3D model defect detection system based on AI identification
By adopting a 3D model defect detection system based on AI recognition in the fire simulation and detection system, integrating a central control room, laboratory and a variety of advanced detection and analysis technologies, the problems of inaccurate data acquisition, difficult real-time monitoring of the test process and insufficient safety protection measures in traditional systems are solved, and comprehensive monitoring and efficient safety protection of the combustion test process are achieved.
Patent Information
- Application Number
- CN202510306712.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-14
- Publication Date
- 2025-06-20
- Estimated Expiration
- 2045-03-14
AI Technical Summary
Traditional fire simulation and detection systems have problems such as inaccurate data collection, difficulty in real-time monitoring of the test process, and insufficient safety protection measures, which limit the effectiveness and safety of combustion tests.
The 3D model defect detection system based on AI recognition is adopted, and the central control room, laboratory, flame analysis system, data acquisition system, AI acquisition system and safety protection system are integrated to achieve real-time monitoring, accurate data acquisition and efficient safety protection of the combustion test process.
It has achieved comprehensive monitoring, accurate data collection and efficient safety protection of the combustion test process, significantly improved the efficiency of fire warning and emergency response, and improved the accuracy and safety of combustion tests.
Smart Images

Figure CN120177704A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of simulation test devices, and particularly to a 3D model defect detection system based on AI recognition. Background Art
[0002] In the fields of fire technology and industrial safety, the research on combustion tests and fire simulations is of crucial importance. These studies not only help to deeply understand the causes, spread processes and impacts of fires, but also provide important experimental bases for the design, performance evaluation and optimization of fire-fighting equipment. However, traditional fire simulation and detection systems have many limitations, such as inaccurate data collection, difficult real-time monitoring of the test process, insufficient safety protection measures, etc. These problems greatly limit the effectiveness and safety of combustion tests.
[0003] With the rapid development of artificial intelligence (AI) technology, its applications in image recognition, data processing and automatic control are becoming increasingly widespread, providing new possibilities for the innovation of fire simulation and detection systems. The intelligent detection system based on AI recognition can achieve real-time monitoring, accurate recognition and rapid response to the fire scene, thus significantly improving the efficiency of fire early warning and emergency response. However, most of the current fire detection systems based on AI recognition on the market focus on the ex post monitoring and analysis of fires that have already occurred, and pay less attention to the fire prevention and control during the combustion test process.
[0004] In addition, during the combustion test process, how to effectively collect and analyze key data such as flame characteristics, combustion products and heat release of the fire source is of great significance for evaluating the performance of fire-fighting equipment and optimizing fire prevention and control strategies. However, traditional data collection systems often have problems such as limited data collection range, insufficient accuracy and poor real-time performance, and are difficult to meet the high standards of modern combustion tests.
[0005] At the same time, the safety protection measures during the combustion test process are also an important aspect that cannot be ignored. Most traditional safety protection systems focus on the physical isolation of the test site and the evacuation of personnel, while the measures in aspects such as blocking the heat in the air, the erosion of chemical substances and ensuring communication during the test process are relatively weak, and it is difficult to comprehensively ensure the safety of test personnel and equipment. Summary of the Invention
[0006] The purpose of the present invention is to provide a 3D model defect detection system based on AI recognition to solve the problems mentioned in the above background technology, so as to achieve comprehensive monitoring, accurate data collection and efficient safety protection of the combustion test process, which is of great significance for improving the overall level of fire simulation and detection technology. This system can realize real-time monitoring, accurate data collection and efficient safety protection of the combustion test process by integrating a central control room, a laboratory and a variety of advanced detection and analysis technologies, providing strong support for the development of the fire protection technology and industrial safety fields.
[0007] To achieve the above object, the present invention provides the following technical solution: A 3D model defect detection system based on AI recognition, characterized in that it includes: a central control room and a laboratory. Inside the central control room, there are a central centralized control and acquisition system, a switch group, a central control system and a safety loudspeaker system. The central centralized control and acquisition system is connected to the central control system through the switch group, and the safety loudspeaker system is directly connected to the central control system. The central control system is connected to a data processing system and two switches. One of the switches is connected to a display system and a conference system set outside the central control room. Inside the laboratory, there are a flame analysis system, a data acquisition system, a heat release acquisition system of the fire source, an AI acquisition system, an ignition system, a safety protection system, a flue gas analysis and treatment system and a fire extinguishing system. The flame analysis system, the data acquisition system and the heat release acquisition system of the fire source are connected through a data acquisition control box set outside the central control room and the central centralized control and acquisition system inside the central control room. The AI acquisition system is connected to the switch group inside the central control room. The safety protection system includes an equipment safety guarantee system and a personnel safety protection system. The flue gas analysis and treatment system includes a ventilation and smoke exhaust system, a flue gas collection system, a flue gas data acquisition system and an exhaust gas treatment system. The ignition system, the safety protection system and the flue gas analysis and treatment system are connected to the other switch set inside the central control room through a central centralized control execution system set outside the central control room.
[0008] Preferably, the AI acquisition system includes: a video acquisition module, a data preprocessing module, a feature extraction module, an intelligent analysis module, and a data storage and transmission module; the video acquisition module consists of multiple high-definition cameras, which are distributed at different positions in the laboratory; the data preprocessing module performs image enhancement processing on the acquired video frames; the feature extraction module uses deep learning algorithms to extract visual features such as the color, shape, texture, and motion trajectory of the flame from the preprocessed video frames; the intelligent analysis module uses machine learning and deep learning models to perform real-time analysis on the video data based on the extracted features to determine whether a fire has occurred and the severity of the fire; the data storage and transmission module stores the acquired original video data, preprocessed data, extracted feature data, analysis results, etc. in a local database for subsequent query, statistics, and analysis.
