Routing inspection auxiliary method based on AR augmented reality

Abstract

The invention relates to an AR (augmented reality)-based inspection auxiliary method. The method comprises a deployment step and an inspection step, the deployment step comprises the following steps: configuring AR equipment for each inspector, planning an inspection route of each inspector, setting equipment to be inspected and an inspection sequence thereof according to the inspection route, and generating an inspection task list of each inspector; the inspection step comprises the following steps: identifying equipment on an inspection route in real time through AR equipment, mapping the position of the equipment to an AR view of inspection personnel in real time, and marking the position as a highlight virtual identifier; the method comprises the following steps: collecting data of each device in real time, comparing the device data collected in real time with preset standard data or evaluating the state of the device through an anomaly detection algorithm, if the device data is abnormal or the device is in a fault state, giving an alarm in real time, and generating a standardized maintenance process through natural language processing to assist inspection personnel in maintenance. Compared with the prior art, the power plant can be provided with few people, and the overall safety and stability of the power plant are enhanced.

Description

Technical Field

[0001] This invention relates to the field of thermal power generation equipment inspection technology, and in particular to an inspection assistance method based on AR augmented reality. Background Technology

[0002] Traditional thermal power generation, as the backbone of the energy industry, plays a vital role in social welfare and economic development. With the continuous introduction of relevant policies, thermal power plants are gradually transforming towards larger units, lower coal consumption, reduced manpower, and intelligent operation. Therefore, effectively improving the overall efficiency of inspections and strengthening effective supervision of operational inspections will be a crucial aspect of this transformation.

[0003] Current power plant operation and inspection rely heavily on manual inspections. Since most thermal power plants were built a long time ago, their level of intelligence is relatively low. Due to cost and work requirements, most thermal power plants cannot be shut down for large-scale modernization. In order to ensure the safety of the plant, a large amount of manpower can only be used for inspections. However, during the inspection, there is a problem that non-technical personnel cannot correctly deal with faulty instruments.

[0004] Some thermal power plants have introduced a QR code-integrated NFC tag-based inspection method, combining data acquisition and inspection check-in. This allows for smart inspections via mobile phone scanning and NFC tag interaction, which can speed up inspections to some extent and provide pre-set handling procedures when encountering faulty equipment. However, the number of QR code NFC tags is limited, allowing only some critical equipment to be marked; inspection personnel cannot access current equipment data for closed-loop monitoring of the entire production process; and for inspection check-in, interaction is limited to tagged equipment, hindering effective planning and supervision of inspection routes, resulting in some missed or under-inspected equipment (equipment without QR code tags). Furthermore, the scanning process also involves communication signal issues. Scanning relies on network access, but poor signal conditions may occur in power plants, making it impossible to obtain sufficient information through scanning during inspections.

[0005] Therefore, there is a need for an auxiliary method that can perform intelligent inspections offline, which can not only connect the production data of the entire process, but also continuously improve the standardization of equipment operation by inspection personnel, and at the same time achieve tracking and recording of the entire inspection process. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the existing technology and provide an AR-based inspection assistance method.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] An AR-based inspection assistance method, the method comprising deployment steps and inspection steps; wherein,

[0009] The deployment steps include:

[0010] Each inspector is equipped with an AR device, their inspection route is planned, and the equipment to be inspected and their inspection order are set according to the inspection route, generating an inspection task list for each inspector.

[0011] The inspection steps include:

[0012] AR devices are used to identify equipment on the inspection route in real time and map the equipment location to the AR field of view of the inspection personnel, marking it as a highlighted virtual icon. Data from each device is collected in real time and compared with preset standard data or evaluated for equipment status through anomaly detection algorithms. If the equipment data is abnormal or the equipment is in a faulty state, an alarm is triggered in real time, and a standardized maintenance process is generated through natural language processing to assist the inspection personnel in maintenance.

[0013] Furthermore, the AR device is integrated into the personal equipment of the inspection personnel, including a safety helmet.

[0014] Furthermore, the process of planning the inspection route includes:

[0015] Based on historical inspection data, obtain the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process. Based on the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process, assign priority to equipment.

