Wagon balance management system based on AI intelligence

Abstract

The invention discloses an AI intelligence-based wagon balance management system, and relates to the technical field of wagon balance management, the system comprises an acquisition module, an analysis module, a passing module and a passing database, historical passing vehicles are monitored and analyzed, damage of different types of vehicles to the wagon balance ground and a connecting piece is analyzed, and the wagon balance ground and the connecting piece are analyzed; and when the vehicle needs to pass through the wagon balance, the speed and the uphill and downhill gradient of the vehicle are controlled, so that the damage of the impact force of the vehicle to the wagon balance surface and the connecting piece is reduced, the stability and the safety of the vehicle are ensured, the phenomenon that the pressure of the vehicle to the wagon balance surface is not uniformly distributed is reduced, and the service life of the vehicle is prolonged. The weighing accuracy is ensured, and the weighing reliability of the wagon balance system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of weighbridge management, and in particular to a weighbridge management system based on AI intelligence. Background Art

[0002] In today's logistics, industrial production and other fields, the scale is a key device for measuring the weight of bulk goods, and its management accuracy and efficiency are crucial. In daily life, a large number of vehicles pass through, stop and weigh on the scale, which increases the damage to the scale surface and reduces the stability of the vehicle. Therefore, in order to ensure the safety of the scale surface and vehicle traffic, it is necessary to manage the speed of vehicles passing the scale and the slope of the scale.

[0003] Prior art, such as the invention patent disclosed in application publication number CN117557175A, is an intelligent control and management system for a weighbridge. The position deviation coefficient of the vehicle is obtained by analysis, and based on the comparison result of the position deviation coefficient, the vehicle's weighing status is determined, thereby improving the accuracy of measurement and making the monitoring and processing process more efficient and rapid. By comparing the empty vehicle weight with the vehicle's previous weighing record weights, the vehicle's compliance with the weighing requirements is obtained and corresponding processing is performed to ensure the compliance of the vehicle with the weighing requirements, which helps to ensure the accurate measurement and compliant transportation of goods. By analyzing the cargo weight accuracy coefficient, it helps to evaluate the accuracy and consistency of the cargo weight, thereby improving the accuracy of cargo weight measurement, cargo transportation efficiency and safety. By analyzing the stability coefficient of the weighbridge, it helps to avoid weighing errors caused by weighbridge problems and ensure the accuracy and credibility of cargo weighing.

[0004] The above scheme discloses the process of weighing the weight of the goods on the scale, but lacks damage monitoring and analysis of the scale surface after the vehicle passes through the scale. As a result, the speed of the vehicle cannot be effectively controlled when the subsequent vehicle enters the scale, thereby failing to reduce the impact force of the vehicle and the wear of the scale surface caused by the impact force. At the same time, the wear of the scale surface seriously affects the stability of the vehicle driving on the scale surface, resulting in uneven pressure distribution of the vehicle on the scale surface, and local deformation, cracking and other problems may occur on the scale surface. At the same time, the instability of the vehicle cannot guarantee the accuracy of weighing, reducing the credibility of the weighing of the scale system.

[0005] In addition, for the uphill and downhill slopes of the scale, if the slope is not appropriate, the engine output power changes when the vehicle climbs the slope, which will cause the vehicle itself to vibrate, and also interfere with the weighing accuracy. Driving on a suitable slope can ensure that the vehicle's braking distance is within a controllable range and ensure the stability of the vehicle. However, the above scheme lacks an analysis of the suitability of the slope of the scale ramp before the vehicle enters the scale, and is unable to adjust the appropriate slope for the vehicle, unable to effectively ensure the stability of the vehicle, reducing the safety of the vehicle on the scale and the accuracy of the weighing results. In addition, it reduces damage to the scale and reduces the cost of scale maintenance. Summary of the invention

[0006] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a weighbridge management system based on AI intelligence.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides an AI-based intelligent weighbridge management system, comprising the following modules: an acquisition module, used to obtain monitoring data of each historical vehicle passing through the weighbridge.

[0008] The analysis module includes a scale surface analysis unit and a ramp analysis unit.

[0009] The weighing surface analysis unit is used to utilize the monitoring data of each historical vehicle passing through the weighing scale, analyze the monitoring data of each historical vehicle passing through the weighing scale, and set a weighing scale damage table.