[0009] Preferably, the fire extinguishing system includes: an embedded water tank housing, a movable fire extinguishing mechanism, a sub-control machine, a helicopter model, fixed nozzles, a hanger, a suspended nozzle pipe, a high-pressure pumping station, and a filtering device; the embedded water tank housing is arranged inside the laboratory floor; the movable fire extinguishing mechanism is arranged at the upper right rear of the embedded water tank housing; the sub-control machine is arranged at the outer front right of the embedded water tank housing, and the sub-control machine is electrically connected to the central centralized control execution system; the helicopter model is installed in the middle of the front side of the top of the embedded water tank housing, and the helicopter model is connected to the ignition system; the number of the fixed nozzles is two groups, and the number of each group of fixed nozzles is four. The two groups of fixed nozzles are installed at intervals from front to back on the top of the embedded water tank housing and on the left and right sides of the helicopter model; the number of the hangers is several, and several hangers are installed at intervals from front to back above the embedded water tank housing; the suspended nozzle pipe is installed in the inner side of several hangers in the front-back direction; the high-pressure pumping station is installed at the outer front left of the embedded water tank housing, and the high-pressure pumping station is connected to the fixed nozzles through a pipeline preset inside the embedded water tank housing, and the high-pressure pumping station is connected to the suspended nozzle pipe through a pipeline. The high-pressure pumping station is electrically connected to the sub-control machine; the filtering device is installed at the outer front right of the embedded water tank housing, and the filtering device is connected to the high-pressure pumping station through a pipeline, the filtering device is connected to the embedded water tank housing through a pipeline, and the filtering device is electrically connected to the sub-control machine.
[0010] Preferably, the movable fire extinguishing mechanism includes: a water tank, a pump body, a unwinding device, a connecting hose, a charging device, and a moving component; there are two water tanks, and the two water tanks are respectively installed on the left and right sides of the top rear of the embedded water tank housing; the pump body is installed at the top of the embedded water tank housing and is located in front of the water tanks, the pump body is connected to the two water tanks through pipelines, and the pump body is electrically connected to the sub-control machine; the unwinding device is installed at the top of the embedded water tank housing and is located in front of the pump body, and the unwinding device is electrically connected to the sub-control machine; the connecting hose is wound around the outer part of the roller of the unwinding device, and one end of the connecting hose is connected to the pump body; the charging device is arranged on the top of the embedded water tank housing and is located on the left side of the pump body, and the charging device is electrically connected to the sub-control machine; the moving component is arranged in front of the charging device.
[0011] Preferably, the moving component includes: a chassis, driving wheels, a sub-control module, a fixing frame, a supporting frame, a pan-tilt head, and a guiding camera; there are four driving wheels, and the four driving wheels are respectively installed at the four inner corners of the chassis; the sub-control module is installed in the middle of the left side of the top of the chassis, the charging device, the driving wheels and the sub-control module are electrically connected, and the sub-control module is remotely network-connected to the sub-control machine; there are two fixing frames, and the two fixing frames are installed in the middle of the rear side of the top of the chassis along the up and down direction; there are two supporting frames, and the two supporting frames are respectively installed at the outer tops of the left and right fixing frames obliquely along the front and rear direction, and the bottom ends of the supporting frames are fixedly connected to the top of the chassis; there are two pan-tilt heads, and the two pan-tilt heads are respectively installed at the outer tops of the left and right fixing frames, and the pan-tilt heads are electrically connected to the sub-control module; there are two guiding cameras, and the two guiding cameras are respectively installed at the mobile tops of the left and right pan-tilt heads, and the guiding cameras are electrically connected to the sub-control module.
[0012] Preferably, the moving component further includes: a first motor, a first sprocket, a tank housing, a second sprocket, a transmission chain, and a limiting tank; there are four first motors, and the four first motors are respectively installed at the outer bottoms of the left and right fixing frames, and the first motors are electrically connected to the sub-control module; the first sprocket is installed outside the rotating ends of the left and right first motors; the tank housing is rotatably connected to the inner tops of the left and right fixing frames through a pin shaft; the second sprocket is fixedly installed on the left side of the axis of the tank housing; the upper and lower inner ends of the transmission chain are respectively meshed with the outside of the second sprocket and the first sprocket; there are two limiting tanks, and the two limiting tanks are respectively opened at the bottom ends of the left and right sides of the tank housing along the up and down direction.
[0013] Preferably, the moving component further includes: a cylinder body, a rotating shaft, a mounting base, a second motor, a gear set, a rotating frame, an electric telescopic rod, a connecting seat, a mounting frame, an annular fixing seat, a third motor and a nozzle; the cylinder body is inserted into the inner cavities of the left and right limiting grooves in the left-right direction; the rotating shaft is rotatably connected to the inner side of the cylinder body through bearings in the left-right direction, and both the left and right ends of the rotating shaft extend out of the cylinder body; the mounting base is mounted on the right side of the top of the outer wall of the cylinder body; the second motor is fixedly mounted on the inner top of the mounting base, and the second motor is electrically connected to the sub-control module; the number of the gear sets is two, and the two gear sets are respectively mounted on the right side of the rotating end of the second motor and the outer right side of the rotating shaft and are meshed with each other; the rotating frame is arranged on the outer side of the rotating shaft; the electric telescopic rod is arranged in the inner top of the housing of the groove body in the up-down direction, and the electric telescopic rod is electrically connected to the sub-control module; the connecting seat is mounted on the bottom of the telescopic end of the electric telescopic rod, and the inner side of the connecting seat is connected to the left side of the outer wall of the cylinder body; the mounting frame is mounted on the front side of the rotating frame; the annular fixing seat is rotatably connected to the inner front end of the mounting frame through a pin shaft; the third motor is mounted on the front right side of the mounting frame, the rotating end of the third motor is connected to the right side of the axis of the annular fixing seat, and the third motor is electrically connected to the sub-control module; the nozzle is detachably mounted on the inner side of the annular fixing seat, and the nozzle is connected to the other end of the connecting hose.
