A steel structure damage identification system and device

Abstract

This invention relates to the technical field of steel structure damage detection, and in particular to a steel structure damage identification system and equipment, including a multimodal sensing acquisition module, an edge intelligent computing module, a wireless communication transmission module, a cloud monitoring and management platform, and a graded early warning module. The multimodal sensing acquisition module employs an electro-ceramic ultrasonic sensor, a micro-strain sensor, a vibration acceleration sensor, an acoustic emission sensor, and a temperature and humidity compensation sensor. During damage identification, these sensors are simultaneously attached to the test points on the steel structure components to acquire ultrasonic guided wave signals, surface micro-strain signals, three-dimensional vibration acceleration signals, damage acoustic emission signals, and environmental temperature and humidity signals. This improves the accuracy and comprehensiveness of steel structure damage identification, provides graded responses according to damage severity, offers all-weather, blind-spot-free early warning, enables autonomous movement on various steel structure surfaces of different shapes, optimizes the limitations of fixed-point monitoring, and adapts to the comprehensive monitoring needs of complex steel structures.

Description

Technical Field

[0001] This invention relates to the technical field of steel structure damage detection, and in particular to a steel structure damage identification system and equipment. Background Technology

[0002] Steel structures, with their core advantages of high strength, light weight, excellent seismic performance, fast construction speed, and recyclability, are widely used in key engineering fields such as high-rise buildings, highway and railway bridges, large-span factories, power transmission towers, port machinery, and marine engineering platforms. During long-term service, these steel structures must continuously withstand multiple complex working conditions, including dynamic loads from vehicles and pedestrians, wind vibration loads, alternating temperature loads, corrosive media erosion, and fatigue cyclic stress. This makes them highly susceptible to fatigue cracks, corrosion thinning, loosening of connections, localized deformation, and even internal hidden damage at weld joints, bolted connections, stress concentration points, and weak sections of components.

[0003] Currently, steel structure damage identification and detection technologies are mainly divided into three categories: traditional manual inspection, conventional non-destructive testing, and fixed structural health monitoring. Each of these technologies has significant limitations in practical engineering applications, making it difficult to meet the current demands for intelligent, real-time, precise, and all-weather damage identification for large and complex steel structures. For example, the existing patent with publication number CN118980743B relates to the field of identification and monitoring device technology, specifically a steel structure damage identification and monitoring device. Its technical solution includes at least two sets of multiple eddy current sensors arranged in a circular array.

[0004] For example, the existing technology announcement number CN218368047U discloses a Mecanum wheel wall-climbing inspection robot based on magnetic suction technology. Its technical solution is: it includes a main body, and two rotating shafts are provided on the front and rear sides of the main body. The rotating shafts are connected to the main body through bearings.

[0005] During the use of existing equipment, it was found that the adsorption reliability of the existing equipment is poor when walking on irregular steel structure surfaces, and it is not convenient to be used on both flat and curved steel structure surfaces at the same time, which increases the limitation of its use. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a steel structure damage identification system and equipment that improves the accuracy and comprehensiveness of steel structure damage identification, provides graded response based on damage severity, offers all-weather, blind-spot-free early warning, enables autonomous movement on the surfaces of various steel structures with different shapes, optimizes the limitations of fixed-point monitoring, and adapts to the full-area monitoring needs of complex steel structures.

[0007] The present invention provides a steel structure damage identification system, comprising a multimodal sensing acquisition module, an edge intelligent computing module, a wireless communication transmission module, a cloud monitoring and management platform, and a hierarchical early warning module; Multimodal sensing and acquisition module: It adopts electro-ceramic ultrasonic sensor, micro-strain sensor, vibration acceleration sensor, acoustic emission sensor and temperature and humidity compensation sensor. During damage identification, it is simultaneously attached to the test point of the steel structure component to collect ultrasonic guided wave signal, surface micro-strain signal, three-dimensional vibration acceleration signal, damage acoustic emission signal and ambient temperature and humidity signal of the steel structure respectively. Edge intelligent computing module: used to complete the entire process of intelligent processing of multimodal sensor data on-site at the device end, so as to realize real-time identification, accurate location and quantitative assessment of damage; Wireless communication transmission module: adopts a dual-mode communication design, using LoRa communication for short-range distributed monitoring and 5G communication for long-range operating conditions; Cloud-based monitoring and management platform: Built on cloud servers, it receives damage data from multiple edge devices in real time, generates a cloud map of steel structure damage distribution and a damage development trend curve, and stores historical monitoring data to provide big data support for structural health assessment and operation and maintenance decisions. Tiered early warning module: In conjunction with the edge intelligent computing module and cloud platform, it sets up a three-level early warning mechanism based on the degree of damage. Minor damage triggers a Level 1 warning, indicated by a local device indicator light and recorded in the cloud. Moderate damage is classified as a Level II warning, with local audio and visual alerts and cloud-based APP push notifications. Severe damage triggers a Level 3 warning system, which includes local strong light and sound alarms, cloud-based SMS and telephone alerts, providing 24 / 7 coverage and preventing damage from worsening and causing safety accidents.

