JP5780916B2 - Creep error compensation function parameter determination method and parameter determination apparatus - Google Patents

Description

  The present invention relates to a parameter determination method and a parameter determination apparatus for a creep error compensation function for compensating for a creep error in a load signal output from a load cell.

  When a load is applied to a load detector, for example, a robust load cell, the load cell bends according to the magnitude of the loaded load, and also causes a creep phenomenon. When the load is removed, the load cell attempts to return to the original state, but at this time, a creep recovery phenomenon occurs. As a result, it is known that a creep error and a creep recovery error occur in the load signal output from the load cell.

  Here, due to the characteristics of each load cell, when the load cell is continuously loaded with a load, a creep error (positive creep error) in which the load signal increases with the passage of time occurs. A creep error (negative creep error) in which the load signal decreases accordingly may occur.

  FIG. 5 is a diagram schematically illustrating an example of a change with time of the load signal when a positive creep error occurs. In this case, when a load is applied to the load cell, the load signal w immediately rises to the load signal value Tw1 corresponding to the initial deflection amount of the load cell, and then the load signal gradually increases with the lapse of time t due to the creep phenomenon. Creep error that increases is generated, and when the load continues to be applied, it finally converges to a certain load signal value Twe. When the load is removed from the load cell, the load signal w first falls by the load value corresponding to the initial return amount of the load cell, and then the load signal w gradually increases with the elapse of time t due to the creep recovery phenomenon. A creep recovery error that decreases to zero, which eventually becomes zero.

  FIG. 6 is a diagram schematically illustrating an example of a change with time of the load signal when a negative creep error occurs. In this case, when a load is applied to the load cell, the load signal w immediately rises to the load signal value Tw1 corresponding to the initial deflection amount of the load cell, and then the load signal gradually increases with the lapse of time t due to the creep phenomenon. A creep error that decreases to a constant value is generated, and when the load continues to be applied, the load signal value Twe eventually converges. When the load is removed from the load cell, the load signal w first falls by the load value corresponding to the initial return amount of the load cell, and then the load signal w gradually increases with the elapse of time t due to the creep recovery phenomenon. Creep error that increases, and this error eventually becomes zero.

  In order to reduce such a creep error, improvements have been made to materials and structures constituting the load cell, but this is not sufficient.

  Therefore, in recent years, a method has been developed in which a load signal including a creep error is input to a digital circuit, and a component of the creep error is removed from the load signal including the creep error using a compensation function (for example, Patent Documents). 1-4).

JP 2011-33374 A JP 2009-53211 A JP 2009-63403 A JP-A-11-2573

  The compensation function for compensating for the above-described creep error uses a value determined based on the creep error saturation amount (creep error maximum generation amount) Ce generated when a known load is applied as a parameter. There are many. In order to determine parameters of such a compensation function, a load cell is connected to a test machine, and a known load (weight) is connected to the load cell, and a predetermined saturation time Te (for example, 30 minutes) determined as a time at which a creep error is saturated. Load), and the load signal values Tw1 and Twe in that case are measured by a testing machine, and the difference between these load signal values Tw1 and Twe is calculated to calculate the creep error saturation amount Ce (for example, Ce = Twe−Tw1). Etc. Here, when a known load (weight) is applied to the load cell, the load signal output from the load cell is not stable including a large vibration component with respect to the creep error, and the creep error saturation amount Ce is obtained. It is difficult to accurately measure the load signal values Tw1 and Twe. In particular, the load signal immediately after the load is applied includes a larger vibration component with respect to the creep error. Possible causes of unstable load signals include, for example, loading a weight by suspending a weight on the load cell. When a load is applied to this load cell, vibration caused by the impact of the load or vibration due to the influence of the surrounding environment Etc. are considered. Vibration due to the influence of the surrounding environment includes vibration due to other equipment operating in the vicinity of the testing machine, and the installation location of the testing machine (the foundation is solid or the first floor of the building It depends on whether it is installed on the upper floor or the like.

  FIG. 7 is a diagram showing a ratio (error ratio) of measured values of creep error and creep recovery error included in an actual load signal when a positive creep error as shown in FIG. 5 occurs.