[0010] The ramp analysis unit is used to set a ramp pass table using the monitoring data of each historical vehicle passing the weighbridge.

[0011] The passage module is used to collect the image of the vehicle when the vehicle reaches the ramp on the scale, and use the scale damage table and ramp passage table to set the vehicle's passage speed and ramp slope data for the scale.

[0012] The beneficial effects of the present invention are as follows: the present invention provides a weighbridge management system based on AI intelligence, which monitors and analyzes the vehicles that have passed through in the past, analyzes the damage caused to the weighbridge surface and connecting parts by different types of vehicles, and the stability of different types of vehicles at different slopes. When a vehicle needs to pass through the weighbridge, the speed of the vehicle and the uphill and downhill slopes are controlled, while reducing the damage to the weighbridge surface and connecting parts caused by the impact force of the vehicle, ensuring the stability and safety of the vehicle, reducing the occurrence of uneven pressure distribution of the vehicle on the weighbridge surface, ensuring the accuracy of weighing, and ensuring the credibility of the weighing of the weighbridge system. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0014] Figure 1 It is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] See also Figure 1 As shown, an AI-based intelligent weighbridge management system includes the following modules: an acquisition module, an analysis module, a passage module and a passage database.

[0017] The acquisition module is used to obtain the monitoring data of each historical vehicle passing through the weighbridge.

[0018] In the above, the monitoring data of each historical vehicle passing through the weighbridge is obtained in the following specific process: before each historical vehicle enters the uphill road, the monitoring equipment is used to collect the weighbridge surface image in front of each historical vehicle.

[0019] When each historical vehicle starts to enter the uphill road, the monitoring equipment is used to collect the driving video of each historical vehicle until each historical vehicle exits the downhill road, and the collection is terminated. At the same time, the speed sensor is used to collect the passing speed of each historical vehicle.

[0020] After each historical vehicle exits the downhill road, the monitoring equipment is used again to collect the scale surface image behind each historical vehicle.

[0021] When each historical vehicle arrives in front of and leaves the weighing surface, the detection values ​​of each sensor under the weighing surface are collected. Thus, the weighing surface image in front of each historical vehicle, the driving video, the passing speed, the weighing surface image after the vehicle, and the detection values ​​of each sensor before and after arriving at the weighing surface and leaving the weighing surface are used as monitoring data of each historical vehicle passing through the weighing surface.

[0022] It should be noted that the sensors under the scale surface are resistance strain sensors, and the internal structure includes an elastic element and a resistance strain gauge attached to the elastic element. When the vehicle drives onto the scale, the weight is transmitted to the sensor through the scale surface, and the elastic element is deformed by the pressure. According to Hooke's law, within the elastic limit, the deformation of the elastic element is proportional to the external force. This deformation causes the resistance strain gauge attached to it to deform, thereby changing the resistance value of the resistance strain gauge. The resistance strain gauge generally forms a Wheatstone bridge circuit, and the change in resistance value will cause the change in the output voltage of the bridge. The output voltage signal is proportional to the pressure applied by the vehicle on the sensor, and after being amplified by the amplifier, it is transmitted to the data acquisition and processing system. The data acquisition and processing system converts the analog voltage signal into a digital signal, and then uses the sensor characteristic curve obtained by pre-calibration to convert the voltage value corresponding to the digital signal into the actual weight value. The calibration process places a standard weight of known weight on the scale surface, records the voltage signal output by the sensor under different weights, and establishes a corresponding relationship curve between voltage and weight. During actual weighing, the corresponding weight value is found from the calibration curve according to the voltage signal output by the sensor, so as to accurately obtain the vehicle weight. The specific content has been made public on the Internet and can be searched on the Internet, so it will not be repeated here.

[0023] The analysis module includes a scale surface analysis unit and a ramp analysis unit.

[0024] The weighing surface analysis unit is used to utilize the monitoring data of each historical vehicle passing through the weighing scale, analyze the monitoring data of each historical vehicle passing through the weighing scale, and set a weighing scale damage table.

[0025] In a specific embodiment, the monitoring data of each historical vehicle passing through the weighing scale is analyzed and a weighing scale damage table is set. The specific process is: using the monitoring data of each historical vehicle passing through the weighing scale, extracting the driving video of each historical vehicle on the weighing surface and the detection values ​​of each sensor before and after arriving at the weighing surface, as well as obtaining the weighing surface image before and after each historical vehicle, and at the same time obtaining the weight, size and travel speed of each historical vehicle.

