JP6061517B2 - Test equipment - Google Patents

Description

  The present invention relates to a test apparatus for performing various tests such as a loading test by applying an external force to a structure to be tested such as a building such as a bridge, a building, and a house.

  Conventionally, as this type of test equipment, behavior data of an actual model test equipment that performs various tests by actually applying an external force to a structure under test that actually exists in physical space, and virtual on a computer A hybrid test is performed on the structure under test that exists in space, using the behavior data when an external force is virtually applied to each other and using the analysis of the actual model test and the virtual model test in conjunction with each other. Yes.

  That is, a simulation test is performed on a structure in which the test result is analyzed by fusing the behavior result of the test piece model in the actual physical space and the test piece model in the virtual space on the computer.

JP-A-2005-171732 JP 2009-250625 A

  FIG. 8 is a schematic view showing the whole test apparatus for such a hybrid test.

  As shown in FIG. 8, the conventional test apparatus 100 performs various tests by actually applying an external force to a structure under test actually existing in a physical space such as a vibration apparatus. An actual model test apparatus 102 is provided.

  An actual model test control device 104 that controls test conditions of the actual model test device 102 and transfers behavior data such as restoring force of the actual model test device 102 is provided. Further, a virtual model control device 106 is provided that calculates the behavior when an external force is virtually applied to a structure under test existing in a virtual space (not shown).

  Further, the test apparatus 100 includes a data control apparatus 108 that controls data exchange between the real model test control apparatus 104 and the virtual model control apparatus 106. In addition, data is exchanged between the virtual model control device 106 and the data control device 108 via, for example, the LAN network 110.

  By the way, in such a conventional test apparatus 100, behavior data such as restoring force of the structure under test actually existing in the physical space in the real model test control apparatus 104 is fed back to the virtual model control apparatus 106.

  As a result, the virtual model control device 106 calculates the behavior when an external force is virtually applied to the structure under test existing in the virtual space, and controls the test conditions in the real model test control device 104. It is configured as follows.

  However, at this time, the behavior of the structure under test actually existing in the physical space in the real model test control device 104 is determined by sampling of the digital controller in the data control device 108 with respect to the input from the virtual model control device 106. Alternatively, a delay is caused by mechanical factors.

  That is, in the conventional test apparatus 100, for example, as shown in the control chart of FIG. 9, a force is input at time t0, and the virtual model control apparatus 106 starts calculation at time t1. . Then, the command is started at the time t2 when the calculation is finished, and the test of the real model test apparatus 102 is started at the time t3 when the command is finished. At this time t3, force input is started again in the next test step.

  Therefore, a control delay Q occurs in the actual model test control apparatus 104 between t0 and time t3.

  Such a control delay of the real model test control device 104 is regarded as “negative attenuation”, and the analysis result of the virtual model may diverge. For this reason, it is necessary to prevent the control delay from affecting the analysis result as much as possible, and the analysis becomes complicated, and a large and high-speed control device is required.

  In view of such a current situation, the present invention has very little control delay in the real model test apparatus, there is no risk of the analysis result of the virtual model being diverged, the analysis is simple, a large and high-speed control apparatus is unnecessary, An object of the present invention is to provide a test apparatus capable of accurate analysis.

  In addition, the present invention provides, for example, a result of behavior of a test piece in an actual physical space and a test piece model in a virtual space on a computer for a structure having a speed dependency such as a seismic isolation device with a damper. It is an object of the present invention to provide a test apparatus capable of analyzing a test result closer to an actual phenomenon.

The present invention was invented in order to achieve the problems and objects in the prior art as described above.
A test apparatus for performing various tests by applying an external force to a structure under test,
An actual model test device that performs various tests by actually applying external force to the structure under test that actually exists in physical space,
A control device for controlling test conditions in the real model test device;
A virtual space computing device for computing the behavior of the virtual model in the virtual space;
With
By the virtual space arithmetic device, the behavior when the external force is virtually applied to the structure under test existing in the virtual space is calculated in advance, and virtual model test prediction data is created,
Virtual by said controller, corresponds to a part of the specimen under test structures existing in the virtual space, relative to the test structure actually present in the physical space, actually the behavior when loaded with external force Pre-calculate based on model test prediction data, create actual model test prediction data,
Based on the virtual model test prediction data and the actual model test prediction data, the control device stores the actual model test prediction data as a control prediction value,
Based on the control predicted value by the control device, with respect to the structure under test actually existing in the physical space corresponding to a part of the test piece of the structure under test existing in the virtual space, Actually start the test with external force applied,
The control device compares the behavior data during the actual test in the actual model test device with the behavior data of the structure under test existing in the virtual space, and based on the comparison result, the test in the actual model test device. It is characterized by being configured to control conditions.

With this configuration , virtual model test prediction data is created by calculating in advance the behavior when an external force is virtually applied to the structure under test existing in the virtual space by the virtual space arithmetic unit. doing.
The virtual model test predicts the behavior when an external force is actually applied to the structure under test that actually exists in the physical space, which corresponds to some test pieces of the structure under test existing in the virtual space. The actual model test prediction data is created by calculating in advance based on the data.
Further, the control device stores the real model test prediction data as a control prediction value based on the virtual model test prediction data and the real model test prediction data, and the control device uses the real model test prediction device based on the control prediction value. The test is actually started by applying an external force to the structure under test actually existing in the physical space, which corresponds to a part of the test piece of the structure under test existing in the virtual space .

Then, the control device compares the behavior data during the actual test in the actual model test device with the behavior data of the structure under test existing in the virtual space. Based on the comparison result, the test in the actual model test device is performed. Control the conditions.

  Therefore, in the real model test apparatus, the control delay is extremely small, the analysis result of the virtual model is not likely to diverge, a large and high-speed control apparatus is unnecessary, and accurate analysis can be performed.