[0008] Preferably, it also includes a wall-climbing motion control module; Wall-climbing motion control module: Installed on the steel structure damage identification equipment, it is used to control the movement, steering, obstacle crossing, and obstacle avoidance of the steel structure damage identification equipment. The wall-climbing motion control module includes a magnetic adsorption control unit, a drive motor control unit, a visual obstacle avoidance unit, and a marking control unit. Magnetic adsorption control unit: used to adjust the magnetic force in real time, adapting to steel structure surfaces of different thicknesses and materials, ensuring firm adsorption and no slippage during movement; Drive motor control unit: adopts dual-loop control to realize the equipment's straight movement, turning, lateral translation, and vertical climbing actions; Visual obstacle avoidance unit: used to observe obstacles ahead in real time and trigger automatic obstacle avoidance commands; Marking control unit: Used to collect detection data from the edge intelligent computing module. When there is damage to the steel structure, the control device marks the location.

[0009] Preferably, it also includes a path planning module; Path planning module: Used to plan the walking path of steel structure damage identification equipment. It supports two modes: automatic planning and remote customization. In automatic mode, a 3D model of the steel structure can be imported to automatically generate a full-coverage inspection path, avoiding non-inspection areas such as welds and stiffeners, and automatically increasing the detection density of stress concentration areas and key stress parts. In custom mode, the inspection path can be modified remotely in the cloud.

[0010] Preferably, the edge intelligent computing module is equipped with an embedded ARM processor and a lightweight deep learning neural network model, and has built-in signal preprocessing unit, feature extraction unit, damage recognition unit and localization and quantitative calculation unit; Signal preprocessing unit: used to perform noise reduction, filtering, and detrending of multimodal raw data, and to remove environmental noise and load interference signals; Feature extraction unit: Based on wavelet packet transform and modal parameter analysis, extracts the time-domain features, frequency-domain features and energy features of the damage signal; Damage identification unit: It adopts a pre-trained lightweight CNN convolutional neural network model to classify and identify the extracted features, distinguishing between various types of damage such as normal working conditions, crack damage, loose connectors, corrosion thinning, and deformation damage. Location and Quantitative Calculation Unit: Based on time difference positioning method and strain distribution inversion algorithm, it accurately calculates the coordinate information of the damage location, and combines signal amplitude and characteristic parameters to quantitatively assess the degree of damage.

[0011] Preferably, the cloud-based monitoring and management platform has functions such as data storage, visualization, historical data query, damage trend analysis, and remote equipment control.

[0012] Preferably, a steel structure damage identification device includes a drive device, a load-bearing device, a marking assembly, a housing, a support wheel, a first traveling wheel, a support shaft, a chassis, a telescopic rod, a first spring, a second traveling wheel, and an electromagnet; The shell contains a compartment where edge intelligent computing modules, wireless communication transmission modules, cloud monitoring and management platforms, hierarchical early warning modules, wall climbing motion control modules, and path planning modules are installed. The compartment is also equipped with a power supply to power the system and equipment. The support wheel is rotatably mounted at the bottom of the front end of the housing; The first traveling wheel is mounted at the rear of the housing via a drive motor; Two sets of support shafts are rotatably mounted on both sides of the housing; The two sets of chassis are respectively mounted on the ends of the two sets of support shafts, swinging up and down and rotating. Two sets of telescopic rods are rotatably installed between two sets of support shafts and two sets of chassis; The two sets of first springs are respectively fitted onto the two sets of telescopic rods; Each pair of second traveling wheels is rotatably mounted on a set of chassis via a drive device; Electromagnets are respectively installed on the second traveling wheel and the first traveling wheel; The carrier device is mounted on the housing and is used to mount the multimodal sensing acquisition module; A marking component is installed on the housing to mark the location of damage to the steel structure. The bottom of the housing is supported by support wheels and a first traveling wheel. A drive device rotates the second traveling wheel, causing it to move the housing. The magnetic force of an electromagnet controls the first and second traveling wheels, enabling the housing to crawl along the steel structure. When the housing travels on irregular steel surfaces, the rotation of the support shaft and the housing ensures maximum contact between the second traveling wheels and the steel surface, improving the device's adhesion and climbing reliability. When traveling on curved steel surfaces, the housings on both sides swing downwards using telescopic rods and the elastic support of the first spring, maximizing the device's contact with the steel surface and preventing the second traveling wheels from becoming suspended due to changes in curvature. When climbing right-angle slopes, the device moves backward, causing the first traveling wheel to first contact the vertical wall. The drive motor then rotates the first traveling wheel, simultaneously rotating the second traveling wheels, thus achieving right-angle climbing and reducing limitations in use, improving adaptability to different scenarios.

