Water-crossing bridge disaster prevention safety inspection equipment and method

Abstract

The invention discloses a water-crossing bridge disaster prevention safety inspection device and method.The inspection device comprises an inspection walking vehicle and a walking track, the walking track stretches across a water channel and is erected on a bridge guardrail, and the inspection walking vehicle is driven by a driving mechanism to walk on the walking track in a reciprocating mode through wheels; a temperature and humidity sensor, a rainfall sensor, a wind speed sensor, a camera and a controller are arranged on the inspection walking vehicle, the temperature and humidity sensor, the rainfall sensor, the wind speed sensor, the camera and the driving mechanism are connected with the controller, and a holder is arranged on the inspection walking vehicle. The camera is erected on the holder to acquire images of a bridge and a water channel in all directions, and the controller is connected with a remote central server through the wireless communication module. By arranging the rail type safety structure, the inspection vehicle runs along the walking rail installed on the outer side of the bridge guardrail, compared with unmanned aerial vehicle inspection, the inspection vehicle is not affected by severe weather, and safety, reliability and stability are high.

Description

Technical Field

[0001] The invention relates to a disaster prevention safety inspection device and a method for a water-spanning bridge. Background Art

[0002] After the construction of a traffic bridge is completed, it is usually necessary to monitor the bridge structure and the environment near the bridge to evaluate the safety of the bridge. Monitor the river conditions under the bridge and respond to sudden floods in a timely manner.

[0003] Most of the current bridge monitoring equipment uses drones with cameras to monitor designated routes. Drones are greatly affected by the environment, and their inspection capabilities will be greatly reduced in bad weather. By installing some fixed cameras for monitoring, the captured image data is directly uploaded to the monitoring center, which will result in single data and inability to make correct judgments in time. At the same time, fixed-position cameras also have a limited shooting field of view. Summary of the invention

[0004] In view of the above problems, the object of the present invention is to provide a disaster prevention and safety inspection device and method for water-spanning bridges.

[0005] In order to achieve the above object, the scheme of the present invention is: A disaster prevention and safety inspection equipment for a water-crossing bridge comprises an inspection vehicle and a travel track. The travel track is pre-installed on a bridge guardrail across a waterway. The inspection vehicle is driven by a driving mechanism through wheels to reciprocate on the travel track. A temperature and humidity sensor, a rain sensor, a wind speed sensor, a camera and a controller are arranged on the inspection vehicle. The temperature and humidity sensor, the rain sensor, the wind speed sensor, the camera and the driving mechanism are connected to the controller. A pan-tilt platform is arranged on the inspection vehicle. The camera is mounted on the pan-tilt platform to obtain images of the bridge and the waterway in all directions. The controller is connected to a remote central server via a wireless communication module.

[0006] The scheme is further as follows: the walking track includes two groove rails arranged parallel to each other and at intervals, the two groove rails are connected and fixed by a support plate, the support plate is used to connect and fix with the bridge guardrail, the groove rail is a "T"-shaped groove rail with a "T"-shaped groove, the middle vertical "1" of the "T"-shaped groove is the opening of the groove rail, and the two parallel "T"-shaped groove rail openings are arranged opposite to each other; the inspection walking vehicle includes a frame, two wheel axles are arranged at intervals at the front and rear of the bottom of the frame, wheels are respectively arranged on the protruding ends of the wheel axles, a temperature and humidity sensor, a rainfall sensor, and a wind speed sensor are fixed on the top of the frame, a pan-tilt seat rod extends out of the frame, the pan-tilt is fixed on the pan-tilt seat rod, the shape of the wheel corresponds to the shape of the "T"-shaped groove, the four wheels of the two axles are placed in the "T"-shaped groove of the groove rail, and are driven by a driving mechanism to drive the frame to reciprocate along the walking track, and the frame is placed between the two groove rails.

[0007] A further solution is: the driving mechanism includes a driving motor and a speed changing mechanism, and the speed changing mechanism is connected to the wheel axle to drive the wheel axle to rotate.

