Multi-modal collaborative intelligent inspection method and system for smart power plant and medium

Abstract

This application provides a multimodal collaborative intelligent inspection method, system, and medium for smart power plants. The method includes: deploying various types of inspection equipment to different areas of the power plant; controlling the inspection equipment to collect multi-dimensional data on the equipment and environment within the power plant according to a set mode based on preset inspection tasks, obtaining multi-dimensional inspection data; comparing the collected multi-dimensional inspection data with preset safety data; comparing operational difference information with a set difference threshold to obtain difference values; analyzing the multi-dimensional inspection data based on the difference values ​​to generate equipment anomaly information and environmental anomaly information; analyzing the equipment anomaly information and environmental anomaly information based on multi-level alarm rules to generate alarm information of different levels; and by deploying different types of inspection equipment to adapt to the inspection needs of different scenarios, collaboratively achieving seamless and comprehensive inspection of key areas of the smart power plant, reducing the labor intensity of maintenance personnel, and improving operational safety.

Description

Technical Field

[0001] This application relates to the field of power plant inspection technology, and more specifically, to a multimodal collaborative intelligent inspection method, system, and medium for smart power plants. Background Technology

[0002] As the core carrier of digital and intelligent transformation in the energy sector, smart power plants have a wide variety of internal equipment with a dispersed layout, covering multiple key areas such as coal conveying corridors, high-voltage power distribution rooms, turbine operating floors, and GIS buildings. The operating environments of the equipment in each area are significantly different—coal conveying corridors have problems such as high dust, high noise, and enclosed spaces; high-voltage power distribution rooms have extremely high requirements for electrical safety monitoring accuracy; and the open areas of the plant need to cope with the mobile inspection needs of complex terrain.

[0003] Currently, smart power plant inspections mostly combine manual inspections with single-type inspection equipment, which presents several pain points: manual inspections are labor-intensive, inefficient, and susceptible to human error leading to missed or false inspections, and pose safety hazards when operating in harsh environments such as high temperatures, low temperatures, and high dust levels; single inspection equipment has limited functionality and cannot adapt to the inspection needs of different areas. For example, track-mounted robots can only inspect within a fixed track range and cannot cover open areas and complex terrains, while quadruped robots cannot meet the high-precision electrical testing requirements of high-voltage substations; there is a lack of effective collaboration between various inspection devices and between the inspection system and the power plant's existing DCS, OA, and power management systems, resulting in prominent data silos. Inspection data cannot be linked with production data for analysis, and anomaly handling processes are cumbersome and slow to respond, making it difficult to achieve intelligent and unmanned closed-loop management of the entire plant's inspections. Summary of the Invention

[0004] The purpose of this application is to provide a multimodal collaborative intelligent inspection method, system and medium for smart power plants. By deploying different types of inspection equipment, it adapts to the inspection needs of different scenarios, and collaboratively achieves blind-spot-free and full-coverage inspection of key areas of smart power plants, reducing the labor intensity of operation and maintenance personnel and improving operational safety.

[0005] This application embodiment also provides a multimodal collaborative intelligent inspection method for smart power plants, including: deploying multiple types of inspection equipment to different areas of the power plant, the multiple types of inspection equipment including a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog and a high-voltage power distribution room track-type lifting inspection robot. Receive preset inspection tasks, and control the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection tasks and the set mode to obtain multi-dimensional inspection data. The multi-dimensional inspection data includes equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. The collected multi-dimensional inspection data is compared with the preset safety data, and the difference information is processed. The difference value is obtained by comparing the operational difference information with the set difference threshold. Based on the difference value, multi-dimensional inspection data is analyzed to generate equipment abnormality information and environmental abnormality information. Based on multi-level alarm rules, the system analyzes equipment and environmental anomalies to generate alarms of different levels. The system then transmits multi-dimensional inspection data and alarm information to the terminal for real-time visualization.