[0013] Preferably, the marking assembly includes a guide sleeve, a crayon, a fixing sleeve, a handle set screw, a bracket, a first guide rod, a second spring, an electric cylinder, and a first guide rail; The guide sleeve is installed on the outer wall of the housing; The crayon slides onto the guide sleeve; The retaining sleeve is attached to the top of the crayon; The handle is fitted onto the retaining sleeve using a set screw. The bracket is slidably mounted on the first guide rail. The first guide rail is set on the outer wall of the housing; The first guide rod is slidably mounted on the bracket, and the bottom end of the first guide rod is connected to the fixed sleeve. The second spring is fitted onto the first guide rod; The electric cylinder is installed on the outer wall of the housing, and the moving end of the electric cylinder is connected to the bracket. The crayon is slid from bottom to top through the guide sleeve and into the fixed sleeve. Then, the handle set screw is rotated to press and fix the crayon. When damage to the steel structure is detected, the housing moves the crayon to the position to be marked. Then, the electric cylinder moves the bracket downward, so that the bracket moves the crayon downward to contact the surface of the steel structure. Then, by controlling the housing to move back and forth, the crayon is used to make marks on the surface of the steel structure.

[0014] Preferably, the bearing device includes a second guide rail, a lifting platform, a support platform, a second guide rod, a third spring, a lead screw, a first motor, and a first camera; The second guide rail is located at the front end of the outer wall of the housing; The lifting platform is slidably mounted on the second guide rail. The second guide rod is slidably installed on the lifting platform, and the bottom end of the second guide rod is connected to the support platform. The multimodal sensing acquisition module is mounted on the support platform; The third spring is fitted onto the second guide rod; The lead screw is rotatably mounted on the inner wall of the second guide rail, and the lifting platform is screwed onto the lead screw. The first motor is mounted on the outer wall of the second guide rail, and the output end of the first motor is connected to the lead screw. The first camera is installed on the outer wall of the lifting platform; the first motor drives the lead screw to rotate, which in turn drives the lifting platform to move up and down, thereby allowing the lifting platform to move up and down the support platform. This facilitates the multimodal sensing acquisition module to contact the steel structure surface for damage identification and detection. By setting a second guide rod and a third spring, the buffering effect when the multimodal sensing acquisition module contacts the steel structure is improved. The first camera takes pictures of the surface of the steel structure, improving the convenience of visually observing damage to the steel structure.

[0015] Preferably, the drive device includes a synchronous pulley, a synchronous belt, and a second motor; Synchronous pulleys are respectively installed on the rotating end of the second traveling wheel; The timing belt is installed between the timing pulleys; The second motor is mounted on the outer wall of the casing, and the output end of the second motor is connected to the second traveling wheel; the second motor provides power to rotate the second traveling wheel, so that the two traveling wheels rotate in coordination to drive the casing.

[0016] Preferably, it also includes an electric rotary table, a second camera, and a third motor; The electric rotary table is mounted on the top of the housing; The second camera is mounted on an electric rotary table. The third motor is mounted on the electric rotary table, and its output is connected to the rotating end of the second camera. The second camera captures images of the working environment, and the captured data is transmitted via a wireless communication module. The electric rotary table and the third motor work together to drive the third motor to rotate and move, improving the flexibility of capturing images from different positions and enhancing the convenience of controlling the movement of the equipment.

