JP2007033127A - Load detection device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To compensate an error based on a bending moment included in an output from a load cell. <P>SOLUTION: Strain gages 6a, 6b provided on a resilient part 4 of a pillar-shaped resilient member on which a load is placed generate a weight measurement signal. The weight measurement signal includes an error component based on the bending moment applied to the resilient member by loading of the load on the strain body. Strain gages 12a, 12b provided on the resilient part 4 output a bending strain measurement signal based on bending strain generated in the resilient member by the bending moment. A CPU 34 inputs the weight measurement signal and the bending strain measurement signal, and compensates the error component included in the weight measurement signal. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a load detection device such as a load cell, and more particularly to a device that compensates for an error based on a bending moment acting on a strain body of a load detection device.

  When a load is detected by a load detection device such as a load cell, a load other than the load to be detected may be applied to the strain body of the load detection device, resulting in an error in the measured value. For example, in Patent Document 1, in a tank vehicle in which a load cell is attached to a vehicle body, a tank is loaded on the load cell, and the liquid weight in the tank is measured, when the stop position of the vehicle body is an inclined surface, an uneven load is applied to the load cell. An example is shown in which an error occurs in the measured quantity due to the action of. In Patent Document 1, correction coefficients corresponding to the inclination angles in the longitudinal direction and the lateral direction of the vehicle body are mapped in advance, the actual inclination angle of the vehicle body is detected by an inclination sensor, and a correction coefficient is obtained. The measurement value is corrected by

  In Patent Document 2, when the load cell is installed at an inclination from the normal posture, the load cell detects the load component, and the detected value is smaller than the original load. In order to correct this, an inclination angle detection sensor is provided, the inclination angle is measured, and the measured load value (load component force) is divided by a value corresponding to the measured angle. The load value is obtained.

JP-A-8-93045 JP 2001-255216 A

  By the way, when a load is loaded on the load cell in a state where the load cell having the columnar strain generating body is inclined, a bending moment is generated in the load cell in the tilt direction of the load cell. As a result, the strain generating body is bent, and bending strain is generated in the strain generating body. Due to this bending strain, a large error occurs in the weight measurement signal from the load detecting element provided in the strain generating body. The technique disclosed in Patent Document 1 corrects an uneven load due to the inclination of the vehicle body, and the technique disclosed in Patent Document 2 corrects a load component and corrects an error caused by a bending moment. It is not possible. Moreover, the amount of error due to the bending moment varies depending on the magnitude of the bending strain, and is larger than the error due to the load component force that is a problem in Patent Document 2. Further, the amount of error varies depending on the direction in which the bending moment is generated.

  An object of this invention is to provide the load detection apparatus which can compensate the error based on a bending moment.

  The load detection device of one embodiment of the present invention includes a column-type strain body on which a load is loaded. This strain body is provided with a weight measuring means. The weight measuring means generates a weight measurement signal based on the strain generated in the strain generating body due to the loading of the load. However, the weight measurement signal includes an error component based on the bending moment when a bending moment acts on the strain-generating body due to the load being loaded on the strain-generating body. For example, a strain gauge or a semiconductor pressure sensor can be used as the weight measuring means. The strain generating body is also provided with a bending strain detecting means. The bending strain detection means outputs a bending strain measurement signal based on the bending strain generated in the strain generating body by the bending moment. For example, a strain gauge or a semiconductor pressure sensor can be used as the bending strain detection means. The weight measurement signal and the bending strain measurement signal are input, and the error compensator compensates the error component included in the weight measurement signal. When the weight measurement signal and the bending strain measurement signal are input to the error compensation means, they can be digitized and input.

  In the load detection device configured as described above, an error component based on the bending moment included in the weight measurement signal is calculated from the bending strain measurement signal, and the weight measurement signal is compensated by the calculated error component. Accordingly, it is possible to detect the load by removing the influence of the error based on the bending moment.

  The magnitude of the error component may vary depending on the direction of the bending strain even when a bending moment of the same magnitude is acting on the strain generating body. In this case, the error compensation means includes means for calculating the magnitude and direction of the bending strain from the bending strain measurement signal, calculation means for calculating the error component from the calculated magnitude and direction of the bending strain, and And removing means for removing the calculated error component from the weight measurement signal.

  In the load detection device configured as described above, an error component according to the magnitude and direction of the bending strain can be calculated, and the weight measurement signal can be compensated based on the calculated error component.

  Further, at least two sets of the bending strain detecting means may be arranged so as to form a predetermined angle around the load direction on the surface of the strain generating body. With this configuration, at least two sets of bending strain detection means can detect components in different directions of bending strain, and by detecting bending strains of different direction components, the direction and magnitude of bending strain can be detected. Can be calculated.