[0014] Preferably, the first motor drives the first sprocket to rotate, and drives the second sprocket to drive the housing of the groove body to turn upwards to a specified position inside the fixed frame under the transmission of the transmission chain, so that the nozzle can extend into the helicopter model.
[0015] Preferably, the third motor drives the annular fixing seat to rotate inside the mounting frame, and makes the annular fixing seat drive the nozzle to rotate to a specified angle, so as to make the nozzle aim at the position of the fire source.
[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. By integrating a flame analysis system, a data acquisition system, a heat release acquisition system for the fire source, and an AI acquisition system, it is possible to monitor and record key data such as flame characteristics, temperature, humidity, and gas concentration during the combustion test in real time and accurately, providing detailed and reliable data support for the subsequent optimization of fire prevention and control strategies and the performance evaluation of fire-fighting equipment. At the same time, the setting of the safety protection system effectively blocks the heat in the air and the erosion of chemical substances during the test, ensuring the safety of the test personnel and equipment, and further improving the accuracy and safety of the combustion test. Using AI technology for image recognition and data analysis, this system can achieve rapid recognition and early warning of the fire scene, significantly improving the accuracy and timeliness of fire early warning. In addition, through the central control system in the central control room, the test personnel can remotely control the fire extinguishing system to quickly control and extinguish the fire source, effectively reducing the damage caused by the fire and enhancing the emergency response ability.
[0017] 2. By designing different working conditions in the laboratory according to the specific requirements of the test tasks, including the type of combustibles, the location of combustibles, fire-fighting means, etc. For example, simulating the accidental combustion of a helicopter fuel tank, controlling the fire situation with high-pressure fine water mist, and selecting a suitable ignition system to meet the ignition requirements under different task requirements. It can be remotely started through the central control system in the central control room on the switch and the central centralized control execution system. The laboratory is equipped with a safety protection system to ensure the safety of the site, equipment and personnel during the test, and can be remotely started through the central control system in the central control room on the switch and the central centralized control execution system. The flue gas analysis and treatment system is used to provide the ventilation conditions required for the test and collect, analyze and process the flue gas generated by combustion. The flame analysis system, data acquisition system and heat release acquisition of the fire source collect combustion data and fire-fighting efficiency data under various working conditions, including data such as the heat field distribution, flue gas component flow rate, heat release of the fire source, and flame characteristics, and transmit the data to the central control system. The AI acquisition system collects the situation during the experiment and transmits the data to the central control system. The central control system centrally manages all data information of the test and displays it in the external monitoring room, laboratory, conference room, etc. through the network communication method in the conference system.
[0018] 3. The sub-control module is remotely started through the sub-control machine. The motor inside the drive wheel drives the drive wheel to rotate, realizing the horizontal movement of the chassis in multiple directions. During the movement of the chassis, the automatic pan-tilt drives the guiding camera to move in multiple angular directions, enabling the guiding camera to collect external image data and send it to the inside of the sub-control module. The sub-control module plans the movement path of the chassis according to the external image data and adjusts the drive wheels at the corresponding positions in real time to change the moving direction. During the movement of the chassis, the motor inside the unwinding device drives the roller to rotate to unwind the connecting hose to cooperate with the movement of the moving parts. After the moving parts move to the specified position outside the helicopter model, the first motor drives the first sprocket to rotate. Driven by the transmission chain, the second sprocket drives the tank housing to flip upward to the specified position inside the fixing frame, enabling the nozzle to extend into the helicopter model. The electric telescopic rod extends and retracts to drive the cylinder body to move along the inner cavity of the limit groove. When the cylinder body moves to the end position of the limit groove, the second motor drives one side of the gear set to rotate, and the other side of the gear set drives the rotating shaft to rotate inside the cylinder body, and the rotating shaft drives the rotating frame to flip to the other direction position. The third motor drives the annular fixing seat to rotate inside the mounting frame, and the annular fixing seat drives the nozzle to rotate to the specified angle, so that the nozzle is aligned with the fire source position. The pump body pressurizes the water stored inside the water tank and supplies it to the inside of the nozzle through the connecting hose, enabling the nozzle to extinguish the fire at the specified position.
[0019] In summary, the present invention can simulate different types of fires caused by different incentives inside a full-size helicopter, extract combustion data, and can verify the fire extinguishing efficiency of different fire-fighting facilities. And it can simulate the situation when a helicopter catches fire, and test the fire extinguishing effects of fixed fire-fighting equipment and movable fire-fighting equipment respectively, so as to optimize the design scheme of fire-fighting equipment. Brief Description of the Drawings
[0020] Figure 1 Schematic diagram of the present invention Figure 2 is Figure 1 Schematic diagram of the fire extinguishing system; Figure 3 is Figure 2 Schematic diagram of the movable fire extinguishing mechanism; Figure 4 is Figure 3 Schematic diagram of the moving parts; Figure 5 is Figure 4 Exploded view of the moving parts; Figure 6 is Figure 5 Enlarged view of part A; Figure 7 is the principle block diagram of the AI acquisition system.