[0026] Using the weight, size and speed of each historical vehicle, we classify each historical vehicle and get the type of each historical vehicle. Then, we obtain the driving video of each historical vehicle of each type on the weighing surface, the detection values ​​of each sensor before and after arriving at the weighing surface, the weighing surface image before and after driving, analyze the damage level of each type of damage to the weighing surface caused by each type of vehicle, and summarize the damage level of each type of damage to the weighing surface to get the weighbridge damage table.

[0027] It should be noted that the traffic database stores weight ranges, size ranges and speed ranges for various types of vehicles. The weight, size and speed of each historical vehicle are compared with the weight range, size range and speed range of each type of vehicle. If the weight, size and speed of a historical vehicle are respectively within the weight range, size range and speed range of a certain type of vehicle, then the type is taken as the type of the vehicle, so as to classify each historical vehicle.

[0028] The types include heavy-duty, large-size, low-speed vehicles and light, small-size, medium-speed vehicles. For example, the weight range of heavy-duty, large-size, low-speed vehicles can be above 3,500 kg, the size range can be above 5.2 meters in length, above 2.0 meters in width and 3-4 meters in height, and the speed range is 0-25 kilometers per hour. The specific values ​​can be adjusted by relevant staff according to actual conditions and are not restricted here.

[0029] In the above, the damage level of each type of damage to the scale surface caused by each type of vehicle is analyzed, and the specific process is: the damage type includes scratch damage type, bending damage type and connector damage type.

[0030] The initial position and initial protruding height of each initial debris in the scale surface are obtained from the scale surface image before each historical vehicle of each type, and the falling position of each new debris, the position of each initial debris and the protruding height of each initial debris are obtained from the scale surface image after each historical vehicle of each type.

[0031] The area where the tires of each historical vehicle travel on the scale surface is obtained from the driving video of each historical vehicle of each type on the scale surface, and then a heat map of the scale surface driving is generated to obtain the high-frequency driving area of ​​the scale surface. If the initial position of an initial debris is not in the high-frequency driving area of ​​the scale surface, but the position of the initial debris is in the high-frequency driving area of ​​the scale surface, the debris is recorded as a first-class dangerous debris, and the initial protrusion height of the first-class dangerous debris is subtracted from the protrusion height to obtain the embedding depth of the first-class dangerous debris, so as to count the number of first-class dangerous debris corresponding to each historical vehicle of each type and the embedding depth of each first-class dangerous debris; each new debris that falls in the high-frequency driving area of ​​the scale surface is recorded as a second-class dangerous debris, and the number of second-class dangerous debris corresponding to each historical vehicle of each type is counted.

[0032] The damage grade of the scratch damage type caused by each type of vehicle on the scale surface is calculated by using the number of Class I hazardous debris, the embedding depth of each Class I hazardous debris and the number of Class II hazardous debris corresponding to each historical vehicle of each type.

[0033] Preferably, the calculation process of the damage level of the scratch damage type caused by each type of vehicle on the scale surface is as follows: the number of Class I dangerous debris and the number of Class II dangerous debris corresponding to each historical vehicle in each type are recorded as and , the embedding depth of each type of dangerous debris corresponding to each historical vehicle in each type is recorded as , where z represents the type number of the vehicle, x represents the number of the historical vehicle, and n represents the number of a class of dangerous debris. z, x and n are all positive integers.

[0034] Using the analytical formula: , get the damage level of the scratch damage type of the z-th type of vehicle on the scale surface , where r represents the damage level, r is a positive integer, R is the maximum damage level, R is a positive integer, r≤R, and They represent the initial debris quantity and new debris quantity of the xth historical vehicle in the zth type, d represents the thickness of the scale surface, e is a natural constant, X represents the number of historical vehicles, N1 represents the total quantity of dangerous debris of a type, , , , They are respectively the upper limit characteristic value of the first damage level, the upper limit characteristic value of the r-1th damage level, the upper limit characteristic value of the rth damage level, and the upper limit characteristic value of the R-1th damage level.