[0017] Compared with existing technologies, the beneficial effects of this invention are as follows: by using multiple types of sensors to simultaneously collect multi-dimensional signals, coupled with temperature and humidity compensation, the accuracy and comprehensiveness of steel structure damage identification are improved. The edge intelligent computing module completes data processing and damage assessment on-site, reducing cloud pressure and realizing real-time damage identification, location and quantitative assessment. Through a three-level hierarchical early warning mechanism linking the edge and the cloud, it responds according to the degree of damage, providing all-weather, blind-spot-free early warning and effectively preventing safety accidents. At the same time, coupled with a wall-climbing mobile device, it can move autonomously on the surface of various steel structures of different shapes, optimizing the limitations of fixed-point monitoring and adapting to the full-domain monitoring needs of complex steel structures. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram showing the connection between the multimodal sensing acquisition module and the edge intelligent computing module, etc. Figure 3 This is a schematic diagram showing the connection between the wall-climbing motion control module and the path planning module, etc. Figure 4 This is an isometric structural diagram of the connection between the housing and the second guide rail, etc. Figure 5 This is an isometric structural diagram showing the connection between the housing and the first traveling wheel, etc. Figure 6 This is a partial isometric structural diagram of the connection between the support shaft and the telescopic rod, etc. Figure 7 This is a partial isometric structural diagram showing the connection between the bracket and the first guide rod, etc. Figure 8 This is a partial isometric structural diagram of the connection between the second guide rail and the lifting platform, etc. Figure 9 This is a partial isometric structural diagram of the connection between the chassis and the second motor, etc. Figure 10 This is an isometric structural diagram showing the connection between the housing and the electric rotary table, etc.

[0019] The attached diagram is labeled as follows: 101, housing; 102, support wheel; 103, first traveling wheel; 104, support shaft; 105, chassis; 106, telescopic rod; 107, first spring; 108, second traveling wheel; 109, electromagnet; 201, guide sleeve; 202, crayon; 203, fixing sleeve; 204, handle set screw; 205, bracket; 206, first guide rod; 207, second spring; 208, electric cylinder; 209, first guide rail; 301, second guide rail; 302, lifting platform; 303, support platform; 304, second guide rod; 305, third spring; 306, lead screw; 307, first motor; 308, first camera; 401, synchronous pulley; 402, synchronous belt; 403, second motor; 501, electric rotary table; 502, second camera; 503, third motor. Detailed Implementation

[0020] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. The present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0021] Example 1 like Figures 1 to 3 As shown, a steel structure damage identification system of the present invention includes a multimodal sensing acquisition module, an edge intelligent computing module, a wireless communication transmission module, a cloud monitoring and management platform, and a hierarchical early warning module; Multimodal sensing and acquisition module: It adopts electro-ceramic ultrasonic sensor, micro-strain sensor, vibration acceleration sensor, acoustic emission sensor and temperature and humidity compensation sensor. During damage identification, it is simultaneously attached to the test point of the steel structure component to collect ultrasonic guided wave signal, surface micro-strain signal, three-dimensional vibration acceleration signal, damage acoustic emission signal and ambient temperature and humidity signal of the steel structure respectively. Edge intelligent computing module: used to complete the entire process of intelligent processing of multimodal sensor data on-site at the device end, so as to realize real-time identification, accurate location and quantitative assessment of damage; Wireless communication transmission module: adopts a dual-mode communication design, using LoRa communication for short-range distributed monitoring and 5G communication for long-range operating conditions; Cloud-based monitoring and management platform: Built on cloud servers, it receives damage data from multiple edge devices in real time, generates a cloud map of steel structure damage distribution and a damage development trend curve, and stores historical monitoring data to provide big data support for structural health assessment and operation and maintenance decisions. Tiered early warning module: In conjunction with the edge intelligent computing module and cloud platform, it sets up a three-level early warning mechanism based on the degree of damage. Minor damage triggers a Level 1 warning, indicated by a local device indicator light and recorded in the cloud. Moderate damage is classified as a Level II warning, with local audio and visual alerts and cloud-based APP push notifications. Severe damage triggers a Level 3 warning system, which includes local strong light and sound alarms, cloud-based SMS and telephone alerts, providing 24 / 7 coverage and preventing damage from worsening and causing safety accidents. It also includes a wall-climbing motion control module; Wall-climbing motion control module: Installed on the steel structure damage identification equipment, it is used to control the movement, steering, obstacle crossing, and obstacle avoidance of the steel structure damage identification equipment. The wall-climbing motion control module includes a magnetic adsorption control unit, a drive motor control unit, a visual obstacle avoidance unit, and a marking control unit. Magnetic adsorption control unit: used to adjust the magnetic force in real time, adapting to steel structure surfaces of different thicknesses and materials, ensuring firm adsorption and no slippage during movement; Drive motor control unit: adopts dual-loop control to realize the equipment's straight movement, turning, lateral translation, and vertical climbing actions; Visual obstacle avoidance unit: used to observe obstacles ahead in real time and trigger automatic obstacle avoidance commands; Marking control unit: Used to collect detection data from the edge intelligent computing module. When there is damage to the steel structure, the control device marks the location. It also includes a path planning module; Path planning module: Used to plan the walking path of steel structure damage identification equipment. It supports two modes: automatic planning and remote customization. In automatic mode, a 3D model of steel structure can be imported to automatically generate a full-coverage inspection path, avoiding non-detection areas such as welds and stiffeners, and automatically increasing the detection density of stress concentration areas and key stress parts. In custom mode, the inspection path can be modified remotely in the cloud. The edge intelligent computing module is equipped with an embedded ARM processor and a lightweight deep learning neural network model, and has built-in signal preprocessing unit, feature extraction unit, damage recognition unit and localization and quantitative calculation unit; Signal preprocessing unit: used to perform noise reduction, filtering, and detrending of multimodal raw data, and to remove environmental noise and load interference signals; Feature extraction unit: Based on wavelet packet transform and modal parameter analysis, extracts the time-domain features, frequency-domain features and energy features of the damage signal; Damage identification unit: It adopts a pre-trained lightweight CNN convolutional neural network model to classify and identify the extracted features, distinguishing between various types of damage such as normal working conditions, crack damage, loose connectors, corrosion thinning, and deformation damage. Location and Quantitative Calculation Unit: Based on time difference positioning method and strain distribution inversion algorithm, it accurately calculates the coordinate information of the damage location, and combines signal amplitude and characteristic parameters to quantitatively assess the degree of damage; The cloud-based monitoring and management platform has functions such as data storage, visualization, historical data query, damage trend analysis, and remote equipment control. In this embodiment, by employing multiple types of sensors to simultaneously collect multi-dimensional signals, coupled with temperature and humidity compensation, the accuracy and comprehensiveness of steel structure damage identification are improved. The edge intelligent computing module completes data processing and damage assessment locally, reducing cloud pressure and enabling real-time damage identification, location, and quantitative assessment. Through a three-level hierarchical early warning mechanism linking the edge and cloud, responses are graded according to the degree of damage, providing all-weather, blind-spot-free early warning and effectively preventing safety accidents. At the same time, coupled with a wall-climbing mobile device, autonomous movement is achieved on the surfaces of various steel structures of different shapes, optimizing the limitations of fixed-point monitoring and adapting to the full-area monitoring needs of complex steel structures.

