Gas field robot inspection method and system based on digital twin model

Abstract

The invention discloses a gas field robot inspection method and system based on a digital twinborn model. The method comprises the steps that the digital twinborn model of a gas field station is constructed in advance; the gas station is inspected through the inspection robot to obtain station operation data, the station operation data is sent to the digital twin model, the inspection result of the gas station is determined through the fault detection model, and a decision basis is provided for emergency response. According to the method, the digital twinning technology, the AI technology and the inspection robot technology are fused, the digital twinning technology is firstly utilized to construct a virtual digital model for the gas station, comprehensive real-time monitoring of facilities such as station equipment and pipelines is achieved, then the inspection robot is utilized to inspect the gas station so as to reduce human input and reduce the inspection risk, and the inspection efficiency of the gas station is improved. And finally, fault detection is performed according to the station operation data acquired by the inspection robot by using an AI technology, so that the accuracy of fault detection is improved.

Description

Technical Field

[0001] This application relates to the technical field of gas inspection, and particularly to a gas field robot inspection method and system based on a digital twin model. Background Art

[0002] With the wide application of gas in the energy field, ensuring the safe operation of gas stations has become crucial. With the increasing demand for gas station inspections, traditional inspections mainly rely on manual inspections and some automated detection equipment. However, traditional manual inspections can only cover a limited area per person per day, and the inspection cycle is long, making it difficult to grasp the operating status of the station in real time. At the same time, manual inspections are difficult to reach in complex terrains or high-altitude areas, resulting in these areas becoming blind spots for inspections. In terms of detection accuracy, manual inspections mainly rely on human senses and simple detection tools, making it difficult to accurately detect minor leaks.

[0003] Although some automated monitoring equipment has improved the inspection efficiency and accuracy to a certain extent, there are still limitations. For example, traditional detection equipment may lack accuracy, especially in detecting minor leaks, and it is difficult to meet the high safety requirements of modern gas stations. In addition, these devices often operate independently, lacking the linkage between systems. On the one hand, they cannot achieve comprehensive real-time monitoring, and on the other hand, their operation is cumbersome, increasing the burden on staff and reducing the speed of emergency response.

[0004] Therefore, the existing technology still needs to be improved. Summary of the Invention

[0005] The technical problem to be solved by this application is to provide a gas field robot inspection method and system based on a digital twin model in view of the deficiencies of the existing technology.

[0006] To solve the above technical problem, the first aspect of this application provides a gas field robot inspection method based on a digital twin model. Specifically, the gas field robot inspection method based on a digital twin model includes: Pre-construct a digital twin model of a gas station; Use an inspection robot to inspect the gas station to obtain the station operation data of the gas station, and send the station operation data to the digital twin model, where the station operation data includes gas equipment operation data and the environmental data of the gas station; Transmit the station operation data through the digital twin model to a trained fault detection model, and determine the inspection result of the gas station through the fault detection model to provide a decision basis for emergency response.

[0007] The described gas field robot inspection method based on a digital twin model, wherein the inspection robot is equipped with a sensing device and audio-visual equipment. Among them, the multiple sensing devices include one or more of a magnetic navigation sensor, a laser obstacle avoidance sensor, a lidar, and an environmental gas detector. The audio-visual equipment includes one or more of a high-definition camera, an infrared camera, an audible and visual alarm, a pick-up, and a loudspeaker.

[0008] The described gas field robot inspection method based on a digital twin model, wherein the inspection of the gas field station by the inspection robot to obtain the operation data of the gas field station specifically includes: Construct an inspection path for the inspection robot, and control the inspection robot to inspect the gas field station according to the inspection path to obtain the operation data of the gas field station; Read the global data of the gas field station at preset time intervals. Among them, the global data includes equipment position information, historical operation data, historical environmental data, and historical inspection paths; According to the obtained global data of the gas field station, use the pre-mounted path planning model through the digital twin model to determine the planned inspection path of the inspection robot. Among them, the path planning model is a neural network model based on deep learning; Take the planned inspection path as the inspection path of the inspection robot, and control the inspection robot to inspect the gas field station according to the inspection path to obtain the operation data of the gas field station.

[0009] The described gas field robot inspection method based on a digital twin model, wherein the control of the inspection robot to inspect the gas field station according to the inspection path to obtain the operation data of the gas field station specifically includes: Control the inspection robot to inspect the gas field station according to the inspection path; Real-time obtain the lidar data and image data in the operation data of the gas field station, and update the inspection path based on the lidar data and image data to provide an obstacle avoidance function for the inspection robot.

[0010] The described gas field robot inspection method based on a digital twin model, wherein the method further includes: Read the historical operation data sequence of the gas field station; Use the pre-mounted fault prediction model through the digital twin model to perform risk prediction on the gas field station through the fault prediction model to obtain a risk prediction result; Perform preventive maintenance operations based on the risk prediction result.