[0016] Based on the priority of each device, an improved ant colony algorithm is used to generate inspection routes with the principle of covering the highest priority device in the shortest time.

[0017] Furthermore, the process by which the AR device attempts to identify equipment data on the inspection route in real time includes:

[0018] If the device has an NFC tag, the AR device will automatically identify the NFC tag through the IoT offline identification TAG and obtain the pre-stored device standard data.

[0019] If the device does not have an NFC tag, the AR device will capture the current actual scene in real time, use image recognition technology to obtain the characteristics of the current device, compare and match them with the preset device characteristics, identify the type of the current device, and obtain the pre-stored device standard data.

[0020] Based on preset equipment characteristics, equipment standard data, and real-time captured current actual images, the outline of the equipment is identified, and the equipment outline is highlighted with a virtual label.

[0021] Furthermore, the real-time acquisition of the device data includes:

[0022] Based on AR devices, the current device's shape is captured by images, and image recognition technology is used to obtain the current device's shape feature data.

[0023] Based on the current equipment type, preset inspection steps are obtained through AR devices to assist inspection personnel in inspecting the equipment and obtaining the current equipment's operating data.

[0024] Furthermore, the process of assessing the device status includes:

[0025] Preset standard data comparison: Compare the current device's external feature data with the preset standard device features. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, compare the remote system monitoring data with the data within the current device. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, analyze the trend of changes in the device's operating parameters through longitudinal comparison to determine whether they exceed the historical fluctuation range. If they exceed the range, the device data is determined to be abnormal. Otherwise, the device data is determined to be normal, and the comparison with the preset standard data is completed.

[0026] Anomaly detection algorithm evaluation: The external feature data and operational data are standardized and associated with relevant equipment data; historical operational data of the equipment are acquired and typical operating modes are extracted, the normal fluctuation range of the operational data is calculated, and a normal operating baseline of the equipment is established in conjunction with the equipment design standards; the standardized external feature data and operational data are compared with the normal operating baseline of the equipment. If the current equipment data exceeds the normal fluctuation range of the operational data, the equipment is determined to be in a numerical fault state; if the current equipment data does not exceed the normal fluctuation range, but the trend of change deviates from the normal operating baseline of the equipment, the equipment is determined to be in a trend fault state; otherwise, the equipment is determined to be fault-free, and the anomaly detection algorithm evaluation is completed.

[0027] Furthermore, the anomaly detection algorithm also classifies the equipment fault status according to the severity, scope of impact, and speed of development of the equipment anomaly.

[0028] Furthermore, the process of generating standardized maintenance procedures through the natural language processing includes:

[0029] Based on the type of equipment requiring repair, obtain the corresponding historical repair records and preset repair steps. Using named entity recognition technology in natural language processing, extract key elements from the historical repair record documents and preset repair steps, including fault characteristics, operation steps, tools and materials, and equipment identification. Through syntactic analysis and semantic network construction, parse the logical relationships between the key elements, and use a deep learning model to understand the contextual semantics and construct the semantic relationships between the key elements. Based on the logical and semantic relationships between the key elements, automatically generate a standardized repair process according to the repair logic.

[0030] Furthermore, if the inspection personnel complete the repair of the faulty equipment according to the standardized maintenance process generated by natural language processing, the standardized maintenance process will be stored in the historical maintenance record.

[0031] Furthermore, the AR device is connected to the digital twin platform. During the inspection process, the AR device collects data from each device in real time and uploads the data to the digital twin platform via wireless connection. When the inspection personnel are conducting inspections, the AR device can be used to obtain the location and movement trajectory of the inspection personnel in real time. The actual inspection path is compared with the planned inspection route in real time on the digital twin platform. If a deviation is found, a reminder is pushed to the manager. If the current inspection personnel are in an area with poor wireless signal, the upload of device data and the real-time comparison of the inspection route are delayed.

