JP2024079045A - Weight measurement device, measurement accuracy determination method and measurement accuracy determination program - Google Patents

Abstract

To perform determination on whether the measurement accuracy of a sensors used to measure the weight of a traveling vehicle is appropriate regardless of the presence/absence of travel of a special vehicle.SOLUTION: A plurality of sensors are arranged side by side on a road surface of a measurement section in the travel direction of a vehicle. Each sensor is connected to a sensor connection part and a measurement signal of a load measured by each sensor is input. A ratio calculation unit calculates a ratio of the total weight obtained by totaling the weight according to the load by a tire of the vehicle that has passed through the sensor and the average weight which is the average of the total weight of each sensor as the weight ratio for each sensor for the vehicle that travels in the measurement section. A determination unit determines whether or not the measurement accuracy of the sensor is appropriate on the basis of the weight ratio calculated for the sensor.SELECTED DRAWING: Figure 2

Description

This invention relates to a technology for determining whether the measurement accuracy of a sensor used to measure the weight of a moving vehicle is appropriate.

Conventionally, there has been a device that measures the weight of a vehicle traveling on a road. This device processes the measurement signals of multiple (two or more) axle load sensors that are lined up in the direction of travel and buried in the road to measure the weight of the traveling vehicle (see Patent Document 1, etc.). As is well known, an axle load sensor is a sensor that is used for each axle of a vehicle to measure the vertical force (axle load) when the wheels attached to that axle pass by.

The axle load is calculated by processing the measurement signals of multiple axle load sensors buried and aligned in the direction of vehicle travel. For example, for each axle, the average value of the axle loads measured by multiple axle load sensors is calculated as the axle load for that axle. The weight of the vehicle is the sum (total) of the axle loads calculated for each axle of the vehicle.

Patent Document 1 also describes a configuration for correcting sensitivity (updating a correction coefficient) for each axle load sensor according to the axle load measurement accuracy without interrupting the measurement of the vehicle weight. Specifically, Patent Document 1 describes a configuration for updating a correction coefficient for correcting sensitivity so that the vehicle weight calculated using measurement data when a predetermined specific vehicle (a vehicle whose vehicle data such as the number of axles and total weight are known) passes by each axle load sensor becomes a true value.

JP 2010-261825 A

However, the specific vehicle in Patent Document 1 is,
(1) The vehicle's total weight (the sum of the axle loads) does not change depending on the usage conditions.
(2) There are no other vehicles with similar numbers of axles, wheelbases, axle loads, etc.
The vehicle must satisfy the following conditions.

Therefore, vehicles such as trucks and trailers that carry cargo do not satisfy the above condition (1). Furthermore, general passenger cars (private and company) do not satisfy the above condition (2). For this reason, specific vehicles have had to be special vehicles, not general vehicles. In Patent Document 1, a self-propelled crane is classified as a specific vehicle.

If the specific vehicle is a special vehicle, the period between the current run of the specific vehicle and the next run of the specific vehicle may be long. With the technology in Patent Document 1, the period for determining whether the measurement accuracy of the axle load sensor has decreased is the period between the current run of the specific vehicle and the next run of the specific vehicle. For this reason, with the technology in Patent Document 1, there may be a situation where it is not possible to determine whether the measurement accuracy of the axle load sensor has decreased over a long period of time.

The objective of this invention is to provide a technology that can determine whether the measurement accuracy of a sensor used to measure the weight of a moving vehicle is appropriate, regardless of whether a special vehicle is being driven.

To achieve the above objective, the weight measuring device of this invention is configured as follows:

Multiple sensors are arranged on the road surface in the measurement section, aligned in the direction of vehicle travel. Each sensor measures the load (pressure) applied by the tires of passing vehicles. Each sensor is connected to the sensor connection section, and the measurement signal of the load measured by each sensor is input.

The sensor may be one that is pressed by tires on both the left and right sides of the vehicle axle, or one that is pressed by a tire on only one side of the vehicle axle.

The ratio calculation unit calculates, for each sensor, the ratio between the total weight, which is the sum of the weights corresponding to the tire loads of the vehicles that passed that sensor, and the average weight, which is the average of the total weights of each sensor, as a weight ratio for the vehicles that traveled through the measurement section.

A moving vehicle vibrates. The load that the tires apply to the road surface is affected by the vehicle vibrations and fluctuates to a certain extent. In other words, the weight calculated from the magnitude of the tire load measured by the sensor is also a value affected by the vehicle vibrations. Therefore, the total weight is a value that is more affected by the vehicle vibrations than the average weight. Also, the more sensors are arranged in a row in the direction the vehicle is traveling, the less the average weight is affected by the vehicle vibrations.

The judgment unit judges whether the measurement accuracy of the sensor is appropriate based on the weight ratio calculated for that sensor.

For example, the determination unit determines whether the measurement accuracy of a sensor is appropriate based on the shape of the frequency distribution of the weight ratio calculated for that sensor. The frequency distribution of the weight ratio can be regarded as the distribution of load fluctuations due to vehicle vibration. Furthermore, the distribution of load fluctuations due to vehicle vibration is a distribution that approximates a normal distribution with a certain degree of variance. Therefore, it is possible to determine whether the measurement accuracy of a sensor is appropriate based on the shape of the frequency distribution of the weight ratio. In other words, it is possible to determine whether the measurement accuracy of a sensor used to measure the weight of a moving vehicle is appropriate, regardless of whether a special vehicle is moving or not.

For example, the determination unit may be configured to determine whether the measurement accuracy of the sensor is appropriate based on the degree of symmetry of the shape of the frequency distribution of the weight ratio based on the frequency peak. In this case, the determination unit may be configured to calculate the degree of symmetry of the shape of the frequency distribution of the weight ratio, and determine whether the measurement accuracy of the sensor is appropriate based on whether the degree calculated here exceeds a predetermined symmetry threshold.

For example, the determination unit may be configured to determine whether the measurement accuracy of the sensor is appropriate based on the magnitude of variance in the frequency distribution of the weight ratio. In this case, for example, the determination unit may be configured to determine whether the measurement accuracy of the sensor is appropriate based on whether the variance in the frequency distribution of the weight ratio is between a predetermined lower threshold and an upper threshold (lower threshold < upper threshold).

In addition, if the sensor is pressed by both tires attached to the axle of a vehicle traveling through the measurement section (i.e., if the sensor is a so-called axle load sensor), the average weight can be obtained as the weight of the vehicle that has traveled through the measurement section.

In addition, if the sensors are so-called wheel load sensors, the sensors can be arranged in two rows in the vehicle width direction. In this case, the sensors in one row can be arranged in a position where they are pressed by one tire attached to the axle of the vehicle traveling in the measurement section, and the sensors in the other row can be arranged in a position where they are pressed by the other tire attached to the axle of the vehicle traveling in the measurement section.

In addition, if the sensor is a so-called wheel load sensor, the sum of the average weight measured by the sensor in one row and the average weight measured by the sensor in the other row can be obtained as the weight of the vehicle that has traveled through the measurement section.

This invention makes it possible to determine whether the measurement accuracy of a sensor used to measure the weight of a moving vehicle is appropriate, regardless of whether a special vehicle is traveling or not.