[0011] The described gas field robot inspection method based on a digital twin model, wherein the historical field operation data sequence includes the historical operation sound data sequence of gas equipment; the fault prediction model includes a fault prediction model; the specific process of using the pre-mounted fault prediction model through the digital twin model to perform risk prediction on the gas field through the fault prediction model to obtain a risk prediction result is as follows: Input the historical operation sound data sequence into the fault prediction model, and analyze the historical operation sound data sequence through the fault prediction model to obtain the predicted operation status of the gas equipment; When the predicted operation status is high abnormal risk, regard the gas equipment failure as the risk prediction result; When the predicted operation status is low abnormal risk, regard the normal gas equipment as the risk prediction result.

[0012] The described gas field robot inspection method based on a digital twin model, wherein the method further includes: Present the field operation data and the inspection result in the digital twin model, and visually present the field operation data and the inspection result to the user through the digital twin model.

[0013] The second aspect of the present application provides a gas field robot inspection system based on a digital twin model, wherein the gas field robot inspection system based on a digital twin model specifically includes: A construction module for pre-constructing a digital twin model of a gas field; An acquisition module for inspecting the gas field through an inspection robot to obtain the field operation data of the gas field, and sending the field operation data to the digital twin model, wherein the field operation data includes gas equipment operation data and environmental data of the gas field; A fault detection module for transmitting the field operation data through the digital twin model to a trained fault detection model, and determining the inspection result of the gas field through the fault detection model to provide a decision-making basis for emergency response.

[0014] The third aspect of the present application provides a computer-readable storage medium storing one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in any of the above-described gas field robot inspection methods based on a digital twin model.

[0015] The fourth aspect of the present application provides a terminal device, which includes: a processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, it implements the steps in any of the above-described gas field robot inspection methods based on the digital twin model.

[0016] Beneficial effects: Compared with the prior art, the present application provides a gas field robot inspection method and system based on a digital twin model. The method includes pre-constructing a digital twin model of a gas field station; inspecting the gas field station by an inspection robot to obtain the operation data of the gas field station, and sending the operation data of the gas field station to the digital twin model; transmitting the operation data of the gas field station to a trained fault detection model through the digital twin model, and determining the inspection result of the gas field station through the fault detection model to provide a decision-making basis for emergency response. The present application integrates digital twin technology, AI technology, and inspection robot technology. First, the digital twin technology is used to construct a virtual digital model for the gas field station to achieve comprehensive real-time monitoring of facilities such as station equipment and pipelines. Then, the inspection robot is used to inspect the gas field station to reduce labor input and inspection risks. Finally, the AI technology is used to perform fault detection based on the operation data collected by the inspection robot, improving the accuracy of fault detection. Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the gas field robot inspection method based on the digital twin model provided by the embodiment of the present application.

[0019] Figure 2 It is a flowchart of the planning process of the inspection path of the inspection robot.

[0020] Figure 3 It is a schematic block diagram of the gas field robot inspection system based on the digital twin model provided by the embodiment of the present application.

[0021] Figure 4 It is a schematic block diagram of the terminal device provided by the embodiment of the present application. Detailed Embodiments

[0022] The embodiments of the present application provide a method and system for robotic inspection of gas fields based on digital twin models. To make the objectives, technical solutions, and effects of the present application clearer and more explicit, the following further elaborates on the present application with reference to the accompanying drawings and by way of examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0023] Those skilled in the art of this technology can understand that, unless specifically stated, the singular forms "a", "an", "the", and "said" used herein may also include the plural forms. It should be further understood that the term "including" used in the specification of the present application means the presence of the described features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or their groups. It should be understood that when we say an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The phrase "and / or" used herein includes all or any unit and all combinations of one or more related listed items.

[0024] Those skilled in the art of this technology can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as the general understanding of those of ordinary skill in the art to which the present application belongs. It should also be understood that terms such as those defined in a general dictionary should be understood to have a meaning consistent with the meaning in the context of the prior art, and will not be interpreted with an idealized or overly formal meaning unless specifically defined as here.

[0025] It should be understood that the sequence numbers and magnitudes of the steps in this embodiment do not imply the order of execution. The order of execution of each process is determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0026] Through research, it has been found that with the widespread application of gas in the energy field, ensuring the safe operation of gas stations has become crucial. With the increasing demand for inspection of gas stations, traditional inspections mainly rely on manual inspections and some automated detection equipment. However, traditional manual inspections can only cover a limited area per person per day, and the inspection cycle is long, making it difficult to grasp the operation status of the station in real time. At the same time, manual inspections are difficult to reach in complex terrains or high-altitude areas, resulting in these areas becoming blind spots for inspections. In terms of detection accuracy, manual inspections mainly rely on human senses and simple detection tools, making it difficult to accurately detect minor leaks and the like.

[0027] Although some automated monitoring devices have improved the inspection efficiency and accuracy to a certain extent, they still have limitations. For example, traditional detection devices may lack precision, especially in detecting minor leaks, and it is difficult to meet the high safety requirements of modern gas stations. In addition, these devices often operate independently, lacking interaction between systems. On the one hand, they cannot achieve comprehensive real-time monitoring. On the other hand, their cumbersome operation increases the burden on staff and reduces the speed of emergency response.