  As shown in FIG. 7, when a creep error occurs, that is, when a load is applied, the load signal contains a lot of vibration components. Therefore, it is impossible to measure the exact value of the load signal value Tw1 at the start of the load and the load signal value Twe at the end of the load shown in FIG. 5, and the creep error saturation amount Ce, which is the difference between them, cannot be accurately measured. It is not easy to ask for.

  Similarly, even when a negative creep error as shown in FIG. 6 occurs, the load signal when a load is applied contains a large amount of vibration components with respect to the creep error, and the creep error saturation amount Ce is accurately determined. It is not easy to ask for.

  If the creep error saturation amount Ce is not accurately obtained, the parameter value of the compensation function determined based on the creep error saturation amount Ce is not accurate, and the compensation amount obtained from the compensation function is not accurate. Creep error cannot be compensated accurately. Further, the creep error saturation amount Ce differs for each load cell and needs to be obtained individually for each load cell.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a parameter determination method and a parameter determination device for a creep error compensation function for accurately compensating a creep error.

  In order to achieve the above object, a method for determining a parameter of a creep error compensation function according to an aspect of the present invention includes a compensation function for calculating a creep error included in a load signal output from a load detector. When determining the parameter used in the equation, which is determined based on a saturation amount of creep error when a predetermined test load is applied to the load detector, the test load is applied to the load detector. The difference between the load signal value immediately after unloading after loading for the time required for saturation of the creep error and the load signal value before loading the test load is calculated, and the difference is calculated as the creep error. The parameter is determined by regarding the amount of saturation.

  According to this parameter determination method, the difference between the load signal value immediately after loading the test load for the time required for saturation of the creep error and unloading it and the load signal value before loading the test load is calculated. The parameter is determined by regarding this difference as a saturation amount of the creep error. Since the difference is calculated using a load signal value at the time of no load when the load signal is stable, an accurate value can be calculated, and a parameter based on the value is also an accurate value. Thus, since the parameter of the compensation function can be determined to an accurate value, the creep error can be accurately compensated by compensating the creep error using the compensation function including this parameter.

  A creep error compensation function parameter determination method according to another aspect of the present invention is a parameter used in a compensation function equation for calculating a creep error included in a load signal output from a load detector, When determining the parameter determined based on the amount of saturation of the creep error when a predetermined test load is applied to the load detector, the time required for the creep error to saturate the test load in the load detector The difference between the load signal value immediately after unloading after loading for a shorter first time and the load signal value before loading the test load is calculated, and the test load is calculated as the difference. The parameter is determined based on a saturation amount of the creep error that is predicted when the creep error is regarded as a time load.

  According to this parameter determination method, the load signal value immediately after unloading the test load after loading the test load for a first time shorter than the time required for the creep error to saturate and the load signal before loading the test load. A difference from the value is calculated, this difference is regarded as a creep error when the test load is applied for the first time, and a parameter is determined based on a saturation amount of the creep error predicted in that case. Since the difference is calculated using the load signal value when the load signal is stable and when there is no load, an accurate value can be calculated, and the creep error saturation amount and parameter predicted based on the value can be calculated. Is an accurate value. Thus, since the parameter of the compensation function can be determined to an accurate value, the creep error can be accurately compensated by compensating the creep error using the compensation function including this parameter. In addition, the time during which the test load is applied can be greatly shortened.

  Further, in this case, the difference is regarded as a creep error when the test load is applied for the first time, and the prediction is made so that the creep error matches the creep error calculated from the equation of the compensation function. It is preferable to calculate the saturation amount of the creep error.

  A creep error compensation function parameter determination device according to another aspect of the present invention is a parameter used in a compensation function equation for calculating a creep error included in a load signal output from a load detector, A parameter determining means for determining the parameter determined based on a saturation amount of a creep error when a predetermined test load is applied to the load detector; and a creep error for saturating the test load in the load detector. Difference calculating means for calculating a difference between a load signal value immediately after unloading after loading for a required time and a load signal value before loading the test load, and the parameter determining means includes the difference The parameter is determined by regarding the difference calculated by the calculation means as a saturation amount of the creep error.

  According to this configuration, the difference between the load signal value immediately after unloading the test load after loading the test load for the time required for the creep error to be saturated and the load signal value before loading the test load is calculated, The parameter is determined by regarding this difference as the amount of saturation of the creep error. Since the difference is calculated using a load signal value at the time of no load when the load signal is stable, an accurate value can be calculated, and a parameter based on the value is also an accurate value. Thus, since the parameter of the compensation function can be determined to an accurate value, the creep error can be accurately compensated by compensating the creep error using the compensation function including this parameter.