[0032] Compared with the prior art, the beneficial effects of the present invention include:

[0033] 1. This invention, through the deep integration of AR technology and intelligent algorithms, constructs a fully closed-loop offline inspection system covering inspection planning, equipment identification, status assessment, fault handling, and process supervision. It realizes the full-process data integration of thermal power plant equipment, operation standard guidance, and traceability of inspection trajectories, fundamentally solving the problems of low efficiency, high missed inspection rate, non-standard fault handling, and lagging supervision in traditional manual inspections, and providing a systematic solution for intelligent transformation.

[0034] 2. This invention integrates AR devices into personal equipment such as safety helmets, freeing up the hands of inspection personnel and enabling an immersive inspection experience of "what you see is what you get"; it highlights the location of equipment in real time, avoiding missed inspections and improving the recognition of inspection targets, especially in complex industrial environments, significantly reducing the cognitive load on personnel.

[0035] 3. The improved ant colony algorithm based on equipment priority in this invention can dynamically balance inspection time and equipment importance, ensuring priority coverage of high-risk and highly related equipment. Compared with traditional manual route planning, inspection time can be shortened by 30%-50%, while reducing invalid path duplication and improving the efficiency of inspection resource allocation.

[0036] 4. This invention breaks through the limitation of traditional QR codes / NFC tags being able to mark only some important equipment. On the basis of the original NFC tags, it also adds image recognition function. Dual-mode recognition enables the identification of all equipment in the plant without blind spots. The offline communication design of AR devices solves the problem of data acquisition in power plant signal blind spots, ensuring that inspection information is available in real time and improving equipment coverage by 100%.

[0037] 5. This invention achieves precise fault location by fusing multimodal data from external features, operating parameters, and historical baselines. It can not only detect numerical anomalies but also identify potential trend-based faults. The fault classification mechanism helps inspection personnel efficiently handle problems according to risk levels, and combined with standardized maintenance process push, it enables non-professionals to quickly perform standardized operations.

[0038] 6. In this invention, the digital twin platform can synchronize inspection trajectories and equipment data in real time, allowing managers to remotely monitor the entire process, detect route deviations and issue warnings in real time, and control the rate of missed inspections; the automatic generation of maintenance processes and the iteration of historical records build an enterprise-level equipment maintenance knowledge base, which helps to pass on experience and continuously optimize, forming a virtuous cycle of detection, processing and optimization. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method of the present invention;

[0040] Figure 2 This is a flowchart of the anomaly detection algorithm of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0042] Example 1

[0043] This embodiment discloses an inspection assistance method based on AR (Augmented Reality), the method being as follows: Figure 1 As shown, it includes two parts: deployment steps and inspection steps.

[0044] The deployment steps include:

[0045] A1, equip each inspection personnel with AR devices and plan their inspection routes;

[0046] A2: Based on the inspection route, set the equipment to be inspected and their inspection order, and generate an inspection task list for each inspection personnel.

[0047] The AR device is integrated into the personnel's personal equipment, including a safety helmet. The main feature is that the AR device has a modular design for easy disassembly and replacement.

[0048] The process of planning inspection routes includes:

[0049] Based on historical inspection data, obtain the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process. Based on the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process, assign priority to equipment.

[0050] Based on the priority of each device, an improved ant colony algorithm is used to generate inspection routes with the principle of covering the highest priority device in the shortest time.

[0051] The inspection process includes:

[0052] B1 uses AR devices to identify equipment on the inspection route in real time and maps the equipment location to the AR field of view of the inspection personnel in real time, marking it as a highlighted virtual icon;

[0053] B2, real-time data collection from various devices;

[0054] B3 compares real-time collected equipment data with preset standard data or evaluates equipment status through anomaly detection algorithms. If the equipment data is abnormal or the equipment is in a faulty state, an alarm is triggered in real time, and a standardized maintenance process is generated through natural language processing to assist inspection personnel in maintenance.

[0055] If no alarm is triggered, return to step B1 and continue the inspection.

[0056] In step B1, the process of the AR device attempting to identify equipment data on the inspection route in real time includes:

[0057] If the device has an NFC tag, the AR device will automatically identify the NFC tag through the IoT offline identification TAG and obtain the pre-stored device standard data.