1 is a schematic diagram showing a vehicle weight measurement system to which an axle load measurement device according to this embodiment is applied. FIG. 2 is a block diagram showing the configuration of the main parts of the axle load measuring device according to this example. FIG. 11 is a diagram illustrating an example of statistical data. 4A to 4C are diagrams showing example histograms of the number of vehicles versus weight ratio. 4 is a flowchart showing a measurement process of the axle load measuring device according to this example. 5 is a flowchart showing a measurement accuracy determination process of the axle load measuring device according to this example. FIG. 13 is a schematic diagram showing a modified vehicle weight measurement system. 8A and 8B are diagrams showing examples of statistical data of deformation theory, in which FIG. 8A shows right-side weight data and FIG. 8B shows left-side weight data.

The following describes an embodiment of the axle load measuring device according to the present invention.

＜1. Application examples＞
Figure 1 is a schematic diagram showing a vehicle weight measurement system to which an axle load measurement device according to this example is applied. The vehicle weight measurement system shown in Figure 1 includes an axle load measurement device 1, three axle load sensors 2 to 4, and two vehicle detection sensors 6 and 7. The axle load sensors 2 to 4 and the vehicle detection sensors 6 and 7 are connected to the axle load measurement device 1. In this example, the axle load measurement device 1 corresponds to the weight measurement device as referred to in this invention. Also, in this example, the axle load sensors 2 to 4 correspond to the sensors as referred to in this invention.

As shown in FIG. 1, vehicle detection sensor 6, axle load sensor 2, axle load sensor 3, axle load sensor 4, and vehicle detection sensor 7 are arranged on the road in this order in the direction of travel of vehicle 100. The measurement section for measuring the axle load of vehicle 100 is the section from vehicle detection sensor 6 to vehicle detection sensor 7. Vehicle detection sensor 6 detects vehicles 100 entering the measurement section. Vehicle detection sensor 7 detects vehicles 100 exiting the measurement section.

The axle load sensors 2-4 are placed on the road surface of the measurement section and measure the load when pressed by the tires of the vehicle 100 traveling on the measurement section. The length of the axle load sensors 2-4 in the vehicle width direction is the length through which both left and right tires (wheels) attached to the axles of the vehicle 100 traveling on this measurement section pass. In other words, when the vehicle 100 travels on the measurement section, the axle load sensors 2-4 are pressed by both left and right tires attached to each axle of the vehicle 100. In other words, the axle load sensors 2-4 output measurement signals to the axle load measurement device 1 that measure the load when pressed by both left and right tires attached to each axle of the vehicle 100 that traveled on the measurement section. The axle load sensors 2-4 may be buried in the road surface or may be partially exposed to the road surface. The axle load sensors 2-4 may be sensors that can measure loads, and are not limited to sensors that use a specific type of element.

The vehicle detection sensors 6 and 7 are, for example, loop coil sensors, and output the change in inductance to the axle load measurement device 1 as a vehicle detection signal (a signal indicating the presence or absence of a vehicle 100).

The vehicle detection sensors 6 and 7 may be optical sensors, radio wave sensors, ultrasonic sensors, etc. that are not buried in the road. Also, the vehicle 100 may be separated based on the detection output of the axle load sensors 2 to 4. In this case, the vehicle detection sensors 6 and 7 may be unnecessary.

Axle load measuring device 1 receives measurement signals from axle load sensor 2, axle load sensor 3, and axle load sensor 4 for each axle of vehicle 100 that has traveled through the measurement section.

In the traveling direction of the vehicle 100, the distance L1 between the axle load sensor 2 and the axle load sensor 3 and the distance L2 between the axle load sensor 3 and the axle load sensor 4 are different lengths. The traveling vehicle 100 vibrates due to the complex interaction of various factors such as road surface unevenness, speed, and tire air pressure. If both the distance L1 between the adjacent axle load sensor 2 and the axle load sensor 3 and the distance L2 between the adjacent axle load sensor 3 and the axle load sensor 4 are close to an integer multiple of the vibration wavelength of the vehicle 100, the measurement error of the axle load may become large. For this reason, in this example, even if one of the distance L1 between the adjacent axle load sensor 2 and the axle load sensor 3 or the distance L2 between the adjacent axle load sensor 3 and the axle load sensor 4 is close to an integer multiple of the vibration wavelength of the vehicle 100, the distance L1 and the distance L2 are made different lengths so that the other does not approximate an integer multiple of the vibration wavelength of the vehicle 100.

The axle load measuring device 1 in this example cumulatively stores a measurement value corresponding to the input measurement signal for each axle load sensor 2 to 4. In this example, this measurement value is described as an axle load calculated using the input measurement signal. However, this measurement value may be any value that corresponds to the input measurement signal. For example, this measurement value may be a digital value of the input measurement signal.

Since the vehicle 100 vibrates while traveling, even if the measurement accuracy of the axle load sensors 2 to 4 is appropriate, there will be variation in the measured axle load values. If the measurement accuracy of the axle load sensors 2 to 4 is appropriate, this variation in the measured axle load values will fall within a certain range (although depending on the magnitude of the vibration of the vehicle 100, it may not fall within the certain range).

The axle load measurement device 1 calculates the total weight for each of the axle load sensors 2 to 4. Specifically, the axle load measurement device 1 calculates the sum of the axle loads of each axle measured by the axle load sensor 2 as the total weight of the axle load sensor 2. The axle load measurement device 1 calculates the sum of the axle loads of each axle measured by the axle load sensor 3 as the total weight of the axle load sensor 3. The axle load measurement device 1 calculates the sum of the axle loads of each axle measured by the axle load sensor 4 as the total weight of the axle load sensor 4.

The axle load measuring device 1 also calculates the average weight as the average of the total weights of the axle load sensors 2 to 4. The axle load measuring device 1 obtains this average weight as the weight of the vehicle 100 that has traveled through the measurement section.

The total weight is the weight of the vehicle 100 without taking into account the effects of vibration of the vehicle 100. The average weight is the weight of the vehicle 100 in which the effects of vibration of the vehicle 100 are reduced compared to the total weight.

In this example, the number of axle load sensors placed in the measurement section is three, but it may be two or more. The more axle load sensors placed in the measurement section, the more the axle load measuring device 1 can obtain the weight (average weight) of the vehicle 100 with reduced effects of vibration of the vehicle 100.

The axle load measuring device 1 calculates a weight ratio, which is the ratio between the total weight of the vehicle 100 measured by each axle load sensor 2 to 4 and the average weight. In this example, the weight ratio is total weight/average weight, but it may also be average weight/total weight.

The axle load measuring device 1 determines whether the measurement accuracy is appropriate for each axle load sensor 2 to 4 based on the shape of a histogram of the weight ratio of each vehicle 100 that traveled through the measurement section during a predetermined period, which is obtained by statistical processing.

As a result, the axle load measuring device 1 can determine whether the measurement accuracy of the axle load sensors 2 to 4 used to measure the weight of the moving vehicle 100 is appropriate, regardless of whether the special vehicle 100 is moving or not.

2. Configuration example
2 is a block diagram showing the configuration of the main parts of the axle load measuring device 1 according to this example. The axle load measuring device 1 according to this example includes a control unit 11, an axle load sensor connection unit 12, a loop coil sensor connection unit 13, a statistical database 14 (statistical DB 14), and an output unit 15.