[0028] To solve the above problems, this application is committed to researching the integration of digital twin technology, AI technology, and inspection robots. By integrating digital twin technology with artificial intelligence and inspection robot technology, it is possible to achieve efficient and precise inspections of gas stations, thereby enhancing the safety and reliability of the stations. Digital twin technology can build a virtual model highly consistent with the actual physical environment for gas stations, enabling comprehensive real-time monitoring of infrastructure such as station equipment and pipelines. By simulating different operating scenarios in the virtual model, potential faults and safety risks can be identified in advance, providing strong guidance for actual inspections. Artificial intelligence technology can deeply analyze and process a large amount of inspection data to achieve intelligent diagnosis and emergency decision-making support. For example, by training historical inspection data with machine learning algorithms, a fault prediction model can be constructed to predict pipeline faults in advance and fundamentally avoid risks. In addition, artificial intelligence can also achieve intelligent control of inspection robots, further improving the efficiency and accuracy of inspections. Introducing inspection robot technology can replace traditional manual inspections, reduce labor requirements, and lower risks during inspections. Inspection robots can be equipped with various sensors, such as methane detectors, temperature sensors, pressure sensors, etc., to comprehensively detect gas stations. At the same time, inspection robots can also achieve autonomous navigation and obstacle avoidance functions, adapting to complex station environments.

[0029] Based on this, in the embodiment of this application, a digital twin model of a gas station is pre-constructed; an inspection robot is used to inspect the gas station to obtain the station operation data of the gas station, and the station operation data is sent to the digital twin model; the digital twin model transmits the station operation data to a trained fault detection model, and the inspection result of the gas station is determined by the fault detection model to provide a decision basis for emergency response. By integrating digital twin technology, AI technology, and inspection robot technology, this application first uses digital twin technology to build a virtual digital model for a gas station to achieve comprehensive real-time monitoring of station equipment, pipelines, and other facilities. Then, inspection robots are used to inspect the gas station to reduce labor input and inspection risks. Finally, AI technology is used to perform fault detection based on the station operation data collected by inspection robots to improve the accuracy of fault detection.

[0030] The following further describes the application content by describing the embodiments in conjunction with the accompanying drawings.

[0031] This embodiment provides a gas field robot inspection method based on a digital twin model, as Figure 1 shown, the method includes: S10. Pre-build a digital twin model of a gas station.

[0032] Specifically, the digital twin model is a digital model built for a gas station, which realizes comprehensive real-time monitoring of infrastructure such as internal equipment and pipelines in the station. For example, using digital twin technology, it is possible to monitor and inspect key equipment such as pipeline pressure, oil and gas pipeline structure, storage tanks, and valves in real time, and at the same time monitor core data such as gas inventory, pipeline flow, and personnel data to ensure the stability and safety of operations. In addition, this digital twin model can also monitor the station environment in real time to provide environmental support for operations.

[0033] Furthermore, the process of building the digital twin model is as follows: First, taking the gas station as a blueprint, collect the basic data of gas equipment and pipelines in the gas station, and then create a static topology model of the gas station and the state models of each gas equipment based on the basic data, and build a basic model of the gas station based on the static topology model and the equipment state model. Second, use the historical operation data of the gas station to develop a fault detection model and a fault prediction model using AI technology. Among them, the fault detection model is used to detect faults based on the operation data of the station collected at the current time, and the fault prediction model is used to predict faults based on the historical operation data sequence within a preset time period. The historical operation data of the gas station can include the historical operation data of gas equipment and the historical environment data of the gas station. Third, use AI technology to develop a path planning model for the global data of the gas station. Among them, the global data of the gas station can include the location information, historical operation data, historical environment data, and historical inspection path of gas equipment. Finally, integrate the basic model, fault detection model, fault prediction model, and path planning model to form a digital twin model of the gas station. After the construction is completed, use this digital twin model to monitor the gas station in real time and obtain real-time monitoring data.

[0034] In the management and operation of gas stations, the collection and collation of basic data is a crucial link. The basic data covers detailed information about gas equipment and pipelines, including equipment location information, equipment operation parameters, pipeline layout information, the connection relationship between gas equipment and gas pipelines, and gas pipeline operation parameters, etc. The static topology model reveals the spatial layout and connection method between gas equipment and gas pipelines. It is constructed based on equipment location information, pipeline layout information, and the connection relationship between gas equipment and gas pipelines. Through the static topology model, the relative positions of each gas equipment and gas pipeline in the physical space and their connection relationships can be clearly seen, facilitating the understanding of the structural layout of the gas station and the real-time monitoring of the gas station.

[0035] The equipment status model shows the real-time operation status of gas equipment and gas pipelines in the gas station. The equipment status model includes the gas equipment status model and the pipeline status model, which are constructed based on equipment operation parameters and gas pipeline operation parameters. Through the equipment status model, the operation status of gas equipment and gas pipelines can be understood in a timely manner, potential problems can be predicted, and corresponding preventive measures can be taken to ensure the stable operation of the entire gas station. Among them, the equipment operation parameters can include pressure, temperature, flow rate, and equipment status, etc.

