input feature map. In the first convolution layer, X is usually the original image data. K is the convolution kernel or filter, which is a small window used to extract features in the convolution layer. Each convolution kernel corresponds to a feature detector, which is used to detect specific features (such as edges, textures, etc.) in the input image. Y i,j : The value of the output feature map (FeatureMap) at position (i, j); the feature map is the result of the convolution operation, which represents the characteristic response of the input image at a certain position after being processed by the convolution kernel; m, n: the index of the convolution kernel, indicating the position of the convolution kernel in space; b: bias term (Bias), used to adjust the result of the convolution operation and increase the flexibility of the model; Y: the output of the convolution layer, usually before applying the activation function; Y activated : Output after applying the activation function; Activation functions (such as ReLU) are used to introduce nonlinearity, enabling the model to learn more complex features. In this way, the convolutional layer can extract useful features from the input image and map these features to the output feature map.

6. The trenchless pipeline repair process according to claim 1, characterized in that: The high-precision radar detection technology in S01 is as follows: Using ground-penetrating radar (GPR): deploying high-frequency radar systems to transmit and receive electromagnetic waves that penetrate the ground, analyzing the reflected signals to identify the precise location and characteristics of underground pipelines; High-resolution data acquisition: High-sensitivity sensors are used for data acquisition to ensure that the acquired radar data has high resolution, so that tiny anomalies can be detected; Signal processing: Use signal processing techniques, such as filtering and denoising algorithms, to improve the quality of radar signals and reduce the impact of environmental noise on detection results; Intelligent data analysis: Apply machine learning algorithms to analyze radar data to automatically identify the material, shape, and possible damage areas of the pipeline; use deep learning models, such as convolutional neural networks (CNNs), to extract complex patterns from radar images and improve the accuracy of identifying pipeline damage; 3D modeling and GIS integration: Convert radar detection data into 3D models to intuitively display the spatial layout and damage of pipelines; integrate detection results with geographic information systems (GIS) to provide accurate geographic location information and environmental background to assist decision-making; Real-time monitoring and intelligent assessment: Implement real-time data processing to quickly respond to possible problems in the pipeline, improve response speed and processing efficiency; develop an intelligent assessment system to automatically assess the damage degree and repair priority of the pipeline, and provide support for maintenance decisions; Predictive maintenance: Build a predictive model to predict pipeline damage trends and potential risks based on historical and real-time data; provide data-driven maintenance decision support to help formulate preventive maintenance plans and reduce unexpected failures and repair costs.

7. The trenchless pipeline repair process according to claim 1, characterized in that: The system logic for automatically evaluating the pipeline damage degree and repair priority in S01 is: Data preparation: Integrate historical and real-time data of pipelines, including age, material type, environmental factors, etc.; clean data, process missing values ​​and outliers, and ensure data quality; Model development: Extract key features from the data, such as pipeline conditions and historical damage records, select appropriate machine learning algorithms, such as random forests or neural networks, train the model using labeled data, and optimize model performance through cross-validation; Assessment and Prioritization: The model predicts the extent of pipeline damage based on input features, combines the extent of damage with pipeline key indicators, calculates the repair priority, and designs an intuitive user interface to display the assessment results and priorities, ensuring that the system can receive and process data in real time and update the assessment; Continuous optimization: Monitor model performance, continuously optimize the model based on user feedback and new data, provide evaluation reports and maintenance recommendations, and assist in decision-making.

8. The trenchless pipeline repair process according to claim 1, characterized in that: The intelligent pipeline monitoring system in S09 mainly includes the following steps: Sensor deployment: Install sensors such as pressure, flow and leakage sensors at key locations in pipelines to collect data in real time; Data transmission: Transmit sensor data to the central monitoring system through wireless communication technology; Data analytics: Using machine learning algorithms to analyze data and identify unusual patterns, such as leaks or pressure anomalies; Real-time warning: Once an abnormality is detected, the system automatically triggers an early warning and notifies relevant personnel through multiple channels (SMS, application push, etc.); GIS integration: Integration with geographic information system (GIS) to accurately locate problem areas and improve response speed; System maintenance: Regularly update analytical models and calibrate sensors to ensure the long-term accuracy and reliability of the monitoring system.

9. A pipeline trenchless repair process according to claim 8, characterized in that: The method for identifying abnormal patterns is: Scaling the data to a uniform scale, such as a range of 0 to 1, or to have zero mean and unit variance; Where X is the original data, μ is the mean, and σ is the standard deviation. Features that are helpful for anomaly detection are extracted from the original data. The principal component analysis (PCA) and recursive feature elimination (RFE) algorithms are used to select the most influential features. The optimal hyperplane is found in the feature space to distinguish normal and abnormal data. The accuracy, recall, F1 score and other indicators are used to evaluate the anomaly detection ability of the model. A threshold is set according to the output of the model, and data points exceeding the threshold are considered abnormal.

Images