[0021] In the figure: 1. Embedded sink housing; 2. Movable fire extinguishing mechanism, where 21 is the water tank, 22 is the pump body, 23 is the unwinding device, 24 is the connecting hose, 25 is the charging device; 3. Moving part, where 31 is the chassis, 32 is the driving wheel, 33 is the sub-control module, 34 is the fixing bracket, 35 is the support bracket, 36 is the automatic pan-tilt head, 37 is the guiding camera, 38 is the first motor, 39 is the first sprocket, 310 is the tank housing, 311 is the second sprocket, 312 is the drive chain, 313 is the limiting tank body, 314 is the cylinder body, 315 is the rotating shaft, 316 is the mounting base, 317 is the second motor, 318 is the gear set, 319 is the rotating frame, 320 is the electric telescopic rod, 321 is the connecting seat, 322 is the mounting frame, 323 is the annular fixing seat, 324 is the third motor, 325 is the nozzle; 4. Sub-control machine; 5. Helicopter model; 6. Fixed nozzle; 7. Hanging bracket; 8. Suspended nozzle pipe; 9. High-pressure pumping station; 10. Filtering device. Detailed implementation mode
[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.
[0023] Please refer to Figures 1-7 , the present invention provides a technical solution: a 3D model defect detection system based on AI recognition, which is characterized in that it includes: a central control room and a laboratory. Inside the central control room, there are a central centralized control and acquisition system, a switch group, a central control system, and a safety intercom system. The central centralized control and acquisition system is connected to the central control system through the switch group, and the safety intercom system is directly connected to the central control system. The central control system is connected to a data processing system and two switches. One of the switches is connected to the display system and a conference system arranged outside the central control room. Inside the laboratory, there are a flame analysis system, a data acquisition system, a fire source heat release acquisition system, an AI acquisition system, an ignition system, a safety protection system, a flue gas analysis and processing system, and a fire extinguishing system.
[0024] The flame analysis system, data acquisition system and fire source heat release acquisition system are connected through a data acquisition control box installed outside the central control room and a central control acquisition system in the central control room. The flame analysis system observes the visual characteristics of flame height, shape, color, etc. through a high-definition AI acquisition system, and collects flame spectrum data information through infrared spectrum detectors, ultraviolet spectrum detectors, etc. The controller inputs the data into the flame data storage and analysis system after pre-processing, and analyzes the obtained data to obtain the basic characteristics of the flame. The data acquisition system adopts a combination of various sensors, which can collect temperature, humidity, carbon dioxide, carbon monoxide, photosensitivity, temperature, smoke and other data in real time. The fire source heat release acquisition system collects temperature and other data during the combustion process of single combustibles and mixed combustibles in the test field in real time. The data acquisition control box summarizes the data collected in the flame analysis system, data acquisition system and fire source heat release acquisition system, and stores them locally or uploads them to the cloud platform through the central control acquisition system through the network and establishes a relevant database. The staff calls the data in the central control acquisition system through the central control system.
[0025] like Figure 7 As shown, the AI acquisition system is connected to the switch group in the central control room. The AI acquisition system includes: a video acquisition module, a data preprocessing module, a feature extraction module, an intelligent analysis module, and a data storage and transmission module; the video acquisition module is composed of multiple high-definition cameras, which are distributed in different locations of the laboratory, including corners, high places, and key areas where fire sources may occur, to ensure that the test site can be covered in all directions without blind spots. The camera has the characteristics of high resolution, wide dynamic range, low illumination, etc., and can clearly capture video images under different lighting conditions; The data preprocessing module performs image enhancement processing on the collected video frames, including brightness adjustment, contrast enhancement, noise removal and other operations to improve the clarity and quality of the image, which is convenient for subsequent feature extraction and analysis; the feature extraction module uses deep learning algorithms to extract the color, shape, texture, and motion trajectory visual features of the flame from the preprocessed video frames. For example, the temperature and burning intensity of the flame can be judged by color features; the type and spread direction of the flame can be identified by shape features; based on the extracted features, the intelligent analysis module uses machine learning and deep learning models to perform real-time analysis of video data to determine whether a fire has occurred and the severity of the fire. Once signs of fire are detected, an early warning signal is immediately issued to notify the central control room and relevant personnel to take corresponding measures; the data storage and transmission module stores the collected original video data, preprocessed data, extracted feature data, and analysis results in the local database for subsequent query, statistics, and analysis.
[0026] The safety protection system includes an equipment safety guarantee system and a personnel safety protection system. The equipment safety guarantee system is used to ensure the safety of the combustion test site and equipment, and can be used alone or in combination. When used as a test item to verify the fire protection efficiency, other systems are equipped as emergency guarantee means to fully ensure the test safety. The personnel safety protection system mainly blocks the heat in the air and the erosion of chemical substances during the test, ensures communication, and maximally guarantees the safety of personnel.
[0027] The flue gas analysis and treatment system includes a ventilation and exhaust system, a flue gas collection system, a flue gas data acquisition system, and an exhaust gas treatment system. The flue gas data collection system transmits the collected flue gas analysis data to the main control system of the laboratory. According to the test requirements, the main control system can control the switch and air volume of the air-conditioning ventilation and exhaust system by remotely controlling the electric valve. The flue gas collection system collects the combustion flue gas and sends it to a professional gas analysis unit for inspection to obtain comprehensive gas analysis data. The gas generated by combustion meets the environmental protection requirements after being treated by the exhaust gas treatment system and is discharged into the atmosphere.
[0028] The ignition system, the safety protection system, and the flue gas analysis and treatment system are connected to another switch installed inside the central control room through the central centralized control execution system installed outside the central control room.