The control unit 11 controls the operation of each part of the main body of the axle load measurement device 1 according to this example. The control unit 11 also has an axle load calculation section 21, a vehicle weight calculation section 22, a ratio calculation section 23, and a judgment section 24. Although not shown in particular, the control unit 11 has a memory provided with a storage area for storing a symmetry threshold value and the like, which will be described later. The details of the axle load calculation section 21, the vehicle weight calculation section 22, the ratio calculation section 23, and the judgment section 24 of the control unit 11 will be described later.

The measurement signals of the connected axle load sensors 2 to 4 are input to the axle load sensor connection unit 12. The axle load sensor connection unit 12 converts the input measurement signals of the axle load sensors 2 to 4 into digital values and outputs them to the control unit 11.

The vehicle detection sensors 6 and 7 are connected to the loop coil sensor connection section 13. The loop coil sensor connection section 13 detects changes in inductance for each of the vehicle detection sensors 6 and 7, and outputs a vehicle detection signal indicating the presence or absence of a vehicle 100 to the control unit 11.

The statistical DB 14 stores statistical data for each vehicle 100 that has traveled through the measured section, associating the vehicle ID, travel date, travel time, axle load, total weight, average weight, and weight ratio (see Figure 3).

The vehicle ID is a number assigned to the vehicle 100 when the vehicle 100 travels through a measured section. The vehicle ID does not identify the vehicle 100. For example, the vehicle ID assigned to a vehicle 100 that has traveled through a measured section is a number obtained by incrementing (+1) the previously assigned vehicle ID. The upper digits of the vehicle ID may also be the date of travel on the measured section. For example, if the date of travel is November 1, 2022, the vehicle ID may be 20221101******. The last digits of the vehicle ID, *****, are a number that is incremented when the vehicle 100 travels through a measured section.

The travel date is the date on which the vehicle 100 traveled through the measurement section.

The travel time is the time when the vehicle 100 travels through the measured section. The travel time may be, for example, the time when the vehicle 100 enters the measured section, the time when the vehicle 100 exits the measured section, or a time intermediate between the time when the vehicle 100 enters the measured section and the time when the vehicle 100 exits the measured section.

The axle load is the weight (axle load) corresponding to the load (load from the left and right tires attached to that axle) measured by each axle load sensor 2-4 for each axle of the vehicle 100 that traveled the measurement section. The statistical data in the example shown in Figure 3 is an example of a vehicle 100 that traveled the measurement section with three axles. For a vehicle 100 with two axles, the statistical data does not include data for the third axle. For a vehicle 100 with four or more axles, the statistical data includes data for the fourth, fifth, etc. axles depending on the number of axles.

3 is the weight of the first axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 2. In other words, MA1 is the axle load of the first axle of the vehicle 100 measured by the axle load sensor 2. MA2 is the axle load of the second axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 2, and MA3 is the axle load of the third axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 2. Similarly, MB1 shown in FIG. 3 is the axle load of the first axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 3, MB2 is the axle load of the second axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 3, and MB3 is the axle load of the third axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 3. MC1 shown in FIG. 3 is the axle load of the first axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 4, MC2 is the axle load of the second axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 4, and MC3 is the axle load of the third axle of the vehicle 100 obtained from the measurement signal of the axle load sensor 4.

Total weight MA is the sum of the axle loads of each axle of vehicle 100 measured by axle load sensor 2, and can be considered to be the weight of vehicle 100 measured by axle load sensor 2. Similarly, total weight MB is the sum of the axle loads of each axle of vehicle 100 measured by axle load sensor 3, and can be considered to be the weight of vehicle 100 measured by axle load sensor 3. Total weight MC is the sum of the axle loads of each axle of vehicle 100 measured by axle load sensor 4, and can be considered to be the weight of vehicle 100 measured by axle load sensor 4.

As shown in FIG. 3, a vehicle 100 having three axles has the following configuration:
The total weight MA=MA1+MA2+MA3,
The total weight MB=MB1+MB2+MB3,
The total weight MC=MC1+MC2+MC3.

The average weight Mav is the average of the total weights M, M, and M. That is,
Average weight Mav = (total weight MA + total weight MB + total weight MC) / 3
It is.

As described above, the total weights MA, MB, and MC are the weights of the vehicle 100 measured by the individual axle load sensors 2-4, and the average weight Mav is the average value of the weights of the vehicle 100 measured by the individual axle load sensors 2-4. Therefore, the total weights MA, MB, and MC are the weights of the vehicle 100 without taking into account the effects of vibration of the vehicle 100. The average weight Mav is the weight of the vehicle 100 in which the effects of vibration of the vehicle 100 are reduced compared to the total weights MA, MB, and MC.

The weight ratio is the ratio between the total weight MA, MB, MC of the vehicle 100 measured by each of the axle load sensors 2 to 4 and the average weight Mav. Specifically, the weight ratio of axle load sensor 2 is MA/Mav, the weight ratio of axle load sensor 3 is MB/Mav, and the weight ratio of axle load sensor 4 is MC/Mav.

When the measurement accuracy of the axle load sensor 2 is appropriate, the weight ratio (MA/Mav) of the axle load sensor 2 can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight (MA) of the vehicle 100 measured by the axle load sensor 2. Similarly, when the measurement accuracy of the axle load sensor 3 is appropriate, the weight ratio (MB/Mav) of the axle load sensor 3 can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight (MB) of the vehicle 100 measured by the axle load sensor 3. Furthermore, when the measurement accuracy of the axle load sensor 4 is appropriate, the weight ratio (MC/Mav) of the axle load sensor 4 can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight (MC) of the vehicle 100 measured by the axle load sensor 4.

The output unit 15 outputs to a higher-level device (not shown) the result of judging whether the measurement accuracy of the axle load sensors 2 to 4 is appropriate, the weight of the vehicle 100 that has traveled through the measurement section (the average weight Mav described above), etc.

Next, we will explain the axle load calculation unit 21, vehicle weight calculation unit 22, ratio calculation unit 23, and judgment unit 24 that the control unit 11 has.

The axle load calculation unit 21 uses the measurement signal of the axle load sensor 2 to calculate the axle load (MA1 to MA3 shown in FIG. 3) of each axle of the vehicle 100 that has traveled through the measurement section. The axle load calculation unit 21 also uses the measurement signal of the axle load sensor 3 to calculate the axle load (MB1 to MB3 shown in FIG. 3) of each axle of the vehicle 100 that has traveled through the measurement section. The axle load calculation unit 21 also uses the measurement signal of the axle load sensor 4 to calculate the axle load (MC1 to MC3 shown in FIG. 3) of each axle of the vehicle 100 that has traveled through the measurement section.

The vehicle weight calculation unit 22 calculates the total weight (MA, MB, MC shown in FIG. 3) and the average weight (Mav shown in FIG. 3).

The ratio calculation unit 23 calculates the weight ratio (MA/Mav, MB/Mav, MC/Mav shown in Figure 3) for each axle load sensor 2 to 4.

The weight ratio may also be Mav/MA, Mav/MB, or Mav/MC.

The determination unit 24 processes the statistical data recorded in the statistical DB 14 and determines whether the measurement accuracy is appropriate for each of the axle load sensors 2 to 4.

The determination unit 24 extracts statistical data (target statistical data) of the vehicle 100 from the statistical DB 14 to be used to determine whether the measurement accuracy of the axle load sensors 2 to 4 is appropriate. For example, the target statistical data may be statistical data of the vehicles 100 that have traveled the measurement section during a set period (e.g., the last week, the last month), or statistical data of a set number of vehicles (e.g., the last 5,000 vehicles, the last 10,000 vehicles).

