A method for identifying vehicle axles based on shaft group type dynamic truck scale
By performing least-squares linear fitting on the weighing sensor data of the axle-group dynamic truck scale, the number of axles of the vehicle can be identified, which solves the problems of complexity and high cost of existing weighing systems and achieves simplified and accurate axle count determination.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAANXI SIWEI WEIGHING APP LIMITED
- Filing Date
- 2022-11-22
- Publication Date
- 2026-07-03
AI Technical Summary
Existing vehicle weighing systems are complex and costly, especially since axle identifiers are prone to damage, increasing maintenance costs, and making it difficult to accurately determine the number and type of axles on a vehicle.
The central control processor processes the weighing sensor data of the axle-type dynamic truck scale, and uses the least squares linear fitting technique to identify peaks of suspected axle loads to determine whether they are real axles, thus simplifying the axle count counting process.
It enables accurate identification of the number of axles in a vehicle while it is in motion, simplifies the structure of the weighing system, reduces costs, and avoids dependence on axle identifiers.
Smart Images
Figure CN115758177B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle weighing technology, and more particularly to a method for identifying the number of axles during vehicle weighing. Background Technology
[0002] Vehicle weighing methods are categorized into axle weighing, axle group weighing, and whole-vehicle weighing. Whole-vehicle weighing is generally the most accurate, but it requires the vehicle to be stationary. Axle weighing and axle group weighing, on the other hand, enable dynamic weighing, allowing weighing to occur while the vehicle is in motion, thus improving efficiency. Furthermore, axle group weighing allows individual axles to travel a distance on the weighing platform, resulting in longer data collection times and the ability to weigh the entire axle assembly, leading to more accurate results compared to axle weighing alone. When determining if a vehicle is overloaded, not only vehicle weight data but also vehicle type and the number of axles are required. Traditional weighing systems typically use axle identifiers to count axles; once a wheel passes the identifier, it generates a signal, and an axle is counted. However, adding axle identifiers to the weighing system increases overall system cost, and these identifiers are susceptible to damage due to constant tire pressure, increasing maintenance costs. Summary of the Invention
[0003] To address the problems of complexity and high cost in existing weighing systems, the technical solution adopted in this invention is as follows:
[0004] A method for identifying vehicle axles based on axle group dynamic truck scales is described. When a vehicle passes over the weighing platform, the data collected by the sensors at the upper end of the weighing platform is processed by the central control processor. The method is characterized by: identifying the peak of suspected axle load, performing least squares linear fitting on the data between two adjacent peaks, and comparing the original data of the suspected axle load peak with the fitted line to determine whether the peak is an axle.
[0005] Furthermore, a method for identifying peaks in suspected axle load patterns.
[0006] S1. Compare the current value collected by the weighing sensor within the filtering time period with the maximum value within the filtering time period. If the current value is greater than the maximum value within the filtering time period, define the start time of the next filtering time with the current value and continue to search for the peak until the current value is less than the maximum value within the filtering time period.
[0007] Based on S1, if the difference between the maximum and minimum values is greater than the minimum axle load, then the maximum value is defined as the peak of the suspected axle load.
[0008] Based on S1, if the difference between the maximum and minimum values is less than the minimum axis weight, continue to search for the maximum value until a current value larger than the previous maximum value is found. Then, take the current value as the starting point of the filter to search for the maximum value until the current value is less than the maximum value in the filter period, and the difference between the maximum and minimum values is greater than the minimum axis weight. Then, define the maximum value as the peak of the suspected axis weight.
[0009] The initial minimum value is 0;
[0010] S2. After finding the maximum value of the suspected axle load peak, find the minimum value of the next peak reference. Take the maximum value of the suspected axle load peak as the starting time of the filtering time. Compare the current value collected by the weighing sensor during the filtering time period with the minimum value during the filtering time period. If the current value is less than the minimum value during the filtering time period, replace the minimum value with the current value. Define the starting time of the filtering time with the current value and continue to find the minimum value until the current value is greater than the minimum value during the filtering time period.
[0011] Based on S2, if the difference between the current value and the minimum value within the filtering time period is greater than the minimum axis weight, then the minimum value is defined as the minimum value to be referenced when finding the next suspected axis weight peak.