[0058] If the device does not have an NFC tag, the AR device will capture the current actual scene in real time, use image recognition technology to obtain the characteristics of the current device, compare and match them with the preset device characteristics, identify the type of the current device, and obtain the pre-stored device standard data.

[0059] Based on preset equipment characteristics, equipment standard data, and real-time captured current actual images, the outline of the equipment is identified, and the equipment outline is highlighted with a virtual label.

[0060] In step B2, the real-time acquisition of device data includes:

[0061] Based on AR devices, the current device's shape is captured by images, and image recognition technology is used to obtain the current device's shape feature data.

[0062] Based on the current equipment type, preset inspection steps are obtained through AR devices to assist inspection personnel in inspecting the equipment and obtaining the current equipment's operating data.

[0063] Step B3, the process of assessing the equipment status includes:

[0064] Preset standard data comparison: Compare the current device's external feature data with the preset standard device features. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, compare the remote system monitoring data with the data within the current device. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, analyze the trend of changes in the device's operating parameters through longitudinal comparison to determine whether they exceed the historical fluctuation range. If they exceed the range, the device data is determined to be abnormal. Otherwise, the device data is determined to be normal, and the comparison with the preset standard data is completed.

[0065] Anomaly detection algorithm evaluation, such as Figure 2 As shown: Standardize the shape feature data and operating data, and associate them with relevant equipment data; acquire historical operating data of the equipment and extract typical operating modes, calculate the normal fluctuation range of the operating data, and establish the normal operating baseline of the equipment in combination with the equipment design standards; compare the standardized shape feature data and operating data with the normal operating baseline of the equipment. If the current equipment data exceeds the normal fluctuation range of the operating data, the equipment is determined to be in a numerical fault state. If the current equipment data does not exceed the normal fluctuation range, but the trend of change deviates from the normal operating baseline of the equipment, the equipment is determined to be in a trend fault state. Otherwise, the equipment is determined to be in a fault-free state, and the anomaly detection algorithm evaluation is completed.

[0066] The anomaly detection algorithm also classifies equipment failure status according to the severity, scope of impact, and speed of development of the anomaly.

[0067] The process of generating standardized maintenance procedures using natural language processing includes:

[0068] Based on the type of equipment requiring repair, obtain the corresponding historical repair records and preset repair steps. Using named entity recognition technology in natural language processing, extract key elements from the historical repair record documents and preset repair steps, including fault characteristics, operation steps, tools and materials, and equipment identification. Through syntactic analysis and semantic network construction, parse the logical relationships between the key elements, and use a deep learning model to understand the contextual semantics and construct the semantic relationships between the key elements. Based on the logical and semantic relationships between the key elements, automatically generate a standardized repair process according to the repair logic.

[0069] If the inspection personnel complete the repair of the faulty equipment according to the standardized maintenance process generated by natural language processing, the standardized maintenance process will be stored in the historical maintenance record.

[0070] AR devices are connected to a digital twin platform. During the inspection process, the AR devices collect data from each device in real time and upload the data to the digital twin platform via wireless connection. When inspectors are conducting inspections, the AR devices can obtain their location and movement trajectory in real time. The actual inspection path is compared with the planned inspection route in real time on the digital twin platform. If a deviation is found, a reminder is pushed to the manager. If the inspector is in an area with poor wireless signal, the upload of device data and the real-time comparison of the inspection route are delayed.

[0071] In addition, the digital twin platform also combines on-site monitoring cameras to use AI to identify violations and trigger real-time alarms during inspections.

[0072] The above methods enable one person to perform multiple roles, achieving a reduction in manpower in power plants. By accessing equipment ledgers, operating procedures, retrospective maintenance records, expert knowledge base guidance, and remote assistance, faults can be diagnosed and resolved locally, enhancing the overall safety and stability of the power plant.

[0073] Below are examples of applying this method in practical situations:

[0074] Deployment steps:

[0075] The power plant's inspection supervisor uses a management platform to configure AR devices integrated into the safety helmets of three inspectors (A, B, and C), and plans inspection routes based on the plant's equipment distribution. The system generates inspection task lists based on equipment importance and production process relationships, using an improved ant colony algorithm.

