Railway bridge multi-point data acquisition system and method

Abstract

The invention relates to a railway bridge multi-point data acquisition system and method, and the system comprises a plurality of sensors which are disposed on a bridge main body; a camera installed at a position where the overall situation of the bridge main body can be observed; and the wireless gateway equipment is wirelessly connected with the plurality of sensors and the cameras. According to the invention, a distributed multi-point monitoring and wireless transmission mode is adopted to carry out real-time data acquisition on a railway bridge state, firstly, a bridge collision state is judged through sensor data, and the severity of a bridge collision accident is graded; the gateway transmits the alarm data and the corresponding video information to a management platform through the Internet of Things, and a manager can inquire the data through the Internet of Things. Meanwhile, for the situation that a management unit needs to deal with safety problems urgently on site, alarm data are pushed to a mobile phone of a manager in real time in the form of a short message and a mobile phone APP, so that the manager arranges on-site confirmation work in time, and secondary accidents are prevented.

Description

Technical Field

[0001] The present invention relates to the field of railway bridge condition monitoring, and particularly to a multi-point data acquisition system and method for railway bridges. Background Art

[0002] A railway bridge is a structure built for a railway to cross a river, lake, valley or other obstacles, and to achieve a grade separation between railway lines or between a railway line and a road. With the rapid development of China's railways, the grade separations between railways and rivers or roads are becoming increasingly dense. At the same time, with the increasing trend of large-scale, heavy-duty and high-speed development of river shipping vessels and road vehicles, the contradictions among bridges, navigable vessels and road vehicles are becoming increasingly prominent. Once a bridge collision accident occurs, in severe cases, it will cause serious consequences such as interruption of railway transportation, loss of personnel and property.

[0003] Currently, in the aspect of pre-alarm inspection and monitoring of railway bridge collisions, most non-contact monitoring methods such as radar technology, video technology and laser technology are used for collision prediction. Their deployment is difficult, the equipment cost is high, and the accuracy of collision alarms needs to be improved. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a multi-point data acquisition system and method for railway bridges in view of the above deficiencies.

[0005] The present invention is realized through the following technical solutions:

[0006] A multi-point data acquisition system for railway bridges, the system comprising:

[0007] A number of sensors, respectively installed at a number of sensor deployment points on the bridge main body;

[0008] A camera, installed at a position where the overall situation of the bridge main body can be observed; and

[0009] A wireless gateway device, wirelessly connected to the number of sensors and the camera;

[0010] The number of sensor deployment points are evenly distributed on both sides of the bridge main body along the length direction of the bridge main body.

[0011] Further, in the multi-point data acquisition system for railway bridges, the sensor is at least one of a vibration monitoring sensor, a displacement sensor, an inclination sensor and a strain sensor.

[0012] Further, in the multi-point data acquisition system for railway bridges, the wireless gateway device is an edge computing gateway device.

[0013] A method for multi-point data acquisition of railway bridges based on the system described in any one of the above. A number of the sensors monitor the dynamic information of the bridge body in real time and transmit it to the wireless gateway device;

[0014] When the data value obtained by at least one of the sensors near the accident location exceeds the threshold, the sensor transmits the obtained dynamic information to the wireless gateway device. The wireless gateway device then collects the dynamic information obtained by other sensors near the accident location, comprehensively analyzes and processes the data information to determine whether a collision accident has occurred, the location of occurrence, and the severity level;

[0015] When the judgment result is a collision accident, the wireless gateway device controls the camera to find the accident location, turns the camera's view to the accident location, records the dynamic of the accident scene, and synchronously uploads it to the background server;

[0016] When the judgment result is not a collision accident, after the wireless gateway device uploads the collected dynamic information to the background server, this round of program ends.

[0017] Furthermore, in the described method for multi-point data acquisition of railway bridges, when the judgment result is a collision accident, the wireless gateway device reports the alarm data to the background server.

[0018] Furthermore, in the described method for multi-point data acquisition of railway bridges, after the camera records the dynamic of the accident scene, the wireless gateway device reports the relevant content to the background server, and relevant duty officers can view the accident video and relevant data information at the management platform.

[0019] Furthermore, in the described method for multi-point data acquisition of railway bridges, the wireless gateway device obtains the dynamic information of several sensors on site and converts the collision level to achieve the corresponding early warning effect; at the same time, intercepts the accident video information recorded by the camera, packages it and reports it to the background server. The background server then performs AI intelligent recognition on the accident video information using an intelligent recognition algorithm; for issues involving safety that urgently require the management unit to arrive at the scene for handling, the alarm data is pushed to the mobile phones of the management personnel in real time in the form of text messages and mobile APPs to notify the management personnel to arrange on-site work.