[0036] The fault detection model covers the gas equipment fault detection model and the environmental detection model. The gas equipment fault detection model aims to conduct fault prediction and detection based on equipment operation data, while the environmental monitoring model uses the environmental data of the gas station for fault prediction. The gas equipment fault detection model is obtained by taking the historical operation data of gas equipment as training samples and conducting in-depth learning training on the neural network model. Similarly, the environmental detection model is formed by taking historical environmental data as training samples and conducting in-depth learning training on the neural network model.

[0037] The path planning model is used to plan the inspection path of the inspection robot to ensure that the inspection robot can efficiently complete the inspection task. The path planning model is developed using AI technology based on equipment location information, historical operation data, historical environmental data, and historical inspection paths. The path planning model takes into account various factors such as the structural layout of the gas station, equipment distribution, and potential risk areas, and plans the optimal inspection path for the inspection robot. During the inspection process, the inspection robot can navigate autonomously according to the planned path, avoiding collisions and obstacles, ensuring the safety and efficiency of the inspection. At the same time, the path planning model can also be dynamically adjusted according to the actual situation to adapt to the changes and requirements of the gas station.

[0038] It can be seen from this that the digital twin model in the embodiments of the present application is constructed by combining artificial intelligence and digital twin technology. The digital twin technology provides rich data sources for AI. Through the digital twin model, AI can obtain the real-time operation data of the gas station, including equipment status, pipeline flow, environmental parameters, etc. These data provide strong support for the analysis and decision-making of AI. Secondly, AI can optimize the digital twin model, which can improve the accuracy and reliability of the digital twin model. For example, AI can predict the fault trend of equipment by analyzing the equipment operation data, and arrange the maintenance plan in advance, so as to reduce the impact of equipment failure on the operation of the station. In addition, the integration of AI and digital twin can also achieve intelligent inspection. The inspection robot, under the control of AI, can plan the inspection route according to the information provided by the digital twin model, improving the inspection efficiency. At the same time, AI can analyze the data collected by the inspection robot in real time, discover abnormal situations in time, and take corresponding measures.

[0039] S20. Inspect the gas station by the inspection robot to obtain the station operation data of the gas station, and send the station operation data to the digital twin model.

[0040] Specifically, the station operation data is collected by the inspection robot and covers the operation data of gas equipment and environmental data in the gas station. In other words, when the inspection robot inspects the gas station, it will collect data to construct the station operation data of the gas station, and the station operation data may include the operation data of gas equipment and the environmental data of the gas station. Among them, the inspection robot can be equipped with a sensing device and audio-visual equipment to assist in collecting the station operation data of the gas station. The sensing device can include one of a magnetic navigation sensor, a laser obstacle avoidance sensor, a lidar, and an environmental gas monitor, but is not limited to a magnetic navigation sensor, a laser obstacle avoidance sensor, a lidar, and an environmental gas detector. The audio-visual equipment can include one or more of a high-definition camera, an infrared camera, an audible and visual alarm, a microphone, and a loudspeaker, but is not limited to including a high-definition camera, an infrared camera, an audible and visual alarm, a microphone, and a loudspeaker.

[0041] Exemplarily, the inspection robot is configured with an explosion-proof chassis, a communication antenna, an audible and visual alarm, a pickup, a loudspeaker, a magnetic navigation sensor, a laser obstacle avoidance sensor, a lidar, a high-definition camera, and an infrared camera. Among them, the explosion-proof chassis is used to provide a stable and reliable working environment for the automatic control device, battery, and drive motor, ensuring safe operation in dangerous environments such as gas stations. The communication antenna is used for real-time communication with the digital twin model to send the station operation data to the digital twin model and receive the control instructions sent by the digital twin model. The pickup and loudspeaker are used for real-time intercom and remote shouting, and the audible and visual alarm is used for audible and visual alarm in case of abnormalities. The magnetic navigation sensor is used to ensure that the inspection robot can accurately navigate in the complex station environment, improving the coverage and accuracy of the inspection. The laser obstacle avoidance sensor is used to detect low-position obstacles, effectively avoiding collisions between the inspection robot and obstacles during the inspection process, ensuring the smooth progress of the inspection work. The lidar is used to construct a three-dimensional map to achieve autonomous obstacle avoidance. The high-definition camera and the infrared camera are used for taking pictures and videos to obtain temperature field images of gas equipment and environmental images, etc. The environmental gas monitor is used to detect the environmental data of the gas station, such as methane, carbon monoxide, ammonia, sulfur dioxide, etc. in the environment.

