Paint floor shrinkage joint repairing device and repairing method thereof

Abstract

The invention belongs to the technical field of building terraces, and particularly relates to a paint terrace shrinkage joint repairing device and a repairing method thereof.The repairing method includes the steps that intelligent recognition of crack features is achieved through a multi-modal sensing network, feature fusion is conducted on multi-source data through a convolutional neural network algorithm, structural cracks and surface cracks are automatically distinguished, and a repairing result is obtained. Obtaining a detection map; based on the detection map, comprehensively considering crack parameters, environmental conditions and construction period requirements, and automatically generating an optimal repairing scheme; the self-adaptive mixing system is driven to complete dynamic modulation of the repairing material; high-precision filling operation is executed through the intelligent mechanical arm; a full-periodicity performance evaluation model is constructed through an Internet of Things monitoring terminal, a repairing result is monitored, an intelligent mechanical arm is adopted, a path is re-planned in real time through an RRT algorithm, and Z-shaped layered filling is executed, so that the filling fullness is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of building floor, and particularly relates to a paint floor contraction joint repairing device and a repairing method thereof. BACKGROUND

[0002] The intermediate green channel of the silicon steel annealing line is a visiting special channel, and the visiting frequency is relatively high. The maintenance standard of the visiting channel area is high. The contraction joint in the green channel is damaged in the change process of factors such as temperature difference caused by seasonal change. After frequent repair of the small repair of the infrastructure, the damage starts again in about three months. Frequent repair cannot effectively solve the problem.

[0003] In the existing paint floor contraction joint repairing practice, the traditional process usually follows the operation process of "manual survey - manual cutting - joint filling construction - manual maintenance".

[0004] Such method has significant technical bottlenecks: first, manual visual detection is greatly influenced by subjective experience, and it is difficult to accurately identify key parameters such as crack width, depth and extension trend; second, material allocation depends on manual experience, and cannot dynamically adjust the ratio according to different working conditions, often leading to insufficient bonding force of the repairing material and the substrate; third, the mechanized construction equipment lacks environmental perception ability, and in the scene of complex floor curved surface or temperature change, it is easy to have problems such as excessive filling of repairing material or solidification shrinkage cracking.

[0005] Therefore, the application provides a paint floor contraction joint repairing device and a repairing method thereof. SUMMARY

[0006] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.

[0007] The technical scheme adopted by the application to solve the technical problems is: the paint floor contraction joint repairing method provided by the application comprises

[0008] Intelligent identification of crack features is realized through a multi-modal sensing network, multi-source data is subjected to feature fusion through a convolutional neural network (CNN) algorithm, micro-cracks of ≤0.3mm are identified, and structural cracks and surface cracks are automatically distinguished, so as to obtain a detection atlas;

[0009] The closed-loop repairing method automatically generates an optimal repairing scheme based on the detection atlas, comprehensively considers crack parameters, environmental conditions and construction period requirements, and automatically generates an optimal repairing scheme;

[0010] An adaptive mixing system is driven to complete dynamic modulation of the repairing material; and a smart mechanical arm is relied on to perform high-precision filling operation;

[0011] A whole-cycle performance evaluation model is constructed through an Internet of Things monitoring terminal to monitor the repairing result.

[0012] Preferably, the intelligent identification of crack features through a multi-modal sensor network includes the following specific processes:

[0013] The laser three-dimensional scanning and infrared thermal imaging fusion technology is adopted, in which the laser scanner obtains the three-dimensional profile data of the crack, the infrared sensor captures the temperature field distribution of the crack area, the multi-source data are fused through the convolution neural network algorithm, the identification of ≤0.3mm micro-cracks and the automatic differentiation of structural cracks and surface cracks are realized.

[0014] Preferably, in the fusion detection process of laser three-dimensional scanning and infrared thermal imaging, the detection robot is controlled to travel at a speed of 0.5m / s, the scanning frequency of the laser scanner is 100Hz, the obtained point cloud data and thermal imaging atlas are transmitted to the vehicle-mounted industrial computer in real time, the three-dimensional visualization atlas containing the crack position, size and direction is generated through the crack feature extraction algorithm, and the crack hazard level is classified.