[0029] As a preferred solution, furthermore, the fire extinguishing system includes: a pre-embedded sink housing 1, a movable fire extinguishing mechanism 2, a sub-control machine 4, a helicopter model 5, fixed nozzles 6, a hanger 7, a suspended nozzle pipe 8, a high-pressure pumping station 9, and a filtering device 10; the pre-embedded sink housing 1 is arranged inside the laboratory floor; the movable fire extinguishing mechanism 2 is arranged at the top right rear of the pre-embedded sink housing 1; the sub-control machine 4 is arranged at the outside right front of the pre-embedded sink housing 1, and the sub-control machine 4 is electrically connected to the central centralized control execution system. A preset program is set inside the sub-control machine 4 and can be remotely started by the central control system through a switch and the central centralized control execution system; the helicopter model 5 is installed in the middle of the front side of the top of the pre-embedded sink housing 1, and the helicopter model 5 is connected to the ignition system; the number of fixed nozzles 6 is two groups, and the number of each group of fixed nozzles 6 is four. The two groups of fixed nozzles 6 are installed at intervals from front to back on the top of the pre-embedded sink housing 1 and on the left and right sides of the helicopter model 5; the number of hangers 7 is several, and the several hangers 7 are installed at intervals from front to back above the pre-embedded sink housing 1; the suspended nozzle pipe 8 is installed in the front and rear directions inside the several hangers 7; the high-pressure pumping station 9 is installed at the outside left front of the pre-embedded sink housing 1. The high-pressure pumping station 9 is connected to the fixed nozzles 6 through a pipeline preset inside the pre-embedded sink housing 1, and the high-pressure pumping station 9 is connected to the suspended nozzle pipe 8 through a pipeline. The high-pressure pumping station 9 is electrically connected to the sub-control machine 4, and the high-pressure pumping station 9 is controlled by the sub-control machine 4. The high-pressure pumping station 9 supplies water source to the inside of the fixed nozzles 6 and the suspended nozzle pipe 8 through the corresponding pipeline; the filtering device 10 is installed at the outside right front of the pre-embedded sink housing 1, and the filtering device 10 is controlled by the sub-control machine 4. The filtering device 10 is connected to the high-pressure pumping station 9 through a pipeline, the filtering device 10 is connected to the pre-embedded sink housing 1 through a pipeline, and the filtering device 10 is electrically connected to the sub-control machine 4. The wastewater after cooling and extinguishing the fire can flow into the filtering device 10 along the drain outlet of the pre-embedded sink housing 1. The filtering device 10 filters the wastewater and then discharges it into the high-pressure pumping station 9 for storage and waiting for recycling.
[0030] As a preferred solution, furthermore, as Figure 3As shown in the figure, the movable fire extinguishing mechanism 2 includes: a water tank 21, a pump body 22, a unwinding device 23, a connecting hose 24, a charging device 25 and a moving component 3; the number of water tanks 21 is two, and the two water tanks 21 are respectively installed on the left and right sides of the top right rear of the embedded water tank housing 1. The water tank 21 can store water inside; the pump body 22 is installed at the top end of the embedded water tank housing 1 and is located in front of the water tank 21. The pump body 22 is connected to the two water tanks 21 through pipelines. The pump body 22 is electrically connected to the sub-control machine 4. The pump body 22 is controlled by the sub-control machine 4 to pressurize the water stored inside the water tank 21 and supply it to the inside of the nozzle 325 through the connecting hose 24; the unwinding device 23 is installed at the top end of the embedded water tank housing 1 and is located in front of the pump body 22. The unwinding device 23 is electrically connected to the sub-control machine 4. The unwinding device 23 is controlled by the sub-control machine 4 to drive the roller to rotate by the internal motor to realize the unwinding or winding operation of the connecting hose 24; the connecting hose 24 is wound around the roller of the unwinding device 23, and one end of the connecting hose 24 is connected to the pump body 22; the charging device 25 is arranged on the top of the embedded water tank housing 1 and is located on the left side of the pump body 22. The charging device 25 is electrically connected to the sub-control machine 4; the moving component 3 is arranged in front of the charging device 25. The charging device 25 is controlled by the sub-control machine 4 and can be connected to the sub-control module 33 to charge it.
[0031] As a preferred solution, furthermore, as Figure 4 and Figure 5As shown in the figure, the moving part 3 includes: a chassis 31, drive wheels 32, a sub-control module 33, a fixing frame 34, a support frame 35, a pan-tilt head 36, a guiding camera 37, a first motor 38, a first sprocket 39, a tank housing 310, a second sprocket 311, a drive chain 312, a limiting tank 313, a cylinder 314, a rotating shaft 315, a mounting base 316, a second motor 317, a gear set 318, a rotating frame 319, an electric telescopic rod 320, a connecting seat 321, a mounting frame 322, an annular fixing seat 323, a third motor 324, and a nozzle 325; the chassis 31 is arranged at the bottommost of the moving part 3; the number of drive wheels 32 is four, and the four drive wheels 32 are respectively installed at the inner four corners of the chassis 31. The drive wheels 32 can be controlled by the sub-control module 33 to move in multiple directions; the sub-control module 33 is installed in the middle of the left side of the top of the chassis 31. The charging device 25 can be docked with the sub-control module 33 for charging. The drive wheels 32 and the sub-control module 33 are electrically connected. The sub-control module 33 is remotely network-connected to the sub-control machine 4. The sub-control module 33 is internally provided with a preset program and a battery to supply power to the internal electrical components of the moving part 3; the number of fixing frames 34 is two, and the two fixing frames 34 are installed in the middle of the rear side of the top of the chassis 31 in the vertical direction; the number of support frames 35 is two, and the two support frames 35 are respectively installed at the outer tops of the left and right fixing frames 34 in an inclined manner in the front-rear direction. The bottom ends of the support frames 35 are fixedly connected to the top of the chassis 31; the number of pan-tilt heads 36 is two, and the two pan-tilt heads 36 are respectively installed at the outer tops of the left and right fixing frames 34. The pan-tilt heads 36 and the sub-control module 33 are electrically connected. The pan-tilt heads 36 can be controlled by the sub-control module 33 to drive the guiding camera 37 to move in multiple angular directions; the number of guiding cameras 37 is two, and the two guiding cameras 37 are respectively installed at the top of the moving ends of the left and right pan-tilt