[0006] Optionally, in the multimodal collaborative intelligent inspection method for smart power plants described in the embodiments of this application, multiple types of inspection equipment are deployed to different areas of the power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, we conducted a targeted analysis of the performance parameters and applicable scenarios of the coal conveying corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

[0007] Optionally, in the multimodal collaborative intelligent inspection method for smart power plants described in the embodiments of this application, the multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

[0008] Optionally, in the multimodal collaborative intelligent inspection method for smart power plants described in the embodiments of this application, the set modes include fully autonomous inspection mode and remote control inspection mode; The fully autonomous inspection mode includes a regular inspection mode and a special inspection mode. The regular inspection mode autonomously completes the inspection based on the preset inspection task content, time, and path parameters. The special inspection mode allows the operator to set inspection points, and the inspection equipment to autonomously complete the corresponding inspection tasks. The remote inspection mode allows operators to manually control the inspection equipment through the back-end management system to complete the inspection.

[0009] Optionally, in the multimodal collaborative intelligent inspection method for smart power plants described in this application embodiment, the method compares operational difference information with a set difference threshold to obtain a difference value, analyzes multi-dimensional inspection data based on the difference value, and generates equipment anomaly information and environmental anomaly information, specifically including: Acquire multi-dimensional inspection data transmitted by inspection equipment, filter out core data related to equipment operating status and environmental status, and eliminate invalid data, abnormal fluctuation data and interference data; Extract historical normal operating data of equipment, industry standard operating data, and power plant preset equipment operating benchmark data to build an operating difference information collection database; Based on the operating characteristics of different areas and different types of equipment in smart power plants, and combined with industry standards, equipment factory parameters and power plant operation and maintenance experience, differential thresholds are set. By comparing the multi-dimensional inspection data with the baseline data in the operational difference information collection database, equipment anomaly information and environmental anomaly information are obtained.

[0010] Optionally, the multimodal collaborative intelligent inspection method for smart power plants described in this application embodiment further includes a self-inspection step for the inspection equipment, specifically including: Acquire the movement trajectory of the inspection equipment, compare the movement trajectory of the inspection equipment with the set standard trajectory, and obtain the trajectory difference; The movement trajectory of the inspection equipment is corrected based on the trajectory difference; Real-time acquisition of the corrected power status information of the inspection equipment; Compare its own battery status information with the set battery threshold; If the device's own power status information is less than the set power threshold, a charging command is generated. Based on the charging command, the training device is controlled to move to the charging position for charging. At the same time, a backup inspection device is added to perform inspections along the unfinished inspection trajectory.

[0011] Secondly, embodiments of this application provide a multimodal collaborative intelligent inspection system for smart power plants. The system includes a memory and a processor. The memory includes a program for a multimodal collaborative intelligent inspection method for smart power plants. When executed by the processor, the program for the multimodal collaborative intelligent inspection method for smart power plants implements the following steps: Multiple types of inspection equipment are deployed to different areas of the power plant. These include a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog, and a high-voltage power distribution room track-type lifting inspection robot. Receive preset inspection tasks, and control the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection tasks and the set mode to obtain multi-dimensional inspection data. The multi-dimensional inspection data includes equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. The collected multi-dimensional inspection data is compared with the preset safety data, and the difference information is processed. The difference value is obtained by comparing the operational difference information with the set difference threshold. Based on the difference value, multi-dimensional inspection data is analyzed to generate equipment abnormality information and environmental abnormality information. Based on multi-level alarm rules, the system analyzes equipment and environmental anomalies to generate alarms of different levels. The system then transmits multi-dimensional inspection data and alarm information to the terminal for real-time visualization.

[0012] Optionally, in the multimodal collaborative intelligent inspection system for smart power plants described in this application embodiment, multiple types of inspection equipment are deployed to different areas of the power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, we conducted a targeted analysis of the performance parameters and suitable scenarios for the coal conveying corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

[0013] Optionally, in the multimodal collaborative intelligent inspection system for smart power plants described in this application embodiment, the multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

[0014] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a program for a multimodal collaborative intelligent inspection method for smart power plants. When the program for the multimodal collaborative intelligent inspection method for smart power plants is executed by a processor, it implements the steps of the multimodal collaborative intelligent inspection method for smart power plants as described in any of the preceding claims.