[0015] Preferably, the dynamic modulation process of the self-adaptive mixing system includes:

[0016] The environmental parameters are collected in real time by the built-in temperature and humidity sensor and the viscometer, the heating module is automatically triggered when the environmental temperature is detected to be lower than 15℃, the gradient filling mode is started for the cracks with a depth exceeding 5mm, the high-flow repair material with a viscosity of 1200-1500cP is used in the bottom layer, and the high-elasticity compound material with a breaking elongation of ≥300% is switched in the surface layer.

[0017] Preferably, during the dynamic modulation of the material, the components A and B are delivered to the dynamic mixer at a flow control precision of ±0.1mL / min through high-precision metering pumps, 50-100 mesh quartz sand fillers are automatically added according to the crack depth data, and the stirring speed is adjusted in real time through the viscosity sensor during the mixing process to ensure that the mixed material is in the best construction state.

[0018] Preferably, the high-precision filling operation of the intelligent mechanical arm includes: the dynamic path planning and force control execution are realized through the end pressure sensor and the visual positioning module, the travel speed and the material extrusion amount are adjusted in real time during the filling process, and the path is re-planned in real time through the RRT algorithm when the base mutation is detected.

[0019] Preferably, the flexible extrusion head at the end of the intelligent mechanical arm maintains a working distance of 5-10mm from the floor, the millimeter-level positioning is realized through the visual positioning system to identify the floor marker points, and the "zigzag" reciprocating filling process is adopted for the cracks with a width of ≥1mm.

[0020] Preferably, the whole-cycle performance evaluation process of the Internet of Things monitoring terminal includes:

[0021] Micro strain sensors and humidity sensors are implanted in the repair area. By collecting interface stress-strain data and curing humidity curves in real time, when the interface stress preset pressure or curing humidity is lower than 50%, early warning information is automatically pushed and maintenance equipment is linked to shorten the time for early defect detection.

[0022] Preferably, the closed-loop repair method generates an optimal solution by integrating crack parameters, environmental conditions and construction period requirements through a multi-objective optimization model.

[0023] Preferably, a paint floor shrinkage joint repair device comprises:

[0024] Multimodal detection unit, integrating a laser 3D scanner and an infrared thermal imager, obtains the 3D contour of the crack through the laser 3D scanner and captures the thermal field distribution through the infrared thermal imager;

[0025] Edge computing unit, with built-in neural network model, performs spatiotemporal calibration, feature extraction and risk classification on detection data;

[0026] The adaptive mixing unit is equipped with a high-precision metering pump and a dynamic mixer. It automatically adjusts the distribution ratio of Group A and Group B and the amount of filler added according to the instructions of the edge computing unit, and has an integrated heating module.

[0027] The intelligent execution unit uses an intelligent robotic arm with a pressure sensor and a flexible extrusion head at the end, and realizes path planning through the RRT algorithm;

[0028] The IoT monitoring unit, which includes a micro strain sensor and a LoRa communication module, collects interface stress and curing humidity in real time, and links with maintenance equipment to perform closed-loop control.

[0029] The beneficial effects of the present invention are as follows:

[0030] 1. The device and method for repairing shrinkage joints in painted floors described in the present invention use multimodal sensing and intelligent recognition technology to achieve millimeter-level quantitative detection of crack characteristics, solving the problem of insufficient accuracy of traditional visual inspection. They also use edge computing and a multi-objective optimization model to achieve real-time generation of repair decisions, solving the problems of cloud-based computing delays and rough decision-making.

[0031] 2. The device and method for repairing shrinkage joints in painted floors, described in this invention, utilize adaptive material mixing and gradient filling mechanisms to achieve dynamic and precise modulation of the repair material. Equipped with a high-precision metering pump and a temperature-viscosity closed-loop system, the device automatically switches between a bottom layer of high-flow material and a surface layer of high-elasticity material based on crack depth, while dynamically adjusting the curing agent ratio based on ambient temperature. This solves the problem of premature cracking caused by incomplete material curing and insufficient adhesion to the substrate.

[0032] 3. The device and method for repairing shrinkage joints in painted floors described in the present invention employ an intelligent robotic arm and, through real-time path replanning using the RRT algorithm, perform zigzag layered filling for wide cracks, thereby improving filling fullness. This addresses the issues of incomplete filling and bubble generation associated with traditional equipment in complex flooring scenarios. The IoT monitoring and closed-loop maintenance system enable full-cycle performance tracking of the repair layer, addressing the high rework rate caused by traditional post-inspection testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below with reference to the accompanying drawings.