[0012] Based on S2, if the difference between the current value and the minimum value is less than the minimum axis weight, then continue searching for a value whose difference between the current value and the minimum value is greater than the minimum axis weight. If the current value is less than the minimum value during the search process, it is replaced. Continue searching for a value whose difference between the current value and the replaced minimum value is greater than the minimum axis weight, until a value is found whose difference between the current value and the minimum value previously replaced is greater than the minimum axis weight. Then, the replaced minimum value is used as the reference minimum value for the next suspected axis weight peak.
[0013] The peaks of each suspected axle load are identified sequentially using the above method until the current value returns to zero.
[0014] Furthermore, the filtering period is within 0.125 seconds, and the minimum axle load is 200kg-300kg.
[0015] Furthermore, for the data between each two adjacent suspected axle weight peaks, a least squares linear fit is performed to calculate the values of k and b, resulting in the linear formula: y = (t - ts)k + b, where y is the weight value at time t corresponding to the fitted line, t is time, ts is the time corresponding to the previous suspected axle weight peak, k is the slope of the line, and b is the initial weight of the fitted line; tn is defined as the time corresponding to the next suspected axle weight peak, and f(t) is the sampled weight value at time t.
[0016] Based on the conditions:
[0017] T1: The b-value of the fitted line, the maximum value (max) and minimum value (min) of the original data within the time interval ts ≤ t < tn, should have min. <b<max。
[0018] T2: The difference between the original data and the corresponding point of the fitted line at the second maximum value is d = (f(tn) - (tn - ts)k + b), where |f(t) - (t - ts)k + b| is the maximum value in the time interval ts ≤ t < tn.
[0019] Determine whether the peak of the suspected axle load is an axle load;
[0020] SⅠ. Fit the peak of the first suspected axis weight to the peak of the second suspected axis weight. If the fitted line should meet T1, the peak of the first suspected axis weight is considered to be the first axis. If the fitted line should not meet T1, then the peak of the second suspected axis weight is fitted to the peak of the third suspected axis weight to continue the judgment, and so on until the fitted line of the peak of the nth suspected axis weight and the peak of the (n+1)th suspected axis weight meets T1. Then the peak of the nth suspected axis weight is determined to be the first axis.
[0021] SⅡ. After determining the first axis, fit the peak of the nth suspected axis weight corresponding to the first axis with the peak of the (n+1)th suspected axis weight;
[0022] Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axis weight with the peak of the (n+2)th suspected axis weight. If it meets T1, then determine the peak of the (n+1)th suspected axis weight as the second axis.
[0023] Based on SⅡ, if the slope k is negative and does not meet T2, then the (n+1)th suspected axis peak is considered interference and is filtered out; the nth suspected axis peak is fitted with the (n+2)th suspected axis peak. If it meets T2, then the (n+2)th suspected axis peak is fitted with the (n+3)th suspected axis peak. If it meets T1, then the (n+3)th suspected axis peak is determined to be the second axis.
[0024] Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axis with the peak of the (n+2)th suspected axis. If it does not meet T1, then the peak of the (n+1)th suspected axis is considered interference and is filtered out. Fit the peak of the nth suspected axis with the peak of the (n+2)th suspected axis. If it meets T2, then fit the peak of the (n+2)th suspected axis with the peak of the (n+3)th suspected axis. If it meets T1, then determine the peak of the (n+3)th suspected axis as the second axis.
[0025] If the slope k is positive based on SⅡ, then the n+1 suspected axis weight peaks are determined as the second axis;
[0026] If the peak of the last suspected axis is the one corresponding to SⅡ, then the peak of the last suspected axis is fitted to the peak of the suspected axis corresponding to the previous axis. If the slope k is positive, the peak of the last suspected axis is considered to be the axis; if the slope k is negative and meets T2, the peak of the last suspected axis is considered to be the axis; if the slope k is negative and does not meet T2, the peak of the last suspected axis is considered to be interference and is filtered out.
[0027] Based on the above conditions, determine whether the peak of each suspected axle load is an axle, and determine the number of axles.
[0028] This technical solution uses an axle-group weighing platform to weigh vehicles and determine the number of axles from the weighing data. This eliminates the need for a separate axle identifier in the weighing system to count the number of axles, simplifying the weighing system and saving costs. Attached Figure Description
[0029] Figure 1 This shows the changes in the weighing sensor data at the upper end of the weighing platform when the six-axle vehicle passes over it.