[0015] As can be seen from the above, the multimodal collaborative intelligent inspection method, system, and medium for smart power plants provided in this application deploy multiple types of inspection equipment to different areas of the power plant. These multiple types of inspection equipment include a track-mounted inspection robot for coal conveying corridors, a four-legged inspection robot dog for the plant area, and a track-mounted lifting inspection robot for high-voltage power distribution rooms. The system receives preset inspection tasks and, based on these tasks, controls the inspection equipment to collect multi-dimensional data about the equipment and environment within the power plant according to a set mode. This multi-dimensional inspection data includes equipment temperature, belt running status, video images, environmental parameters, audio information, and equipment operating parameters. The collected multi-dimensional inspection data is then compared with the preset inspection data... The system compares established safety data to identify operational discrepancies. It then compares these discrepancies with set threshold values ​​to obtain difference values. Based on these difference values, it analyzes multi-dimensional inspection data to generate equipment and environmental anomaly information. Using multi-level alarm rules, it analyzes these anomalies to generate alarms of different levels. The multi-dimensional inspection data and alarm information are then transmitted to the terminal in real time for visualization. By deploying different types of inspection equipment, it adapts to the inspection needs of various scenarios, such as enclosed corridors, open plant areas, and high-voltage power distribution rooms, collaboratively achieving comprehensive, blind-spot-free inspections of key areas in the smart power plant. This reduces the workload of maintenance personnel and improves operational safety. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a multimodal collaborative intelligent inspection method for smart power plants provided in this application embodiment; Figure 2 A flowchart illustrating the method for analyzing equipment anomaly information and environmental anomaly information in a multimodal collaborative intelligent inspection method for smart power plants, as provided in this application embodiment. Figure 3 A flowchart illustrating the self-inspection method of the inspection equipment for a multimodal collaborative intelligent inspection method for smart power plants, provided in this application embodiment. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0019] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] Please refer to Figure 1 , Figure 1 This is a flowchart of a multimodal collaborative intelligent inspection method for smart power plants, as described in some embodiments of this application. This multimodal collaborative intelligent inspection method for smart power plants is used in terminal equipment and includes the following steps: S101 deploys various types of inspection equipment to different areas of the power plant. These inspection equipment include a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog, and a high-voltage power distribution room track-type lifting inspection robot. S102 receives a preset inspection task and controls the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection task and the set mode, so as to obtain multi-dimensional inspection data, including equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. S103 compares the collected multi-dimensional inspection data with the preset safety data and processes the difference information; S104: Based on the operational difference information, compare it with the set difference threshold to obtain the difference value. Based on the difference value, analyze multi-dimensional inspection data to generate equipment abnormality information and environmental abnormality information. The S105 analyzes equipment and environmental anomaly information based on multi-level alarm rules, generates alarm information of different levels, and transmits multi-dimensional inspection data and alarm information to the terminal for visualization in real time.

[0021] According to embodiments of the present invention, multiple types of inspection equipment are deployed to different areas of a power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, we conducted a targeted analysis of the performance parameters and applicable scenarios of the coal conveying corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