[0034] Figure 1 This is a flow chart of the method of embodiment 1 of the present invention. DETAILED DESCRIPTION

[0035] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0036] Example 1

[0037] like Figure 1 As shown, a method for repairing shrinkage joints of a paint floor according to an embodiment of the present invention comprises:

[0038] Intelligent recognition of crack features is achieved through a multimodal sensing network. A convolutional neural network (CNN) algorithm is used to fuse features from multiple source data, identifying microcracks ≤ 0.3 mm and automatically distinguishing between structural cracks and surface cracks to generate a detection map.

[0039] During the inspection process, a fusion of laser 3D scanning and infrared thermal imaging technology is employed. The laser scanner acquires 3D crack profile data with a spatial resolution of 0.1mm, while an onboard infrared sensor captures the temperature distribution in the crack area. Using a convolutional neural network (CNN) algorithm, this multi-source data is integrated to accurately identify microcracks ≤ 0.3mm and automatically distinguish between structural cracks and surface cracks. Compared to traditional visual inspection, this system improves crack identification accuracy and detection efficiency.

[0040] Based on the detection map, a multi-objective optimization model is activated to automatically generate the optimal repair plan by comprehensively considering crack parameters, environmental conditions (real-time temperature and humidity, substrate moisture content) and construction period requirements;

[0041] Drive the adaptive mixing system to dynamically modulate the repair material; rely on the intelligent robotic arm to perform high-precision filling operations;

[0042] An intelligent mixing system based on fuzzy logic control is designed, with built-in temperature and humidity sensors and viscometers to collect environmental parameters in real time. When the ambient temperature is detected to be below 15°C, the system automatically triggers the heating module and increases the curing agent ratio (with an adjustment accuracy of ±0.5%). If the crack depth exceeds 5mm, a gradient filling mode is activated, with a high-flowability repair material (viscosity controlled at 1200-1500cP) used as the bottom layer and a high-elasticity compound material (elongation at break ≥300%) switched to the surface layer.

[0043] The system uses an intelligent robotic arm as the execution unit, equipped with a pressure sensor and a visual positioning module at its end. During the filling process, the robotic arm automatically adjusts its speed and material extrusion rate based on real-time feedback from contact pressure. If a sudden change in the substrate (such as a localized hollow) is detected, the system uses an improved RRT* algorithm to replan the path in real time, avoiding the rigid collision issues common in traditional mechanical construction.

[0044] Build a full-cycle performance evaluation model through the IoT monitoring terminal to monitor the repair results.

[0045] An IoT monitoring system based on LoRa technology is constructed, and micro strain sensors and humidity sensors are implanted in the repair area to collect stress-strain data and curing humidity curves of the repair interface in real time. The data is pre-processed through the edge computing gateway. When the interface stress exceeds the threshold (set to 1.2MPa) or the curing humidity is lower than 50%, the system automatically pushes an early warning message to the management terminal and links the maintenance equipment to initiate moisturizing measures.

[0046] In this embodiment, the intelligent recognition of crack characteristics is achieved through a multimodal sensing network, and the specific process includes:

[0047] Laser 3D scanning and infrared thermal imaging fusion technology are used. The laser scanner obtains 3D contour data of cracks with a spatial resolution of 0.1mm, and the infrared sensor captures the temperature field distribution in the crack area. The multi-source data is fused with features through the convolutional neural network algorithm to realize the identification of microcracks ≤0.3mm and the automatic distinction between structural cracks and surface cracks.

[0048] In this embodiment, during the fusion detection process of laser three-dimensional scanning and infrared thermal imaging, the detection robot is controlled to move at a speed of 0.5m / s, the laser scanner has a scanning frequency of 100Hz, and the acquired point cloud data and thermal imaging map are transmitted to the on-board industrial computer in real time. A three-dimensional visualization map including the crack position, size, and direction is generated through a self-developed crack feature extraction algorithm, and the crack hazard level is graded according to the BSEN13813 standard.

[0049] In this embodiment, the dynamic modulation process of the adaptive mixing system includes:

[0050] Environmental parameters are collected in real time through the built-in temperature and humidity sensors and viscometer. When the ambient temperature is detected to be lower than 15°C, the heating module is automatically triggered and the curing agent ratio is adjusted with an accuracy of ±0.5%. For cracks with a depth exceeding 5mm, the gradient filling mode is activated, and the bottom layer uses a high-flowability repair material with a viscosity of 1200-1500cP, and the surface layer is switched to a high-elasticity compound material with an elongation at break ≥300%.