[0042] In addition, the inspection robot can be configured with an autonomous obstacle avoidance function and a path planning function, that is, the above path planning model can be deployed in the digital twin model or can be deployed in the inspection robot at the same time. The inspection robot can determine whether to perform path planning through the digital twin model or through the path planning model carried by itself according to its operating load. For example, when the operating load is greater than the preset load threshold, path planning is performed through the digital twin model. When the operating load is less than or equal to the preset load threshold, path planning is performed through the path planning model carried by itself. This method can flexibly select the path planning method according to the actual load situation, improving the inspection efficiency. When the operating load is heavy, using the digital twin model for path planning can more accurately determine the optimal path, reducing the energy consumption and time cost of the inspection robot. When the operating load is light, planning through the path planning model carried by the inspection robot itself can reduce the dependence on the digital twin model and improve the autonomy and flexibility of the inspection robot.

[0043] Of course, in practical applications, other methods can also be used to determine the execution entity of path planning. For example, it can be determined based on the running load and remaining computing resources. Specifically, first, it is determined whether the remaining computing resources meet the running requirements of the path planning model. When the requirements are not met, path planning is performed through the digital twin model. When the requirements are met, the inspection task level configured for the inspection robot is obtained, and a preset load threshold is selected according to the inspection task level and the remaining computing resources. Finally, the running load is compared with the selected preset load threshold. When the running load is less than or equal to the preset load threshold, path planning is performed through the path planning model carried by itself. When the running load is greater than the preset load threshold, path planning is preferentially performed through the digital twin model. In this way, a more intelligent selection of the appropriate path planning method can be made according to the actual situation of the inspection robot. If the remaining computing resources are insufficient, directly using the digital twin model for path planning can ensure the accuracy and reliability of path planning and avoid planning failures or poor planning results caused by insufficient computing resources. If the remaining computing resources are sufficient, the preset load threshold can be flexibly adjusted according to the inspection task level and the remaining computing resources, so as to improve the autonomy and flexibility of the inspection robot as much as possible on the premise of ensuring the quality of path planning. This path planning method that comprehensively considers the running load, remaining computing resources, and inspection task level can further improve the inspection efficiency and intelligent level of the inspection robot.

[0044] The following takes path planning using the path planning model configured through the twin data model as an example for illustration.

[0045] Exemplarily, as Figure 2 shown, the inspection of the gas station by the inspection robot to obtain the station operation data of the gas station specifically includes: S21. Construct an inspection path for the inspection robot and control the inspection robot to inspect the gas station according to the inspection path to obtain the station operation data of the gas station; S22. Read the global data of the gas station at preset time intervals; S23. According to the obtained global data of the gas station, use the pre-carried path planning model through the digital twin model to determine the planned inspection path of the inspection robot, where the path planning model is a neural network model based on deep learning; S24. Use the planned inspection path as the inspection path of the inspection robot and control the inspection robot to inspect the gas station according to the inspection path to obtain the station operation data of the gas station.

[0046] Specifically, the inspection path is the path adopted when the inspection robot starts the inspection, that is, the initial inspection path of the inspection robot. It can be constructed based on the global data at the initial inspection or preset by the user, etc. After the inspection robot starts the inspection, it will perform the inspection according to this inspection path and collect the global data of the gas station in real time.

[0047] During the inspection process of the inspection robot, the global data of the gas station will be read every preset time interval. The global data is the data within a preset data time period before the acquisition time. The global data includes but is not limited to equipment location information, historical operation data, historical environmental data, and historical inspection paths. Then, based on the global data, path planning is carried out through the digital twin model using the pre-mounted path planning model to obtain the planned inspection path of the inspection robot. This application uses a path planning model based on deep learning technology for path planning and can comprehensively consider various factors such as the structural layout of the gas station, equipment distribution, and potential risk areas to plan the optimal inspection path for the inspection robot. This can not only improve the inspection efficiency but also ensure the safe operation of the inspection robot in a complex environment.

[0048] Furthermore, after obtaining the planned inspection path of the inspection robot, the planned inspection path can be directly used as the inspection path of the inspection robot, or it can be determined whether to use this planned inspection path as the inspection path of the inspection robot through interaction with the user. It can also be to first detect whether there is a user-specified planned path. When there is a user-specified planned path, this planned inspection path can be abandoned and the user-specified planned path can be continued to be executed; or the latest inspection result can be read. If there is no risk area, the user-specified planned path can be continued to be executed. If there is a risk area and the planned inspection path includes the risk area, the planned inspection path will be used as the inspection path. If there is a risk area but the planned inspection path does not include the risk area, then with the inspection nodes in the risk area as the constraints of the inspection path, a new inspection path will be planned for the inspection robot.

[0049] When re-planning the inspection path, various factors such as the location of the risk area, the current location of the inspection robot, the remaining battery power, and the urgency of the inspection task can be comprehensively considered to ensure that the inspection path can cover all important risk areas and efficiently complete the inspection task. At the same time, the re-planned inspection path also needs to consider the kinematic constraints and environmental constraints of the inspection robot to ensure that the inspection robot can operate safely and stably during the actual inspection process. After determining the inspection path, the inspection robot will perform inspections according to the new inspection path, enabling the inspection robot to prioritize inspections of risk areas, and then quickly perform re-fault detection on the risk areas to clarify whether there are risks, enabling users to quickly adopt emergency responses to the risk areas and improving the accuracy of emergency responses.