heads 36. The guiding cameras 37 and the sub-control module 33 are electrically connected. The guiding cameras 37 are controlled by the sub-control module 33. The guiding cameras 37 collect external image data and send it to the inside of the sub-control module 33. The sub-control module 33 plans the moving path of the chassis 31 according to the external image data and adjusts the drive wheels 32 at the corresponding positions in real time to change the moving direction; the number of first motors 38 is four, and the four first motors 38 are respectively installed at the outer bottoms of the left and right fixing frames 34. The first motors 38 and the sub-control module 33 are electrically connected. The first motors 38 are controlled by the sub-control module 33 to drive the first sprockets 39 to rotate; the first sprockets 39 are installed on the outer sides of the rotating ends of the left and right first motors 38; the tank housing 310 is rotatably connected to the inner tops of the left and right fixing frames 34 by pins; the second sprocket 311 is fixedly installed on the left side of the axis of the tank housing 310; the upper and lower inner ends of the drive chain 312 are respectively meshed with the outside of the second sprocket 311 and the first sprocket 39;There are two limiting groove bodies 313, and the two limiting groove bodies 313 are respectively arranged at the bottom ends on the left and right sides of the groove body housing 310 along the up and down direction; the cylinder body 314 is inserted into the inner cavities of the left and right limiting groove bodies 313 along the left and right direction; the rotating shaft 315 is rotationally connected to the inside of the cylinder body 314 through bearings along the left and right direction, and the left and right ends of the rotating shaft 315 extend out of the cylinder body 314; the mounting base 316 is installed at the top right side of the outer wall of the cylinder body 314; the second motor 317 is fixedly installed at the top inside of the mounting base 316, the second motor 317 is electrically connected to the sub-control module 33, and the second motor 317 is controlled by the sub-control module 33 to drive the rotation of the gears in one side gear set 318; the number of gear sets 318 is two, and the two gear sets 318 are respectively installed on the right side of the rotating end of the second motor 317 and the outer right side of the rotating shaft 315 and are meshed with each other; the rotating frame 319 is arranged on the outside of the rotating shaft 315; the electric telescopic rod 320 is arranged along the up and down direction at the top inside of the groove body housing 310, the electric telescopic rod 320 is electrically connected to the sub-control module 33, and the electric telescopic rod 320 is controlled by the sub-control module 33 to drive the connecting seat 321 to move through its own elongation and shortening; the connecting seat 321 is installed at the bottom of the telescopic end of the electric telescopic rod 320, and the inside of the connecting seat 321 is connected to the left outer wall of the cylinder body 314; the mounting frame 322 is installed on the front side of the rotating frame 319; the annular fixed seat 323 is rotationally connected to the front end inside of the mounting frame 322 through a pin shaft; the third motor 324 is installed at the front right side of the mounting frame 322, the rotating end of the third motor 324 is connected to the right side of the axis of the annular fixed seat 323, the third motor 324 is electrically connected to the sub-control module 33, and the third motor 324 is controlled by the sub-control module 33 to drive the annular fixed seat 323 to rotate; the nozzle 325 is detachably installed inside the annular fixed seat 323, and the nozzle 325 is connected to the other end of the connecting hose 24.;
[0032] The working principle is as follows: Step 1: According to the specific requirements of the test task, different working conditions are designed in the laboratory, including the type of combustibles, the location of combustibles, fire-fighting means, etc. For example, simulate the accidental combustion of a helicopter fuel tank, control the fire with high-pressure fine water mist, select a suitable ignition system to meet the ignition requirements under different task requirements. It can be remotely started through the central control system in the central control room on the switch and the central centralized control execution system. A safety protection system is equipped in the laboratory to ensure the safety of the site, equipment and personnel during the test. It can be remotely started through the central control system in the central control room on the switch and the central centralized control execution system. The flue gas analysis and treatment system is used to provide the ventilation conditions required for the test and collect, analyze and process the flue gas generated by combustion. The flame analysis system, data acquisition system and heat release acquisition system of the fire source collect the combustion data and fire-fighting efficiency data under various working conditions, including data such as the thermal field distribution, the flow rate of flue gas components, the heat release of the fire source, and the flame characteristics, and transmit the data to the central control system. The AI acquisition system collects the situation during the experiment and transmits the data to the central control system; The central control system centrally processes all data information of the test and displays it in the external monitoring room, laboratory, conference room, etc. through the network communication method in the conference system; Step 2: The ignition system ignites inside the helicopter model 5 according to the fire situation requirements to simulate the occurrence of a fire, and collects the combustion data of the helicopter model 5. After the combustion is completed, the central control system remotely starts the sub-control machine 4 on the switch and the central centralized control execution system. The pre-set system inside the sub-control machine 4 controls the start of the high-pressure pump station 9 and the filtering equipment 10. The high-pressure pump station 9 supplies water to the inside of the fixed nozzle 6 and the suspended nozzle pipe 8 through the corresponding pipelines, so that the fixed nozzle 6 and the suspended nozzle pipe 8 respectively perform fire extinguishing and cooling operations on the side wall and the top of the helicopter model 5. The waste water after cooling and fire extinguishing flows into the inside of the filtering equipment 10 along the drain outlet of the embedded water tank shell 1. The filtering equipment 10 filters the waste water and discharges it into the high-pressure pump station 9 for storage and waiting for recycling; Step 3: The sub-control machine 4 controls the remote start of the sub-control module 33. The sub-control module 33 internally pre-sets the system to control the start of the driving wheel 32, the automatic pan-tilt 36, and the guiding camera 37. The motor inside the driving wheel 32 drives the driving wheel to rotate to achieve the horizontal movement of the chassis 31 in multiple directions. During the movement of the chassis 31, the automatic pan-tilt 