[0051] In this embodiment, when the materials are dynamically modulated, components A and B are delivered to the dynamic mixer through a high-precision metering pump with a flow control accuracy of ±0.1 mL / min. At the same time, 0-5% of 50-100 mesh quartz sand filler is automatically added according to the crack depth data. During the mixing process, the viscosity sensor is used to adjust the stirring speed of 500-1500 rpm in real time to ensure that the mixture is in the optimal construction state.

[0052] In this embodiment, the high-precision filling operation of the intelligent robotic arm includes: dynamic path planning and force control execution through the end pressure sensor and visual positioning module, real-time adjustment of the travel speed and material extrusion volume with a pressure control accuracy of ±0.1N during the filling process, and real-time path re-planning through the improved RRT* algorithm when a substrate mutation is detected, ensuring that the material filling fullness reaches more than 99.3% when operating on curved surfaces with a curvature radius ≥100mm.

[0053] In this embodiment, the flexible extrusion head at the end of the intelligent robotic arm maintains a working distance of 5-10mm from the floor, and the floor marking points are identified by the visual positioning system to achieve millimeter-level positioning. A "zigzag" reciprocating filling process is adopted for wide cracks ≥1mm. The filling thickness of each layer is controlled at 0.5-1mm, the filling pressure range is 0.5-2MPa, and the moving speed is 0.1-0.3m / min.

[0054] In this embodiment, the full-cycle performance evaluation process of the IoT monitoring terminal includes:

[0055] Micro strain sensors and humidity sensors are implanted in the repair area, and the interface stress-strain data and curing humidity curve are collected in real time through LoRa technology. When the interface stress preset pressure (the preset pressure is obtained based on historical experience and is generally 1.2MPa) or the curing humidity is lower than 50%, an early warning message is automatically pushed and the maintenance equipment is linked to shorten the early defect detection time to within 4 hours.

[0056] In this embodiment, the closed-loop repair method generates an optimal solution by integrating crack parameters, environmental conditions and construction period requirements through a multi-objective optimization model, realizing a technical upgrade from two-dimensional visual detection to three-dimensional intelligent perception, from manual experience control to data-driven closed-loop regulation, and real-time monitoring of the entire cycle from post-detection.

[0057] Example two

[0058] Comparative example one, wherein another embodiment of the present invention is:

[0059] In this embodiment, a paint floor shrink joint repair device adopts the above-mentioned paint floor shrink joint repair method, comprising:

[0060] The multi-modal detection unit integrates a laser three-dimensional scanner and an infrared thermal imager, acquires a three-dimensional profile of the crack through the laser three-dimensional scanner, and captures a thermal field distribution through the infrared thermal imager.

[0061] The edge computing unit is built-in with a neural network model, performs spatio-temporal calibration, feature extraction and risk grading on the detection data.

[0062] The adaptive mixing unit is configured with a high-precision metering pump and a dynamic mixer, automatically adjusts the distribution ratio of A group and B group and the amount of filler added according to the instruction of the edge computing unit, and is integrated with a heating module.

[0063] The intelligent execution unit adopts an intelligent mechanical arm, the end of which is equipped with a pressure sensor and a flexible extrusion head, and realizes path planning through an RRT algorithm.

[0064] The Internet of Things monitoring unit contains a micro-strain sensor and a LoRa communication module, and collects interface stress and curing humidity in real time, and executes closed-loop control in conjunction with the maintenance equipment.

[0065] The above shows and describes the basic principles, main features and advantages of the present application. It should be understood by those skilled in the art that the present application is not limited to the above examples, and the above examples and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for repairing shrinkage joints in painted floors, characterized by: include Intelligent recognition of crack features is achieved through a multimodal sensing network. Feature fusion of multi-source data is performed using a convolutional neural network algorithm, and structural cracks and surface cracks are automatically distinguished to obtain a detection map. Based on the detection map, a closed-loop repair method is used to comprehensively consider crack parameters, environmental conditions and construction period requirements to automatically generate the optimal repair plan; Drive the adaptive mixing system to dynamically modulate the repair material; rely on the intelligent robotic arm to perform high-precision filling operations; Build a full-cycle performance evaluation model through the IoT monitoring terminal to monitor the repair results.