  A creep error compensation function parameter determination device according to another aspect of the present invention is a parameter used in a compensation function equation for calculating a creep error included in a load signal output from a load detector, A parameter determining means for determining the parameter determined based on a saturation amount of a creep error when a predetermined test load is applied to the load detector; and a creep error for saturating the test load in the load detector. Difference calculating means for calculating a difference between a load signal value immediately after unloading after loading for a first time shorter than a required time and a load signal value before loading the test load; The determining means is predicted when the difference calculated by the difference calculating means is regarded as a creep error when the test load is loaded for the first time. Based on the saturation amount of the serial creep error is configured to determine the parameters.

  According to this configuration, the load signal value immediately after unloading the test load after loading the test load for a first time shorter than the time required for the creep error to be saturated, and the load signal value before loading the test load, The difference is calculated, and this difference is regarded as a creep error when the test load is applied for the first time, and the parameter is determined based on the saturation amount of the creep error predicted in that case. Since the difference is calculated using the load signal value when the load signal is stable and when there is no load, an accurate value can be calculated, and the creep error saturation amount and parameter predicted based on the value can be calculated. Is an accurate value. Thus, since the parameter of the compensation function can be determined to an accurate value, the creep error can be accurately compensated by compensating the creep error using the compensation function including this parameter. In addition, the time during which the test load is applied can be greatly shortened.

  In this case, the parameter determination means regards the difference as a creep error when the test load is applied for the first time, and the creep error coincides with a creep error calculated from the equation of the compensation function. Thus, it is preferable to be configured to calculate a saturation amount of the predicted creep error.

  The present invention has the above-described configuration, and has an effect that it is possible to provide a parameter determination method and a parameter determination device for a creep error compensation function for accurately compensating a creep error.

It is a block diagram which shows one structural example of the test apparatus used in order to determine the parameter of the creep error compensation function of embodiment of this invention. It is a figure which shows typically an example of a time-dependent change of the load signal in case the positive creep error used in order to demonstrate the parameter determination method of the creep error compensation function in embodiment of this invention arises. It is a figure which shows typically an example of a time-dependent change of the load signal in case the negative creep error used for demonstrating the parameter determination method of the creep error compensation function in embodiment of this invention occurs. (A) is a flowchart which shows an example of operation | movement of the test apparatus for enforcing the parameter determination method of the creep error compensation function by a 1st method, (B) is the creep error compensation function of a 2nd method. It is a flowchart which shows an example of operation | movement of the test apparatus for implementing the parameter determination method. It is a figure which shows typically an example of the time-dependent change of the load signal in case a positive creep error arises. It is a figure which shows typically an example of the time-dependent change of the load signal in case a negative creep error arises. It is a figure which shows the ratio (error ratio) of the measured value of the creep error contained in the actual load signal when a positive creep error arises, and a creep recovery error.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. In the following description, the same or corresponding elements are denoted by the same reference symbols throughout all the drawings, and redundant description thereof is omitted. Further, the present invention is not limited to the following embodiment.

(Embodiment)
FIG. 1 is a block diagram showing a configuration example of a test apparatus (parameter determination apparatus) used for determining parameters of a creep error compensation function according to an embodiment of the present invention.

  The test apparatus includes a control device 3, a display device 4, and an operation input unit 5. However, the amplification and A / D conversion unit 2 may or may not be included in the test device. Also good. That is, the amplification and A / D converter 2 may be built in the test apparatus or may be provided attached to the load cell 1.

  A control device 3 is connected to a load cell 1 that is a load detector via an amplification and A / D conversion unit (amplification unit and A / D conversion unit) 2. The load cell 1 amplifies and outputs an analog load signal having a voltage corresponding to the applied load to the A / D converter 2. The load cell 1 is, for example, a strain gauge type load cell, and is a metal strain generating body, for example, four sets of strain gauges attached to the strain generating body, and a bridge output circuit in which the four sets of strain gauges are bridge-connected. And is configured to output a voltage corresponding to the amount of strain of the strain generating body due to the load being applied.