[0035] It should be noted that , , , They are reference values ​​for confirming the damage level and are set by relevant engineers. The thickness of the scale surface is obtained from the scale management backend.

[0036] The flatness of the weighing surface is obtained from the images of the weighing surface before and after each historical vehicle of each type, and the curvature of the weighing surface of each historical vehicle of each type is calculated; and the damage level of the bending damage type and the connector damage type of each type of vehicle to the weighing surface is calculated by using the detection values ​​of each sensor before and after each historical vehicle of each type arrives in front of the weighing surface and leaves the weighing surface.

[0037] Preferably, the calculation process of the damage level of the bending damage type of each type of vehicle to the weighing surface is as follows: obtain the length of the weighing surface from the scale management, obtain the height difference of the weighing surface from the weighing surface image before and after each historical vehicle of each type, divide the height difference of the weighing surface in the weighing surface image before and after each historical vehicle of each type by the length of the weighing surface, and obtain the curvature before and after of each historical vehicle of each type, subtract the curvature after and the curvature before of each historical vehicle of each type, and obtain the curvature of the weighing surface of each historical vehicle of each type, cluster the curvature of the weighing surface of each historical vehicle of each type, and obtain the curvature of the weighing surface of each type of vehicle, and compare them with the weighing surface curvature intervals corresponding to each damage level of the bending damage type in the traffic database to obtain the damage level of the bending damage type of each type of vehicle to the weighing surface; wherein the weighing surface curvature intervals corresponding to each damage level of the bending damage type are similar in meaning and setting process to the weight intervals, size intervals and travel speed intervals of each type of vehicle, and will not be repeated here.

[0038] Preferably, the calculation process of the damage level of each type of vehicle connector damage type is as follows: subtract the detection value of each sensor corresponding to each historical vehicle of each type after leaving the weighing surface from the detection value of each sensor before arriving at the weighing surface, and obtain the difference value of each sensor corresponding to each historical vehicle of each type, and then select the maximum difference value as the difference value corresponding to each historical vehicle of each type, and cluster the difference values ​​corresponding to each historical vehicle of each type to obtain the difference value of each type of vehicle, and then compare it with the difference value interval of each damage level of the connector damage type in the traffic database to obtain the damage level of the connector damage type of each type of vehicle; wherein the difference value interval of each damage level of the connector damage type is similar to the meaning and setting process of the weight interval, size interval and travel speed interval of each type of vehicle, and will not be repeated here.

[0039] Debris on the scale surface, such as stones, metal shavings, etc., will be embedded in the scale surface when the vehicle is driving, causing scratches. In the subsequent rolling, the scratches will gradually expand and deepen. Long-term accumulation may also affect the weighing accuracy. When the scale surface is deformed, such as local depressions or bulges, it affects the stability of the vehicle on the scale. The scale surface is firmly connected to the supporting structure by connectors. Loose connectors cause the scale surface to be unstable, and the weight distribution of the vehicle on the scale surface changes, causing the pressure signal received by the corresponding sensor to be distorted, and ultimately causing deviations in the weighing data. Therefore, when the vehicle passes, it is necessary to analyze the scratch damage, bending damage and connector damage on the scale surface to reduce the damage to the scale and the danger to the vehicle when subsequent vehicles enter the scale, and ensure the accuracy of weighing.

[0040] The ramp analysis unit is used to set a ramp pass table using the monitoring data of each historical vehicle passing the weighbridge.

[0041] In a specific embodiment, the ramp traffic table is set up, and the specific process is as follows: obtaining driving videos from the monitoring data of each historical vehicle passing through the weighing scale, intercepting uphill videos and downhill videos from the driving videos, summarizing the uphill videos and downhill videos of each historical vehicle of each type, obtaining the uphill surface slope, the number of new debris on the uphill surface and the vehicle stability from the uphill videos of each historical vehicle of each type, thereby calculating the traffic characteristic values ​​of each type of vehicle driving on each uphill surface slope, and similarly calculating the traffic characteristic values ​​of each type of vehicle on each downhill surface slope.

[0042] Preferably, the uphill video of each historical vehicle of each type is used, and the driving trajectory of each historical vehicle of each type is constructed using GPS, and the lateral offset in the driving trajectory of each historical vehicle of each type is used, and the inverse of the lateral offset in the driving trajectory of each historical vehicle of each type is used as the vehicle stability.