[0022] Specifically, this includes: Analysis of inspection needs for the smart power plant area—a comprehensive survey of key areas of the smart power plant to clarify the spatial layout, equipment distribution, operating environment, and inspection priorities of each area: For example, the coal conveying corridor will focus on inspecting belt conveyors, drive units, and other equipment, characterized by high dust levels, noise levels, enclosed spaces, and slopes in some areas; the plant area (steam turbine operating floor, GIS building) will focus on inspecting steam turbine equipment and GIS equipment, characterized by open spaces, complex terrain, and obstacles in some areas; the high-voltage distribution room will focus on inspecting 10KV and above electrical equipment, characterized by relatively enclosed spaces, high requirements for electrical testing accuracy, and the need for multi-height inspections. Simultaneously, existing inspection pain points, equipment operation risks, and required inspection frequency for each area will be identified, resulting in a regional inspection needs report. Inspection Equipment Selection and Adaptation Analysis – Based on the inspection needs reports for each region, a targeted analysis was conducted on the performance parameters and suitable scenarios of the track-mounted inspection robot for coal conveying corridors, the quadrupedal inspection robot for the plant area, and the track-mounted lifting inspection robot for high-voltage power distribution rooms. For the coal conveying corridors, the track-mounted inspection robot was selected, with a focus on verifying its protection level (≥IP66), climbing ability (≥45°), dust tolerance, and track adaptability to ensure stable operation in enclosed corridors and multi-slope environments. For the plant area, the quadrupedal inspection robot was selected, with a focus on verifying its navigation accuracy (repeat positioning accuracy ±5cm), obstacle crossing ability (≥25cm), wading depth (≥25cm), and flexible movement ability to ensure coverage of complex terrain and obstacle areas. For the high-voltage power distribution rooms, the track-mounted lifting inspection robot was selected, with a focus on verifying its positioning accuracy (repeat positioning error ≤±10mm), lifting stroke (≥1500mm), electrical detection accuracy, and anti-electric shock capability to ensure it meets the needs of multi-height, high-precision electrical inspections. Deployment Plan Planning and Design – Combining regional layout, equipment distribution, and inspection paths, specific deployment plans are planned for three types of inspection equipment: A coal conveyor corridor track-type inspection robot, with high-strength spliced ​​tracks arranged on both sides of each coal conveyor belt. The tracks cover all belt conveyors and drive units in the coal bunker and enclosed trestle areas. The minimum bending radius of the tracks is ≤600mm, and the load-bearing capacity is no less than 1.5 times the robot's weight. Non-contact wireless distributed charging stations are deployed at suitable locations in the corridor to ensure the robot's autonomous charging needs. A plant area quadrupedal inspection robot dog is deployed on the 17-meter level of the turbine operating floor and within the GIS building. Optimal inspection paths are planned to avoid equipment obstacles. Autonomous charging piles are deployed at key nodes to ensure convenient charging. The LiDAR SLAM navigation module is also tested to ensure navigation accuracy. A high-voltage power distribution room track-type lifting inspection robot is deployed in the 10KV power distribution rooms of Units #5 and #6. Dedicated tracks are used, and a sliding contact line power supply method is adopted to ensure 24-hour uninterrupted inspection. The track layout avoids hazardous areas of electrical equipment, and the lifting stroke covers the height range of all electrical equipment. Deployment, Implementation, and Debugging – Following the planned deployment scheme, the inspection equipment was deployed in stages: First, the installation and debugging of tracks, charging piles, and power supply devices in each area were completed to ensure that the tracks were securely fixed, the power supply was stable, and the charging piles were well-suited; then, the three types of inspection equipment were installed in their respective areas, and the equipment was fixed, its position calibrated, and network access was completed to ensure that all inspection equipment could stably access the back-end management system via wireless communication; finally, the equipment linkage was debugged to test the accuracy of the inspection path, the completeness of data collection, the mobility of movement, and the reliability of charging for each inspection device. Any issues discovered during debugging, such as track deviation, navigation error, or communication interruption, were promptly optimized and adjusted to ensure deployment quality. Deployment Effectiveness Verification and Optimization – After deployment, a continuous trial run of no less than 72 hours will be conducted to verify the deployment adaptability of the three types of inspection equipment: the coal conveying corridor robot will verify its operational stability and inspection coverage in a dusty environment; the plant area quadruped robot dog will verify its mobility and inspection efficiency in complex terrain; and the high-voltage power distribution room robot will verify its high-precision electrical testing capabilities and 24-hour uninterrupted operational reliability. Simultaneously, the inspection coverage rate, data transmission success rate, and equipment failure rate of each device will be statistically analyzed. By comparing regional inspection requirements, the shortcomings of the deployment plan will be analyzed, and the track layout, inspection paths, and equipment parameters will be optimized to ensure that the deployment plan meets the inspection needs of each region and achieves seamless coverage of key areas of the smart power plant.