  Furthermore, the at least two sets of bending strain detection means can be arranged orthogonal to each other around the load application direction. In this case, the means for calculating the magnitude and direction of the bending strain obtains the magnitude of the bending strain by the square root of the square sum of the bending strain measurement signals from the at least two sets of bending strain detection means, and the at least 2 The direction of the bending strain is determined from the ratio of the bending strain measurement signals from the pair of bending strain detecting means. By detecting the bending strain components orthogonal to each other in this way, the magnitude can be obtained from the square root of the sum of squares of each component and the ratio of the bending components whose directions are orthogonal.

  Further, the means for calculating the error component may store the error component in various bending strain directions in a state where bending strains having different magnitudes are generated in the strain generating body. it can. In this case, the means for calculating is based on the stored error components, the calculated bending strain magnitude, and the calculated bending strain direction, and the calculated bending strain magnitude and The error component corresponding to the direction is calculated. With this configuration, there is no need to store error components in advance corresponding to various magnitudes and directions of bending strain, and there is no need to install a large-capacity storage unit.

  As described above, according to the present invention, it is possible to compensate for an error based on a bending moment that may occur in a load detection device using a columnar strain body, and to detect a load with high accuracy.

  The load detection device according to an embodiment of the present invention is a load cell 2 and includes, for example, a columnar strain body. The strain generating body has a strain generating portion 4 as shown in FIG. The strain generating portion 4 has, for example, a substantially square cross section. When a load is applied to the center of the cross section of the strain generating section 4 in a direction perpendicular to the cross section, the strain generating section 4 is compressed in the vertical direction and stretched in a direction perpendicular to the vertical method. The

  In order to detect the compression force and the extension force, weight measuring means, for example, a compression detection strain gauge 6a and an extension detection strain gauge 6b are provided in the strain generating section 4. That is, a total of four strain gauges 6a are adhered to the four side surfaces of the strain generating portion 4 along the compression direction, that is, the vertical direction. In addition, a total of four strain gauges 6 b are attached to the four side surfaces of the strain-generating portion 4 along the extension direction, that is, the direction perpendicular to the vertical direction.

  As shown in FIG. 2, these strain gauges 6a and 6b are Wheatstone bridges having a series circuit in which two strain gauges 6a are connected in series as opposite sides and a series circuit in which two strain gauges 6b are connected in series as opposite sides. The circuit 7 is configured. A voltage V is applied between the opposite vertices of the bridge circuit 7, and a weight measurement signal, for example, a load output Vw is taken out between the other vertices.

  When this load cell 2 is used for a track scale, for example, it is disposed at each of the four corners of the back surface of the weighing table 8 as shown in FIG. When a truck 10 that is large in weight and is loaded with a large load gets on the weighing table 8, the weighing table 8 bends depending on the size of the weight and the position where the weight is applied. Accordingly, the load cell 2 may be inclined as shown in FIGS. 4 (a) and 4 (b). A bending moment is generated in the strain generating portion 4 of the load cell 2 along the inclination direction, the strain generating portion 4 is bent, and a bending strain is generated in the strain generating portion 4. Due to this bending strain, an error component is included in the load output Vw.

  The inclination direction differs depending on how the weighing table 8 bends. As shown in FIG. 5, when the axis passing through the center of the strain generating portion 4 is the z axis, the load cell 2 is inclined in any direction around the z axis. . Here, an x-axis and a y-axis that are orthogonal to the z axis and orthogonal to each other are defined. Specifically, an axis that is orthogonal to two opposing side surfaces of the strain-generating portion 4 and that is also orthogonal to the z axis is defined as x. An axis perpendicular to the x-axis and the z-axis is defined as a y-axis, and an angle formed between the tilt direction of the load cell 2 and the x-axis is defined as a tilt angle α.

  In order to compensate for an error included in the load output Vw due to the bending strain, a bending strain detecting means, for example, an x-axis direction bending strain detecting strain gauge 12a and a y-axis direction bending strain detecting strain gauge 12b. Are provided in the strain generating portion 4. The strain gauges 12a for detecting the bending strain in the x-axis direction are attached to two side surfaces of the strain generating section 4 that are orthogonal to the x-axis and that face each other. The strain gauges 12b for detecting the bending strain in the y-axis direction are attached to two side surfaces of the strain-generating part 4 that are orthogonal to the y-axis and that face each other. As shown in FIG. 1, the strain gauges 12a and 12b are arranged above the compression detection strain gauge 6a so as to be positioned on the same straight line as the compression gauge strain gauge 6a.