[0043] In a specific embodiment, the specific process of calculating the traffic characteristic values ​​of each type of vehicle traveling on each uphill slope is as follows: using the uphill slope of each historical vehicle of each type, the number of new debris on the uphill surface and the vehicle stability, counting the number of new debris on each uphill surface and the stability of each vehicle when each type of vehicle passes through each uphill slope, clustering the number of new debris on each uphill surface and the stability of each vehicle when each type of vehicle passes through each uphill slope, obtaining the number of new debris on the uphill surface and the stability of the vehicle when each type of vehicle passes through each uphill slope, and performing normalization processing, and the processed values ​​are marked as and , where q represents the number of the slope of each uphill surface, and q is a positive integer.

[0044] Using the calculation formula: Get the traffic characteristic value of the z-th type of vehicle traveling on the q-th uphill slope .

[0045] The traffic characteristic values ​​of each type of vehicle traveling on each uphill slope and each downhill slope are summarized to obtain a ramp traffic table.

[0046] The passage module is used to collect the image of the vehicle when it reaches the ramp of the scale, and use the scale damage table and ramp passage table to set the vehicle's passage speed and ramp slope data for the scale. The ramp slope data includes the uphill slope and the downhill slope.

[0047] In a specific embodiment, the speed and ramp gradient data of the vehicle passing the weighing scale are set, and the specific process is: identify the vehicle's license plate number from the vehicle's image, obtain the vehicle's passage record from the passage database, if the vehicle has a passage record, then obtain the weight of the vehicle at each historical passage from the vehicle's passage record, cluster the weight of the vehicle at each historical passage, and use the result as the reference weight of the vehicle.

[0048] The dimensions of the vehicle are obtained from the image of the vehicle, and the dimensions and reference weight of the vehicle are compared with the dimension range and weight range corresponding to each type in the traffic database. If the dimensions and reference weight of the vehicle are respectively within the dimension range and weight range corresponding to a certain type, the type is used as the marking type, and the damage level of the scratch damage type, the damage level of the bending damage type and the damage level of the connector damage type of each marking type are obtained from the damage table of the weighbridge. Then, the damage severity value of each marking type is calculated, and the marking type with the smallest damage severity value is selected as the type of the vehicle. The travel speed range of the corresponding type of the vehicle is obtained, and any travel speed in the travel speed range of the corresponding type of the vehicle can be used as the travel speed of the vehicle passing the weighbridge.

[0049] It should be noted that if there is no passage record for the vehicle, a feature extraction algorithm is used to calculate the similarity between the image of the vehicle and the images of each historical vehicle in the passage record of the vehicle in the passage database. Each historical vehicle with a similarity greater than a preset similarity is regarded as a marked vehicle, and the weight of each marked vehicle at each historical passage is obtained. The weight of each marked vehicle at each historical passage is then clustered, and the result is used as the reference weight of the vehicle.

[0050] Preferably, the damage severity value of each mark type is calculated by recording the damage level of the scratch damage type, the damage level of the bending damage type and the damage level of the connector damage type of each mark type as , and , using the calculation formula: , get the damage severity value of the pth marker type ,in, , , They are the important weight of scratch damage type, the important weight of bending damage type, and the important weight of connector damage type; , , ;

[0051] In the formula, , , They represent the mean damage level of the scratch damage type, the mean damage level of the bending damage type, and the mean damage level of the connector damage type for each mark type. , , They are respectively the maximum difference between the damage levels of the scratch damage types of each mark type, the maximum difference between the damage levels of the bending damage types, and the maximum difference between the damage levels of the connector damage types.

[0052] Obtain the traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill slope and each downhill slope from the ramp traffic table, and simultaneously obtain the current slope of the uphill surface of the scale, and analyze the necessity of replacing the uphill surface slope of the scale. If the necessity of replacing the uphill surface slope of the scale is insufficient, use the current slope of the uphill surface of the scale as the uphill surface slope in the ramp slope data; if the necessity of replacing the uphill surface slope of the scale is sufficient, select the uphill surface slope with the largest traffic characteristic value among the corresponding type of vehicles as the uphill surface slope in the ramp slope data, and analyze the downhill surface slope in the ramp slope data in the same way.