[0022] Example 2 like Figures 4 to 10 As shown, a steel structure damage identification device of the present invention includes a driving device, a bearing device, a marking assembly, a housing 101, a support wheel 102, a first traveling wheel 103, a support shaft 104, a chassis 105, a telescopic rod 106, a first spring 107, a second traveling wheel 108, and an electromagnet 109. The housing 101 contains a compartment where the edge intelligent computing module, wireless communication transmission module, cloud monitoring and management platform, hierarchical early warning module, wall climbing motion control module and path planning module are installed. The compartment is also equipped with a power supply for powering the system and equipment. Support wheel 102 is rotatably mounted on the bottom front end of housing 101; The first traveling wheel 103 is mounted on the rear of the housing 101 via a drive motor; Two sets of support shafts 104 are rotatably mounted on both sides of the housing 101; The two sets of chassis 105 are respectively mounted on the ends of the two sets of support shafts 104, swinging up and down and rotating. Two sets of telescopic rods 106 are rotatably installed between two sets of support shafts 104 and two sets of chassis 105, respectively. The two sets of first springs 107 are respectively fitted onto the two sets of telescopic rods 106; Each pair of second walking wheels 108 is rotatably mounted on a set of chassis 105 via a drive device; Electromagnets 109 are respectively mounted on the second traveling wheel 108 and the first traveling wheel 103; The support device is mounted on the housing 101 and is used to mount the multimodal sensing acquisition module; A marking component is disposed on the housing 101, and the marking component is used to mark the location of damage to the steel structure; The marking assembly includes a guide sleeve 201, a crayon 202, a fixing sleeve 203, a handle set screw 204, a bracket 205, a first guide rod 206, a second spring 207, an electric cylinder 208, and a first guide rail 209. Guide sleeve 201 is installed on the outer wall of housing 101; Crayon 202 is slidably mounted on guide sleeve 201; Fixing sleeve 203 is fitted onto the top of crayon 202; The handle set screw 204 is fitted onto the fixing sleeve 203; The bracket 205 is slidably mounted on the first guide rail 209. The first guide rail 209 is disposed on the outer wall of the housing 101; The first guide rod 206 is slidably mounted on the bracket 205, and the bottom end of the first guide rod 206 is connected to the fixed sleeve 203. The second spring 207 is fitted onto the first guide rod 206; The electric cylinder 208 is installed on the outer wall of the housing 101, and the moving end of the electric cylinder 208 is connected to the bracket 205; The drive device includes a synchronous pulley 401, a synchronous belt 402, and a second motor 403; Synchronous pulleys 401 are respectively installed on the rotating end of the second traveling wheel 108; Synchronous belt 402 is installed between synchronous pulleys 401; The second motor 403 is mounted on the outer wall of the housing 105, and the output end of the second motor 403 is connected to the second traveling wheel 108. It also includes an electric rotary table 501, a second camera 502, and a third motor 503; The electric rotary table 501 is mounted on the top of the housing 101; The second camera 502 is rotatably mounted on the electric rotary table 501; The third motor 503 is mounted on the electric rotary table 501, and the output end of the third motor 503 is connected to the rotating end of the second camera 502. In this embodiment, the bottom of the housing 101 is supported by the support wheel 102 and the first traveling wheel 103. The second traveling wheel 108 is rotated by the drive device, causing the second traveling wheel 108 to move the housing 101. By controlling the magnetic force of the electromagnet 109, the first traveling wheel 103 and the second traveling wheel 108 drive the housing 101 to crawl on the steel structure. When the housing 101 travels on an irregular steel structure surface, the rotation of the support shaft 104 and the housing 105 facilitates the maximum contact between the second traveling wheels 108 on both sides and the surface of the steel structure, improving the reliability of the device's adsorption and wall climbing. When it needs to travel on a curved steel structure surface, the housings 105 on both sides swing downwards through the elastic support of the telescopic rod 106 and the first spring 107, so that the device fits the surface of the steel structure to the maximum extent and avoids the second traveling wheels 108 on both sides being suspended due to changes in curvature. When a right-angle climb is required, the control device moves backward, causing the first traveling wheel 103 to first contact the vertical wall. Then, the drive motor drives the first traveling wheel 103 to rotate, while the second traveling wheels 108 on both sides rotate, thereby realizing the right-angle climbing movement of the device, reducing the limitations of use and improving the adaptability of the scene. The crayon 202 slides from bottom to top through the guide sleeve 201 and extends into the fixed sleeve 203. Then, the handle set screw 204 is rotated and moved to press and fix the crayon 202. When damage to the steel structure is detected, the housing 101 drives the crayon 202 to move to the position to be marked. Then, the electric cylinder 208 drives the bracket 205 to move downward, so that the bracket 205 drives the crayon 202 to move downward and contact the surface of the steel structure. Then, by controlling the housing 101 to move back and forth, the crayon 202 is used to make marks on the surface of the steel structure.