  The amplification and A / D conversion unit 2 amplifies the analog load signal input from the load cell 1 in the amplification unit, and samples the amplified analog load signal in a predetermined sampling period (d) in the A / D conversion unit. , Converted into a digital load signal and output to the control device 3.

  The control device 3 is composed of, for example, a microcontroller and functions as a parameter determination unit, a difference calculation unit, and the like. As the control unit 31, for example, a CPU of a microcontroller or the like is used. As the storage unit 32, for example, an internal memory of a microcontroller is used. The control unit 31 and the storage unit 32 are connected to each other. The storage unit 32 stores a control program (the CPU execution program). The control unit 31 executes processing such as calculation and controls the display device 4 by executing a control program stored in the storage unit 32.

  The display device 4 is a display having a display screen, and displays, for example, a graph showing a change over time of the load signal input from the amplification and A / D conversion unit 2 under the control of the control device 3 (control unit 31). can do.

  The operation input unit 5 includes, for example, a keyboard, and can input various instructions to the test apparatus such as an instruction to start the operation of the test apparatus to the control device 3 (control unit 31). Based on the instruction input from the operation input unit 5, the operation of the test apparatus including the control device 3 is controlled.

  In the following description, the load signal is a digital load signal input to the control device 3 (control unit 31).

  A method for determining a parameter of the creep error compensation function in the present embodiment will be described with reference to FIGS.

  FIG. 2 is a diagram schematically showing an example of a change in load signal with time when a positive creep error occurs, and FIG. 3 schematically shows an example of a change in load signal with time when a negative creep error occurs. FIG. 2 and 3 each show an example when a test load is applied when determining a parameter (creep ratio β) of the compensation function.

  The compensation function for compensating for the creep error of the load cell 1 applies, for example, the load signal value increment wi immediately after an arbitrary load (a load to be measured) is applied to the load cell 1 and the same load. It can be shown as a function f (wi, t) with time (load load time) t as a variable. Note that the amount of increase wi of the load signal value immediately after the load of the measurement target is applied is the load signal value (w1) immediately after the load of the measurement target is applied, from the load signal value (w1) immediately before the load is applied. It is calculated by subtracting the load signal value (w0) (wi = w1-w0).

If the load signal before creep error compensation at load load time t is wa and the load signal after compensation is wx,
wx = wa−f (wi, t)
It becomes. That is, the compensation function f (wi, t) is a function for calculating the creep error, and the load signal value (wx) in which the creep error is compensated can be obtained by the above equation.

  Here, as an example of the compensation function f (wi, t), as disclosed in Patent Document 1 (Japanese Patent Laid-Open No. 2011-33374), for example, the compensation function f (wi, t) is linear. As a result of weighted addition of the function g (wi, t) and the saturation function h (wi, t), for example, the following expression (1) is given.

f (wi, t) = k.g (wi, t) + (1-k) .h (wi, t) (1)
here,
g (wi, t) = C (wi) · (t / Te) (2)
h (wi, t) = C (wi) · {1-exp (−t / τ)} (3)
K is a weighting factor (0 <k <1), τ is a time constant of the first-order lag response of the creep error, and Te is a time required for the creep error to be saturated after the load is applied. This is a predetermined saturation time (for example, 30 minutes) set in advance. Here, a step response function of a first-order lag system is used as the saturation function h (wi, t), but is not limited to this, and the absolute value of the amount of change per unit time decreases with time. Functions can be used. In this specification, as also shown in the above (3), for example, the ^{e y,} denoted as exp (y).

  Note that the load load time t corresponds to a time until a certain sampling time (0 ≦ t ≦ Te) starting from the load load start time (t = 0), and this sampling time (t) is a sampling period (sampling). If the interval is d and the sampling number from the beginning of load loading (n = 0) is n (n = 0, 1, 2, 3,...), It can also be expressed as t = n · d.

  Here, C (wi) is a creep error saturation amount (maximum amount of creep error) when a load to be measured is applied, and is represented by the following equation, for example.

C (wi) = β · wi = β (w1-w0)
Here, β is a creep ratio, and is a parameter for calculating a creep error saturation amount (maximum amount of creep error) when a load to be measured is applied. This creep ratio β is a ratio of a creep error saturation amount to a load to be applied (a load not including a creep error) wi, and may be considered to be constant regardless of the magnitude of the applied load.

  This creep ratio β is defined by the following equation, for example.