[0053] In the above, the necessity of analyzing the slope change of the uphill surface on the scale is specifically analyzed as follows: using the traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill surface slope, the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill surface slopes is calculated, and at the same time, the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale is extracted. If the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill surface slopes is less than or equal to the set difference rate threshold or the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale is greater than or equal to the set traffic characteristic value threshold, it indicates that the necessity of changing the slope of the uphill surface on the scale is insufficient; if the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill surface slopes is greater than the set difference rate threshold and the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale is less than the set traffic characteristic value threshold, it indicates that the necessity of changing the slope of the uphill surface on the scale is sufficient.

[0054] It should be noted that the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill slopes is the result of average calculation of the difference in traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill slope divided by the traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill slope.

[0055] The set difference rate threshold is a reference value for judging whether the stability difference of vehicles of a type when driving on different slopes is large. The specific value is set by engineers based on actual usage scenarios. When the difference rate is greater than or equal to the difference rate threshold, it indicates that the stability difference of vehicles of a type when driving on different slopes is large, and it is necessary to select a suitable slope to ensure the stability of the vehicle. Otherwise, it indicates that the stability difference of vehicles of a type when driving on different slopes is small, and it is not necessary to select a suitable slope.

[0056] The traffic characteristic value threshold is a reference value for judging whether the vehicle is stable when climbing a slope. The specific value is set by engineers based on actual usage scenarios. When the traffic characteristic value is greater than or equal to the traffic characteristic value threshold, it indicates that the vehicle is relatively stable when climbing a slope. Otherwise, it indicates that the vehicle is unstable when climbing a slope.

[0057] The traffic database is used to store weight ranges, size ranges and travel speed ranges for various types of vehicles, store the pound surface curvature ranges corresponding to each damage level of the bending damage type and the difference value ranges for each damage level of the connecting part damage type, and also store the vehicle's traffic records.

[0058] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they shall all fall within the protection scope of the present invention.

Claims

1. A weighbridge management system based on AI intelligence, characterized in that: Includes the following modules: The acquisition module is used to obtain the monitoring data of each historical vehicle passing through the weighbridge; The analysis module includes a scale surface analysis unit and a ramp analysis unit; The weighing surface analysis unit is used to analyze the monitoring data of each historical vehicle passing through the weighing scale and set a weighing scale damage table; The ramp analysis unit is used to set a ramp pass table using monitoring data of each historical vehicle passing the weighbridge; The passage module is used to collect the image of the vehicle when the vehicle reaches the ramp on the scale, and use the scale damage table and ramp passage table to set the vehicle's passage speed and ramp slope data for the scale.

2. According to claim 1, the AI-based weighbridge management system is characterized in that: The specific process of obtaining the monitoring data of each historical vehicle passing through the weighbridge is as follows: Before each historical vehicle enters the uphill road, use monitoring equipment to collect images of the scale surface in front of each historical vehicle; When each historical vehicle starts to drive into the uphill road, the monitoring equipment is used to collect the driving video of each historical vehicle until each historical vehicle drives out of the downhill road, and the collection is ended. At the same time, the speed sensor is used to collect the passing speed of each historical vehicle; When each historical vehicle exits the downhill road, the monitoring equipment is used again to collect the scale surface image behind each historical vehicle; When each historical vehicle arrives in front of and leaves the weighing surface, the detection values ​​of each sensor under the weighing surface are collected. Thus, the weighing surface image in front of each historical vehicle, the driving video, the passing speed, the weighing surface image after the vehicle, and the detection values ​​of each sensor before and after arriving at the weighing surface and leaving the weighing surface are used as monitoring data of each historical vehicle passing through the weighing surface.

3. The AI-based weighbridge management system according to claim 1 is characterized in that: The monitoring data of each historical vehicle passing through the weighbridge is analyzed to set a weighbridge damage table. The specific process is as follows: Using the monitoring data of each historical vehicle passing through the weighbridge, extract the driving video of each historical vehicle on the weighbridge and the detection values ​​of each sensor before and after arriving at the weighbridge, as well as obtain the weighbridge image before and after each historical vehicle, and at the same time obtain the weight, size and passing speed of each historical vehicle; Using the weight, size and speed of each historical vehicle, we classify each historical vehicle and get the type of each historical vehicle. Then, we obtain the driving video of each historical vehicle of each type on the weighing surface, the detection values ​​of each sensor before and after arriving at the weighing surface, the weighing surface image before and after driving, analyze the damage level of each type of damage to the weighing surface caused by each type of vehicle, and summarize the damage level of each type of damage to the weighing surface to get the weighbridge damage table.