[0023] Example 3 Based on Embodiment 2, the supporting device includes a second guide rail 301, a lifting platform 302, a support platform 303, a second guide rod 304, a third spring 305, a lead screw 306, a first motor 307, and a first camera 308. The second guide rail 301 is disposed at the front end of the outer side wall of the housing 101; The lifting platform 302 is slidably mounted on the second guide rail 301. The second guide rod 304 is slidably mounted on the lifting platform 302, and the bottom end of the second guide rod 304 is connected to the support platform 303. The multimodal sensing acquisition module is mounted on the support platform 303; The third spring 305 is fitted onto the second guide rod 304; The lead screw 306 is rotatably mounted on the inner wall of the second guide rail 301, and the lifting platform 302 is screwed onto the lead screw 306. The first motor 307 is mounted on the outer wall of the second guide rail 301, and the output end of the first motor 307 is connected to the lead screw 306. The first camera 308 is installed on the outer wall of the lifting platform 302. The first motor 307 drives the lead screw 306 to rotate, which in turn drives the lifting platform 302 to move up and down. This, in turn, drives the support platform 303 to move up and down, making it easier for the multimodal sensing acquisition module to come into contact with the steel structure surface for damage identification and detection. By setting the second guide rod 304 and the third spring 305, the buffering effect when the multimodal sensing acquisition module comes into contact with the steel structure is improved. The first camera 308 takes pictures of the surface of the steel structure, improving the convenience of visually observing damage to the steel structure.

[0024] The main functions achieved by this invention are: 1. Improve the accuracy and comprehensiveness of steel structure damage identification. The edge intelligent computing module completes data processing and damage assessment locally, reducing cloud pressure and realizing real-time damage identification, location and quantitative assessment. The edge and cloud are linked through a three-level hierarchical early warning mechanism to respond according to the degree of damage. 2. It can move autonomously on various steel structure surfaces of different shapes, including flat, curved, and irregular surfaces, reducing limitations in use and improving adaptability to different scenarios.