β = Ce / (Tw1-Tw0)
Here, Ce is a creep error saturation amount (maximum amount of creep error) when a known test load (for example, a 10 kg weight) is applied to the load cell 1. Tw1 is a load signal value immediately after the test load is applied, and is regarded as not including a creep error. Further, Tw0 is a load signal value when there is no load before the test load is applied. The creep error saturation amount Ce is expressed by the following equation, for example.

Ce = Twe-Tw1
Here, Twe is a load signal value when the same test load is applied for a predetermined creep error saturation time Te and the creep error is saturated or the creep error is considered to be saturated. These Tw1, Tw0, Twe, and Ce are also shown in FIGS.

  Here, as described above, since the load signal changes so as to vibrate while the load is being applied and is not stable, it is difficult to accurately measure the load signal values Tw1 and Twe, and the creep error is saturated. It is not easy to accurately determine the quantity Ce. Further, the creep ratio β is determined based on the creep error saturation amount Ce when a test load is applied according to the above-described definition formula. Therefore, if the creep error saturation amount Ce is not an accurate value, the creep ratio β is also accurate. It's not a good value.

  Therefore, in this embodiment, when a load is applied to the load cell, a creep error occurs with time, but immediately after unloading, a creep recovery error corresponding to the immediately preceding creep error occurs (in other words, Note that the previous creep error remains), and that the load signal is stable when the load is not loaded (including immediately after unloading) (for example, in FIG. 7). Focusing on the creep recovery error portion), the creep error saturation amount Ce and the creep ratio β are obtained by the parameter determination methods (first method and second method) described below.

  The test apparatus according to the present embodiment is configured to be able to select and carry out either the first method or the second method described below by operating the operation input unit 5. It may be configured to perform only the method.

  As a parameter determination method in the present embodiment, first, in the first method, after a test load is loaded on the load cell 1 for a predetermined creep error saturation time Te, the load is unloaded, and the creep recovery error De immediately after this unloading is calculated. The creep ratio β is obtained by regarding the value as the value of the creep error saturation Ce. Here, the creep recovery error De immediately after unloading is a residual component of the creep error (here, the creep error saturation amount Ce) that occurs immediately before unloading. Therefore, the creep recovery error De is defined as the creep error saturation amount Ce. Can be considered.

  That is, in the first method, the load signal value Tw0 before the test load is applied and the test load is applied for a predetermined saturation time Te and the creep error is saturated. The load signal value Tw2 immediately after unloading is obtained, and the load signal value Tw0 before loading is subtracted from the load signal value Tw2 immediately after unloading to obtain the creep recovery error De (= Tw2-Tw0) immediately after unloading.

  The creep recovery error De thus obtained is regarded as the creep error saturation amount Ce, and the creep ratio β is calculated by the following equation.

β = De / (Tw1-Tw0)
Here, as the load signal value Tw1, an average value of a plurality of continuous load signal values sampled immediately after the test load is applied is calculated, and the average value is used. Even if the load signal value Tw1 includes a creep error, the creep error is very small compared to the value of Tw1−Tw0, and there is no practical problem in calculating the creep ratio β. For example, as shown in FIG. 7, the creep error is a very small value of about 0.01% of the value of Tw1−Tw0 if the saturation amount is averaged.

  FIG. 4A is a flowchart showing an example of an operation (test operation) of a test apparatus for implementing the creep error compensation function parameter determination method according to the first method. This operation is realized by processing of the control device 3. Here, all the information to be stored is stored in the storage unit 32.

  When the test apparatus starts a test operation, a load signal is input from the amplification and A / D converter 2 to the control apparatus 3 at a predetermined sampling period (for example, 1 ms).

  First, after starting the test, the control device 3 acquires and stores the load signal value Tw0 before the test load is applied to the load cell 1 (step S1).

  Here, the person in charge of the test loads a test load on the load cell 1 and unloads the load after a predetermined saturation time Te has elapsed.

  Next, the control device 3 determines whether or not the sequentially inputted load signal value is equal to or greater than the load load determination threshold wt (step S2), and the test load is applied to exceed the load load determination threshold wt. Then, the load signal value Tw1 at that time is stored (step S3). Here, the load load determination threshold wt is a set value stored in the storage unit 32 in advance. For example, when the test load is set to 10 kg, the load load determination threshold value wt corresponds to about 20 g. Corresponds to a small value. In addition, since the load signal when the load is applied varies greatly, it is continuously input immediately after the test load is applied (that is, immediately after the load signal value becomes equal to or greater than the load load determination threshold wt). An average value of a plurality of load signal values is calculated, and the average value is stored as a load signal value Tw1 immediately after the test load is applied.