  The two strain gauges 12a for detecting the bending strain in the x-axis direction constitute a Wheatstone bridge circuit 15 together with two dummy resistors 14, as shown in FIG. In this bridge circuit 15, the strain gauges 12 a are arranged so as to make opposite sides of the dummy resistor 14, a voltage V is applied between the opposite vertices of the bridge circuit 15, and a bending strain measurement signal is transmitted from between the other vertices. A signal, for example an x-axis bending strain measurement signal load output Vx, is taken. Although not shown, the strain gauge 12b for detecting the y-axis direction bending strain also forms a Wheatstone bridge circuit 17 together with the dummy resistor, and outputs a bending strain measurement signal signal, for example, a y-axis direction bending strain measurement signal load output Vy. ing.

The x-axis direction bending strain measurement signal load output Vx represents the x-axis direction component of the bending moment, and the y-axis direction bending strain measurement signal output Vy represents the y-axis direction component of the bending moment. Therefore, the angle α can be obtained by obtaining the ratio Vy / Vx between the two, and the angle α is represented by tan ⁻¹ (Vy / Vx). The magnitude of the bending moment is represented by (Vx ² + Vy ² ) ^1/2 .

  Since the direction and magnitude of the bending moment can be determined in this way, the relationship between these and the error component E included in the load output Vw is examined in advance, and the direction of the bending moment when the load cell 2 is used. And the error component E can be compensated using the above relationship.

  For example, in the column-type load cell 2, if a load output Vw when a certain load is applied while the strain-generating body of the load cell 2 is almost upright is Vtrue, the strain-generating portion 4 of the column-type load cell 2 has an α degree. If bending strain is generated in the direction and the same load is applied as in the upright state, the load output Vw is obtained by adding an error component E to the true value Vtrue. In addition, the error component E differs depending on the direction in which the bending moment acts (the direction in which the load cell 2 is inclined) even if the amount of bending strain is the same. Further, the error component E varies depending on the magnitude of the bending strain.

  Therefore, the load cell 2 is preliminarily applied with a rated load, and a plurality of predetermined inclination angles with respect to the z-axis, for example, a total of two types of inclination angles of the first and second types are maintained, respectively. It is made to make one round while rotating at a predetermined angle from α = 0 to 360 degrees. At this time, the load output Vw is measured for each predetermined angle, and the load output Vw when the rated load is applied in a state where the load cell 2 is upright is subtracted from each load output Vw to obtain an error component for each predetermined angle. E is determined.

At the same time, the x-axis direction bending strain measurement signal load output Vx and the y-axis direction bending strain measurement signal load output Vy are measured for each predetermined angle, and the amount of bending strain (Vx ² + Vy ² ) for each predetermined angle. ^{Find 1/2} . When the amount of bending strain is also measured with respect to the actual load cell 2, it varies according to the change in the rotation angle α. However, the fluctuation is relatively small. Therefore, the average value of the amount of bending strain for each predetermined angle when the first tilt angle is given is defined as M1, and the amount of bending strain for each predetermined angle when the second tilt angle is given. The average value is M2, and these M1 and M2 are representative values of bending strain amounts with respect to two types of tilt amounts with respect to the z axis of the load cell 2. FIG. 7 shows an error curve representing the relationship between the error component E and α in the case of the bending strain amounts M1 and M2.

In FIG. 7, two equations representing the relationship between the error component in the bending strain amounts M1 and M2 and α are determined using, for example, the least square method, and the bending moment when the load cell 2 is actually tilted is determined. Using the magnitude of M, the above two equations are modified, the angle α when the load cell 2 is actually tilted is obtained by tan ⁻¹ (Vy / Vx), and that α is the modified equation. The error component E can be obtained by substituting for.

  However, in this embodiment, in order to simplify the calculation, a certain angle range, for example, α = 0 to 90 degrees, 90 to 180 degrees, 180 to 270 degrees, 270 to 0 degrees in FIG. Compensation is performed by regarding each interval as a linear change. The error component when the bending strain amount at α = 0 ° is M1 is e01, and the error component when M = 2 at α = 0 ° is e02, and the measurement is performed with the track 10 actually on the weighing table 8. Assuming that the bending strain amount is Mm, an error component Em0 corresponding to the bending strain amount Mm at α = 0 degree is obtained by linear approximation as shown in FIG. Similarly, error components Em90, Em180, and Em270 at the bending strain amount Mm at α = 90 degrees, 180 degrees, and 270 degrees are obtained.