[0025] The electromagnet 109, electric cylinder 208, first motor 307, first camera 308, second motor 403, electric rotary table 501, second camera 502 and third motor 503 of the steel structure damage identification system and equipment of the present invention are commercially available. Technical personnel in this industry only need to install and operate it according to the accompanying instruction manual, without requiring any creative work from those skilled in the art.

[0026] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A steel structure damage identification system, characterized in that, It includes a multimodal sensing and acquisition module, an edge intelligent computing module, a wireless communication transmission module, a cloud monitoring and management platform, and a hierarchical early warning module; Multimodal sensing and acquisition module: It adopts electro-ceramic ultrasonic sensor, micro-strain sensor, vibration acceleration sensor, acoustic emission sensor and temperature and humidity compensation sensor. During damage identification, it is simultaneously attached to the test point of the steel structure component to collect ultrasonic guided wave signal, surface micro-strain signal, three-dimensional vibration acceleration signal, damage acoustic emission signal and ambient temperature and humidity signal of the steel structure respectively. Edge intelligent computing module: used to complete the entire process of intelligent processing of multimodal sensor data on-site at the device end, so as to realize real-time identification, accurate location and quantitative assessment of damage; Wireless communication transmission module: adopts a dual-mode communication design, using LoRa communication for short-range distributed monitoring and 5G communication for long-range operating conditions; Cloud-based monitoring and management platform: Built on cloud servers, it receives damage data from multiple edge devices in real time, generates a cloud map of steel structure damage distribution and a damage development trend curve, and stores historical monitoring data to provide big data support for structural health assessment and operation and maintenance decisions. Tiered early warning module: In conjunction with the edge intelligent computing module and cloud platform, it sets up a three-level early warning mechanism based on the degree of damage. Minor damage triggers a Level 1 warning, indicated by a local device indicator light and recorded in the cloud. Moderate damage is classified as a Level II warning, with local audio and visual alerts and cloud-based APP push notifications. Severe damage triggers a Level 3 warning system, which includes local strong light and sound alarms, cloud-based SMS and telephone alerts, providing 24 / 7, all-weather, comprehensive warning coverage.

2. The steel structure damage identification system as described in claim 1, characterized in that, It also includes a wall-climbing motion control module; Wall-climbing motion control module: Installed on the steel structure damage identification equipment, it is used to control the movement, steering, obstacle crossing, and obstacle avoidance of the steel structure damage identification equipment. The wall-climbing motion control module includes a magnetic adsorption control unit, a drive motor control unit, a visual obstacle avoidance unit, and a marking control unit. Magnetic adsorption control unit: used to adjust the magnetic force in real time, adapting to steel structure surfaces of different thicknesses and materials, ensuring firm adsorption and no slippage during movement; Drive motor control unit: adopts dual-loop control to realize the equipment's straight movement, turning, lateral translation, and vertical climbing actions; Visual obstacle avoidance unit: used to observe obstacles ahead in real time and trigger automatic obstacle avoidance commands; Marking control unit: Used to collect detection data from the edge intelligent computing module. When there is damage to the steel structure, the control device marks the location.

3. The steel structure damage identification system as described in claim 1, characterized in that, It also includes a path planning module; Path planning module: Used to plan the walking path of steel structure damage identification equipment. It supports two modes: automatic planning and remote customization. In automatic mode, a 3D model of the steel structure can be imported to automatically generate a full-coverage inspection path, avoiding non-inspection areas such as welds and stiffeners, and automatically increasing the detection density of stress concentration areas and key stress parts. In custom mode, the inspection path can be modified remotely in the cloud.

4. The steel structure damage identification system as described in claim 1, characterized in that, The edge intelligent computing module is equipped with an embedded ARM processor and a lightweight deep learning neural network model, and has built-in signal preprocessing unit, feature extraction unit, damage recognition unit and localization and quantitative calculation unit; Signal preprocessing unit: used to perform noise reduction, filtering, and detrending of multimodal raw data, and to remove environmental noise and load interference signals; Feature extraction unit: Based on wavelet packet transform and modal parameter analysis, extracts the time-domain features, frequency-domain features and energy features of the damage signal; Damage identification unit: It adopts a pre-trained lightweight CNN convolutional neural network model to classify and identify the extracted features, distinguishing between various types of damage such as normal working conditions, crack damage, loose connectors, corrosion thinning, and deformation damage. Location and Quantitative Calculation Unit: Based on time difference positioning method and strain distribution inversion algorithm, it accurately calculates the coordinate information of the damage location, and combines signal amplitude and characteristic parameters to quantitatively assess the degree of damage.