  Next, it is determined whether or not the sequentially inputted load signal value is less than the load load determination threshold wt (step S4). When the test load is unloaded and becomes less than the load load determination threshold wt, The load signal value Tw2 is stored (step S5). Since the load signal immediately after unloading is stable, the load signal value input first immediately after unloading is stored as Tw2.

  Next, the creep recovery error De immediately after unloading is calculated from the load signal value Tw2 immediately after unloading the test load and the load signal value Tw0 before being loaded (step S6). Here, Tw2−Tw0 = De.

Next, the creep ratio β as a parameter is set to β = De / (Tw1−Tw0).
Is calculated and stored (step S7).

  In this first method, a creep recovery error De (= Tw2−Tw0) immediately after the test load is applied for the creep error saturation time Te and then unloaded is obtained, and this creep recovery error De is calculated as the creep error saturation amount Ce. It is assumed that the creep ratio β is calculated. Here, since the load signal values Tw2 and Tw0 used for calculation of the creep recovery error De are load signal values at the time of no load, the load signal is stable and an accurate value can be acquired, and accurate creep recovery is performed. The error De can be calculated. Therefore, the correct creep ratio β can be calculated by calculating the creep ratio β using the creep recovery error De immediately after unloading instead of the creep error saturation amount Ce. Further, by using a compensation function including the creep ratio β calculated as described above as a parameter, the creep error can be accurately compensated.

  Next, in the second method, a test load is applied to the load cell 1 for a time Ts shorter than the creep error saturation time Te (0 <Ts <Te), then unloaded, and a creep recovery error Ds immediately after this unloading. (= Tws−Tw0) is obtained, and the creep error saturation amount Ce and the creep ratio β are calculated using this creep recovery error Ds and a creep error saturation amount calculation formula described later. A load signal when the test load is unloaded after the above time Ts is indicated by a one-dot chain line in FIGS. The time Ts (hereinafter referred to as “test load loading time Ts”) may not be a predetermined time.

  First, a method for deriving a creep error saturation amount calculation formula and the like will be described.

Substituting Equations (2) and (3) into Equation (1) above,
f (wi, t) = C (wi). [k (t / Te) + (1-k) {1-exp (-t / τ)}]
It becomes.

  Here, as described above, the creep error saturation time Te is a predetermined time (for example, 30 minutes) that is predetermined as a time during which the creep error is saturated, and the weighting coefficient k and the time constant τ have already been determined. The weighting factor k and the time constant τ are based on, for example, a measurement record of a change over time of a load signal when a test load is applied to a plurality (about 10) of load cells similarly manufactured using the same component. Is determined as an average value. In other words, the weighting factor k and the time constant τ can be applied uniformly to each load cell manufactured using the same component, but the creep error saturation amount (maximum generation amount) differs for each load cell. Is. There are various methods for determining the weighting factor k and the time constant τ as described in, for example, Patent Document 1 (Japanese Patent Laid-Open No. 2011-33374), but this is not the gist of the present invention. The details are omitted.

Here, if the increase amount of the load signal value immediately after the test load is applied is Twi (= Tw1−Tw0), and the creep error occurring in the test load load time Ts is Cs, a compensation function is used.
Cs = f (Twi, Ts)
= C (Twi). [K (Ts / Te) + (1-k) {1-exp (-Ts / τ)}]
It can be expressed as.

When this equation is transformed into an equation for calculating C (Twi),
C (Twi) = Cs / [k (Ts / Te) + (1-k) {1-exp (-Ts / τ)}]
It becomes.

This C (Twi) is a creep error saturation amount Ce when a test load is applied. Therefore,
Ce = Cs / [k (Ts / Te) + (1-k) {1-exp (-Ts / τ)}] (4)
It becomes. This equation (4) can be said to be a prediction equation of the creep error saturation amount Ce derived from the compensation function.

  Here, since it is not easy to accurately obtain the creep error Cs at the test load load time Ts (time ts) from the load signal having a lot of vibration components, the test load load time Ts (0 <Ts) is used instead of the creep error Cs. <Te) The creep recovery error Ds (= Tws−Tw0) immediately after the lapse is calculated and used.