  Then, the bending strain angle αm measured in a state where the track is actually on the track scale is 0 degree or more and less than 90 degree, 90 degree or more and less than 180 degree, 180 degree or more and less than 270 degree, 270 degree or more and 0 Judge which section is less than degree. For example, if αm is 0 degree or more and less than 90 degrees, a bending strain amount Em0 at α = 0 degree and a bending strain amount Em90 at α = 90 degree are selected, and linear approximation is performed as shown in FIG. The error component Em at the bending strain amount Mm and the angle αm is calculated. Then, the true value Vtrue is calculated by subtracting the error component Em from the load output Vw at that time. In addition, although divided into four sections, it is also possible to determine which section αm belongs to by dividing into smaller sections.

  The load cell 2 is configured as shown in FIG. 10 so that such processing can be performed. That is, the load output Vw of the bridge circuit 7 is amplified by the amplifier 20 and converted to the digital load output DVw by the A / D converter 22. The x-axis direction bending strain measurement signal load output Vx of the bridge circuit 15 and the y-axis bending strain measurement signal load output Vy of the bridge circuit 17 are also amplified by the amplifiers 24 and 26, respectively, and are digitally converted by the A / D converters 28 and 30, respectively. It is converted into an x-axis direction bending strain measurement signal load output DVx and a digital y-axis bending strain measurement signal load output DVy, and input to an error compensation means, for example, the CPU 34 via the input / output circuit 32.

  The CPU 34 performs the above-described calculation while using the memory 36 as a working area according to a storage unit, for example, a program stored in the memory 36.

  For this reason, the memory 36 is preliminarily digitized with error components e01, e02, e11, e12... At α = 0 degrees, 90 degrees, 180 degrees, 270 degrees in the case of the average bending strain amounts M1, M2. It is remembered.

  In a state where the truck 10 is on the weighing platform 8, the digital load output DVw, the digital x-axis direction bending strain measurement signal load output DVx, and the digital y-axis bending strain measurement signal load output DVy are input to the CPU 34.

  The CPU 34 calculates an angle αm representing the direction in which the bending moment acts from the digital x-axis direction bending strain measurement signal load output DVx and the digital y-axis bending strain measurement signal load output DVy. The CPU 34 thus functions as a means for calculating the direction of bending strain from the bending strain measurement signal. The CPU 34 determines whether the calculated αm belongs to a range of 0 ° to less than 90 °, 90 ° to less than 180 °, 180 ° to less than 270 °, and 270 ° to less than 0 °.

  Next, the CPU 34 calculates a bending strain amount Mm from the digital x-axis bending strain measurement signal load output DVx and the digital y-axis bending strain measurement signal load output DVy. Thus, the CPU 34 functions as means for calculating the magnitude of the bending strain from the bending strain measurement signal.

  Then, the CPU 34 reads from the memory 36 error components corresponding to the bending strain amounts M1 and M2 at the angle of one boundary of the determined range. The CPU 34 determines an error component corresponding to the bending strain amount Mm at one boundary from the bending strain amounts M1 and M2 and error components corresponding to the bending strain amounts M1 and M2 by linear approximation. Similarly, an error component corresponding to the bending strain amount Mm at the other boundary is determined by linear approximation.

  The error component Em at the angle αm is calculated using the angle α at both the boundaries, the error component corresponding to the bending strain amount Mm at each of the boundaries, and the previously calculated angle αm. Thus, the CPU 34 functions as an error calculation unit.

  Then, by subtracting this Em from the digital load output DVw, a true value DVtrue from which the influence of bending strain is removed is calculated. That is, the CPU 34 functions as a removing unit.

  Although only one load cell 2 is shown in FIG. 10, actually, the load output Vw, the x-axis direction bending strain measurement signal load output Vx from the bridge circuits 7, 15, 17 of the four load cells 2, The y-axis direction bending strain measurement signal load output Vy is amplified and digitized, supplied to the CPU 34, and the above-described processing is performed for each load cell. Therefore, the memory 36 is obtained by digitizing the error components e01, e02, e11, e12... At α = 0 degrees, 90 degrees, 180 degrees, 270 degrees when the average bending strain amounts M1, M2 are. Stored for each load cell 2. The true values DVtrue calculated in this way are added together, and the added value is displayed on the weight indicator 38.