[0050] In one implementation, controlling the inspection robot to perform inspections on the gas station according to the inspection path to obtain the operation data of the gas station specifically includes: Controlling the inspection robot to perform inspections on the gas station according to the inspection path; Real-time obtaining the lidar data and image data in the operation data of the gas station, and updating the inspection path based on the lidar data and image data to provide an obstacle avoidance function for the inspection robot.

[0051] Specifically, the lidar data can be collected by the lidar carried by the inspection robot, and the image data is obtained by the high-definition camera and / or infrared camera carried by the inspection robot. Based on the lidar data and image data, obstacles in the current scene where the inspection robot is located can be detected. Then, updating the inspection path according to the lidar data and image data can enable the inspection robot to avoid obstacles in the current scene. Among them, the process of updating the inspection path based on the lidar data and image data can be: performing obstacle detection based on the lidar data and image data to obtain obstacle position information; then detecting whether the obstacle position information is on the inspection path. When it is on the inspection path, taking the obstacle as the center, re-planning an inspection path segment that bypasses the obstacle, and using this inspection path segment to replace the corresponding inspection path segment in the inspection path to update the inspection path, and controlling the inspection robot to perform inspections according to the updated inspection path, which can not only avoid obstacles but also maintain the original inspection efficiency and coverage as much as possible. At the same time, the inspection robot will also update its internal environmental map in real time and add the newly detected obstacle information to the map to better perform obstacle avoidance and path planning in future inspection tasks.

[0052] S30. Transmit the station operation data through the digital twin model to the trained fault detection model, and determine the inspection result of the gas station through the fault detection model, so as to provide a decision-making basis for emergency response.

[0053] Specifically, the fault detection model is used to perform fault detection on the station operation data collected at the current moment to obtain the inspection result. Among them, the inspection result may include the fault situation of gas equipment, the risk area in the gas station yard, and the fault situation of gas pipelines. That is to say, through the fault detection model, the gas equipment, gas pipelines with faults and the risk areas with fault risks (for example, areas where the sulfur dioxide concentration is higher than the preset concentration threshold, etc.) in the gas station yard can be detected. In this application, the inspection result of the gas station is determined through the fault detection model, which can monitor the operation status of the gas station in real time and timely discover fault problems, providing a decision-making basis for emergency response.

[0054] In one implementation, the gas station yard robot inspection method based on the digital twin model further includes: Read the historical station operation data sequence of the gas station; Through the digital twin model, use the pre-mounted fault prediction model, and perform risk prediction on the gas station through the fault prediction model to obtain the risk prediction result; Perform preventive maintenance operations based on the risk prediction result.

[0055] Specifically, the historical station operation data sequence may include the historical station operation data between the current reading moment and the previous reading time, or may be the historical station operation data within multiple preset times before the current reading moment. Among them, the historical station operation data may include historical operation data and historical environmental data. The fault prediction model is pre-constructed and used to perform risk prediction on the gas station. Among them, the trigger condition for using the pre-mounted fault prediction model through the digital twin model may be that the inspection result is normal, that is, there are no faults in gas equipment and gas pipelines, and there are no risk areas. That is to say, when the inspection result obtained by using the fault detection model is normal, then use the fault prediction model to perform fault prediction on the gas station, pre-judge the gas equipment faults, gas pipeline faults and risk areas in advance, discover potential faults and safety hazards, and thus avoid risks from the source.

[0056] Of course, in practical applications, in order to reduce the computing resources required by the digital twin model, when the inspection result obtained by using the fault detection model is normal, risk assessment can be carried out based on the operation data of the station (for example, comparing each data item in the operation data of the station with its corresponding risk threshold). When the assessment result of the risk assessment indicates the existence of risks, the historical operation data sequence of the gas station is read, and fault prediction is performed through the fault prediction model to further confirm whether there are potential fault hazards. This can minimize unnecessary fault prediction operations on the premise of ensuring the safe operation of the gas station, thereby reducing the consumption of computing resources of the digital twin model. In addition, when the time interval between the current time and the execution time of the previous execution of the fault prediction model reaches a preset duration, the historical operation data sequence of the gas station can be triggered to be read, so as to perform a risk prediction through the fault prediction model to obtain a risk prediction result.

[0057] In one implementation, the fault prediction model uses the historical operation sound data of the gas equipment for risk prediction. In other words, the historical operation data sequence of the station includes the historical operation sound data sequence of the gas equipment. Correspondingly, the digital twin model uses the pre-mounted fault prediction model to perform risk prediction on the gas station through the fault prediction model to obtain the risk prediction result specifically as follows: Input the historical operation sound data sequence into the fault prediction model, and analyze the historical operation sound data sequence through the fault prediction model to obtain the predicted operation state of the gas equipment; When the predicted operation state is a high abnormal risk, the gas equipment failure is used as the risk prediction result; When the predicted operation state is a low abnormal risk, the normal operation of the gas equipment is used as the risk prediction result.