That is, when the creep recovery error Ds is regarded as the creep error Cs, the following equation (5) is obtained from the equation (4).
Ce = Ds / [k (Ts / Te) + (1-k) {1-exp (-Ts / τ)}] (5)
The creep error saturation amount Ce is calculated using the equation (5) as a creep error saturation amount calculation formula. That is, the creep error saturation amount Ce calculated here is a predicted value calculated by regarding the creep recovery error Ds immediately after the test load loading time Ts as the creep error Cs. The test load loading time Ts is measured by the control device 3. Here, the creep recovery error Ds immediately after unloading is a residual component of the creep error (here, the creep error Cs) occurring immediately before unloading, and therefore, the creep recovery error Ds is regarded as the creep error saturation amount Cs. Can do.

  And creep ratio (beta) is computable by the following (6) Formula or (7) Formula.

β = Ce / (Tw1-Tw0) (6)
β = Ds / [k (Ts / Te) + (1-k) {1-exp (-Ts / τ)}] (Tw1-Tw0)
... (7)
Here, as the load signal value Tw1, an average value of a plurality of continuous load signal values sampled immediately after the test load is applied is calculated, and the average value is used. Even if the load signal value Tw1 includes a creep error, the creep error is very small compared to the value of Tw1−Tw0, and there is no practical problem in calculating the creep ratio β.

  FIG. 4B is a flowchart showing an example of the operation (test operation) of the test apparatus for implementing the creep error compensation function parameter determination method according to the second method. This operation is realized by processing of the control device 3. Here, all the information to be stored is stored in the storage unit 32.

  When the test device starts a test operation, a load signal is input from the amplification and A / D converter 2 to the control device 3 at a predetermined time interval (for example, 1 ms).

  Steps S11 to S14 are the same as steps S1 to S4 in the case of FIG. However, here, the person in charge of the test loads the test load on the load cell 1, sets the test load loading time Ts to be shorter than the predetermined creep error saturation time Te, and unloads after loading for about 5 to 10 minutes, for example. The control device 3 is configured to measure the test load loading time Ts.

  Next, when the test load is unloaded and the load signal value becomes less than the load load determination threshold wt, the control device 3 stores the load signal value at that time, that is, the load signal value Tws immediately after unloading (step). S15). At this time, after the test is started, the control device 3 determines the time from when the load signal value becomes equal to or higher than the load load determination threshold wt to less than the load load determination threshold wt, for example, ts−t1 as the test load load. Calculated as time Ts.

  Next, the creep recovery error Ds immediately after unloading is calculated from the load signal value Tws immediately after unloading the test load and the load signal value Tw0 before being loaded (step S16). Here, Tws−Tw0 = Ds.

  Next, the creep error saturation amount Ce is calculated using the creep error saturation amount calculation formula of the above-described equation (5) (step S17).

Next, the creep ratio β as a parameter is expressed by β = Ce / (Tw1−Tw0) in the equation (6).
Is calculated and stored (step S18).

  Instead of calculating the creep error saturation amount Ce (step S17), in step S18, the creep ratio β may be calculated using the above-described equation (7).

  In this second method, a creep recovery error Ds (= Tws−Tw0) immediately after unloading after applying a test load for a time shorter than the creep error saturation time Te (test load load time Ts) is obtained. The creep recovery error Ds is regarded as the creep error Cs, the creep error saturation amount Ce is calculated, and the creep ratio β is calculated. Since the load signal values Tws and Tw0 used for the calculation of the creep recovery error Ds are the load signal values when there is no load, the load signal is stable and an accurate value can be obtained, and the accurate creep recovery error Ds can be obtained. Can be calculated. Therefore, it is possible to calculate an accurate creep error saturation amount Ce and an accurate creep ratio β. Further, by using a compensation function including the creep ratio β calculated as described above as a parameter, the creep error can be accurately compensated.

  According to the second method, the test time (the load time of the test load) for determining the parameter (creep ratio β) can be significantly shortened compared to the first method.

  In the present embodiment, the load load determination threshold wt may be calculated by adding a predetermined value to the load signal value Tw0 before the test load is applied.