[0023] According to an embodiment of the present invention, multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

[0024] According to embodiments of the present invention, the setting modes include a fully autonomous inspection mode and a remote inspection mode; The fully autonomous inspection mode includes a regular inspection mode and a special inspection mode. The regular inspection mode autonomously completes the inspection based on the preset inspection task content, time, and path parameters. In the special inspection mode, the operator sets the inspection points, and the inspection equipment autonomously completes the corresponding inspection tasks. In the remote inspection mode, operators manually control the inspection equipment through the back-end management system to complete the inspection.

[0025] Please refer to Figure 2 , Figure 2This is a flowchart illustrating a method for analyzing equipment anomaly information and environmental anomaly information in a multimodal collaborative intelligent inspection method for smart power plants, as described in some embodiments of this application. According to embodiments of the present invention, based on operational difference information, a difference value is obtained by comparing it with a set difference threshold. Multi-dimensional inspection data is then analyzed based on the difference value to generate equipment anomaly information and environmental anomaly information, specifically including: S201: Acquire multi-dimensional inspection data transmitted by the inspection equipment, filter out core data related to equipment operating status and environmental status, and remove invalid data, abnormal fluctuation data and interference data. S202, extract historical normal operation data of equipment, industry standard operation data and power plant preset equipment operation benchmark data, and construct an operation difference information collection database; S203 sets difference thresholds based on the operating characteristics of different areas and types of equipment in smart power plants, combined with industry standards, equipment factory parameters and power plant operation and maintenance experience. S204 compares the multi-dimensional inspection data with the baseline data in the operation difference information collection library to obtain equipment abnormality information and environmental abnormality information.

[0026] It should be noted that the operation difference information collection and preprocessing process involves the back-end management system receiving multi-dimensional inspection data transmitted from the inspection equipment, filtering out core data related to equipment operating status and environmental status, eliminating invalid data, abnormal fluctuation data, and interference data, and completing missing data. The preprocessed inspection data is then used as the basic operation data. At the same time, historical normal operation data of the equipment, industry standard operation data, and power plant-preset equipment operation benchmark data are extracted to construct an operation difference information collection library. The collection dimensions and frequency of operation difference information are clearly defined to ensure the completeness and accuracy of the operation difference information.

[0027] Differential threshold setting—Based on the operating characteristics of different areas and types of equipment in the smart power plant, combined with industry standards, equipment factory parameters, and power plant operation and maintenance experience, differential thresholds are set for various dimensions such as equipment temperature, belt operation status, equipment operating parameters, environmental parameters (dust concentration, temperature and humidity, noise, gas concentration), and electrical parameters. Differential thresholds include normal thresholds, early warning thresholds, and danger thresholds. The normal threshold is the allowable fluctuation range for normal equipment operation, the early warning threshold is the trigger point for abnormal hazards, and the danger threshold is the trigger point for emergency response. At the same time, a dynamic threshold adjustment mechanism is established to periodically calibrate the differential thresholds based on the equipment's operating years, environmental changes, and operation and maintenance feedback to ensure the adaptability and scientific nature of the thresholds.

[0028] Difference value calculation - Compare the preprocessed basic operation data with the benchmark data in the operation difference information collection library, calculate the operation difference information of each dimension data through a preset algorithm, and then quantitatively compare the operation difference information with the corresponding set difference threshold to obtain the specific difference value of each dimension data; among them, the absolute value comparison method and the ratio comparison method are combined for difference value calculation. The absolute value comparison method is used to calculate the difference value for continuous data such as temperature and gas concentration, and the ratio comparison method is used to calculate the difference value for discrete data such as equipment operation parameters and belt operation status to ensure the accuracy of difference value calculation.

[0029] Multi-dimensional patrol data analysis - Based on the calculated difference values of each dimension, classify and analyze the multi-modal patrol data: for equipment-related data, focus on analyzing the correlation between the equipment temperature difference value, operation parameter difference value, electrical parameter difference value and equipment failures, and combine the historical abnormal data of the equipment to judge the abnormal type and severity corresponding to the difference value; for environment-related data, focus on analyzing the correlation between the difference values of each environmental parameter and the impact of the environmental difference value on the equipment operation status, and judge the diffusion range and potential risks of environmental abnormalities; at the same time, integrate auxiliary data such as video images and audio information to verify the analysis results of the difference values and eliminate misjudgments.