The determination unit 24 generates a histogram of the number of vehicles 100 versus weight ratio (corresponding to the frequency distribution of weight ratio in this invention) for each axle load sensor 2-4. Figures 4 (A)-(C) are diagrams showing example histograms of the number of vehicles versus weight ratio. Figure 4 (A) is an example histogram of the number of vehicles versus weight ratio when the measurement accuracy of the axle load sensor is appropriate. Figures 4 (B) and (C) are example histograms of the number of vehicles versus weight ratio when the measurement accuracy of the axle load sensor is not appropriate.

The total weight measured by the axle load sensors 2-4 is the weight affected by the vibration of the vehicle 100. The frequency distribution of the magnitude of the effect of the vibration of the vehicle 100 on the axle load sensors 2-4 approximates a normal distribution with a certain degree of variance. The frequency distribution of the weight ratio of each axle load sensor 2-4 can be regarded as the distribution of the fluctuation of the load due to the vibration of the vehicle 100. Therefore, for axle load sensors 2-4 with appropriate measurement accuracy, the shape of the frequency distribution of the weight ratio approximates a normal distribution with a certain degree of variance. Conversely, for axle load sensors 2-4 with inappropriate measurement accuracy, the shape of the frequency distribution of the weight ratio does not approximate a normal distribution with a certain degree of variance.

For example, as shown in Figure 4 (A), when the measurement accuracy of the axle load sensor is appropriate, the histogram will have a roughly symmetrical shape between the left and right sides based on the weight ratio where the number of vehicles 100 peaks. On the other hand, as shown in Figures 4 (B) and (C), when the measurement accuracy of the axle load sensor is not appropriate, the histogram will have an asymmetrical shape between the left and right sides based on the weight ratio where the number of vehicles 100 peaks.

The measurement accuracy of the axle load sensor does not drop from the appropriate first level to the inappropriate second level all at once, but drops gradually from the first level to the second level. Therefore, even if the measurement accuracy of axle load sensors 2 to 4 is not appropriate, the target statistical data extracted to generate the frequency distribution of the weight ratio includes statistical data when the measurement accuracy was appropriate. For this reason, for axle load sensors with inappropriate measurement accuracy, the shape of the frequency distribution of the weight ratio does not approximate a normal distribution with a certain degree of variance, as shown in Figures 4 (B) and (C).

The determination unit 24 determines whether the measurement accuracy is appropriate for each axle load sensor 2-4 based on the shape of the frequency distribution of the weight ratio for that axle load sensor 2-4.

The control unit 11 of the axle load measuring device 1 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the measurement accuracy determination program of the present invention, it operates as an axle load calculation unit 21, a vehicle weight calculation unit 22, a ratio calculation unit 23, and a determination unit 24. The memory also has an area for expanding the measurement accuracy determination program of the present invention, and an area for temporarily storing data generated when the measurement accuracy determination program is executed. The control unit 11 may be an LSI that integrates the hardware CPU, memory, etc. Also, the hardware CPU is a computer that executes the measurement accuracy determination method of the present invention.

<3. Operation example>
The axle load measuring device 1 according to this example performs a measurement process for measuring the axle load, etc. of the vehicle 100 that has traveled through a measurement section, and a measurement accuracy determination process for determining whether the measurement accuracy of the axle load sensors 2 to 4 is appropriate.

Figure 5 is a flowchart showing the measurement process of the axle load measuring device according to this example. The axle load measuring device 1 waits for the vehicle 100 to enter the measurement section (s1). The control unit 11 detects that the vehicle 100 has entered the measurement section based on a change in inductance of the vehicle detection sensor 6 connected to the loop coil sensor connection section 13.

When the axle load calculation unit 21 detects that the vehicle 100 has entered the measurement section, it starts the process of storing the measurement signals of the axle load sensors 2 to 4 connected to the axle load sensor connection unit 12 in memory (s2). The time interval for storing the measurement signals of the axle load sensors 2 to 4 is, for example, several tens of msec to several hundreds of msec.

When the axle load calculation unit 21 detects that the vehicle 100 has left the measurement section (s3), it ends the process of storing the measurement signals of the axle load sensors 2 to 4 in memory, which started in s2 (s4). The axle load calculation unit 21 processes the measurement signals cumulatively stored in memory for each of the axle load sensors 2 to 4, and calculates the axle load of each axle (s5).

The measurement signal of each axle load sensor changes when the axle (wheel) of the vehicle 100 passes by. Therefore, the axle load calculation unit 21 can obtain the number of axles of the vehicle 100 by counting the points where the measurement signals of the axle load sensors 2 to 4 change. The axle load calculation unit 21 can also extract the measurement signals when the wheels of each axle pass by the axle load sensors 2 to 4, and calculate the axle loads (for example, MA1 to MA3, MB1 to MB3, MC1 to MC3 shown in Figure 3).

The vehicle weight calculation unit 22 calculates the total weights MA, MB, and MC of the axle load sensors 2-4 for the vehicle 100 that has passed through the current measurement section, and calculates the average weight Mav (s6). The total weight MA of the axle load sensor 2 is the sum of the axle weights of the axles of the vehicle 100 that has passed through the current measurement section, measured by the axle load sensor 2. The total weight MB of the axle load sensor 3 is the sum of the axle weights of the axles of the vehicle 100 that has passed through the current measurement section, measured by the axle load sensor 3. The total weight MC of the axle load sensor 4 is the sum of the axle weights of the axles of the vehicle 100 that has passed through the current measurement section, measured by the axle load sensor 4.

The average weight Mav is the average of the total weights MA, MB, and MC of each axle load sensor 2 to 4.

The ratio calculation unit 23 calculates the weight ratio for each of the axle load sensors 2 to 4 (s7). The axle load measurement device 1 generates the statistical data shown in FIG. 3 for the vehicle 100 that has traveled through the current measurement section, stores the generated statistical data in the statistical DB 14 (s8), and returns to s1.

In this way, the axle load measuring device 1 of this example generates the statistical data shown in FIG. 3 for each vehicle 100 that has traveled through the measurement section. The statistical DB 14 stores the statistical data generated for each vehicle 100 that has passed through the measurement section.

The axle load measuring device 1 may also output the weight of the vehicle 100 (the average weight Mav calculated in s6) at the output unit 15 each time the vehicle 100 travels through the measurement section.

The statistical data stored in the statistical DB14 can also be used as useful information for determining whether or not road repair work is required.

Next, the measurement accuracy determination process of the axle load measuring device 1 according to this example will be described. This measurement accuracy determination process is a process for determining whether the measurement accuracy of the axle load sensors 2 to 4 is appropriate. Figure 6 is a flowchart showing the measurement accuracy determination process.

The determination unit 24 determines whether it is time to determine the measurement accuracy of the axle load sensors 2 to 4 (s11). The determination timing may be, for example, 0:00 AM every Monday, 0:00 AM on the 1st of every month, 0:00 AM on the 1st of every month, or when the number of vehicles 100 that have traveled the measurement section since the end of the previous measurement accuracy determination process reaches a set number (500 vehicles, 1000 vehicles), or when instructed by an administrator, or it may be some other timing.