  The compensation function f (wi, t) used in the present embodiment is a linear function g (wi, t) and a saturation function h in which the absolute value of the amount of change per unit time decreases with time. (Wi, t) is weighted and added. In this case, since the compensation function is calculated by weighting and adding the linear function g (wi, t) and the saturation function h (wi, t), the weighting coefficient k is determined according to the creep error characteristic of each load cell 1. By setting the value, the creep error can be compensated with higher accuracy. For example, for a load cell that outputs a load signal having a creep error characteristic with a large increase rate characteristic immediately after load application and a small gradual increase characteristic, the weight (k) of the linear function g (wi, t) is reduced and the saturation function By increasing the weight (1-k) of h (wi, t), the generated creep error can be approximated with high accuracy. For a load cell that outputs a load signal having a creep error characteristic that has a small increase rate characteristic immediately after load application and a large incremental characteristic, the weight (k) of the linear function g (wi, t) is increased and the saturation function is increased. By reducing the weight (1-k) of h (wi, t), the generated creep error can be approximated with high accuracy.

  In the present embodiment, the compensation function f (wi, t) obtained by weighted addition of the linear function and the saturation function is used as the compensation function, but the present invention is not limited to this. The present invention is a compensation function for calculating the creep error included in the load signal, and is a compensation function having a parameter determined based on the creep error saturation amount Ce when a predetermined test load is applied. The present invention can be applied as a method and a device for determining the parameter.

  INDUSTRIAL APPLICABILITY The present invention is useful as a parameter determination method and parameter determination device for a creep error compensation function for accurately compensating a creep error.

DESCRIPTION OF SYMBOLS 1 Load cell 2 Amplification and A / D conversion part 3 Control apparatus 4 Display apparatus 5 Operation input part

Claims (6)

A parameter used in a compensation function formula for calculating a creep error included in a load signal output from a load detector, and a saturation amount of a creep error when a predetermined test load is applied to the load detector. In determining the parameter determined based on:
Calculate the difference between the load signal value immediately after unloading the test load after the test load is loaded for the time required for the creep error to saturate and the load signal value before loading the test load. The parameter is determined by regarding the difference as the amount of saturation of the creep error.
Method for determining parameters of creep error compensation function. A parameter used in a compensation function formula for calculating a creep error included in a load signal output from a load detector, and a saturation amount of a creep error when a predetermined test load is applied to the load detector. In determining the parameter determined based on:
A load signal value immediately after unloading the test load after loading the test load for a first time shorter than a time required for saturation of a creep error and a load signal value before loading the test load And determining the parameter based on a saturation amount of the creep error that is predicted when the difference is regarded as a creep error when the test load is loaded for the first time period.
Method for determining parameters of creep error compensation function. The difference is regarded as a creep error when the test load is applied for the first time, and the creep error is calculated so that the creep error matches the creep error calculated from the equation of the compensation function. Calculate saturation,
The creep error compensation function parameter determination method according to claim 2. A parameter used in a compensation function formula for calculating a creep error included in a load signal output from a load detector, and a saturation amount of a creep error when a predetermined test load is applied to the load detector. Parameter determining means for determining the parameter determined based on:
Calculate the difference between the load signal value immediately after unloading the test load after the test load is loaded for the time required for the creep error to saturate and the load signal value before loading the test load. Difference calculating means,
The parameter determination means is configured to determine the parameter by regarding the difference calculated by the difference calculation means as a saturation amount of the creep error.
A parameter determination device for creep error compensation function. A parameter used in a compensation function formula for calculating a creep error included in a load signal output from a load detector, and a saturation amount of a creep error when a predetermined test load is applied to the load detector. Parameter determining means for determining the parameter determined based on:
A load signal value immediately after unloading the test load after loading the test load for a first time shorter than a time required for saturation of a creep error and a load signal value before loading the test load Difference calculating means for calculating the difference between and
The parameter determination means is based on a saturation amount of the creep error predicted when the difference calculated by the difference calculation means is regarded as a creep error when the test load is loaded for the first time. Configured to determine the parameter;
A parameter determination device for creep error compensation function. The parameter determining means regards the difference as a creep error when the test load is applied for the first time, and makes the prediction so that the creep error matches the creep error calculated from the equation of the compensation function. Configured to calculate a saturation amount of the creep error to be
The creep error compensation function parameter determination apparatus according to claim 5.

Images