4. The AI-based weighbridge management system according to claim 3 is characterized in that: The specific process of analyzing the damage level of each type of damage to the scale surface by each type of vehicle is as follows: The damage types include scratch damage type, bending damage type and connector damage type; The initial position and initial protrusion height of each initial debris in the scale surface are obtained from the scale surface image before each historical vehicle of each type, and the falling position of each new debris, the position of each initial debris and the protrusion height of each initial debris are obtained from the scale surface image after each historical vehicle of each type; From the driving videos of each historical vehicle of each type on the scale surface, the driving area of ​​the tires of each historical vehicle on the scale surface is obtained, and then the scale surface driving heat map is generated to obtain the high-frequency driving area of ​​the scale surface. If the initial position of an initial debris is not in the high-frequency driving area of ​​the scale surface, but the position of the initial debris is in the high-frequency driving area of ​​the scale surface, the debris is recorded as a first-class dangerous debris, and the initial protrusion height of the first-class dangerous debris is subtracted from the protrusion height to obtain the embedding depth of the first-class dangerous debris, so as to count the number of first-class dangerous debris corresponding to each historical vehicle of each type and the embedding depth of each first-class dangerous debris; each new debris dropped in the high-frequency driving area of ​​the scale surface is recorded as a second-class dangerous debris, and the number of second-class dangerous debris corresponding to each historical vehicle of each type is counted; Using the number of Class I hazardous debris, the embedding depth of each Class I hazardous debris, and the number of Class II hazardous debris corresponding to each historical vehicle of each type, calculate the damage level of the scratch damage type on the scale surface caused by each type of vehicle; The flatness of the weighing surface is obtained from the images of the weighing surface before and after each historical vehicle of each type, and the curvature of the weighing surface of each historical vehicle of each type is calculated; and the damage level of the bending damage type and the connector damage type of each type of vehicle to the weighing surface is calculated by using the detection values ​​of each sensor before and after each historical vehicle of each type arrives in front of the weighing surface and leaves the weighing surface.

5. The AI-based weighbridge management system according to claim 4 is characterized in that: The calculation process of the damage level of the scratch damage type caused by each type of vehicle to the scale surface is as follows: The number of Class I hazardous debris and the number of Class II hazardous debris corresponding to each historical vehicle in each type are recorded as and , the embedding depth of each type of dangerous debris corresponding to each historical vehicle in each type is recorded as , where z represents the type number of the vehicle, x represents the number of the historical vehicle, and n represents the number of a type of dangerous debris. z, x and n are all positive integers; Using the analytical formula: , get the damage level of the scratch damage type of the z-th type of vehicle on the scale surface , where r represents the damage level, r is a positive integer, R is the maximum damage level, R is a positive integer, r≤R, and They represent the initial debris quantity and new debris quantity of the xth historical vehicle in the zth type, d represents the thickness of the scale surface, e is a natural constant, X represents the number of historical vehicles, N1 represents the total quantity of dangerous debris of a type, , , , They are respectively the upper limit characteristic value of the first damage level, the upper limit characteristic value of the r-1th damage level, the upper limit characteristic value of the rth damage level, and the upper limit characteristic value of the R-1th damage level.

6. The AI-based weighbridge management system according to claim 5 is characterized in that: The specific process of setting the ramp access table is as follows: Obtain driving videos from monitoring data of each historical vehicle passing through the weighbridge, intercept uphill videos and downhill videos from the driving videos, summarize uphill videos and downhill videos of each historical vehicle of each type, obtain uphill surface slope, the number of new debris on the uphill surface, and vehicle stability from the uphill videos of each historical vehicle of each type, thereby calculating the traffic characteristic values ​​of each type of vehicle driving on each uphill surface slope, and similarly calculate the traffic characteristic values ​​of each type of vehicle on each downhill surface slope; The traffic characteristic values ​​of each type of vehicle traveling on each uphill slope and each downhill slope are summarized to obtain a ramp traffic table.