[0008] The solution is further that: the inspection walking vehicle is also provided with a solar panel and a power supply battery, the solar panel is connected to the power supply battery, and the power supply battery provides power for the controller, the driving mechanism, the sensor and the camera.

[0009] A further solution is that the outer edge of the wheel is covered with a rubber sleeve.

[0010] A method for inspecting the disaster prevention safety of a water-spanning bridge is an inspection method for the inspection equipment for inspecting the disaster prevention safety of a water-spanning bridge. The method comprises: obtaining signals of a temperature and humidity sensor, a rainfall sensor, and a wind speed sensor in real time, judging whether the ambient temperature and humidity or rainfall or wind speed exceeds a set threshold value, and if the set threshold value is exceeded, starting an inspection vehicle to inspect back and forth on a pre-installed walking track of a bridge guardrail. The steps of executing the inspection comprise: Step 1: Turn on the camera and obtain images of the bridge structure and the waterway under the bridge in real time while the inspection vehicle is moving back and forth; Step 2: Process the acquired image to extract the image’s characteristic image data, compare the characteristic image data with the standard image data model, evaluate the comparison results to determine whether there is a safety hazard, and issue an alarm based on the level of the safety hazard; Wherein: the standard image data model is a standard image data model obtained by pre-processing and storing image data of the same position under the safe state of the bridge structure and the waterway and the unsafe state that has occurred.

[0011] The solution is further that: the standard image data model is a neural network comparison model, which is a neural network comparison model that extracts feature image data from image data of the same position in advance under a safe state and a simulated unsafe state of the bridge structure and the waterway, and performs feature image data learning processing; in the second step, the feature image data is compared with the standard image data model, and the feature image data is input into the neural network comparison model to obtain an evaluation result.

[0012] The solution is further that: the unsafe state includes an actually occurring unsafe state and a simulated unsafe state, and the simulated unsafe state is an unsafe state that is simulated by modifying the image.

[0013] The solution is further as follows: in the first step, the inspection vehicle moves in a process of alternating between stopping and moving, and the images of the bridge structure and the waterway under the bridge are acquired when the inspection vehicle stops moving.

[0014] The solution is further as follows: the alternating stop and go means stopping after walking a set distance, completing the acquisition of the image of the bridge structure and the image of the waterway under the bridge during the stopping time, and then continuing to walk.

[0015] The beneficial effects of the present invention are: 1. Inspection operations respond more promptly. Once the sensors on the inspection vehicle detect severe weather such as heavy rain or strong winds, the inspection vehicle will be triggered to start. The inspection vehicle will immediately start running and carry out inspection tasks, quickly inspect the river environment under the bridge and report the analysis results in a timely manner.

[0016] 2. Inspection tasks are completed better and more efficiently. Once the equipment is installed and running, it can be put into use. It can automatically carry out inspection tasks, and it can also remotely control the inspection vehicle to achieve regular inspections. The power supply equipment of the inspection vehicle is a rechargeable battery on the vehicle. The rechargeable battery is charged by the solar panel installed on the inspection vehicle, which reduces energy consumption and the use of power lines.

[0017] 3. The monitoring results are more accurate. Use pre-trained models for image analysis. The models are used to determine the flow of rivers. Pre-trained large models can be used for water area analysis and to extract the water surface range.

[0018] 4. Good safety protection. By setting up a track-type safety structure, compared with drone inspections, it is not affected by bad weather, and has high safety, reliability and stability. When not in operation, the inspection vehicle can be parked by locking the tires.