[0030] Figure 2 A schematic diagram for finding suspected axle load peaks and minimum values;
[0031] Figure 3 This is a partial schematic diagram illustrating the relationship between changes in weighing sensor data and the fitted linear line;
[0032] Figure 4 This is a partial schematic diagram illustrating the relationship between changes in weighing sensor data and the fitted linear line;
[0033] Figure 5 This is a partial schematic diagram illustrating the relationship between changes in weighing sensor data and the fitted linear line;
[0034] Figure 6 This is a partial schematic diagram illustrating the relationship between changes in weighing sensor data and the fitted linear line. Detailed Implementation
[0035] The method of this invention is to process and calculate the number of axles of a vehicle by means of data collected by the weighing sensor at the upper end of the weighing platform when the vehicle passes through the weighing platform of the axle group dynamic truck scale system. Since the weighing sensor will produce fluctuations when the vehicle passes through the weighing platform, the upper end will form a rapid rise and slow fall fluctuation. In addition to the fluctuation caused by the change in the number of axles on the weighing platform when the vehicle is moving, the vibration generated by the vehicle during driving will also affect the fluctuation of the data collected by the weighing sensor. If the influence of vibration can be eliminated, then the remaining fluctuation will only be the fluctuation caused by the change in the number of axles. Thus, the number of axles can be determined by these changes. This invention is based on this theory and uses the central control processor to process the data to eliminate interference and obtain the number of axles. Specifically, it includes the following three steps: finding the peak of suspected axle load, performing least squares linear fitting on the data between two adjacent peaks in turn, and comparing the original data of the suspected axle load peak with the fitted line to determine whether the peak is an axle.
[0036] The method of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:
[0037] like Figure 1 The image shows the changes in the load cell data at the upper end of the weighing platform as a six-axle vehicle passes over it. The following explanation uses the upper end of the weighing platform as an example to illustrate this method:
[0038] The method for identifying peaks suspected to be due to axle load is as follows:
[0039] S1. Compare the current value collected by the weighing sensor within the filtering time period with the maximum value within the filtering time period. If the current value is greater than the maximum value within the filtering time period, define the start time of the next filtering time with the current value and continue to search for the peak until the current value is less than the maximum value within the filtering time period.
[0040] like Figure 2 As shown, starting from the initial moment, the weighing sensor data rises rapidly from 0. When the sensor at the upper end of the weighing device receives a signal, the filtering time begins at time t1. During this filtering period (t2-t1), a maximum value x1 is generated. From the diagram, x2 appears to be the maximum value. In reality, due to vibration, x2 represents the maximum value recorded during the t1-t2 period. At the final cutoff time t2 of the filtering period, a current value x2 is generated. This current value is compared with the maximum value. If the current value x2 is greater than x1, then the time point of the current value x2 is used as the starting point of the new filtering period, and the search for peaks continues using the same method. For example, if x3 is the maximum value during the filtering period from t2 to t3, then the search for peaks continues starting from t3 as the next filtering time point. During the period from t3 to t4, the maximum value is xm. At this point, the current value x4 is less than this maximum value xm, thus a peak is found.
[0041] At this point, it is necessary to determine the difference between the peak xm and the minimum value x1. If there are no vehicles on the weighing platform at the beginning and the minimum value is 0, it is determined that the difference between xm and 0 exceeds the minimum axle load, which is 300kg. If the difference between xm and 0 exceeds the minimum axle load, then xm is recorded as a suspected axle load peak.
[0042] If the difference between xm and 0 does not exceed the minimum axle load, it is considered an interference, and the search for a suspected axle load peak continues. This process first requires finding a value greater than xm, such as... Figure 2 If the value x5 at time t5 is greater than xm, then take time t5 as the starting point of filtering and continue to search for peaks according to the above steps until xmn is found, and the difference between xmn and 0 is greater than the minimum axle weight, then it is recorded as a suspected axle weight peak.
[0043] After finding the suspected axle load peak, the first step in finding the next suspected axle load peak is to determine the minimum reference value for the next suspected axle load peak. Since there may be axles on the weighing platform that have not yet been removed from the weighing platform, the minimum reference value will not be 0.
[0044] The method for finding the minimum value is as follows:
[0045] After finding the maximum value of the peak of the suspected axle load, find the minimum value of the next peak as a reference. The maximum value of the found suspected axle load peak is used as the start time of the filtering time. During the filtering time period, the current value collected by the weighing sensor is compared with the minimum value of the filtering time period. If the current value is less than the minimum value of the filtering time period, the current value is used to replace the minimum value, and the start time of the filtering time is defined with the current value. Continue to find the minimum value until the current value is greater than the minimum value of the filtering time period.