[0020] Furthermore, in the described method for multi-point data acquisition of railway bridges, the collision level is calculated by calculating the standard deviation of the triaxial acceleration values generated when the sensor is collided respectively, and taking the maximum value as the collision level evaluation value; the wireless gateway device sets the trigger signal output according to the collision level and reports it to the background server; the background server gives corresponding collision warnings to the management personnel according to the collision level.

[0021] Further, in the described method for collecting multi-point data of a railway bridge, the intelligent recognition algorithm is to verify and confirm the license plate of the vehicle when a collision occurs and whether the collision accident reaches the level requiring on-site handling; license plate recognition and confirmation is to recognize the vehicle type and license plate in chronological order, then select the vehicle closest in time to the collision moment, and select the most likely collision vehicle by comparing vehicle types; if vehicles of the same type are present in a very short period, then compare the heights, and the tallest one is the collision vehicle.

[0022] Further, in the described method for collecting multi-point data of a railway bridge, the main body of the bridge is segmented through instance segmentation, and it is judged whether the bridge has undergone large deformation and whether the accident reaches the level requiring on-site handling by comparing whether the center point and angle of the obtained mask exceed a preset threshold.

[0023] The advantages and effects of the present invention are as follows:

[0024] 1. The present invention provides a multi-point data collection system and method for a railway bridge, which uses a distributed multi-point monitoring method that combines sensing and vision to collect real-time data of the railway bridge, improving the response speed of the management department to accidents and providing a basis for judgment.

[0025] 2. The present invention directly monitors the contact between an object and the main body of the railway bridge using sensors, adjusts the camera angle in real time according to the perception situation, and then effectively realizes high-precision judgment of collisions through real-time warning via the Internet of Things technology, reducing unnecessary errors caused by human factors and environmental factors.

[0026] 3. The present invention provides a multi-point data collection system and method for a railway bridge, which uses a method of mutual cooperation and complementary optimization between edge computing and cloud computing. While ensuring that all alarm data is uploaded, AI intelligent recognition of video information is performed through a deep learning algorithm. For accidents that urgently require the management unit to arrive at the scene for handling, the alarm data is pushed to the mobile phones of the management personnel in real time in the form of text messages and mobile phone APPs, so that the management personnel can arrange on-site confirmation work in a timely manner to prevent the occurrence of secondary accidents. It improves the management efficiency of the management department for bridge collision accidents.

[0027] 4. The multi-point data collection system for a railway bridge provided by the present invention uses multi-point wireless deployment, reducing the construction difficulty, saving construction costs, and having a relatively low equipment cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Shows the deployment schematic diagram of the multi-point data collection system for a railway bridge provided by the present invention;

[0029] Figure 2 Shows the flowchart of the method for collecting multi-point data of a railway bridge provided by the present invention.

[0030] Description of the drawing reference numerals: 1 - bridge main body, 2 - sensor, 3 - camera, 4 - wireless gateway device, 5 - railway, 6 - road surface. Detailed implementation manners

[0031] To make the purpose, technical solutions and advantages of the implementation of the present invention clearer, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as a limitation to the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present invention. The embodiments of the present invention will be described in detail below with reference to the accompanying drawings:

[0032] In the description of the present invention, it should be understood that unless otherwise specified, the meaning of "a plurality of" is two or more; the terms "center", "longitudinal", "transverse", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the scope of protection of the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and should not be construed as indicating or implying relative importance. In the description of the present invention, it should also be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected, or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0033] Figure 1Shows the deployment schematic diagram of the multi-point data acquisition system for railway bridges provided by the present invention. The multi-point data acquisition system includes a number of sensors 2, cameras 3, and wireless gateway devices 4. The number of sensors are evenly and densely distributed and installed at a number of sensor deployment points on the bridge main body 1. The number of sensor deployment points are evenly distributed on both sides of the bridge main body 1 along the length direction of the bridge main body, so that any abnormality at any position of the bridge main body 1 can be sensed in the first time. When a vehicle or ship causes a collision accident to the bridge main body, the sensor can timely sense the collision part, which is of great help to bridge maintenance and personnel safety. Specifically, the sensor 2 is at least one of a vibration monitoring sensor, a displacement sensor, an inclination sensor, and a strain sensor, and can be densely deployed inside the bridge culvert, and adopts an internal battery and a wireless deployment method, without being troubled by wire pulling. The camera is installed at a position where the overall situation of the bridge main body 1 can be observed. The wireless gateway device 4 can be deployed together with the camera 3 in a place suitable for seeing the overall situation of the bridge main body and the culvert. The wireless gateway device 4 is wirelessly connected to a number of sensors 2, and it can be wired or wirelessly connected to the camera 3.