36 drives the guiding camera 37 to move in multiple angular directions, so that the guiding camera 37 collects external image data and sends it to the inside of the sub-control module 33. The sub-control module 33 plans the movement path of the chassis 31 according to the external image data and adjusts the driving wheel 32 at the corresponding position in real time to change the moving direction. During the movement of the chassis 31, the pre-set program inside the sub-control machine 4 controls the start of the unwinding device 23. The motor inside the unwinding device 23 drives the roller to rotate to unwind the connecting hose 24 to cooperate with the movement of the moving part 3. After the moving part 3 moves to the specified position outside the helicopter model 5, the pre-set program inside the sub-control module 33 controls the start of the first motor 38, the electric telescopic rod 320, the second motor 317, and the third motor 324. The first motor 38 drives the first sprocket 39 to rotate, and under the drive of the transmission chain 312, drives the second sprocket 311 to drive the tank housing 310 to turn upwards to the specified position inside the fixed frame 34, so that the nozzle 325 can extend into the helicopter model 5. The electric telescopic rod 320 extends and retracts to drive the cylinder body 314 to move along the inner cavity of the limit groove body 313. When the cylinder body 314 moves to the end position of the limit groove body 313, the second motor 317 drives the gear set 318 on one side to rotate, and the gear set 318 on the other side drives the rotating shaft 315 to rotate inside the cylinder body 314, and the rotating shaft 315 drives the rotating frame 319 to turn to the other side position. The third motor 324 drives the annular fixing seat 323 to rotate inside the mounting frame 322, and the annular fixing seat 323 drives the nozzle 325 to rotate to the specified angle so that the nozzle 325 is aligned with the fire source position. The pre-set program inside the sub-control machine 4 controls the start of the pump body 22. The pump body 22 pressurizes the water source stored inside the water tank 21 and supplies it to the inside of the nozzle 325 through the connecting hose 24, so that the nozzle 325 extinguishes the fire at the specified position.
[0033] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A 3D model defect detection system based on AI recognition, characterized in that: include: The central control room and laboratory are equipped with a central control acquisition system, a switch group, a central control system and a safety shouting system. The central control acquisition system is connected to the central control system through the switch group, and the safety shouting system is directly connected to the central control system. The central control system is connected to a data processing system and two switches, one of which is connected to a display system and a conference system set up outside the central control room. The laboratory is equipped with a flame analysis system, a data acquisition system, a fire source heat release acquisition system, an AI acquisition system, an ignition system, a safety protection system, a smoke analysis and processing system and a fire extinguishing system. The analysis system, data acquisition system and fire source heat release acquisition system are connected through a data acquisition control box arranged outside the central control room and a central control acquisition system in the central control room. The AI acquisition system is connected to a switch group in the central control room. The safety protection system includes an equipment safety assurance system and a personnel safety protection system. The flue gas analysis and processing system includes a ventilation and smoke exhaust system, a flue gas collection system, a flue gas data acquisition system and a waste gas treatment system. The ignition system, the safety protection system and the flue gas analysis and processing system are connected to another switch arranged inside the central control room through a central control execution system arranged outside the central control room.
2. The 3D model defect detection system based on AI recognition according to claim 1, characterized in that: The AI acquisition system includes: The video acquisition module consists of multiple high-definition cameras distributed in different locations in the laboratory; A data preprocessing module performs image enhancement processing on the collected video frames; The feature extraction module uses deep learning algorithms to extract the color, shape, texture, and motion trajectory visual features of the flame from the preprocessed video frames; Intelligent analysis module, based on the extracted features, uses machine learning and deep learning models to analyze video data in real time to determine whether a fire has occurred and the severity of the fire; The data storage and transmission module stores the collected original video data, pre-processed data, extracted feature data, and analysis results in the local database for subsequent query, statistics, and analysis.
3. The 3D model defect detection system based on AI recognition according to claim 1, characterized in that: The fire extinguishing system comprises: A pre-buried water tank housing (1) is arranged inside the laboratory floor; A movable fire extinguishing mechanism (2) is arranged at the right rear of the top end of the embedded water tank housing (1); A sub-control machine (4) is arranged on the outside right front of the embedded water tank housing (1), and the sub-control machine (4) is electrically connected to the central control execution system; A helicopter model (5) is mounted on the middle portion of the front top of the embedded water tank housing (1), and the helicopter model (5) is connected to an ignition system; Fixed nozzles (6), the number of the fixed nozzles (6) being two groups, the number of the fixed nozzles (6) in each group being four, the two groups of fixed nozzles (6) being installed at intervals from front to back on the top of the embedded water tank housing (1) and located on the left and right sides of the helicopter model (5); A hanger (7), wherein the number of the hangers (7) is a plurality, and the plurality of hangers (7) are installed at intervals from front to back above the embedded water tank housing (1); A suspended nozzle pipe (8) is installed on the inner side of a plurality of hangers (7) in the front-to-back direction; A high-pressure pump station (9) is installed at the front left of the exterior of the embedded water tank housing (1); the high-pressure pump station (9) is connected to the fixed nozzle (6) via a pipeline preset inside the embedded water tank housing (1); the high-pressure pump station (9) and the suspended nozzle pipe (8) are connected via the pipeline; and the high-pressure pump station (9) and the sub-control machine (4) are electrically connected; The filter device (10) is installed on the front right outside of the embedded water tank housing (1); the filter device (10) and the high-pressure pump station (9) are connected via a pipeline; the filter device (10) and the embedded water tank housing (1) are connected via a pipeline; and the filter device (10) and the sub-control machine (4) are electrically connected.