[0076] A: Boiler body → superheater → feedwater pump (high priority, covering key high-temperature and high-pressure equipment);

[0077] B: Steam turbine cylinder block → speed control system → lubrication oil station (must be checked in the order of the process);

[0078] C: Generator stator → rotor → cooling system (involves both electrical and mechanical testing).

[0079] AR devices synchronously receive task lists, device standard data, and preset maintenance process libraries.

[0080] For signal blind spots in the factory area (such as underground pipeline layers), the AR device downloads the preset feature data (shape outline, NFC tag information), standard operating parameters and historical maintenance records of all devices in advance to ensure that identification and fault diagnosis can still be performed offline.

[0081] Inspection steps:

[0082] A, wearing an AR safety helmet, enters the boiler area. The equipment automatically matches the current position through inertial navigation and visual positioning. A three-dimensional plant map is generated in real time in the AR field of view, and the path of the first inspection equipment "boiler body" is marked with a yellow highlighted arrow, while the distance of the equipment is displayed.

[0083] For devices with NFC tags, when device A approaches the sensor, the AR device automatically reads the NFC tag using IoT offline identification technology, obtains the standard pressure range of the sensor, and simultaneously takes a picture of the sensor's appearance, using image recognition to detect the position of the dial pointer.

[0084] For devices without NFC tags, the AR device captures images of the pipe weld area, identifies it as a "main steam pipe" through a pre-set feature library, and automatically retrieves the standard shape parameters of that pipe. Simultaneously, following the inspection steps prompted by the AR view, device A collects sound data via its microphone, and the system analyzes the spectrum in real time for any anomalies.

[0085] The system compares the real-time collected pressure value of 1.3 MPa with the standard range of 0.8-1.2 MPa and determines it to be a numerical anomaly. At the same time, it analyzes the pressure trend of the sensor over the past 30 days and finds that it has been rising continuously and exceeds the historical normal fluctuation range, which is further determined to be an "emergency fault". The sensor location flashes a red border in the AR field of view, and a text alarm pops up at the same time: "Pressure exceeds the limit! It is recommended to reduce the pressure immediately" and the fault level and impact are announced in voice.

[0086] The system automatically retrieves historical maintenance records and preset steps based on the fault type, generates a standardized process through natural language processing, and instructs User A to repair the boiler pressure sensor. Upon completion, the AR device automatically photographs the repaired pressure gauge and packages the maintenance process and pre- and post-repair data, uploading them to the digital twin platform. The platform updates the device status to "normal" and stores the maintenance record in the knowledge base for future reference in similar faults.

[0087] After the inspection is completed, the system automatically generates an inspection report based on the full-process data collected by the AR device, marking the equipment that was missed, the handling of anomalies, and the efficiency analysis.

[0088] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An inspection assistance method based on AR (Augmented Reality), characterized in that, The method includes deployment steps and inspection steps; wherein... The deployment steps include: Each inspector is equipped with an AR device, their inspection route is planned, and the equipment to be inspected and their inspection order are set according to the inspection route, generating an inspection task list for each inspector. The inspection steps include: AR devices are used to identify equipment on the inspection route in real time and map the equipment location to the AR field of view of the inspection personnel, marking it as a highlighted virtual icon. Data from each device is collected in real time and compared with preset standard data or evaluated for equipment status through anomaly detection algorithms. If the equipment data is abnormal or the equipment is in a faulty state, an alarm is triggered in real time, and a standardized maintenance process is generated through natural language processing to assist the inspection personnel in maintenance.

2. The inspection assistance method based on AR (Augmented Reality) according to claim 1, characterized in that, The AR device is integrated into the personal equipment of the inspection personnel, which includes a safety helmet.

3. The inspection assistance method based on AR (Augmented Reality) according to claim 1, characterized in that, The process of planning the inspection route includes: Based on historical inspection data, obtain the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process. Based on the importance of equipment, the failure risk rate of equipment, and the correlation between equipment and production process, assign priority to equipment. Based on the priority of each device, an improved ant colony algorithm is used to generate inspection routes with the principle of covering the highest priority device in the shortest time.