[0046] For specific references Figure 2 As shown:
[0047] Taking the time point tmn as the starting time of the filtering time, within the filtering time period, refer to Figure 2 If the current value x6 corresponding to the end time t6 within the filtering time period is less than the minimum value within the filtering time period, then the search for the minimum value continues from the time point t6 as the starting time point until the current time point x7 is greater than the minimum value xi within the filtering time period t6-t7, and the difference between the current value x7 and this minimum value xi is greater than the minimum axle load, then the minimum value xi is defined as the minimum value referenced by the next suspected axle load peak;
[0048] If the difference between x7 and xi is less than the minimum axis weight, then continue searching for a current value x8 such that the minimum difference between x8 and xi is greater than the minimum axis weight. This minimum value xi is defined as the minimum reference value for the next potential axis weight peak. If a value smaller than xi is found during the search for the current value, it is replaced. For example, if xin is less than xi, the minimum value is replaced by xin. Then, continue searching for a value whose difference between the current value and xin is greater than the minimum axis weight. For example, if the difference between x9 and xin is greater than the minimum axis weight, this minimum value xin is defined as the minimum reference value for the next peak. Similarly, if, during the search for a current value whose difference between xin and x8 is greater than the minimum axis weight, and a value smaller than xin is found before a corresponding current value is found, then xin is replaced. This continues until a value is found whose difference between the current value and the most recently replaced minimum value is greater than the minimum axis weight. This most recently replaced minimum value is then used as the minimum reference value for the next peak.
[0049] Using the method described above for finding peaks and minimum values, find the peaks of each suspected axle load in turn until the current value returns to zero.
[0050] In the above process, according to the speed limit requirement of the weighing system of 36km / h, which corresponds to 10m / s, and the minimum distance between the two wheelbases is 1.25 meters, it can be determined that the filtering time is set within 0.125 seconds; the minimum axle load is generally greater than 300kg according to statistics of the trolley axle load, and this is taken as 200kg-300kg to exclude the interference of people walking on the weighing platform.
[0051] Once a peak with suspected axle weight is found, a straight line is fitted between two adjacent peaks. Specifically, the least squares method is used for straight line fitting, and the values of k and b are calculated to obtain the straight line formula: y = (t - ts)k + b, where y is the weight value at time t corresponding to the fitted line, t is time, ts is the time corresponding to the previous peak with suspected axle weight, k is the slope of the straight line, and b is the initial weight of the fitted line; tn is defined as the time corresponding to the next peak with suspected axle weight, and f(t) is the sampled weight value at time t.
[0052] Determine whether a peak in a suspected axle load is an axle load based on the following conditions:
[0053] T1: The b-value of the fitted line, the maximum value (max) and minimum value (min) of the original data within the time interval ts ≤ t < tn, should have min. <b<max。
[0054] like Figure 3 In the figure, max1 and max2 are two suspected peak values, min is the minimum value of this data, and b is the fitted line value at time t1, which is greater than min and less than max2.
[0055] T2: The difference between the original data and the corresponding point of the fitted line at the second maximum value is d = (f(tn) - (tn - ts)k + b), where d is the maximum value in the time interval ts ≤ t < tn.
[0056] exist Figure 3 In the figure, d represents the difference between the original data and the fitted line at time t2. As can be seen from the figure, the difference d between the sampled data and the fitted line is the largest at time t2.
[0057] When making a judgment:
[0058] SⅠ. Fit the peak of the first suspected axis weight to the peak of the second suspected axis weight. If the fitted line should meet T1, the peak of the first suspected axis weight is considered to be the first axis. If the fitted line should not meet T1, then the peak of the second suspected axis weight is fitted to the peak of the third suspected axis weight to continue the judgment, and so on until the fitted line of the peak of the nth suspected axis weight and the peak of the (n+1)th suspected axis weight meets T1. Then the peak of the nth suspected axis weight is determined to be the first axis.
[0059] Since there is no contrast peak on the front side of the first axis, the judgment can only be made based on condition T1.
[0060] exist Figure 3 In the equation, the fitted line between the suspected peak max1 and the subsequent max2 satisfies T1 at time t1, with max1 being the first axis.
[0061] SⅡ. After determining the first axis, fit the peak of the nth suspected axis weight corresponding to the first axis with the peak of the (n+1)th suspected axis weight;
[0062] Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axis weight with the peak of the (n+2)th suspected axis weight. If it meets T1, then determine the peak of the (n+1)th suspected axis weight as the second axis.
