JP3208826B2 - Control device for structural testing machine - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control apparatus for a structural test machine for performing a static strength test, a fatigue test, and the like of a plurality of components constituting a structure such as an aircraft and a building. It relates to correction of a load signal during feedback control.

[0002]

2. Description of the Related Art Conventionally, as a control device of a structural test machine for correcting a load signal during this kind of feedback control,
The one shown in FIG. 5 is known. Hereinafter, description will be made with reference to the drawings.

[0003] Reference numeral 21 denotes a signal converter for converting a response signal measured from each component of the structure into a digital signal.
Detects a peak (crest) and a trough (trough) of the response signal converted into a digital signal by the signal conversion unit 21, obtains an amplitude of the response signal from the detected peak and the trough, and calculates a difference between the amplitude of the target signal ( response signal analyzer for detecting ε), 23
Is a correction operation unit that performs a correction calculation of a load signal to be output next time based on the difference ε obtained by the response signal analysis unit 22,
Reference numeral 24 denotes a storage device in which the correction coefficient (n) used for the correction calculation of the correction calculation unit 23 is stored. Reference numeral 25 denotes a load that outputs the load signal calculated by the correction calculation unit 23 to each component of the structure. It is a signal output unit.

FIG. 6 is a diagram showing a difference between a measured response signal and a target signal. The waveform of the solid line in FIG. 6 is the target signal, and the waveform of the dotted line is the response signal. FIG. 6 (a)
FIG. 6 shows a case where the response signal is smaller than the target signal.
(B) shows a case where the response signal is larger than the target signal.

[0005] The above-described response signal analyzing unit 22 obtains an amplitude R _esp from the peak and trough of the measured response signal, determines the difference between the target signal amplitude _{R ange ε ((R ange -R} esp) / 2) . Based on the correction coefficient n stored in the storage device 24 and ε obtained by the response signal analysis unit 22, the correction calculation unit 23 calculates the correction calculation represented by the following equation (1) as the peak of the response signal. Perform each time a trough is detected. Load signal = Target signal + (1 / n) × ε (1) That is, the correction operation unit 23 gradually corrects the deviation between the response signal and the target signal, and obtains a response signal close to the target signal. Thus, the load signal is obtained by the above-described correction calculation.

[0006]

However, the prior art having such a structure has the following problems.

That is, in a test of this type of structure, target signals having different amplitudes are continuously applied to the constituent elements, and if the previous target signal is different, the current target signal is different. It is empirically recognized that a large difference occurs between the measured response signal and the target signal even if the values are the same. For example, a response signal measured when a target signal corresponding to 7 t is added as the previous load, and then a target signal corresponding to 10 t is added as the current load, and a target signal corresponding to 6 t as the previous load are calculated. In addition, the difference between the target signal corresponding to 10 t and the response signal measured when a target signal corresponding to 10 t is subsequently added as the current load is different. Although the difference between the measured response signal and the target signal differs due to the difference in the combination pattern (hereinafter referred to as a load pattern) between the previous target signal and the current target signal, As described above, since the target signal is corrected with a constant correction coefficient regardless of the load pattern to generate the load signal, there is a problem that a load faithful to the target signal cannot be given to the load point. is there.

Further, in this type of structure test, since loads are applied in parallel to the load points of the components, the load points interfere with each other. The interference between the load points may cause a difference between the response signal and the target signal. In addition, since the interference between the load points is not always constant, even if the load signal is created by correcting the target signal with a fixed correction coefficient as in the conventional example, the load signal is faithful to the target signal. It is difficult to apply a load.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and has been made without being affected by differences in load patterns and mutual interference between load points in a test on a plurality of components of a structure. An object of the present invention is to provide a control device for a structural test machine capable of accurately correcting a load signal and improving the accuracy of feedback control at each load point.

[0010]

The present invention has the following configuration to achieve the above object. That is, the present invention provides load signal output means for individually outputting a load signal to each of a plurality of load points of a structure under test, and individually input response signals from a plurality of load points of the structure. Response signal input means, correction coefficient storage means for storing a correction coefficient for correcting a load signal for each load pattern which is a combination of the previous target signal and the current target signal, and input from the response signal input means. A new correction coefficient is calculated based on the difference between the obtained response signal and the corresponding target signal, and a correction coefficient updating unit that updates the correction coefficient stored in the correction coefficient storage unit; and a correction coefficient storage unit. And a calculating means for calculating a load signal according to a load pattern to be generated next time using the stored correction coefficient.

[0011]

The operation of the present invention is as follows. When outputting a load signal of a certain type of load pattern, the calculating means reads a correction coefficient corresponding to the load pattern from the correction coefficient storage means, and corrects the target signal using the correction coefficient to obtain a load pattern. Calculate the corresponding load signal. This load signal is output to each load point of the structure via the load signal output means. Response signals measured at each load point of the structure are taken in from response signal input means.
The correction coefficient updating means calculates a new correction coefficient based on the difference between the input response signal and the corresponding target signal, and updates the correction coefficient stored in the correction coefficient storage means. Next, when outputting a load signal of the same load pattern, the load signal is calculated using the new correction coefficient stored in the correction coefficient storage means. Hereinafter, each time a response signal is input, the above-described correction coefficient is updated.