[0019] The present invention is described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the structure of the present invention; Figure 2 It is a schematic diagram of the walking track structure of the present invention. DETAILED DESCRIPTION

[0021] Embodiment 1: A disaster prevention safety inspection equipment for a bridge across water, such as Figure 1 and Figure 2As shown, the inspection equipment includes an inspection vehicle 1 and a walking track 2. The walking track is pre-installed on a bridge guardrail (not shown in the figure) across the waterway, for example, on the outside of the bridge guardrail. The inspection vehicle is driven by a driving mechanism 102 through wheels 101 to reciprocate on the walking track 2. A temperature and humidity sensor 3, a rain sensor 5, a wind speed sensor 6, a camera 4 and a controller 7 are arranged on the inspection vehicle; the temperature and humidity sensor, the rain sensor, the wind speed sensor, the camera 4 and the driving mechanism 102 are connected to the controller 7. A pan-tilt head 8 is arranged on the inspection vehicle 1. The camera 4 is mounted on the pan-tilt head 8 to obtain images of the bridge and the waterway in all directions. The controller 7 is connected to a remote central server (not shown) through a wireless communication module. The controller 7 transmits the received data to the remote central server, and the central server processes the data accordingly.

[0022] Wherein: the walking track 2 includes two groove rails 201 arranged parallel to each other and spaced apart, the two groove rails 201 are connected and fixed by a support plate 202, the support plate 202 is used to connect and fix with the bridge guardrail, the two rails are welded in advance with the support plate, and then the support plate is punched, the support plate and the bridge guardrail are connected, the metal guardrail is fixed with bolts, the concrete guardrail is punched, and then fixed with expansion screws; the groove rail 201 is a "T"-shaped groove rail with a "T"-shaped groove 201-1, the middle vertical "1" of the "T"-shaped groove is the opening 201-2 of the groove rail, and the two parallel "T"-shaped groove rail openings 201-2 are arranged oppositely (that is, the "T"-shaped groove is laid horizontally); the inspection walking vehicle includes a frame 103, two wheel axles 104 are arranged at intervals at the front and rear bottom of the frame, and wheels 101 are respectively arranged at both ends of the wheel axle 104 protruding from the frame 103, and temperature and humidity sensors 3 and The rain sensors 5 are respectively arranged at the top of the frame 103, the camera 4 extends out of the frame 103 through a pan-tilt seat rod 105 and the side frame is arranged on the outside of the frame 103. The pan-tilt 8 is fixed on the pan-tilt seat rod 105 to control the rotation of the camera. A connecting rod 106 extends outward from the top of the frame, and the wind speed sensor 6 is fixed on the connecting rod 106. The shape of the wheel 101 corresponds to the "T"-shaped groove shape of the groove rail. The four wheels 101 of the two shaft rods 104 are placed in the "T"-shaped groove of the groove rail 201 and are driven by the driving mechanism 102 to drive the frame 103 to move back and forth along the walking track 2. The frame 103 is placed between the two groove rails; three of the sensors: temperature and humidity sensor, rain sensor, and wind speed sensor are in direct contact with the air. Once a large change in temperature and humidity, wind speed, and rainfall is detected, it will be immediately started and the inspection operation will begin. This can not only extend the service life of the inspection vehicle but also increase the work efficiency of the inspection vehicle.

[0023] In the embodiment: the driving mechanism 102 includes a driving motor and a speed changing mechanism, and the speed changing mechanism is connected to the wheel shaft 104 to drive the wheel shaft to rotate. In order to make the wheel center run smoothly without slipping, the outer edge of the wheel 101 is wrapped with a rubber sleeve.

[0024] Among them, in order to ensure that the inspection vehicle can work without power failure for a long time: the inspection vehicle is also provided with a solar panel 9 and a power supply battery 10. The solar panel 9 is 100W, and different powers can be selected as needed. The solar panel 9 is connected to the power supply battery 10, and the power supply battery 10 provides power for the controller 7, the driving mechanism 102, the temperature and humidity sensor 3, the rain sensor 5, the wind speed sensor 6 and the camera 4. The solar panel is installed on the side of the inspection vehicle (the angle can be adjusted manually according to the light direction). The solar panel generates electricity for an average of 6 hours a day. If the conversion efficiency of the inverter is 90%, 100W / 90%*6h=666Wh can be generated every day, that is, 0.666 degrees of electricity. The central control equipment uses 5V 3A power supply, and consumes 360KWh or 0.36 degrees of electricity a day. By comparing the charging and power consumption, it can be seen that using solar panels as charging equipment for the inspection vehicle can meet the power supply and demand requirements of the inspection vehicle.