4. The 3D model defect detection system based on AI recognition according to claim 3 is characterized in that: The movable fire extinguishing mechanism (2) comprises: A water tank (21), wherein the number of the water tanks (21) is two, and the two water tanks (21) are respectively installed on the left and right sides of the top rear right of the embedded water tank housing (1); A pump body (22) is installed at the top of the embedded water tank housing (1) and is located in front of the water tank (21); the pump body (22) is connected to the two water tanks (21) via a pipeline; and the pump body (22) is electrically connected to the sub-controller (4); An unwinding device (23) is installed at the top of the embedded water tank housing (1) and is located at the front side of the pump body (22), and the unwinding device (23) is electrically connected to the sub-control machine (4); A connecting hose (24) is wound around the outside of the roller of the unwinding device (23), and one end of the connecting hose (24) is connected to the pump body (22); A charging device (25) is arranged on the top of the embedded water tank housing (1) and located on the left side of the pump body (22), and the charging device (25) is electrically connected to the sub-control machine (4); The moving component (3) is arranged in front of the charging device (25).
5. The 3D model defect detection system based on AI recognition according to claim 4 is characterized in that: The moving part (3) comprises: chassis (31); Driving wheels (32), the number of the driving wheels (32) being four, and the four driving wheels (32) being respectively mounted at four inner corners of the chassis (31); The sub-control module (33) is installed in the middle of the left side of the top of the chassis (31), and the charging device (25), the driving wheel (32) and the sub-control module (33) are electrically connected to the sub-control module (33) and the sub-control machine (4) via a remote network connection; A fixing frame (34), wherein the number of the fixing frames (34) is two, and the two fixing frames (34) are installed in the middle of the rear side of the top end of the chassis (31) along the up-down direction; Support frames (35), the number of the support frames (35) being two, the two support frames (35) being respectively installed at the outer tops of the left and right fixing frames (34) in an inclined manner along the front-back direction, and the bottom ends of the support frames (35) being fixedly connected to the top of the chassis (31); An automatic pan-tilt platform (36), wherein the number of the automatic pan-tilt platforms (36) is two, and the two automatic pan-tilt platforms (36) are respectively mounted on the outer tops of the left and right fixing frames (34), and the automatic pan-tilt platforms (36) are electrically connected to the sub-control module (33); A guiding camera (37), wherein the number of the guiding cameras (37) is two, and the two guiding cameras (37) are respectively mounted on the top of the mobile ends of the left and right automatic pan-tilt platforms (36), and the guiding camera (37) is electrically connected to the sub-control module (33).
6. The 3D model defect detection system based on AI recognition according to claim 5, characterized in that: The moving component (3) further comprises: A first motor (38), wherein the number of the first motors (38) is four, and the four first motors (38) are respectively mounted on the outer bottoms of the left and right fixing frames (34), and the first motors (38) are electrically connected to the sub-control module (33); A first sprocket (39) mounted on the outer sides of the rotating ends of the left and right first motors (38); The tank shell (310) is rotatably connected to the inner tops of the left and right fixing frames (34) via a pin; A second sprocket (311) is fixedly mounted on the left side of the axis of the tank housing (310); A transmission chain (312), wherein the inner upper and lower ends of the transmission chain (312) are respectively meshed with the outer ends of the second sprocket (311) and the first sprocket (39). The limiting groove bodies (313) are two in number, and the two limiting groove bodies (313) are respectively opened at the bottom ends of the left and right sides of the groove body shell (310) along the up-down direction.
7. The 3D model defect detection system based on AI recognition according to claim 6, characterized in that: The moving component (3) further comprises: The cylinder (314) is inserted into the inner cavities of the two left and right limiting grooves (313) along the left-right direction; A rotating shaft (315) is rotatably connected to the inner side of the cylinder (314) via a bearing in a left-right direction, and left and right ends of the rotating shaft (315) extend out of the cylinder (314); A mounting base (316) mounted on the right side of the top end of the outer wall of the cylinder (314); A second motor (317) is fixedly mounted on the inner top of the mounting base (316), and the second motor (317) is electrically connected to the sub-control module (33); A gear set (318), wherein the number of the gear sets (318) is two, and the two gear sets (318) are respectively mounted on the right side of the rotating end of the second motor (317) and the right side of the outside of the rotating shaft (315) and mesh with each other; A rotating frame (319) is arranged outside the rotating shaft (315); An electric telescopic rod (320) is arranged at the top of the inner cavity of the tank shell (310) in the up-down direction, and the electric telescopic rod (320) is electrically connected to the sub-control module (33); A connecting seat (321) is installed at the bottom of the telescopic end of the electric telescopic rod (320), and the inner side of the connecting seat (321) is connected to the left side of the outer wall of the cylinder (314); A mounting frame (322) mounted on the front side of the rotating frame (319); An annular fixing seat (323) is rotatably connected to the inner front end of the mounting frame (322) via a pin; A third motor (324) is mounted in front of the right side of the mounting frame (322), a rotating end of the third motor (324) is connected to the right side of the axis of the annular fixing seat (323), and the third motor (324) is electrically connected to the sub-control module (33); The spray head (325) is detachably mounted on the inner side of the annular fixing seat (323), and the spray head (325) is connected to the other end of the connecting hose (24).
8. The 3D model defect detection system based on AI recognition according to claim 7, characterized in that: The first motor (38) drives the first sprocket (39) to rotate, and under the drive of the transmission chain (312), drives the second sprocket (311) to drive the tank shell (310) to flip upward to a specified position on the inner side of the fixing frame (34), so that the nozzle (325) can extend into the interior of the helicopter model (5).
9. The 3D model defect detection system based on AI recognition according to claim 8, characterized in that: The third motor (324) drives the annular fixing seat (323) to rotate inside the mounting frame (322), and causes the annular fixing seat (323) to drive the spray head (325) to rotate to a specified angle, so that the spray head (325) is aligned with the fire source.
Citation Information
Patent Citations
Tunnel guide rail advancing fire-fighting robot system and using method
CN112043991A
Fire extinguishing test platform for simulating ship cabin
CN118857818A
Equipment for cultivation of vine plants that are easy to maintain
KR1020220114949A