[0063] Figure 4 In the figure, for point max2, the slope k of the fitted line with point max1 is negative, and the difference between the sampled data and the fitted line at time t2 is the largest as shown in d, which satisfies condition T2; the fitted line 2 of max2 and max3 satisfies condition T1 at time t2. It can be seen from the figure that the peak of the suspected axis weight of max2 is the second axis.
[0064] Based on SⅡ, if the slope k is negative and does not meet T2, then the (n+1)th suspected axis peak is considered interference and is filtered out; the nth suspected axis peak is fitted with the (n+2)th suspected axis peak. If it meets T2, then the (n+2)th suspected axis peak is fitted with the (n+3)th suspected axis peak. If it meets T1, then the (n+3)th suspected axis peak is determined to be the second axis.
[0065] Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axis with the peak of the (n+2)th suspected axis. If it does not meet T1, then the peak of the (n+1)th suspected axis is considered interference and is filtered out. Fit the peak of the nth suspected axis with the peak of the (n+2)th suspected axis. If it meets T2, then fit the peak of the (n+2)th suspected axis with the peak of the (n+3)th suspected axis. If it meets T1, then determine the peak of the (n+3)th suspected axis as the second axis.
[0066] Figure 5 In the equation, the slope of the fitted line 1 between max2 and max1 is negative and satisfies condition T2. The fitted line 2 between max2 and max3 has a value less than min2 at time t2. The peak of the suspected axle load in max2 is an interference.
[0067] If the slope k is positive based on SⅡ, then the n+1 suspected axis weight peaks are determined as the second axis;
[0068] Since there are potential axis weight peaks on both sides of the central peak, it is necessary to fit the data to the peaks on both sides separately, and then judge the results using T1 and T2 respectively. Because satisfying T2 may not satisfy T1, this situation needs to be filtered out.
[0069] If the peak of the last suspected axis is the one corresponding to SⅡ, then the peak of the last suspected axis is fitted to the peak of the suspected axis corresponding to the previous axis. If the slope k is positive, the peak of the last suspected axis is considered to be the axis; if the slope k is negative and meets T2, the peak of the last suspected axis is considered to be the axis; if the slope k is negative and does not meet T2, the peak of the last suspected axis is considered to be interference and is filtered out.
[0070] Since there are no subsequent peaks to fit the final suspected axle load peak, the judgment can only be made by fitting it with the preceding peaks and based on the previous conditions.
[0071] Figure 6 In the figure, max2 is the last suspected peak of the axle load, and the fitted line with max1 has a negative slope. The data in the figure shows that the weight difference between the sampled data and the fitted line is not the largest at time t1, and max2 is the interference.
[0072] Based on the above conditions, determine whether the peak of each suspected axle load is an axle, and determine the number of axles.
Claims
1. A method for identifying vehicle axles based on axle-group dynamic truck scales, wherein when a vehicle passes over the weighing platform, a central control processor processes data collected by sensors at the upper end of the weighing platform, characterized in that: Find the peaks that are suspected to be axle loads, and perform least squares linear fitting on the data between two adjacent peaks in turn. Compare the original data of the peaks that are suspected to be axle loads with the fitted lines to determine whether the peak is an axle load. For the data between each two adjacent suspected axle weight peaks, a least squares straight line is fitted, and the values of k and b are calculated to obtain the straight line formula: y = (t - ts)k + b, where y is the weight value at time t corresponding to the fitted line, t is time, ts is the time corresponding to the previous suspected axle weight peak, k is the slope of the straight line, and b is the initial weight of the fitted line; tn is defined as the time corresponding to the next suspected axle weight peak, and f(t) is the sampled weight value at time t. Based on the conditions: T1: The b-value of the fitted line, the maximum value (max) and minimum value (min) of the original data within the time interval ts ≤ t < tn, should have min. <b<max; T2: The difference between the original data and the corresponding point of the fitted line at the second maximum value is d = (f(tn) - (tn - ts)k + b), where |f(t) - (t - ts)k + b| is the maximum value in the time interval ts ≤ t < tn. Determine whether the peak of the suspected axle load is an axle load; SⅠ. Fit the peak of the first suspected axis weight to the peak of the second suspected axis weight. If the fitted line should meet T1, the peak of the first suspected axis weight is considered to be the first axis. If the fitted line should not meet T1, then the peak of the second suspected axis weight is fitted to the peak of the third suspected axis weight to continue the judgment, and so on until the fitted line of the peak of the nth suspected axis weight and the peak of the (n+1)th suspected axis weight meets T1. Then the peak of the nth suspected axis weight is determined to be the first axis. SⅡ. After determining the first axis, fit the peak of the nth suspected axis weight corresponding to the first axis with the peak of the (n+1)th