When the determination unit 24 determines that it is time for determination, it extracts target statistical data from the statistical data stored in the statistical DB 14 (s12). For example, the determination unit 24 may extract, as the target statistical data, statistical data of the vehicles 100 that have traveled the measured section within the last week or month, or may extract, as the target statistical data, statistical data of the last 5,000 vehicles 100 or the last 10,000 vehicles 100.

The determination unit 24 determines the axle load sensor to be determined whether the measurement accuracy is appropriate (s13). In s13, one of the axle load sensors 2-4 connected to the axle load sensor connection unit 12 of the axle load measuring device 1 and for which the measurement accuracy has not been determined to be appropriate in the current measurement accuracy determination process (undetermined axle load sensors 2-4) is determined to be the axle load sensor to be determined.

The determination unit 24 generates a frequency distribution of weight ratios for the axle load sensor to be determined in s13 (s14). In s14, a frequency distribution of weight ratios is generated for the target statistical data extracted in s12.

The determination unit 24 determines whether the frequency distribution of the weight ratio generated in s14 for the axle load sensor being determined can be regarded as a normal distribution (s15).

In s15, for example, the degree of symmetry between the left distribution shape and the right distribution shape, which are obtained by dividing the frequency distribution generated in s14 using the peak (the weight ratio at which the number of vehicles 100 was the greatest) as the reference (center), is calculated. The determination unit 24 calculates, for example, the similarity between the left distribution shape and the left distribution shape flipped left side as the degree of symmetry between the left distribution shape and the right distribution shape. If the calculated degree of symmetry exceeds a predetermined symmetry threshold, the determination unit 24 determines that the frequency distribution of the weight ratios generated in s14 can be regarded as a normal distribution. Conversely, if the calculated degree of symmetry does not exceed the predetermined symmetry threshold, the determination unit 24 determines that the frequency distribution of the weight ratios generated in s14 cannot be regarded as a normal distribution.

The determination unit 24 may also calculate the variance ^δ2 of the frequency distribution generated in s14, and determine that the frequency distribution of the weight ratios generated in s14 can be regarded as a normal distribution if the calculated variance ^δ2 is a value between a predetermined upper threshold and a predetermined lower threshold. Conversely, the determination unit 24 determines that the frequency distribution of the weight ratios generated in s14 cannot be regarded as a normal distribution if the calculated variance ^δ2 is not a value between a predetermined upper threshold and a predetermined lower threshold.

The determination unit 24 may calculate the standard deviation δ instead of the variance δ ² , and determine whether the frequency distribution of the weight ratios generated in s14 can be regarded as a normal distribution based on the standard deviation δ. Also, the kurtosis, skewness, or the like of the frequency distribution of the weight ratios may be used to determine whether this frequency distribution can be regarded as a normal distribution.

The determination unit 24 may be configured to determine that the frequency distribution of the weight ratios generated in s14 can be regarded as a normal distribution when the degree of symmetry of the distribution shape exceeds the symmetry threshold and the variance ^δ2 is a value between a predetermined upper threshold and a predetermined lower threshold. In other words, the determination unit 24 may be configured to determine that the frequency distribution of the weight ratios generated in s14 cannot be regarded as a normal distribution when the degree of symmetry of the distribution shape does not exceed the symmetry threshold or when the variance ^δ2 is not a value between a predetermined upper threshold and a predetermined lower threshold. Furthermore, the kurtosis, skewness, etc. of the frequency distribution of the weight ratios may be added to determine whether this frequency distribution can be regarded as a normal distribution.

When the determination unit 24 determines that the frequency distribution of the weight ratio can be regarded as a normal distribution, it determines that the measurement accuracy of the axle load sensor being evaluated is appropriate (s16). Conversely, when the determination unit 24 determines that the frequency distribution of the weight ratio cannot be regarded as a normal distribution, it determines that the measurement accuracy of the axle load sensor being evaluated is not appropriate (s17).

The determination unit 24 determines whether there are any axle load sensors whose measurement accuracy has not yet been determined (s18). If there are any axle load sensors whose measurement accuracy has not yet been determined, the determination unit 24 returns to s13 and repeats the above process. If there are any axle load sensors whose measurement accuracy has not yet been determined, the determination unit 24 associates each axle load sensor 2-4 with the current determination result as to whether the measurement accuracy is appropriate, outputs the result to the higher-level device (s19), and returns to s1.

In this way, the axle load measuring device 1 in this example can determine whether the measurement accuracy of the axle load sensors 2 to 4 used to measure the weight of the moving vehicle 100 is appropriate, regardless of whether a special vehicle is being driven.

In addition, the measurement accuracy of the axle load sensors 2 to 4 can be judged periodically or at a time specified by the administrator.

4. Modifications
Next, a modified vehicle weight measurement system will be described with reference to Fig. 7, which is a schematic diagram showing the modified vehicle weight measurement system.

This modified vehicle weight measurement system differs from the above example in that the axle load sensor 2 is composed of a pair of wheel load sensors 2R and 2L, the axle load sensor 3 is composed of a pair of wheel load sensors 3R and 3L, and the axle load sensor 4 is composed of a pair of wheel load sensors 4R and 4L.

The axle load sensor 2 is a pair of wheel load sensors 2R, 2L arranged in the vehicle width direction of the vehicle 100 traveling on the road. Similarly, the axle load sensor 3 is a pair of wheel load sensors 3R, 3L arranged in the vehicle width direction of the vehicle 100 traveling on the road, and the axle load sensor 4 is a pair of wheel load sensors 4R, 4L arranged in the vehicle width direction of the vehicle 100 traveling on the road. The wheel load sensors 2R to 4R are arranged at a position where the right tire of the vehicle 100 passes. The wheel load sensors 2L to 4L are arranged at a position where the left tire of the vehicle 100 passes. The wheel load sensors 2R to 4R and 2L to 4L are, for example, piezoelectric sensors, and output a measurement signal to the axle load measurement device 1 according to the pressing force when the tire passes.

The axle load measuring device 1 of this modified example has the same configuration as the above example, as shown in FIG. 2. However, it differs from the above example in that wheel load sensors 2R-4R and 2L-4L are connected to the axle load sensor connection section 12.

In addition, the axle load measurement device 1 of this modified example stores right side weight data and left side weight data as a pair for each vehicle 100 in the statistical DB 14 as statistical data. Figure 8 (A) is a diagram showing the right side weight data, and Figure 8 (B) is a diagram showing the left side weight data.

The pair of right side weight data and left side weight data have the same vehicle ID, travel date, and travel time. The vehicle ID, travel date, and travel time of the right side weight data and left side weight data are the same as in the example above.

The right-side weight data is generated based on the measurement signals of the wheel load sensors 2R to 4R. The left-side weight data is generated based on the measurement signals of the wheel load sensors 2L to 4L. The wheel load of the right-side weight data is the weight (wheel load) corresponding to the load (load from the right-side tire attached to the axle) measured by each wheel load sensor 2R to 4R for each axle of the vehicle 100 that traveled the measurement section. The right-side weight data in the example shown in FIG. 8(A) and the left-side weight data in the example shown in FIG. 8(B) are examples of a vehicle 100 that traveled the measurement section with three axles. In the case of a vehicle 100 with two axles, the right-side weight data and the left-side weight data do not include data on the third axle. In the case of a vehicle 100 with four or more axles, the right-side weight data and the left-side weight data include data on the fourth axle, fifth axle, etc. depending on the number of axles.