  In the above embodiment, the strain-generating portion 4 of the columnar load cell 2 has a square cross-sectional shape. However, the present invention is not limited to this, and the cross-sectional shape as shown in FIG. The distortion part 4a can also be used. Even in this case, the axis passing through the center of the strain generating portion 4a is defined as the z axis, the x axis and the y axis perpendicular to the z axis and perpendicular to each other are defined, and the bending strain in the x axis direction is detected so as to be orthogonal to the x axis. The strain gauge 12a for y-axis direction bending strain is bonded to the strain-generating portion 4a so that the strain gauge 12a for use is perpendicular to the y-axis.

  In the above embodiment, the error curve representing the value of the error component corresponding to the amount of bending strain when the track 10 is placed on the weighing platform 8 is determined, and the bending strain on the determined error curve is determined. In order to determine the error component corresponding to the direction (angle α), linear approximation is used. However, the present invention is not limited to this, and various known methods such as the least square method and the Newton interpolation method are used. be able to. However, in that case, it is necessary to increase the number of error curves more than two in the above embodiment.

  In the above-described embodiment, two x-axis direction bending strain detecting strain gauges and two y-axis direction bending strain detecting strain gauges are provided, but at least one each may be provided.

  In the above embodiment, the strain gauge is used to detect the load output Vw, the x-axis direction, and the y-axis direction bending strain. However, the present invention is not limited to this. For example, the output is generated by applying a load such as a semiconductor pressure sensor. As long as it is generated, various types can be used.

  In the above embodiment, the load cell according to the present invention is used for the track scale, but the present invention is not limited to this, and the load cell is a weight measuring device that may be bent and tilted by receiving a bending moment. It can be used for various things.

It is a partial abbreviation perspective view of a distortion part of a load cell of one embodiment of the present invention. FIG. 2 is a circuit diagram of a bridge circuit for load detection of the load cell of FIG. 1. It is a bottom view of the track scale using the load cell of FIG. It is the front view and side view which show the state in which the track | truck was mounted on the track scale of FIG. FIG. 2 is a plan view of a strain generating part in the load cell of FIG. 1. FIG. 2 is a circuit diagram of a bridge circuit for detecting an x-axis direction bending strain of the load cell of FIG. 1. It is an error curve figure which shows the relationship between the bending strain direction and error component in the load cell of FIG. It is explanatory drawing of the determination method of the error component with respect to the arbitrary bending distortion amounts in the load cell of FIG. It is explanatory drawing of the determination method of the error component with respect to the arbitrary bending strain directions in the load cell of FIG. It is a block diagram of the load cell of FIG. It is a top view of the strain generating part of the modification of the load cell of FIG.

Explanation of symbols

2 Load cell 4 Strain portion 6a 6b Strain gauge (weight measuring means)
12a 12b strain gauge (bending strain detection means)
34 CPU (error compensation means)

Claims (5)

A column-type strain body on which a load is loaded;
A weight measurement signal is generated on the strain body, and a weight measurement signal is generated based on the strain generated in the strain body due to the loading of the load. The weight measurement signal is generated by loading the load load on the strain body. A weight measuring means including an error component based on a bending moment acting on the strain body;
A bending strain detecting means provided in the strain generating body and outputting a bending strain measurement signal based on the bending strain generated in the strain generating body by the bending moment;
Error compensation means for inputting the weight measurement signal and the bending strain measurement signal and compensating the error component included in the weight measurement signal;
Load detecting device provided. In the load detection device according to claim 1, the magnitude of the error component varies depending on the direction of the bending strain, even when a bending moment of the same magnitude is acting on the strain body.
The error compensation means includes means for calculating the magnitude and direction of the bending strain from the bending strain measurement signal, calculation means for calculating the error component from the calculated magnitude and direction of the bending strain, and the weight measurement. Removing means for removing the calculated error component from the signal;
Load detecting device provided.   3. The load detection device according to claim 2, wherein at least two sets of the bending strain detection means are arranged so as to form a predetermined angle around the load load direction on the surface of the strain generating body.   4. The load detection device according to claim 3, wherein the at least two sets of bending strain detection means are arranged orthogonally to each other around the load direction, and the means for calculating the magnitude and direction of the bending strain includes: The magnitude of the bending strain is obtained by the square root of the square sum of the bending strain measurement signals from at least two sets of bending strain detection means, and the bending strain is calculated from the ratio of the bending strain measurement signals from the at least two sets of bending strain detection means. Load detection device that determines the direction of the.   5. The load detecting device according to claim 4, wherein the means for calculating the error component includes the error component in various bending strain directions in a state where bending strains having different sizes are generated in the strain generating body. And the calculated bending strain magnitude and direction based on the stored error component, the calculated bending strain magnitude, and the calculated bending strain direction. A load detection device for calculating the corresponding error component.

Images