4. The inspection assistance method based on AR (Augmented Reality) according to claim 1, characterized in that, The process by which the AR device attempts to identify equipment data on the inspection route in real time includes: If the device has an NFC tag, the AR device will automatically identify the NFC tag through the IoT offline identification TAG and obtain the pre-stored device standard data. If the device does not have an NFC tag, the AR device will capture the current actual scene in real time, use image recognition technology to obtain the characteristics of the current device, compare and match them with the preset device characteristics, identify the type of the current device, and obtain the pre-stored device standard data. Based on preset equipment characteristics, equipment standard data, and real-time captured current actual images, the outline of the equipment is identified, and the equipment outline is highlighted with a virtual label.

5. The inspection assistance method based on AR (Augmented Reality) according to claim 1, characterized in that, The real-time acquisition of the device data includes: Based on AR devices, the current device's shape is captured by images, and image recognition technology is used to obtain the current device's shape feature data. Based on the current equipment type, preset inspection steps are obtained through AR devices to assist inspection personnel in inspecting the equipment and obtaining the current equipment's operating data.

6. The inspection assistance method based on AR (Augmented Reality) according to claim 5, characterized in that, The process of assessing the equipment status includes: Preset standard data comparison: Compare the current device's external feature data with the preset standard device features. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, compare the remote system monitoring data with the data within the current device. If they are inconsistent, the device data is determined to be abnormal. If they are consistent, analyze the trend of changes in the device's operating parameters through longitudinal comparison to determine whether they exceed the historical fluctuation range. If they exceed the range, the device data is determined to be abnormal. Otherwise, the device data is determined to be normal, and the comparison with the preset standard data is completed. Anomaly detection algorithm evaluation: The external feature data and operational data are standardized and associated with relevant equipment data; historical operational data of the equipment are acquired and typical operating modes are extracted, the normal fluctuation range of the operational data is calculated, and a normal operating baseline of the equipment is established in conjunction with the equipment design standards; the standardized external feature data and operational data are compared with the normal operating baseline of the equipment. If the current equipment data exceeds the normal fluctuation range of the operational data, the equipment is determined to be in a numerical fault state; if the current equipment data does not exceed the normal fluctuation range, but the trend of change deviates from the normal operating baseline of the equipment, the equipment is determined to be in a trend fault state; otherwise, the equipment is determined to be fault-free, and the anomaly detection algorithm evaluation is completed.

7. The inspection assistance method based on AR (Augmented Reality) according to claim 6, characterized in that, The anomaly detection algorithm also classifies equipment fault states based on the severity, scope of impact, and speed of development of the anomaly.

8. The inspection assistance method based on AR (Augmented Reality) according to claim 1, characterized in that, The process of generating standardized maintenance procedures through natural language processing includes: Based on the type of equipment requiring repair, obtain the corresponding historical repair records and preset repair steps. Using named entity recognition technology in natural language processing, extract key elements from the historical repair record documents and preset repair steps, including fault characteristics, operation steps, tools and materials, and equipment identification. Through syntactic analysis and semantic network construction, parse the logical relationships between the key elements, and use a deep learning model to understand the contextual semantics and construct the semantic relationships between the key elements. Based on the logical and semantic relationships between the key elements, automatically generate a standardized repair process according to the repair logic.

9. The inspection assistance method based on AR (Augmented Reality) according to claim 8, characterized in that, If the inspection personnel complete the repair of the faulty equipment according to the standardized maintenance process generated by natural language processing, the standardized maintenance process will be stored in the historical maintenance record.

10. The AR-based inspection assistance method according to claim 1, characterized in that, The AR device is connected to the digital twin platform. During the inspection process, the AR device collects data from each device in real time and uploads the data to the digital twin platform via wireless connection. When the inspection personnel are conducting inspections, the AR device can obtain the location and movement trajectory of the inspection personnel in real time. The actual inspection path is compared with the planned inspection route in real time on the digital twin platform. When a deviation is found, a reminder is pushed to the manager. If the current inspection personnel are in an area with poor wireless signal, the uploading of equipment data and the real-time comparison of the inspection route will be delayed.

Images