[0034] The wireless gateway device is an edge computing gateway device, which has functions such as wireless data acquisition, controlling the camera pan-tilt, edge computing according to the data health model, data uploading, and data monitoring model updating. The algorithm of the wireless gateway device is updated remotely in real time according to the on-site situation and different indicators of each site. And each indicator can be determined by the parameters of multiple sensors. Edge computing technology is a distributed open platform that integrates network, computing, storage, and application core capabilities on the network edge side close to things or data sources, and provides edge intelligent services nearby. That is, edge computing analyzes the data collected from the terminal directly in the local device or network close to where the data is generated, without the need to transmit the data to the cloud data processing center.

[0035] As Figure 1 , which shows that the multi-point data acquisition system for railway bridges provided by the present invention is applied to the bridge environment on the road surface, and it can also be applied to the bridge environment on the river surface.

[0036] Based on the above multi-point data acquisition method of the multi-point data acquisition system for railway bridges, the multi-point data acquisition method includes:

[0037] A number of sensors 2 continuously monitor the dynamic information of the bridge main body 1 and transmit it to the wireless gateway device 4. The sensor 2 can filter out the vibration of the train on the railway through an algorithm.

[0038] As Figure 2As shown, when a collision accident occurs, the data value obtained by at least one sensor near the accident site exceeds the threshold. The sensor transmits the obtained dynamic information to the wireless gateway device, and the wireless gateway device then collects the dynamic information obtained by other sensors near the accident site, comprehensively analyzes and processes the data information, and determines the accident location and severity level.

[0039] When the judgment result is a collision accident, the wireless gateway device issues an accident occurrence alarm notification (the first notification), and controls the camera to search for the accident site, turns the camera view to the accident site, quickly adjusts the angle and focal length of the camera, records the dynamic of the accident scene, and synchronously uploads it to the background server. The wireless gateway device pre-sets the position control model for each sensor deployment point. When an alarm is confirmed, the wireless gateway device will control the camera angle and focal length according to the control model of the sensor deployment point. After the camera 3 records the dynamic of the accident scene, the wireless gateway device 4 reports the relevant content to the background server, and then the background server notifies the relevant personnel (the second notification) to watch the on-site accident video and relevant data information for the relevant personnel to make reference and judgment on the accident situation. The relevant content includes the relevant data information near the accident site transmitted by the sensor and the on-site accident video, etc.

[0040] When the judgment result is not a collision accident, the wireless gateway device uploads the collected dynamic information to the background server and ends this round of program.

[0041] In an embodiment, at a certain moment, when the data value obtained by the strain sensor exceeds the set threshold, the strain sensor transmits the data information to the wireless gateway device, and the wireless gateway device then collects the data information of displacement sensors, tilt sensors, vibration sensors, etc. near the location. The wireless gateway device combines the overall performance of these data to judge whether a bridge collision accident has occurred. If it meets the requirements of the set model judgment, the wireless gateway device alarms, controls the camera to take videos of the specified position, uploads them to the background server, and alarms that a collision accident has occurred.

[0042] The wireless gateway obtains the data of on-site sensor devices and converts the collision level to achieve the warning effect. At the same time, it intercepts the accident video information recorded by multi-directional cameras, packages it, and reports it to the background server. The server then uses intelligent recognition algorithms to perform AI intelligent recognition on the video information of the alarm data. For safety issues that urgently require the management unit to arrive at the scene for handling, the alarm data is pushed to the mobile phones of the management personnel in real time in the form of text messages and mobile APPs, so that the management personnel can arrange on-site confirmation work in a timely manner to prevent the occurrence of secondary accidents. The collision level is calculated by calculating the standard deviation of the triaxial acceleration values generated when the sensor is collided, and the maximum value is taken as the collision level evaluation value. The evaluation levels include less than 100 as ignored, 100 - 450 as mild, 450 - 950 as moderate, and greater than 950 as severe. The wireless gateway has a built-in digital output and supports setting the trigger signal output according to the collision level, that is, an external audible and visual warning device can be connected. Moreover, regardless of the severity of the collision, it will immediately report to the background server. The background server can prompt the corresponding collision warning to the central personnel in the management platform through means such as message flashing and sound prompts, and then decide whether on-site emergency disposal is required after AI intelligent analysis and judgment. The intelligent recognition algorithm mainly verifies and confirms the vehicle license plate at the time of the collision and whether the collision accident reaches the recognition level that requires on-site disposal. Among them, license plate recognition and confirmation identify the vehicle type and license plate in chronological order, then select the nearest vehicle according to the collision moment, and select the most likely collision vehicle according to the vehicle type comparison. If the same type of vehicle exists in a very short period of time, then compare the heights, and the highest one is the collision vehicle. Identifying whether the accident reaches the level that requires on-site disposal is to segment the bridge through instance segmentation, and compare whether the center point and angle of the obtained mask mask exceed the preset threshold to judge whether the bridge has undergone large deformation. The real-time performance of edge computing is used to achieve timely warning, and then the accuracy of cloud computing is used to realize the integration with management operations, realizing a reasonable and efficient accident emergency handling system, which reflects the value of the collaboration between edge computing and cloud computing.