[0012]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram illustrating a schematic configuration of a control device of the structural test machine according to the present embodiment.

In FIG. 1, reference numeral 1 denotes an A / D (Analog to Digital) converter for converting a response signal input from a structural test analog controller (not shown) into a digital signal. The A / D converter 1 has a plurality of channels corresponding to each load point of the structure. This A / D converter 1
This corresponds to the response signal input means of the present invention.

Reference numeral 2 denotes a D / A (Digital to Digital Output) which converts a load signal stored as a digital signal into an analog signal and outputs the analog signal to a structural test analog controller (not shown).
Analog) converter. The D / A converter 2 has a plurality of channels corresponding to each load point of the structure. This D / A converter 2 corresponds to the load signal output means of the present invention.

Reference numeral 10 denotes a control device of the structural test machine. The control device 10 of the structural test machine includes an auxiliary storage device 3,
CPU (Central Processing Unit)
4, RAM (Ramdom Access Memory) 5, CRT (Cath
ode Ray Tube), an input / output device 6 composed of a K / B (Key Board), and an input / output interface 7.

The auxiliary storage device 3 previously stores a procedure (program) of a load signal correction process described later, data on a target signal used for a structural test, data on a correction coefficient for correcting the load signal, and the like. ing. The CPU 4 performs calculation processing of a load signal, update processing of a correction coefficient, and the like according to a program stored in the auxiliary storage device 3, and corresponds to a calculation means and a correction coefficient update means of the present invention. The RAM 5 stores the correction coefficient and the like read from the auxiliary storage device 3, and corresponds to the correction coefficient storage unit of the present invention.

CPU 4, RAM 5, A / D converter 1,
The D / A converter 2, the auxiliary storage device 3, and the input / output device 6 are connected by an input / output interface 7.

Next, a correction process of the control device of the structural test machine having the above-described configuration will be described with reference to FIG. The response signal measured from each component of the structure is converted into a digital signal by the A / D converter 1 and taken into the control device 10 of the structure test machine (step S1).

Next, the peak and the trough of the response signal taken in step S1 are detected, and the amplitude R of the response signal shown in FIG.
_esp is obtained (step S2). In addition, the obtained amplitude R
The amplitude of _esp and the corresponding target signal is
Will be displayed.

Referring to the display on the CRT 6, the operator:
It is determined whether or not the correction coefficient is to be updated, and input from the input / output device 6. This determination will be described later. In accordance with an input instruction from the input / output device 6, it is determined whether or not to update the correction coefficient (step S3). If the correction coefficient is to be updated, a correction coefficient update process of step S4 is executed to update the correction coefficient. If not, the correction calculation processing of step S5 is executed.

Here, the correction coefficients stored in the RAM 5 will be described with reference to FIG. The correction coefficient for correcting the load signal is stored in a matrix corresponding to a combination (load pattern) of a target load corresponding to a certain load and a target load output one time before the output of the target load. I have. As described above, it is empirically recognized that, as described above, if the previous target signal is different, even if the current target signal is the same, a large difference occurs between the measured response signal and the target signal. Because. For example, as shown in FIG. 4A, when a target load corresponding to 4t is output from a target load output corresponding to 3t, a load signal for outputting a target load corresponding to 4t is obtained. the uses a M ₃₄ as correction coefficient, as shown in FIG. 4 (b), when outputting the target load corresponding to 4t from the output of the target load corresponding to 2t, the target load corresponding to 4t the correction factor for obtaining a load signal for outputting another correction coefficient M ₂₄ is used.

Hereinafter, a description will be given of a determination as to whether or not to update the correction coefficient. The correction coefficient is used in the correction calculation of the load signal to be output next time during the test as described in step S5 below, and is updated by the response signal and the target signal as described in step S4 below. , The difference gradually approaches the target signal. However, the purpose of the correction is to adjust the output load signal so that a load signal such that a load corresponding to the target signal is applied to the load point is output. Therefore, based on the amplitude of the response signal and the amplitude of the target signal displayed on the input / output device 6 described above, it is determined that a load signal such that a load corresponding to the target load is applied to the load point has already been output. In this case, no further update of the correction coefficient is required. That is, it is possible to determine that the optimum correction is being performed by performing the correction calculation based on the currently stored correction coefficient. Also, when the update is performed in spite of the optimal correction coefficient, if it is in a normal state,
As is clear from the following equation (2), (R _ange −
R _esp ) becomes 0 and the value of the correction coefficient is not changed. However, when sudden mutual interference or the like occurs between the load points, the correction coefficient in the optimum state is affected by the influence and the like.
The correction coefficient calculated under the temporary influence is updated. Therefore, by instructing not to update the correction coefficient when it is determined that the correction is performed in an optimal state, it is possible to eliminate the influence of sudden interference and the like.