RMA1 shown in FIG. 8(A) is the wheel load of the right tire of the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2R. In other words, RMA1 is the wheel load of the right tire of the first axle of the vehicle 100 measured by the wheel load sensor 2R. RMA2 is the wheel load of the right tire of the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2R, and RMA3 is the wheel load of the right tire of the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2R. Similarly, RMB1 shown in FIG. 8(A) is the wheel load of the right tire of the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3R, RMB2 is the wheel load of the right tire of the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3R, and RMB3 is the wheel load of the right tire of the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3R. RMC1 shown in FIG. 8(A) is the wheel load of the right tire on the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4R, RMC2 is the wheel load of the right tire on the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4R, and RMC3 is the wheel load of the right tire on the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4R.

The right total weight RMA is the sum of the wheel loads of the right tires on each axle of the vehicle 100 measured by the wheel load sensor 2R, and can be considered to be the weight of the right side of the vehicle 100 measured by the wheel load sensor 2R. Similarly, the right total weight RMB is the sum of the wheel loads of the right tires on each axle of the vehicle 100 measured by the wheel load sensor 3R, and can be considered to be the weight of the right side of the vehicle 100 measured by the wheel load sensor 3R. The right total weight RMC is the sum of the wheel loads of the right tires on each axle of the vehicle 100 measured by the wheel load sensor 4R, and can be considered to be the weight of the right side of the vehicle 100 measured by the wheel load sensor 4R.

As shown in FIG. 8A, a vehicle 100 having three axles has the following configuration:
Right total weight RMA=RMA1+RMA2+RMA3,
Right total weight RMB=RMB1+RMB2+RMB3,
Right total weight RMC=RMC1+RMC2+RMC3.

The right average weight RMav is the average of the right total weight RMA, the right total weight RMB, and the right total weight RMC. That is,
Right average weight RMav
= (Right total weight RMA + Right total weight RMB + Right total weight RMC) / 3
It is.

As described above, the right total weights RMA, RMB, RMC are the weights of the right side of the vehicle 100 measured by each of the wheel load sensors 2R-4R, and the right average weight RMav is the average value of the weights of the right side of the vehicle 100 measured by each of the wheel load sensors 2R-4R. Therefore, the right total weights RMA, RMB, RMC are the weights of the right side of the vehicle 100 that do not take into account the effects of vibrations of the vehicle 100. The right average weight RMav is the weight of the right side of the vehicle 100 in which the effects of vibrations of the vehicle 100 are reduced compared to the right total weights RMA, RMB, RMC.

The right weight ratio is the ratio between the right total weight RMA, RMB, RMC of the vehicle 100 measured by each of the wheel load sensors 2R to 4R and the right average weight RMav. Specifically, the right weight ratio of wheel load sensor 2R is RMA/RMav, the right weight ratio of wheel load sensor 3R is RMB/RMav, and the right weight ratio of wheel load sensor 4R is RMC/RMav.

When the measurement accuracy of the wheel load sensor 2R is appropriate, the right weight ratio (RMA/RMav) of the wheel load sensor 2R can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight of the right side of the vehicle 100 (RMA) measured by the wheel load sensor 2R. Similarly, when the measurement accuracy of the wheel load sensor 3R is appropriate, the right weight ratio (RMB/RMav) of the wheel load sensor 3R can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight of the right side of the vehicle 100 (RMB) measured by the wheel load sensor 3R. Furthermore, when the measurement accuracy of the wheel load sensor 4R is appropriate, the right weight ratio (RMC/RMav) of the wheel load sensor 4R can be regarded as the magnitude of the effect that the vibration of the vehicle 100 has on the weight of the right side of the vehicle 100 (RMC) measured by the wheel load sensor 4R.

Also, LMA1 shown in FIG. 8(B) is the wheel load of the left tire of the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2L. In other words, LMA1 is the wheel load of the left tire of the first axle of the vehicle 100 measured by the wheel load sensor 2L. Also, LMA2 is the wheel load of the left tire of the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2L, and LMA3 is the wheel load of the left tire of the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 2L. Similarly, LMB1 shown in FIG. 8(B) is the wheel load of the left tire of the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3L, LMB2 is the wheel load of the left tire of the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3L, and LMB3 is the wheel load of the left tire of the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 3L. LMC1 shown in FIG. 8(B) is the wheel load of the left tire on the first axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4L, LMC2 is the wheel load of the left tire on the second axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4L, and LMC3 is the wheel load of the left tire on the third axle of the vehicle 100 obtained from the measurement signal of the wheel load sensor 4L.

The left total weight LMA is the sum of the wheel loads of the left tires on each axle of the vehicle 100 measured by the wheel load sensor 2L, and can be considered to be the weight of the left side of the vehicle 100 measured by the wheel load sensor 2L. Similarly, the left total weight LMB is the sum of the wheel loads of the left tires on each axle of the vehicle 100 measured by the wheel load sensor 3L, and can be considered to be the weight of the left side of the vehicle 100 measured by the wheel load sensor 3L. The left total weight LMC is the sum of the wheel loads of the left tires on each axle of the vehicle 100 measured by the wheel load sensor 4L, and can be considered to be the weight of the left side of the vehicle 100 measured by the wheel load sensor 4L.

As shown in FIG. 8B, the vehicle 100 has three axles.
The left total weight LMA=LMA1+LMA2+LMA3,
The left total weight LMB=LMB1+LMB2+LMB3,
The left total weight LMC=LMC1+LMC2+LMC3.

The left average weight LMav is the average of the left total weight LMA, the left total weight LMB, and the left total weight LMC. That is,
Left average weight LMav
= (left total weight LMA + left total weight LMB + left total weight LMC) / 3
It is.

As described above, the left total weights LMA, LMB, LMC are the weights of the left side of the vehicle 100 measured by each of the wheel load sensors 2L to 4L, and the left average weight LMav is the average value of the weights of the right side of the vehicle 100 measured by each of the wheel load sensors 2L to 4L. Therefore, the left total weights LMA, LMB, LMC are the weights of the left side of the vehicle 100 that do not take into account the effects of vibrations of the vehicle 100. The left average weight LMav is the weight of the left side of the vehicle 100 in which the effects of vibrations of the vehicle 100 are reduced compared to the left total weights LMA, LMB, LMC.

The left weight ratio is the ratio between the left total weight LMA, LMB, LMC of the vehicle 100 measured by each of the wheel load sensors 2L to 4L and the left average weight LMav. Specifically, the weight ratio of wheel load sensor 2L is LMA/LMav, the weight ratio of wheel load sensor 3L is LMB/LMav, and the weight ratio of wheel load sensor 4L is LMC/LMav.

When the measurement accuracy of the wheel load sensor 2L is appropriate, the left weight ratio (LMA/LMav) of the wheel load sensor 2L can be regarded as the magnitude of the effect of the vibration of the vehicle 100 on the weight of the left side of the vehicle 100 (LMA) measured by the wheel load sensor 2L. Similarly, when the measurement accuracy of the wheel load sensor 3L is appropriate, the left weight ratio (LMB/LMav) of the wheel load sensor 3L can be regarded as the magnitude of the effect of the vibration of the vehicle 100 on the weight of the left side of the vehicle 100 (LMB) measured by the wheel load sensor 3L. Furthermore, when the measurement accuracy of the wheel load sensor 4L is appropriate, the left weight ratio (LMC/LMav) of the wheel load sensor 4L can be regarded as the magnitude of the effect of the vibration of the vehicle 100 on the weight of the left side of the vehicle 100 (LMC) measured by the wheel load sensor 4L.