[0030] Abnormal information generation and classification - According to the results of multi-dimensional patrol data analysis, classify and judge the situations where the difference value exceeds the normal threshold, and generate corresponding equipment abnormal information and environmental abnormal information: when the difference value is between the normal threshold and the warning threshold, generate warning-type abnormal information to prompt the operation and maintenance personnel to focus on and conduct regular inspections; when the difference value is between the warning threshold and the danger threshold, generate general abnormal information to clarify the abnormal handling priority and basic handling suggestions; when the difference value exceeds the danger threshold, generate emergency abnormal information, synchronously trigger audible and visual alarms, and clarify the emergency handling process and safety precautions; at the same time, label the core information such as the difference value, abnormal occurrence time, abnormal location, and associated data for each abnormal information to ensure the traceability of the abnormal information.

[0031] Abnormal information verification and output - Secondarily verify the generated equipment abnormal information and environmental abnormal information, and combine the operation status of the patrol equipment and the multi-system linkage data to confirm the authenticity and accuracy of the abnormal information, and eliminate false alarm information caused by data transmission errors and threshold deviations; after passing the verification, classify and sort the abnormal information according to the abnormal type and severity, and synchronously transmit it to the abnormal management module of the background management system, the digital twin system and relevant operation and maintenance terminals to provide comprehensive and accurate support for abnormal handling, and at the same time store the abnormal information and analysis results synchronously for subsequent operation and maintenance analysis and threshold optimization Please refer to Figure 3 , Figure 3This is a flowchart illustrating a self-testing method for inspection equipment in a multimodal collaborative intelligent inspection method for smart power plants, as described in some embodiments of this application. According to embodiments of the present invention, it further includes a self-testing step for the inspection equipment, specifically comprising: S301, acquire the movement trajectory of the inspection equipment, compare the movement trajectory of the inspection equipment with the set standard trajectory, and obtain the trajectory difference; S302, Correcting the movement trajectory of the inspection equipment based on the trajectory difference; S303, real-time acquisition of the corrected power status information of the inspection equipment; S304 compares its own battery status information with the set battery threshold. S305 If its own power status information is less than the set power threshold, a charging command is generated. Based on the charging command, the training device is controlled to move to the charging position for charging. At the same time, a backup inspection device is added to perform inspection along the unfinished inspection trajectory.

[0032] Secondly, embodiments of this application provide a multimodal collaborative intelligent inspection system for smart power plants. The system includes a memory and a processor. The memory includes a program for a multimodal collaborative intelligent inspection method for smart power plants. When the program for the multimodal collaborative intelligent inspection method for smart power plants is executed by the processor, it implements the following steps: Multiple types of inspection equipment are deployed to different areas of the power plant. These include a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog, and a high-voltage power distribution room track-type lifting inspection robot. Receive preset inspection tasks, and control the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection tasks and the set mode to obtain multi-dimensional inspection data, including equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. The collected multi-dimensional inspection data is compared with the preset safety data, and the difference information is processed. The difference value is obtained by comparing the operational difference information with the set difference threshold. Based on the difference value, multi-dimensional inspection data is analyzed to generate equipment abnormality information and environmental abnormality information. Based on multi-level alarm rules, the system analyzes equipment and environmental anomalies to generate alarms of different levels. The system then transmits multi-dimensional inspection data and alarm information to the terminal for real-time visualization.

[0033] According to embodiments of the present invention, multiple types of inspection equipment are deployed to different areas of a power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, a targeted analysis was conducted on the performance parameters and suitable scenarios of the coal conveyor corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

[0034] According to an embodiment of the present invention, multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

[0035] A third aspect of the present invention provides a computer-readable storage medium including a program for a multimodal collaborative intelligent inspection method for a smart power plant. When the program is executed by a processor, it implements the steps of the multimodal collaborative intelligent inspection method for a smart power plant as described above.