5. A steel structure damage identification system as described in claim 1, characterized in that, The cloud-based monitoring and management platform has functions such as data storage, visualization, historical data query, damage trend analysis, and remote equipment control.

6. A steel structure damage identification device, characterized in that, It includes a drive unit, a load-bearing device, a marking assembly, a housing (101), a support wheel (102), a first traveling wheel (103), a support shaft (104), a chassis (105), a telescopic rod (106), a first spring (107), a second traveling wheel (108), and an electromagnet (109). The housing (101) is equipped with a compartment, in which the edge intelligent computing module, wireless communication transmission module, cloud monitoring and management platform, hierarchical early warning module, wall climbing motion control module and path planning module are installed. The compartment is also equipped with a power supply for powering the system and equipment. The support wheel (102) is rotatably mounted on the bottom front end of the housing (101); The first traveling wheel (103) is mounted on the rear of the housing (101) by a drive motor; Two sets of support shafts (104) are rotatably mounted on both sides of the housing (101); The two sets of chassis (105) are respectively mounted on the ends of the two sets of support shafts (104) by swinging up and down and rotating. Two sets of telescopic rods (106) are rotatably installed between two sets of support shafts (104) and two sets of chassis (105); The two sets of first springs (107) are respectively fitted onto the two sets of telescopic rods (106); Each pair of second walking wheels (108) is rotatably mounted on a set of chassis (105) via a drive device; Electromagnets (109) are respectively mounted on the second traveling wheel (108) and the first traveling wheel (103); The carrier is mounted on the housing (101) and is used to mount the multimodal sensing acquisition module; The marking assembly is disposed on the housing (101) and is used to mark the location of damage to the steel structure.

7. A steel structure damage identification device as described in claim 6, characterized in that, The marking assembly includes a guide sleeve (201), a crayon (202), a fixing sleeve (203), a handle set screw (204), a bracket (205), a first guide rod (206), a second spring (207), an electric cylinder (208), and a first guide rail (209). The guide sleeve (201) is installed on the outer wall of the housing (101); The crayon (202) is slidably mounted on the guide sleeve (201); The retaining sleeve (203) is fitted onto the top of the crayon (202); The handle set screw (204) is fitted onto the fixing sleeve (203); The bracket (205) is slidably mounted on the first guide rail (209); The first guide rail (209) is set on the outer wall of the housing (101); The first guide rod (206) is slidably mounted on the bracket (205) and the bottom end of the first guide rod (206) is connected to the fixed sleeve (203); The second spring (207) is fitted onto the first guide rod (206); The electric cylinder (208) is installed on the outer wall of the housing (101), and the moving end of the electric cylinder (208) is connected to the bracket (205).

8. A steel structure damage identification device as described in claim 6, characterized in that, The supporting device includes a second guide rail (301), a lifting platform (302), a support platform (303), a second guide rod (304), a third spring (305), a lead screw (306), a first motor (307), and a first camera (308). The second guide rail (301) is located at the front end of the outer side wall of the housing (101); The lifting platform (302) is slidably mounted on the second guide rail (301); The second guide rod (304) is slidably mounted on the lifting platform (302), and the bottom end of the second guide rod (304) is connected to the support platform (303); The multimodal sensing acquisition module is mounted on the support platform (303); The third spring (305) is fitted onto the second guide rod (304); The lead screw (306) is rotatably mounted on the inner wall of the second guide rail (301), and the lifting platform (302) is screwed onto the lead screw (306); The first motor (307) is mounted on the outer wall of the second guide rail (301), and the output end of the first motor (307) is connected to the lead screw (306); The first camera (308) is installed on the outer wall of the lifting platform (302).

9. A steel structure damage identification device as described in claim 6, characterized in that, The drive device includes a synchronous pulley (401), a synchronous belt (402), and a second motor (403). Synchronous pulleys (401) are respectively installed on the rotating end of the second traveling wheel (108); The timing belt (402) is installed between the timing pulleys (401); The second motor (403) is mounted on the outer wall of the chassis (105), and the output end of the second motor (403) is connected to the second traveling wheel (108).

10. A steel structure damage identification device as described in claim 6, characterized in that, It also includes an electric rotary table (501), a second camera (502) and a third motor (503); An electric rotary table (501) is mounted on top of the housing (101); The second camera (502) is rotatably mounted on the electric rotary table (501); The third motor (503) is mounted on the electric rotary table (501), and the output end of the third motor (503) is connected to the rotating end of the second camera (502).

Images