In this modified axle load measuring device 1, the axle load calculation unit 21 uses the measurement signal of the wheel load sensor 2R to calculate the wheel load of the right tire (RMA1 to RMA3 shown in FIG. 8(A)) of each axle of the vehicle 100 that has traveled through the measurement section. The axle load calculation unit 21 also uses the measurement signal of the wheel load sensor 2L to calculate the wheel load of the left tire (LMA1 to LMA3 shown in FIG. 8(B)) of each axle of the vehicle 100 that has traveled through the measurement section.

The axle load calculation unit 21 also uses the measurement signal of the wheel load sensor 3R to calculate the wheel load of the right tire (RMB1 to RMB3 shown in FIG. 8(A)) of each axle of the vehicle 100 that has traveled through the measurement section. The axle load calculation unit 21 also uses the measurement signal of the wheel load sensor 3L to calculate the wheel load of the left tire (LMB1 to LMB3 shown in FIG. 8(B)) of each axle of the vehicle 100 that has traveled through the measurement section.

Furthermore, the axle load calculation unit 21 uses the measurement signal of the wheel load sensor 4R to calculate the wheel load of the right tire of each axle of the vehicle 100 that has traveled through the measurement section (RMC1 to RMC3 shown in FIG. 8(A)). Also, the axle load calculation unit 21 uses the measurement signal of the wheel load sensor 4L to calculate the wheel load of the left tire of each axle of the vehicle 100 that has traveled through the measurement section (LMC1 to LMC3 shown in FIG. 8(B)).

The vehicle weight calculation unit 22 calculates the above-mentioned right total weight (RMA, RMB, RMC shown in FIG. 8(A)), left total weight (LMA, LMB, LMC shown in FIG. 8(B)), right average weight (RMav shown in FIG. 8(A)), and left average weight (LMav shown in FIG. 8(B)). In this example, the vehicle weight calculation unit 22 obtains the sum of the right average weight RMav and the left average weight LMav (RMav+LMav) of the vehicle 100 that has traveled the measurement section as the weight of this vehicle 100.

The axle load measuring device 1 of this modified example can determine the uneven loading state of the cargo on the vehicle 100 by comparing the right average weight RMav and the left average weight LMav.

The ratio calculation unit 23 calculates the right weight ratio (RMA/RMav, RMB/RMav, RMC/RMav shown in FIG. 8(A)) for each of the wheel load sensors 2R to 4R, and calculates the left weight ratio (LMA/LMav, LMB/LMav, LMC/LMav shown in FIG. 8(B)) for each of the wheel load sensors 2L to 4L.

As in the above example, the right weight ratio may be RMav/RMA, RMav/RMB, or RMav/RMC, and the left weight ratio may be LMav/LMA, LMav/LMB, or LMav/LMC.

The determination unit 24 processes the statistical data recorded in the statistical DB 14 and determines whether the measurement accuracy is appropriate for each of the wheel load sensors 2R-4R and 2L-4L.

The determination unit 24 extracts statistical data (target statistical data) of the vehicle 100 from the statistical DB 14 to be used to determine whether the measurement accuracy of the wheel load sensors 2R-4R and 2L-4L is appropriate. For example, the target statistical data may be statistical data of the vehicles 100 that have traveled the measurement section during a set period, as in the above example, or it may be statistical data of a set number of vehicles.

The determination unit 24 generates a histogram of the number of vehicles 100 relative to the right weight ratio for each of the wheel load sensors 2R to 4R. The determination unit 24 also generates a histogram of the number of vehicles 100 relative to the left weight ratio for each of the wheel load sensors 2L to 4L.

As explained in the above example, in a wheel load sensor with appropriate measurement accuracy, the shape of the frequency distribution of the weight ratio (right weight ratio or left weight ratio) approximates a normal distribution with a certain degree of variance. Conversely, in a wheel load sensor with inappropriate measurement accuracy, the shape of the frequency distribution of the weight ratio (right weight ratio or left weight ratio) does not approximate a normal distribution with a certain degree of variance.

The determination unit 24 determines whether the measurement accuracy is appropriate for each wheel load sensor 2R-4R, 2L-4L based on the shape of the frequency distribution of the weight ratio for that wheel load sensor 2R-4R, 2L-4L.

The axle load measuring device 1 of this modified example also executes the measurement process shown in Figure 5 and the measurement accuracy determination process shown in Figure 6, similar to the above example. However, in this modified example, the process for the axle load sensors 2 to 4 in the above example is replaced with the process for the wheel load sensors 2R to 4R and 2L to 4L.

In this modified axle load measuring device 1, as in the above example, it is possible to determine whether the measurement accuracy of the axle load sensors 2-4 (wheel load sensors 2R-4R, 2L-4L) used to measure the weight of the moving vehicle 100 is appropriate, regardless of whether a special vehicle is moving or not. In addition, it is possible to determine whether the measurement accuracy of the wheel load sensors 2R-4R, 2L-4L is appropriate periodically, or at a timing specified by the administrator.

In addition, this modified axle load measuring device 1 can also determine uneven loading of cargo on the vehicle 100 by comparing the right average weight RMav with the left average weight LMav.

Also, in the above example, the configuration is such that whether the measurement accuracy of the axle load sensors 2 to 4 (wheel load sensors 2R to 4R, 2L to 4L) is appropriate is determined based on whether the frequency distribution can be regarded as a normal distribution. The axle load measuring device 1 may determine whether the measurement accuracy of the axle load sensors 2 to 4 (wheel load sensors 2R to 4R, 2L to 4L) is appropriate based on whether the weight ratio X, which is a peak in the frequency distribution, is between a predetermined upper limit ratio Y (e.g., Y=1.1) and a lower limit ratio Z (e.g., Z=0.9). Specifically, the determining unit 24:
Upper limit ratio Y > weight ratio X > lower limit ratio Z
In other words, the determination unit 24 may determine that the measurement accuracy is appropriate.
Upper limit ratio Y > weight ratio X > lower limit ratio Z
If not, it may be determined that the measurement accuracy is not appropriate.

The determination unit 24 may also determine whether the measurement accuracy of the axle load sensors 2 to 4 (wheel load sensors 2R to 4R, 2L to 4L) is appropriate by using both whether the frequency distribution is a normal distribution and the weight ratio X that is a peak in the frequency distribution.

The above modified example may also be configured to calculate the right weight ratio by dividing the right total weight measured by each of the wheel load sensors 2R to 4R by the sum of the right average weight RMav and the left average weight LMav (RMav+LMav). Similarly, the above modified example may also be configured to calculate the left weight ratio by dividing the left total weight measured by each of the wheel load sensors 2L to 4L by the sum of the right average weight RMav and the left average weight LMav (RMav+LMav).