[0036] This invention discloses a multimodal collaborative intelligent inspection method, system, and medium for smart power plants. It involves deploying various types of inspection equipment to different areas of the power plant, including a track-mounted inspection robot for coal conveying corridors, a four-legged inspection robot for the plant area, and a track-mounted lifting inspection robot for high-voltage power distribution rooms. The system receives preset inspection tasks and, based on these tasks, controls the inspection equipment to collect multi-dimensional data on the equipment and environment within the power plant according to a set mode. This multi-dimensional inspection data includes equipment temperature, belt conveyor operating status, video images, environmental parameters, audio information, and equipment operating parameters. The collected multi-dimensional inspection data is then compared with preset safety parameters. The system compares data to identify operational discrepancies. Based on these discrepancies, it compares them with set threshold values ​​to obtain discrepancy values. Multi-dimensional inspection data is then analyzed based on these discrepancy values ​​to generate equipment and environmental anomaly information. Multi-level alarm rules are used to analyze the equipment and environmental anomaly information, generating alarms of different levels. This multi-dimensional inspection data and alarm information are transmitted to the terminal in real time for visualization. By deploying different types of inspection equipment, the system adapts to the inspection needs of various scenarios, such as enclosed corridors, open plant areas, and high-voltage power distribution rooms. This collaborative approach achieves comprehensive, blind-spot-free inspections of key areas in the smart power plant, reducing the workload of maintenance personnel and improving operational safety.

[0037] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0038] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0039] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0040] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0041] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A multimodal collaborative intelligent inspection method for smart power plants, characterized in that, include: Multiple types of inspection equipment are deployed to different areas of the power plant. These include a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog, and a high-voltage power distribution room track-type lifting inspection robot. Receive preset inspection tasks, and control the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection tasks and the set mode to obtain multi-dimensional inspection data. The multi-dimensional inspection data includes equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. The collected multi-dimensional inspection data is compared with the preset safety data, and the difference information is processed. The difference value is obtained by comparing the operational difference information with the set difference threshold. Based on the difference value, multi-dimensional inspection data is analyzed to generate equipment abnormality information and environmental abnormality information. Based on multi-level alarm rules, the system analyzes equipment and environmental anomalies to generate alarms of different levels. The system then transmits multi-dimensional inspection data and alarm information to the terminal for real-time visualization.

2. The multimodal collaborative intelligent inspection method for smart power plants according to claim 1, characterized in that, Deploying various types of inspection equipment to different areas of the power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, we conducted a targeted analysis of the performance parameters and applicable scenarios of the coal conveying corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

3. The multimodal collaborative intelligent inspection method for smart power plants according to claim 2, characterized in that, The multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

4. The multimodal collaborative intelligent inspection method for smart power plants according to claim 3, characterized in that, The settings include a fully autonomous inspection mode and a remote inspection mode; The fully autonomous inspection mode includes a regular inspection mode and a special inspection mode. The regular inspection mode autonomously completes the inspection based on the preset inspection task content, time, and path parameters. The special inspection mode allows the operator to set inspection points, and the inspection equipment to autonomously complete the corresponding inspection tasks. The remote inspection mode allows operators to manually control the inspection equipment through the back-end management system to complete the inspection.

5. The multimodal collaborative intelligent inspection method for smart power plants according to claim 4, characterized in that, By comparing operational discrepancies with set discrepancy thresholds, discrepancy values ​​are obtained. Multi-dimensional inspection data is then analyzed based on these discrepancy values ​​to generate equipment anomaly and environmental anomaly information, specifically including: Acquire multi-dimensional inspection data transmitted by inspection equipment, filter out core data related to equipment operating status and environmental status, and eliminate invalid data, abnormal fluctuation data and interference data; Extract historical normal operating data of equipment, industry standard operating data, and power plant preset equipment operating benchmark data to build an operating difference information collection database; Based on the operating characteristics of different areas and types of equipment in smart power plants, and combined with industry standards, equipment factory parameters and power plant operation and maintenance experience, differential thresholds are set. By comparing the multi-dimensional inspection data with the baseline data in the operational difference information collection database, equipment anomaly information and environmental anomaly information are obtained.