[0025] Embodiment 2: A method for inspecting the disaster prevention and safety of a water-spanning bridge is based on the inspection method of the inspection equipment for inspecting the disaster prevention and safety of a water-spanning bridge described in Example 1. Therefore, the content of Example 1 is applicable to this embodiment. The method includes: defining temperature and humidity thresholds, defining rainfall thresholds, and defining wind speed thresholds, acquiring signals from temperature and humidity sensors, rainfall sensors, and wind speed sensors in real time, and determining whether the ambient temperature and humidity or rainfall or wind speed exceeds any one of the three defined thresholds. If the set threshold is exceeded, the inspection vehicle is started to inspect back and forth on the pre-installed walking track of the bridge guardrail. If the set threshold is not exceeded, the inspection vehicle is controlled to stay at a set position, such as the bridge head, and the camera is turned off. The steps of executing the inspection include: Step 1: Turn on the camera and obtain images of the bridge structure and the waterway under the bridge in real time while the inspection vehicle is moving back and forth; Step 2: Process the acquired image to extract the image’s characteristic image data, compare the characteristic image data with the standard image data model, evaluate the comparison results to determine whether there is a safety hazard, and issue an alarm based on the level of the safety hazard; Wherein: the standard image data model is a standard image data model obtained by pre-processing and storing image data of the same position under the safe state of the bridge structure and the waterway and the unsafe state that has occurred.

[0026] The standard image data model is a neural network comparison model, which is a neural network comparison model that extracts feature image data from image data of the same position in advance when the bridge structure and waterway are in a safe state and in an unsafe state, and performs feature image data learning processing; in the second step, the feature image data is compared with the standard image data model, and the feature image data is input into the neural network comparison model to obtain an evaluation result. The neural network comparison model uses a convolutional neural network and a Tensorflow deep learning framework for model training. The data set contains a labeled picture corresponding to each actual river photo, and these two types of pictures (1080 pictures per category) are used for model training. The model is trained by the original image and the image with a mask, and the original image is annotated with the river and bridge structure to produce a masked image. The model uses a convolutional neural network to achieve image segmentation, and uses the masked image as the target to segment the image, and segment the range of the river flow and the range of the bridge structure. The image segmentation technology of this model is to analyze the image according to each pixel, process the original image according to the mask image, and obtain an analysis result that is similar to the mask image after classifying the original image according to pixels. The image is processed using an encoder and a decoder, which can handle small sample data sets well.

[0027] Since the actual number of unsafe conditions that have occurred may not be enough to reflect the full picture, some have not occurred. For example, if a disaster really occurred, it would have already happened and it is impossible to obtain it in real time. Therefore, the amount of unsafe condition data is insufficient for comparison: Therefore, the unsafe condition includes the actual unsafe condition and the simulated unsafe condition. The simulated unsafe condition is a model image simulated by a laboratory model, or an unsafe condition simulated by modifying the captured image, such as simulating the bending and breaking of bridge components, simulating increased river flow, etc.

[0028] In the first step, the inspection vehicle moves in a process of alternating between stopping and moving, and the images of the bridge structure and the waterway under the bridge are acquired when the inspection vehicle stops moving.

[0029] The stop and go method is to stop after walking a set distance, complete the acquisition of the image of the bridge structure and the image of the waterway under the bridge during the stop time, and then continue walking.

[0030] The above-mentioned embodiment of the disaster prevention safety inspection equipment and method for water-crossing bridges fully describes the technical solution of the present invention, which can automatically perform inspection tasks and remotely control the inspection vehicle to achieve regular inspections. Compared with drone inspections, which are not affected by bad weather, they are safer, more reliable and more stable.

Claims

1. A disaster prevention and safety inspection equipment for water-spanning bridges, characterized in that: The inspection equipment includes an inspection vehicle and a walking track. The walking track is pre-installed on the bridge guardrail across the waterway. The inspection vehicle is driven by a driving mechanism through wheels to reciprocate on the walking track. The inspection vehicle is provided with a temperature and humidity sensor, a rain sensor, a wind speed sensor, a camera and a controller. The temperature and humidity sensor, the rain sensor, the wind speed sensor, the camera and the driving mechanism are connected to the controller. The inspection vehicle is provided with a pan-tilt platform. The camera is mounted on the pan-tilt platform to obtain images of the bridge and the waterway in all directions. The controller is connected to the remote central server through a wireless communication module.