7. The AI-based weighbridge management system according to claim 6 is characterized in that: The specific process of calculating the traffic characteristic values ​​of various types of vehicles traveling on various uphill slopes is as follows: Using the uphill slope of each historical vehicle of each type, the number of new debris on the uphill surface and the vehicle stability, the number of new debris on each uphill surface and the stability of each vehicle when each type of vehicle passes through each uphill slope are counted, and the number of new debris on each uphill surface and the stability of each vehicle when each type of vehicle passes through each uphill slope are clustered to obtain the number of new debris on the uphill surface and the stability of the vehicle when each type of vehicle passes through each uphill slope, and normalize them. The processed values ​​are marked as and , where q represents the number of each uphill slope, and q is a positive integer; Using the calculation formula: Get the traffic characteristic value of the z-th type of vehicle traveling on the q-th uphill slope .

8. The AI-based weighbridge management system according to claim 1 is characterized in that: The specific process of setting the speed and slope data of the vehicle passing the weighbridge is as follows: Identify the vehicle license plate number from the vehicle image, obtain the vehicle's travel record from the travel database, and if the vehicle has a travel record, obtain the vehicle's weight at each historical travel from the vehicle's travel record, cluster the vehicle's weight at each historical travel, and use the result as the vehicle's reference weight; The size of the vehicle is obtained from the image of the vehicle, and the size and reference weight of the vehicle are compared with the size interval and weight interval corresponding to each type in the traffic database. If the size and reference weight of the vehicle are respectively within the size interval and weight interval corresponding to a certain type, the type is used as the marking type, and the damage level of the scratch damage type, the damage level of the bending damage type and the damage level of the connector damage type of each marking type are obtained from the damage table of the weighbridge. Then, the damage severity value of each marking type is calculated, and the marking type with the smallest damage severity value is selected as the type of the vehicle. The passing speed interval of the corresponding type of the vehicle is obtained, and any passing speed in the passing speed interval of the corresponding type of the vehicle can be used as the passing speed of the vehicle passing the weighbridge; Obtain the traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill slope and each downhill slope from the ramp traffic table, and simultaneously obtain the current slope of the uphill surface of the scale, and analyze the necessity of replacing the uphill surface slope of the scale. If the necessity of replacing the uphill surface slope of the scale is insufficient, use the current slope of the uphill surface of the scale as the uphill surface slope in the ramp slope data; if the necessity of replacing the uphill surface slope of the scale is sufficient, select the uphill surface slope with the largest traffic characteristic value among the corresponding type of vehicles as the uphill surface slope in the ramp slope data, and analyze the downhill surface slope in the ramp slope data in the same way.

9. The AI-based weighbridge management system according to claim 8, characterized in that: The specific process of calculating the damage severity value of each mark type is as follows: The damage level of the scratch damage type, the damage level of the bending damage type and the damage level of the connector damage type of each marked type are recorded as , and , using the calculation formula: , get the damage severity value of the pth marker type ,in, , , They are the important weight of scratch damage type, the important weight of bending damage type, and the important weight of connector damage type; , , ; In the formula, , , They represent the mean damage level of the scratch damage type, the mean damage level of the bending damage type, and the mean damage level of the connector damage type for each mark type. , , They are respectively the maximum difference between the damage levels of the scratch damage types of each mark type, the maximum difference between the damage levels of the bending damage types, and the maximum difference between the damage levels of the connector damage types.

10. The AI-based weighbridge management system according to claim 8, characterized in that: The necessity of changing the slope of the slope on the scale is analyzed, and the specific analysis process is as follows: Using the traffic characteristic values ​​of the corresponding type of vehicle traveling on each uphill slope, calculate the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill slopes, and extract the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale. If the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill slopes is less than or equal to the set difference rate threshold or the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale is greater than or equal to the set traffic characteristic value threshold, it indicates that there is insufficient necessity to change the slope of the uphill surface on the scale. If the difference rate of the traffic characteristic values ​​of the corresponding type of vehicle traveling on different uphill slopes is greater than the set difference rate threshold and the traffic characteristic value of the corresponding type of vehicle on the current slope of the uphill surface on the scale is less than the set traffic characteristic value threshold, it indicates that there is sufficient necessity to change the slope of the uphill surface on the scale.

Images