6. The multimodal collaborative intelligent inspection method for smart power plants according to claim 5, characterized in that, It also includes the self-inspection steps for the inspection equipment, specifically including: Acquire the movement trajectory of the inspection equipment, compare the movement trajectory of the inspection equipment with the set standard trajectory, and obtain the trajectory difference; The movement trajectory of the inspection equipment is corrected based on the trajectory difference; Real-time acquisition of the corrected power status information of the inspection equipment; Compare its own battery status information with the set battery threshold; If the device's own power status information is less than the set power threshold, a charging command is generated. Based on the charging command, the training device is controlled to move to the charging position for charging. At the same time, a backup inspection device is added to perform inspections along the unfinished inspection trajectory.

7. A multimodal collaborative intelligent inspection system for smart power plants, characterized in that, The system includes a memory and a processor. The memory contains a program for a multimodal collaborative intelligent inspection method for smart power plants. When the program for the multimodal collaborative intelligent inspection method for smart power plants is executed by the processor, it performs the following steps: Multiple types of inspection equipment are deployed to different areas of the power plant. These include a coal conveying corridor track-type inspection robot, a plant area four-legged inspection robot dog, and a high-voltage power distribution room track-type lifting inspection robot. Receive preset inspection tasks, and control the inspection equipment to collect multi-dimensional data on the equipment and environment in the power plant according to the preset inspection tasks and the set mode to obtain multi-dimensional inspection data. The multi-dimensional inspection data includes equipment temperature, belt running status, video images, environmental parameters, audio information and equipment operating parameters. The collected multi-dimensional inspection data is compared with the preset safety data, and the difference information is processed. The difference value is obtained by comparing the operational difference information with the set difference threshold. Based on the difference value, multi-dimensional inspection data is analyzed to generate equipment abnormality information and environmental abnormality information. Based on multi-level alarm rules, the system analyzes equipment and environmental anomalies to generate alarms of different levels. The system then transmits multi-dimensional inspection data and alarm information to the terminal for real-time visualization.

8. The multimodal collaborative intelligent inspection system for smart power plants according to claim 7, characterized in that, Deploying various types of inspection equipment to different areas of the power plant, specifically including: A comprehensive survey of all key areas of the smart power plant was conducted to clarify the spatial layout, equipment distribution, operating environment, and key inspection points of each area. Based on the inspection needs reports from various regions, a targeted analysis was conducted on the performance parameters and suitable scenarios of the coal conveyor corridor track-type inspection robot, the plant area quadruped inspection robot dog, and the high-voltage power distribution room track-type lifting inspection robot. Based on the regional layout, equipment distribution, and inspection routes, three different planning and deployment schemes for inspection equipment were developed. According to the planned deployment scheme, the inspection equipment was deployed step by step, and the deployment results were obtained. Based on the deployment results, the inspection equipment is continuously tested to verify its deployment adaptability, and the adaptation results are obtained. Based on the adaptation results, the deployment position of the inspection equipment is adjusted.

9. The multimodal collaborative intelligent inspection system for smart power plants according to claim 8, characterized in that, The multi-dimensional inspection data collection specifically includes: Equipment temperature detection: Based on infrared thermal imager, the temperature of the equipment body and belt conveyor rollers is measured to obtain temperature data in real time and identify over-temperature anomalies; Belt running status detection: Detects belt tears, misalignment, and large foreign objects, and promptly identifies abnormal belt operation; Video surveillance: Based on high-definition visible light cameras, it captures on-site video images to monitor equipment appearance, personnel behavior, and on-site environment in real time; Environmental parameter monitoring: Monitoring on-site dust concentration, temperature and humidity, noise and gas concentration, including O2, O3, CO, SF6 and coal dust concentration; Audio information acquisition: Based on the audio acquisition equipment, the noise of the equipment is collected to enable two-way voice communication between the control room and the site; Equipment operating parameter detection: Identify readings of pressure gauges, indicator lights, thermometers, and level gauges; detect the operating status of circuit breakers, disconnect switches, and pressure plate equipment.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a multimodal collaborative intelligent inspection method program for smart power plants. When the multimodal collaborative intelligent inspection method program for smart power plants is executed by a processor, it implements the steps of the multimodal collaborative intelligent inspection method for smart power plants as described in any one of claims 1 to 6.

Images