Furthermore, in this modified example, similarly to the above example, it may be possible to judge whether the measurement accuracy is appropriate for each of the axle load sensors 2 to 4. Specifically, each of MA1, MA2, MA3, MB1, MB2, MB3, MC1, MC2, and MC3 shown in FIG.
MA1 = RMA1 + LMA1, MA2 = RMA2 + LMA2, MA3 = RMA3 + LMA3,
MB1 = RMB1 + LMB1, MB2 = RMB2 + LMB2, MB3 = RMB3 + LMB3,
MC1 = RMC1 + LMC1, MC2 = RMC2 + LMC2, MC3 = RMC3 + LMC3,
The value shall be calculated as follows.

In this case, similar to the example above, the total weight and weight ratio may be calculated for each axle load sensor 2 to 4 to determine whether the measurement accuracy is appropriate.

For example, if any of the axle load sensors 2-4 are determined to have inappropriate measurement accuracy, the above-mentioned frequency distribution may be generated for the pair of wheel load sensors that make up the axle load sensors 2-4, and the wheel load sensor with inappropriate measurement accuracy may be determined. For example, the right total weight and the left total weight may be compared between the axle load sensor 2-4 determined to have inappropriate measurement accuracy and the other axle load sensors 2-4 (axle load sensors 2-4 determined to have appropriate measurement accuracy) to determine which wheel load sensor has inappropriate measurement accuracy. Furthermore, the wheel load sensor with inappropriate measurement accuracy may be determined by using a method other than the above for the axle load sensors 2-4 determined to have inappropriate measurement accuracy.

In the above example, the statistical data includes the total weight (in the modified example, right total weight, left total weight), weight ratio (in the modified example, right weight ratio, left weight ratio), and average weight (in the modified example, right average weight, left average weight), but it may be configured not to include these. In this case, the measurement accuracy determination process may be configured to calculate the total weight (in the modified example, right total weight, left total weight), weight ratio (in the modified example, right weight ratio, left weight ratio), and average weight (in the modified example, right average weight, left average weight) for each target statistical data extracted in s12. With this configuration, the storage capacity of the statistical DB 14 can be reduced.

Also, some of the components shown in FIG. 2 (e.g., the determination unit 24 and the ratio calculation unit 23) may be provided in the higher-level device, and the axle load measurement device 1 and the higher-level device may cooperate to perform the above-mentioned processing. Also, the statistical data may be configured not to include any of the items related to the vehicle ID, travel date, and travel time.

Note that this invention is not limited to the above-described embodiment as it is, and in the implementation stage, the components can be modified and embodied without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in the above-described embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, components from different embodiments may be appropriately combined. In addition, the measurement process shown in FIG. 5 and the measurement accuracy determination process shown in FIG. 6 are merely examples, and the order of each step may be changed as appropriate.

Furthermore, the correspondence between the configuration according to the present invention and the configuration according to the above-described embodiment can be described as follows.
<Additional Notes>
a sensor connection unit (12) to which a measurement signal of the load caused by the tires of a passing vehicle (100) measured by a plurality of sensors (2 to 4) arranged in a line in the traveling direction of the vehicle (100) on the road surface of the measurement section is input;
a ratio calculation unit (23) that calculates, for each of the sensors (2 to 4) of the vehicle (100) that has traveled through the measurement section, a ratio between a total weight obtained by adding up weights corresponding to the loads applied by the tires of the vehicle (100) that have passed each of the sensors (2 to 4) and an average weight that is an average of the total weights of the sensors (2 to 4);
and a determination unit (24) that determines whether the measurement accuracy of the sensor (2-4) is appropriate based on the weight ratio calculated for that sensor (2-4).

1...Axle load measuring device 2 to 4...Axle load sensors 2R to 4R, 2L to 4L...Wheel load sensors 6, 7...Vehicle detection sensor 11...Control unit 12...Axle load sensor connection section 13...Loop coil sensor connection section 14...Statistical database (statistical DB)
15... output section 21... axle load calculation section 22... vehicle weight calculation section 23... ratio calculation section 24... determination section 100... vehicle

Claims (12)

a sensor connection unit to which a measurement signal of the load of the tires of a passing vehicle, the measurement signal being measured by a plurality of sensors arranged in a line in the vehicle's traveling direction on the road surface of the measurement section, is input;
a ratio calculation unit that calculates, for each of the sensors, a ratio between a total weight obtained by adding up weights corresponding to the loads of the tires of the vehicles that have traveled through the measurement section, and an average weight that is an average of the total weights of the sensors; and
a determination unit that determines whether the measurement accuracy of the sensor is appropriate based on the weight ratio calculated for the sensor. The weight measuring device according to claim 1, wherein the determination unit determines whether the measurement accuracy of the sensor is appropriate based on the shape of the frequency distribution of the weight ratio calculated for that sensor. The weight measuring device according to claim 2, wherein the determination unit determines whether the measurement accuracy of the sensor is appropriate based on the degree of symmetry of the shape of the frequency distribution of the weight ratio based on the frequency peak. The weight measuring device according to claim 3, wherein the determination unit determines whether the measurement accuracy of the sensor is appropriate based on whether the degree of symmetry of the shape of the frequency distribution of the weight ratio exceeds a predetermined symmetry threshold. The weight measuring device according to claim 2, wherein the determination unit determines whether the measurement accuracy of the sensor is appropriate based on the magnitude of variance of the frequency distribution of the weight ratio. The weight measuring device according to claim 5, wherein the determination unit determines whether the measurement accuracy of the sensor is appropriate based on whether the variance of the frequency distribution of the weight ratio is between a predetermined lower threshold and an upper threshold. The weight measuring device according to any one of claims 1 to 6, wherein the sensor is pressed by tires on both sides attached to the axle of a vehicle traveling in the measurement section. The weight measuring device according to claim 7, further comprising an acquisition unit that acquires the average weight as the weight of the vehicle that has traveled through the measurement section. The sensors are arranged in two rows in the vehicle width direction,
The sensor in one row is pressed by a tire on one side attached to an axle of a vehicle traveling in the measurement section,
7. The weight measuring device according to claim 1, wherein the sensor in the other row is pressed by a tire on the other side attached to an axle of a vehicle traveling in the measurement section. The weight measuring device according to claim 9, further comprising an acquisition unit that acquires the sum of the average weight measured by the sensor in the one row and the average weight measured by the sensor in the other row as the weight of the vehicle that has traveled through the measurement section. a ratio calculation step of processing measurement signals of the load caused by the tires of passing vehicles measured by a plurality of sensors arranged in a line in the traveling direction of the vehicle on the road surface of the measurement section, which are input to the sensor connection unit, and calculating, for each of the vehicles that have traveled through the measurement section, a ratio between a total weight obtained by adding up weights corresponding to the load caused by the tires of vehicles that have passed that sensor and an average weight that is an average of the total weights of the sensors, as a weight ratio;
a determining step of determining whether the measurement accuracy of the sensor is appropriate based on the weight ratio calculated for the sensor, the method being executed by a computer. a ratio calculation step of processing measurement signals of the load caused by the tires of passing vehicles measured by a plurality of sensors arranged in a line in the traveling direction of the vehicle on the road surface of the measurement section, which are input to the sensor connection unit, and calculating, for each of the vehicles that have traveled through the measurement section, a ratio between a total weight obtained by adding up weights corresponding to the load caused by the tires of vehicles that have passed that sensor and an average weight that is an average of the total weights of the sensors, as a weight ratio;
and a determining step of determining whether the measurement accuracy of the sensor is appropriate based on the weight ratio calculated for the sensor.

Images