2. The inspection device according to claim 1, characterized in that: The walking track includes two groove rails arranged parallel to each other and spaced apart, the two groove rails are connected and fixed by a support plate, the support plate is used to connect and fix with the bridge guardrail, the groove rail is a "T"-shaped groove rail with a "T"-shaped groove, the middle vertical "1" of the "T"-shaped groove is the opening of the groove rail, and the two parallel "T"-shaped groove rail openings are arranged opposite to each other; the inspection walking vehicle includes a frame, two wheel axles are arranged at intervals at the front and rear of the bottom of the frame, wheels are respectively arranged on the protruding ends of the wheel axles, a temperature and humidity sensor, a rainfall sensor, and a wind speed sensor are fixed on the top of the frame, a pan-tilt seat rod extends out of the frame, the pan-tilt is fixed on the pan-tilt seat rod, the shape of the wheel corresponds to the shape of the "T"-shaped groove, the four wheels of the two axles are placed in the "T"-shaped groove of the track, and the frame is driven by a driving mechanism to reciprocate along the walking track, and the frame is placed between the two groove rails.

3. The inspection device according to claim 2, characterized in that: The driving mechanism comprises a driving motor and a speed changing mechanism, and the speed changing mechanism is connected to the wheel shaft to drive the wheel shaft to rotate.

4. The inspection device according to claim 1, characterized in that: The inspection vehicle is also provided with a solar panel and a power supply battery. The solar panel is connected to the power supply battery, and the power supply battery provides power for the controller, the driving mechanism, the sensor and the camera.

5. The inspection device according to claim 1, characterized in that: The outer edge of the wheel is covered with a rubber sleeve.

6. A method for inspecting the disaster prevention safety of a water-spanning bridge, based on the inspection method of the disaster prevention safety inspection equipment for a water-spanning bridge according to claim 1, the method comprising: Acquire signals from temperature and humidity sensors, rainfall sensors, and wind speed sensors in real time to determine whether the ambient temperature and humidity, rainfall, or wind speed exceeds a set threshold. If the set threshold is exceeded, start the inspection vehicle to conduct reciprocating inspections on the pre-installed walking track of the bridge guardrail. The steps of executing the inspection include: Step 1: Turn on the camera and obtain images of the bridge structure and the waterway under the bridge in real time while the inspection vehicle is moving back and forth; Step 2: Process the acquired image to extract the image’s characteristic image data, compare the characteristic image data with the standard image data model, evaluate the comparison results to determine whether there is a safety hazard, and issue an alarm based on the level of the safety hazard; Wherein: the standard image data model is a standard image data model obtained by pre-processing and storing image data of the same position under the safe state of the bridge structure and the waterway and the unsafe state that has occurred.

7. The method according to claim 6, characterized in that The standard image data model is a neural network comparison model, which is a neural network comparison model that extracts feature image data from image data of the same position in a safe state and in an unsafe state of the bridge structure and the waterway in advance and performs feature image data learning processing; in the second step, the feature image data is compared with the standard image data model, and the feature image data is input into the neural network comparison model to obtain an evaluation result.

8. The method according to claim 6 or 7, characterized in that: The unsafe state includes an actually occurring unsafe state and a simulated unsafe state. The simulated unsafe state is an unsafe state that is simulated by modifying an image.

9. The method according to claim 6, characterized in that In the first step, the inspection vehicle moves in a process of alternating between stopping and moving, and the images of the bridge structure and the waterway under the bridge are acquired when the inspection vehicle stops moving.

10. The method according to claim 9, characterized in that The stop and go method is to stop after walking a set distance, complete the acquisition of the image of the bridge structure and the image of the waterway under the bridge during the stop time, and then continue walking.

Images