2. A method for repairing shrinkage joints of a painted floor according to claim 1, characterized in that: The intelligent identification of crack characteristics is achieved through the multimodal sensing network, and the specific process includes: Laser 3D scanning and infrared thermal imaging fusion technology are used, in which the laser scanner obtains the 3D contour data of the crack, and the infrared sensor captures the temperature field distribution in the crack area. The multi-source data is fused with features through the convolutional neural network algorithm to realize the identification of microcracks ≤0.3mm and the automatic distinction between structural cracks and surface cracks.

3. A method for repairing shrinkage joints of a painted floor according to claim 2, characterized in that: During the fusion detection process of laser three-dimensional scanning and infrared thermal imaging, the detection robot is controlled to move at a speed of 0.5m / s, the laser scanner has a scanning frequency of 100Hz, and the acquired point cloud data and thermal imaging map are transmitted to the on-board industrial computer in real time. A three-dimensional visual map containing the crack location, size, and direction is generated through the crack feature extraction algorithm, and the crack hazard level is graded.

4. A method for repairing shrinkage joints of a painted floor according to claim 1, characterized in that: The dynamic modulation process of the adaptive mixing system includes: Environmental parameters are collected in real time through the built-in temperature and humidity sensors and viscometer. When the ambient temperature is detected to be lower than 15°C, the heating module is automatically triggered. For cracks deeper than 5mm, a gradient filling mode is activated. The bottom layer uses a high-flowability repair material with a viscosity of 1200-1500cP, and the surface layer is switched to a high-elasticity compound material with an elongation at break ≥300%.

5. A method for repairing shrinkage joints of a painted floor according to claim 4, characterized in that: When the materials are dynamically modulated, components A and B are delivered to the dynamic mixer through a high-precision metering pump with a flow control accuracy of ±0.1mL / min. At the same time, 50-100 mesh quartz sand filler is automatically added according to the crack depth data. During the mixing process, the stirring speed is adjusted in real time through a viscosity sensor to ensure that the mixture is in the best construction state.

6. A method for repairing shrinkage joints of a painted floor according to claim 1, characterized in that: The high-precision filling operation of the intelligent robotic arm includes: dynamic path planning and force control execution through the end pressure sensor and visual positioning module, real-time adjustment of the travel speed and material extrusion volume during the filling process, and real-time path replanning through the RRT algorithm when a substrate mutation is detected.

7. A method for repairing shrinkage joints of a painted floor according to claim 6, characterized in that: The flexible extrusion head at the end of the intelligent robotic arm maintains a 5-10mm operating distance from the floor, and recognizes floor marking points through a visual positioning system to achieve millimeter-level positioning. A "zigzag" reciprocating filling process is used for cracks ≥1mm wide.

8. The method for repairing shrinkage joints of a painted floor according to claim 1, characterized in that: The full-cycle performance evaluation process of the IoT monitoring terminal includes: Micro strain sensors and humidity sensors are implanted in the repair area. By collecting interface stress-strain data and curing humidity curves in real time, when the interface stress preset pressure or curing humidity is lower than 50%, early warning information is automatically pushed and maintenance equipment is linked to shorten the time for early defect detection.

9. A method for repairing shrinkage joints of a painted floor according to claim 8, characterized in that: The closed-loop repair method generates an optimal solution by integrating crack parameters, environmental conditions and construction period requirements through a multi-objective optimization model.

10. A paint floor shrinkage joint repair device, using the paint floor shrinkage joint repair method according to claim 8, characterized in that: include: Multimodal detection unit, integrating a laser 3D scanner and an infrared thermal imager, obtains the 3D contour of the crack through the laser 3D scanner and captures the thermal field distribution through the infrared thermal imager; Edge computing unit, with built-in neural network model, performs spatiotemporal calibration, feature extraction and risk classification on detection data; The adaptive mixing unit is equipped with a high-precision metering pump and a dynamic mixer. It automatically adjusts the distribution ratio of Group A and Group B and the amount of filler added according to the instructions of the edge computing unit, and has an integrated heating module. The intelligent execution unit uses an intelligent robotic arm with a pressure sensor and a flexible extrusion head at the end, and realizes path planning through the RRT algorithm; The IoT monitoring unit, which includes a micro strain sensor and a LoRa communication module, collects interface stress and curing humidity in real time, and links with maintenance equipment to perform closed-loop control.

Images