Next, the process of updating the correction coefficient performed in step S4 will be described. In step S4, step S2
A new correction coefficient M _new is calculated based on the amplitude R _esp of the response signal obtained in the above, the amplitude R _{ange of the} target signal output this time, the proportionality constant G, and the current correction coefficient M _old read from the RAM 5. It is determined by the following equation (2). M _new = M _old + G × ( _Range −R _esp ) / R _ange (2) Here, the proportionality constant G is a constant multiplied by the difference between R _ange and R _esp, and for each tester. This is the determined correction weight.
Further, as the current correction coefficient M _old , for example, a response signal is input by outputting a target load corresponding to 3 t last time, and then a response signal is input by outputting a target signal corresponding to 4 t. If, when updating the correction coefficients for the response signal to 4t target signal of the correction coefficient M ₃₄ in FIG. 3 is used as the current correction factor M _old. Then, the calculated new correction coefficient M _new is written in the location where M _old , which was read out from the RAM 5, in the previous example, M ₃₄ was stored (step S 4).

The correction calculation process in step S5 is performed as follows. In other words, the correction calculation processing is performed based on the combination of the target signal output this time and the target signal to be output next time (the initial value in the case where the correction signal has never been updated is:
1.0) is read from the RAM 5. For example, this time 4t
In the example of FIG. 3, when the target signal corresponding to the target load has already been output and the target load corresponding to 5 t is to be output next time, the target load corresponding to 4 t output immediately before and the target load 5 t output corresponding to the M ₄₅ is a correction coefficient of a load pattern for R
Read from AM5. The load signal _Command to be output next time is calculated by the following equation (3) based on the read correction coefficient M, the amplitude R _{ange of the} target signal to be output next time, and the mapping coefficient MP. _{C ommand = MP × M × R} ange ......... (3) , where the mapping coefficients MP is a constant for adjusting the response of the entire test apparatus.

The load signal calculated in step S5 is output from the D / A converter 2 when the output timing comes (step S6). Then, returning to the processing of step S1, for example, the load signal _Command corrected for the target signal corresponding to 5t in the above example is output, and when the response signal is input, the processing of steps S2 and S4 is performed. it allows one pre-target load 4t outputted to be updated based on the above equation M ₄₅ is a correction coefficient of a load pattern corresponding to the target load 5t for outputting (2).

By repeatedly executing the above processing, the correction coefficient for each load pattern stored in the RAM 5 is updated one after another. Therefore, when the same load pattern appears next time, the load signal is calculated using the updated correction coefficient, and the error between the target signal and the response signal due to the influence of interference between the load points of the structure. And a load signal faithful to the target signal can be given to each load point.

[0027]

As is apparent from the above description, according to the present invention, a correction coefficient determined for each load pattern, which is a combination of a target signal to be output and a target signal output immediately before, is used. Since the load signal to be output is calculated, the response signal does not greatly deviate from the target signal due to the influence of the load signal output immediately before as in the conventional example, and the load signal faithful to the target signal is obtained. At the load point.

Further, according to the present invention, the correction coefficient for each load pattern is sequentially updated based on the measured response signal, and when the same load pattern appears next, the updated correction coefficient is used. Since the load signal is calculated, even if there is interference between the load points and an error temporarily occurs between the target signal and the response signal, the error gradually decreases and a faithful load is applied to the target signal. Will be able to do it.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of a control device of a structural test machine according to an embodiment.

FIG. 2 is a flowchart illustrating a correction procedure of the control device of the structural test machine according to the present embodiment.

FIG. 3 is a schematic diagram of a correction coefficient for each load pattern stored in a RAM.

FIG. 4 is an explanatory diagram of a load pattern and its correction coefficient.

FIG. 5 is a block diagram showing a schematic configuration of a conventional technique.

FIG. 6 is a diagram illustrating a difference between a measured response signal and a target signal.

[Explanation of symbols]

Reference Signs List 1 A / D (Analog to Digital) converter 2 D / A (Digital to Analog) converter 3 Auxiliary storage device 4 CPU (Central Processing Unit)
5) RAM (Random Access Memory) 6 I / O device 7 I / O interface 10 Control device for structural test machine

Claims (1)

(57) [Claims] 1. Load signal output means for individually outputting load signals to a plurality of load points of a structure under test, and response signals individually input from a plurality of load points of the structure. Response signal input means, correction coefficient storage means for storing a correction coefficient for correcting a load signal for each load pattern which is a combination of the previous target signal and the current target signal, and input from the response signal input means. Correction coefficient updating means for calculating a new correction coefficient based on the difference between the response signal and the corresponding target signal, and updating the correction coefficient stored in the correction coefficient storage means; and storing the correction coefficient in the correction coefficient storage means. And a calculating means for calculating a load signal according to a load pattern to be generated next time using the corrected correction coefficient.