suspected axis weight; Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axis weight with the peak of the (n+2)th suspected axis weight. If it meets T1, then determine the peak of the (n+1)th suspected axis weight as the second axis. Based on SⅡ, if the slope k is negative and does not meet T2, then the (n+1)th suspected axis peak is considered interference and is filtered out; the nth suspected axis peak is fitted with the (n+2)th suspected axis peak. If it meets T2, then the (n+2)th suspected axis peak is fitted with the (n+3)th suspected axis peak. If it meets T1, then the (n+3)th suspected axis peak is determined to be the second axis. Based on SⅡ, if the slope k is negative and meets T2, then fit the peak of the (n+1)th suspected axle weight with the peak of the (n+2)th suspected axle weight. If it does not meet T1, then the peak of the (n+1)th suspected axle weight is considered to be interference and is filtered out. Fit the peak of the nth suspected axis weight to the peak of the (n+2)th suspected axis weight. If T2 is satisfied, then fit the peak of the (n+2)th suspected axis weight to the peak of the (n+3)th suspected axis weight. If T1 is satisfied, then determine the peak of the (n+3)th suspected axis weight as the second axis. If the slope k is positive based on SⅡ, then the n+1 suspected axis weight peaks are determined as the second axis; Based on SⅡ, if the (n+1)th peak is the last suspected axis peak, then based on the fitting of the last suspected axis peak with the peak of the suspected axis corresponding to the previous axis, if the slope k is positive, then the last suspected axis peak is considered an axis; if the slope k is negative and meets T2, then the last suspected axis peak is considered an axis; if the slope k is negative and does not meet T2, then the last suspected axis peak is considered an axis. T2 considers the last suspected axle load peak to be interference and filters it out. Based on the above conditions, determine whether the peak of each suspected axle load is an axle, and determine the number of axles.
2. The method for identifying vehicle axles based on axle group dynamic truck scale according to claim 1, characterized in that: Method for identifying peaks in suspected axle load S1. Compare the current value collected by the weighing sensor within the filtering time period with the maximum value within the filtering time period. If the current value is greater than the maximum value within the filtering time period, define the start time of the next filtering time with the current value and continue to search for the peak until the current value is less than the maximum value within the filtering time period. Based on S1, if the difference between the maximum and minimum values is greater than the minimum axle load, then the maximum value is defined as the peak of the suspected axle load. Based on S1, if the difference between the maximum and minimum values is less than the minimum axis weight, continue to search for the maximum value until a current value larger than the previous maximum value is found. Then, take the current value as the starting point of the filter to search for the maximum value until the current value is less than the maximum value in the filter period, and the difference between the maximum and minimum values is greater than the minimum axis weight. Then, define the maximum value as the peak of the suspected axis weight. The initial minimum value is 0; S2. After finding the maximum value of the suspected axle load peak, find the minimum value of the next peak reference. Take the maximum value of the suspected axle load peak as the starting time of the filtering time. Compare the current value collected by the weighing sensor during the filtering time period with the minimum value during the filtering time period. If the current value is less than the minimum value during the filtering time period, replace the minimum value with the current value. Define the starting time of the filtering time with the current value and continue to find the minimum value until the current value is greater than the minimum value during the filtering time period. Based on S2, if the difference between the current value and the minimum value within the filtering time period is greater than the minimum axis weight, then the minimum value is defined as the minimum value to be referenced when finding the next suspected axis weight peak. Based on S2, if the difference between the current value and the minimum value is less than the minimum axis weight, then continue to search for a value where the difference between the current value and the minimum value is greater than the minimum axis weight. If the current value is less than the minimum value during the search process, then replace it. Continue to search for a value where the difference between the current value and the replaced minimum value is greater than the minimum axis weight, until a value is found where the difference between the current value and the minimum value that was previously replaced is greater than the minimum axis weight. Then, the replaced minimum value is used as the reference minimum value for the next suspected axis weight peak. The peaks of each suspected axle load are identified sequentially using the above method until the current value returns to zero.
3. The method for identifying vehicle axles based on axle group dynamic truck scale according to claim 2, characterized in that: The filtering period is within 0.125 seconds, and the minimum axle load is 200kg-300kg.