[0058] Specifically, the historical operation sound data sequence is collected by a pickup carried by a robot, which is the sound data generated during the operation of the gas equipment. By analyzing this sound data, it can be determined whether the gas equipment has a fault. Furthermore, risk prediction can be carried out based on the historical operation sound data sequence to obtain the predicted operation state of the gas equipment, and the risk prediction result will be determined according to the predicted operation state. Specifically, when the predicted operation state is a high abnormal risk, the gas equipment failure is used as the risk prediction result, and when the predicted operation state is a low abnormal risk, the normal operation of the gas equipment is used as the risk prediction result.

[0059] It should be noted that the fault prediction model can also perform risk prediction based on other data of the station operation data. For example, by analyzing historical gas monitoring data, a gas leakage prediction model is established. When new gas monitoring data is input, the model will judge whether there is a risk of gas leakage according to the change trend of gas concentration and other relevant parameters; or, combining image recognition technology and temperature monitoring data, using the image data collected by high-definition cameras and the temperature data detected by infrared cameras to perform risk prediction on the operation status of equipment to determine whether there is a fault risk in gas equipment. For this reason, in a typical implementation manner, the fault prediction model uses multi-modal data for risk prediction. The multi-modal data includes historical operation sound data sequences, various gas historical concentration sequences, historical image data sequences, and historical temperature data. Among them, the historical operation sound data sequences, historical image data sequences, and historical temperature data are for the same gas equipment, and the various gas historical concentration sequences are gas concentration data within a preset range centered on the gas equipment.

[0060] Furthermore, when the risk prediction result is a gas equipment fault, the gas equipment can be repaired or updated. In this way, the gas equipment that may have a fault can be repaired or replaced to reduce the probability of faults and improve the safety and stability of the gas station. At the same time, the gas station can also be comprehensively analyzed according to the risk prediction result to determine more reasonable preventive maintenance operations. For example, when the predicted operation status of a gas equipment is a fault, the gas equipment and other gas equipment connected to it can be repaired synchronously, and the inspection robot and manual collaborative inspection can be carried out within a preset range centered on the gas equipment, etc., to further improve the operation efficiency and safety of the gas station and provide a strong guarantee for the long-term stable operation of the gas station.

[0061] In one implementation manner, after obtaining the station operation data, in addition to performing fault detection based on the station operation data, the station operation data can also be synchronously presented in the digital twin model. For this reason, the gas field robot inspection method based on the digital twin model further includes: Presenting the station operation data and the inspection result in the digital twin model, and visually presenting the station operation data and the inspection result to the user through the digital twin model.

[0062] Specifically, the digital twin model is a visualization model. Through the digital twin model, the station operation data can be presented to the user in real time, enabling the user to intuitively understand the operation status of the station through the digital twin platform and timely discover potential problems. For example, through the three-dimensional visualization model, the layout and operation of equipment can be clearly seen, as well as the flow direction and pressure change of pipelines, improving the efficiency and accuracy of management.

[0063] In summary, the gas field robot inspection method based on the digital twin model in the embodiments of the present application realizes comprehensive real-time monitoring and efficient inspection of gas stations by integrating digital twin technology, AI technology, and inspection robot technology. The AI technology endows the system with the ability of intelligent analysis and judgment, and can quickly and accurately identify various abnormal situations. The introduction of robot technology replaces traditional manual inspections, reducing labor costs and inspection risks. For example, through the digital twin model, AI can obtain rich data sources and intelligently control the operation of inspection robots to achieve efficient and accurate inspections. At the same time, the inspection robot can simultaneously obtain various combustible gases, the appearance information of gas equipment, the temperature information of gas equipment, and the operating sound information of gas equipment, etc., and can comprehensively monitor the operating status of gas stations, improving the accuracy and reliability of inspections. At the same time, the inspection robot can also independently plan inspection routes and automatically avoid obstacles, improving the safety and timeliness of inspections.

[0064] Based on the above gas field robot inspection method based on the digital twin model, this embodiment provides a gas field robot inspection system based on the digital twin model, as Figure 3 shown. The gas field robot inspection system based on the digital twin model specifically includes: A construction module 100, configured to pre-construct a digital twin model of a gas station; An acquisition module 200, configured to inspect the gas station through an inspection robot to obtain the station operation data of the gas station, and send the station operation data to the digital twin model, where the station operation data includes gas equipment operation data and environmental data of the gas station; A fault detection module 300, configured to transmit the station operation data through the digital twin model to a trained fault detection model, and determine the inspection result of the gas station through the fault detection model, so as to provide a decision-making basis for emergency response.

[0065] Based on the above gas field robot inspection method based on the digital twin model, the present application also provides a terminal device, as Figure 4 shown. It includes at least one processor 20; a display screen 21; and a memory 22, and may further include a communication interface 23 and a bus 24. Among them, the processor 20, the display screen 21, the memory 22, and the communication interface 23 can communicate with each other through the bus 24. The display screen 21 is set to display a preset user guidance interface in the initial setting mode. The communication interface 23 can transmit information. The processor 20 can call the logical instructions in the memory 22 to execute the method in the above embodiments.