[0043] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of implementation of the present invention. Any equivalent changes and modifications made within the protection scope of the present invention shall be considered to fall within the protection scope of the present invention.

Claims

1. A railway bridge multi-point data acquisition system, characterized in that: The system includes: A plurality of sensors are installed correspondingly at a plurality of sensor deployment points of the bridge body; A camera installed at a position capable of observing the overall condition of the bridge body; and A wireless gateway device, wirelessly connected to the plurality of sensors and the camera; A plurality of sensor deployment points are evenly distributed on both sides of the bridge body along the length direction of the bridge body.

2. A railway bridge multi-point data acquisition system according to claim 1, characterized in that: The sensor is at least one of a vibration monitoring sensor, a displacement sensor, a tilt sensor and a strain sensor.

3. A railway bridge multi-point data acquisition system according to claim 1, characterized in that: The wireless gateway device is an edge computing gateway device.

4. A railway bridge multi-point data collection method based on the system according to any one of claims 1 to 3, characterized in that: The plurality of sensors monitor the dynamic information of the bridge body in real time and transmit the information to the wireless gateway device; When the data value obtained by at least one of the sensors near the accident site exceeds the threshold, the sensor transmits the obtained dynamic information to the wireless gateway device, and the wireless gateway device then collects the dynamic information obtained by other sensors near the accident site, conducts comprehensive analysis and processing on the data information, and determines whether a collision accident has occurred, the location of occurrence, and the severity level; When the judgment result is a collision accident, the wireless gateway device controls the camera to find the accident location, rotates the camera angle to the accident location, records the dynamics of the accident scene, and synchronously uploads it to the background server; When the judgment result is not a collision accident, the wireless gateway device uploads the collected dynamic information to the background server and ends this round of procedures.

5. A railway bridge multi-point data collection method according to claim 4, characterized in that: When the judgment result is a collision accident, the wireless gateway device reports the alarm data to the background server.

6. A railway bridge multi-point data collection method according to claim 4, characterized in that: After the camera completes recording the dynamics of the accident scene, the wireless gateway device reports the relevant content to the backend server, and the relevant on-duty personnel can view the on-site accident video and related data information through the management platform.

7. A railway bridge multi-point data collection method according to claim 4, characterized in that: The wireless gateway device obtains the dynamic information of several sensors on site, and converts the collision level to achieve the corresponding early warning effect; at the same time, the accident video information recorded by the camera is intercepted, packaged and reported to the background server, and the background server then uses an intelligent recognition algorithm to perform AI intelligent recognition on the accident video information; for safety issues that urgently require the management unit to be on site for processing, the alarm data is pushed to the management personnel's mobile phone in real time through text messages and mobile phone APPs, notifying the management personnel to arrange on-site work.

8. A railway bridge multi-point data collection method according to claim 7, characterized in that: The collision level is calculated based on the standard deviation of the three-axis acceleration values ​​generated when the sensor is collided, and the maximum value is taken as the collision level judgment value; the wireless gateway device sets the trigger signal output according to the collision level and reports it to the background server; the background server prompts the management personnel to issue the corresponding collision warning according to the collision level.

9. The method for collecting multi-point data of a railway bridge according to claim 7, characterized in that: The intelligent recognition algorithm verifies and confirms the vehicle license plate at the time of the collision and whether the collision accident requires on-site treatment; the license plate recognition confirmation identifies the vehicle type and license plate in chronological order, then takes the nearest vehicle according to the collision time, and takes the most likely collision vehicle according to the vehicle type comparison; If there are vehicles of the same type in a close period of time, their heights will be compared and the highest one will be the vehicle that collided.

10. A railway bridge multi-point data collection method according to claim 9, characterized in that: The bridge body is segmented by examples, and the obtained mask center point and angle are compared to see whether they exceed the preset threshold, so as to determine whether a large deformation of the bridge occurs and identify whether the accident requires on-site treatment.

Images