[0066] In addition, when the logic instructions in the above-mentioned memory 22 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0067] As a computer-readable storage medium, the memory 22 can be set to store software programs and computer-executable programs, such as the program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 20 executes functional applications and data processing by running the software programs, instructions or modules stored in the memory 22, that is, implements the methods in the above embodiments.

[0068] The memory 22 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the terminal device, etc. In addition, the memory 22 may include high-speed random access memory and may also include non-volatile memory. For example, various media that can store program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks or optical discs can also be transient storage media.

[0069] In addition, the specific processes of loading and executing multiple instructions by the above-mentioned storage medium and the instruction processor in the terminal device have been described in detail in the above method, and will not be repeated here one by one.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A gas field robot inspection method based on a digital twin model, characterized in that, The described gas field robot inspection method based on a digital twin model specifically includes: Pre - construct a digital twin model of a gas station; Use an inspection robot to inspect the gas station to obtain the station operation data of the gas station, and send the station operation data to the digital twin model, where the station operation data includes gas equipment operation data and environmental data of the gas station; The digital twin model transmits the station operation data to a trained fault detection model, and the fault detection model determines the inspection result of the gas station to provide a decision - making basis for emergency response.

2. The gas field robot inspection method based on the digital twin model according to claim 1, wherein, The inspection robot is equipped with a sensing device and audio - visual equipment. Among them, multiple sensing devices include one or more of a magnetic navigation sensor, a laser obstacle avoidance sensor, a lidar, and an environmental gas detector, and the audio - visual equipment includes one or more of a high - definition camera, an infrared camera, an audible and visual alarm, a microphone, and a loudspeaker.

3. The method for robot inspection and patrol of a gas field based on a digital twin model according to claim 1, wherein The process of using the inspection robot to inspect the gas station to obtain the station operation data of the gas station specifically includes: Construct an inspection path for the inspection robot, and control the inspection robot to inspect the gas station according to the inspection path to obtain the station operation data of the gas station; Read the global data of the gas station at preset time intervals, where the global data includes equipment position information, historical operation data, historical environmental data, and historical inspection paths; According to the obtained global data of the gas station, the digital twin model uses a pre - installed path planning model to determine the planned inspection path of the inspection robot, where the path planning model is a neural network model based on deep learning; Use the planned inspection path as the inspection path of the inspection robot, and control the inspection robot to inspect the gas station according to the inspection path to obtain the station operation data of the gas station.

4. The method for robot inspection of gas field based on digital twin model according to claim 3, wherein The process of controlling the inspection robot to inspect the gas station according to the inspection path to obtain the station operation data of the gas station specifically includes: Control the inspection robot to inspect the gas station according to the inspection path; Real - time obtain the lidar data and image data in the station operation data of the gas station, and update the inspection path based on the lidar data and image data to provide an obstacle avoidance function for the inspection robot.

5. The method for inspecting a gas field robot based on a digital twin model according to claim 1, wherein, The method also includes: Read the historical station operation data sequence of the gas station; The digital twin model uses a pre - installed fault prediction model, and the fault prediction model performs risk prediction on the gas station to obtain a risk prediction result; Perform preventive maintenance operations based on the risk prediction result.

6. The method for robot inspection and patrol of gas field based on digital twin model according to claim 5, characterized in that, The historical station operation data sequence includes the historical operation sound data sequence of gas equipment; the fault prediction model includes a fault prediction model; the process of the digital twin model using a pre - installed fault prediction model and the fault prediction model performing risk prediction on the gas station to obtain a risk prediction result is specifically: Input the historical operation sound data sequence into the fault prediction model, and analyze the historical operation sound data sequence through the fault prediction model to obtain the predicted operation state of the gas equipment; When the predicted operation state is a high abnormal risk, regard the gas equipment failure as the risk prediction result; When the predicted operation state is a low abnormal risk, regard the normal operation of the gas equipment as the risk prediction result.

7. The method for robot inspection of gas field based on digital twin model according to claim 1, characterized in that The method further includes: Present the station operation data and the inspection results in the digital twin model, and visually present the station operation data and the inspection results to the user through the digital twin model.

8. A gas field robot inspection system based on a digital twin model, characterized in that, The gas field robot inspection system based on the digital twin model specifically includes: A construction module for pre-constructing a digital twin model of a gas station; An acquisition module for inspecting the gas station through an inspection robot to obtain the station operation data of the gas station, and sending the station operation data to the digital twin model, wherein the station operation data includes gas equipment operation data and environmental data of the gas station; A fault detection module for transmitting the station operation data through the digital twin model to a trained fault detection model, and determining the inspection results of the gas station through the fault detection model to provide a decision basis for emergency response.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the gas field robot inspection method based on the digital twin model according to any one of claims 1-7.

10. A terminal device, characterized in that, Including: A processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, it implements the steps in the gas field robot inspection method based on the digital twin model according to any one of claims 1-7.

Images