  In addition, for example, for a structure having a speed dependency such as a seismic isolation device with a damper, the behavior results of the test specimen in the actual physical space and the test specimen model in the virtual space on the computer are merged. It is possible to analyze the test results closer to the actual phenomenon.

  Thereby, it can apply to the simulation test which performs arbitrary input in the middle of a test, and feeds back the reaction to the test object which can calculate behavior to some extent beforehand.

  The test apparatus of the present invention is characterized in that the control of the test conditions in the control apparatus is performed by PID control, feedforward control, and control amount compensation feedback control.

  With this configuration, normal PID control and feedforward control are performed, and further, the control amount is compensated as needed by control amount compensation feedback control. Thereby, it is possible to prevent the influence of the control delay from affecting the model calculation of the virtual model test.

  Further, in the test apparatus of the present invention, the control of the test conditions in the control apparatus is based on the behavior data during the actual test step in the real model test apparatus, and the behavior data of the structure under test existing in the virtual space is obtained. According to the present invention, the test condition of the next test step in the actual model test apparatus is controlled based on the calculation result.

  With this configuration, the behavior data of the structure under test existing in the virtual space is calculated based on the behavior data during the actual test step in the actual model test apparatus, and the actual data is calculated based on the calculation result. Since the test conditions for the next test step in the model test device are controlled, there is very little control delay in the actual model test device, there is no risk of the analysis result of the virtual model being diverged, and there is no need for a large, high-speed control device. Accurate analysis is possible.

In addition, the test apparatus of the present invention is
The control device is
An actual model test control device for controlling test conditions in the actual model test device;
A virtual model control device for calculating the behavior when an external force is virtually applied to the structure under test existing in the virtual space;
A data control device for controlling the exchange of data between the real model test control device and the virtual model control device;
It is characterized by providing.

  By configuring in this way, the control device is controlled from the real model test control device, the virtual model control device, and the data control device that controls the exchange of data between the real model test control device and the virtual model control device. It is composed.

  Therefore, the behavior data during the actual test in the actual model test apparatus is compared with the behavior data of the structure under test existing in the virtual space, and the test conditions in the actual model test apparatus are controlled based on the comparison result. Therefore, in the real model test apparatus, the control delay is extremely small, the analysis result of the virtual model is not likely to diverge, and a large and high-speed control apparatus is not required and accurate analysis can be performed.

According to the present invention, the virtual space arithmetic device pre-calculates the behavior when the external force is virtually applied to the structure under test existing in the virtual space, and creates virtual model test prediction data. Yes.
The virtual model test predicts the behavior when an external force is actually applied to the structure under test that actually exists in the physical space, which corresponds to some test pieces of the structure under test existing in the virtual space. The actual model test prediction data is created by calculating in advance based on the data.
Further, the control device stores the real model test prediction data as a control prediction value based on the virtual model test prediction data and the real model test prediction data, and the control device uses the real model test prediction device based on the control prediction value. The test is actually started by applying an external force to the structure under test actually existing in the physical space, which corresponds to a part of the test piece of the structure under test existing in the virtual space .

Then, the control device compares the behavior data during the actual test in the actual model test device with the behavior data of the structure under test existing in the virtual space. Based on the comparison result, the test in the actual model test device is performed. Control the conditions.

  Therefore, in the real model test apparatus, the control delay is extremely small, the analysis result of the virtual model is not likely to diverge, a large and high-speed control apparatus is unnecessary, and accurate analysis can be performed.

  In addition, for example, for a structure having a speed dependency such as a seismic isolation device with a damper, the behavior results of the test specimen in the actual physical space and the test specimen model in the virtual space on the computer are merged. It is possible to analyze the test results closer to the actual phenomenon.

  Thereby, it can apply to the simulation test which performs arbitrary input in the middle of a test, and feeds back the reaction to the test object which can calculate behavior to some extent beforehand.

FIG. 1 is a schematic view showing the entire test apparatus of the present invention. FIG. 2 is a schematic diagram for explaining a use state of the test apparatus of FIG. FIG. 3 is a schematic diagram showing a control method of the test apparatus of the present invention. FIG. 4 is a block diagram showing a method for controlling the test apparatus of the present invention. FIG. 5 is a graph showing the relationship between displacement and time in a conventional test apparatus. FIG. 6 is a graph showing the relationship between displacement and time in the test apparatus of the present invention. FIG. 7 is a graph showing the relationship between displacement and time when a model test is performed on a ramen viaduct assuming an actual seismic wave in the test apparatus of the present invention. FIG. 8 is a schematic view showing the entire conventional test apparatus. FIG. 9 is a schematic diagram for explaining a use state of the conventional test apparatus of FIG.

  Hereinafter, embodiments (examples) of the present invention will be described in more detail with reference to the drawings.

  FIG. 1 is a schematic diagram illustrating the entire test apparatus according to the present invention, FIG. 2 is a schematic diagram illustrating a use state of the test apparatus in FIG. 1, and FIG. 3 is a schematic diagram illustrating a method for controlling the test apparatus according to the present invention. FIG. 4 is a block diagram showing a control method of the test apparatus of the present invention.

  1 and 2, reference numeral 10 generally indicates a test apparatus of the present invention.

  As shown in FIGS. 1 to 2, the test apparatus 10 of the present invention actually applies an external force to the structure under test C <b> 1 that actually exists in the physical space A such as a vibration apparatus. A real model test apparatus 12 for performing various tests is provided.

  An actual model test control device 14 is provided for controlling test conditions of the actual model test device 12 and transferring behavior data such as restoring force of the actual model test device 12. In addition, as shown in FIG. 1, it functions as a virtual space computing device that calculates the behavior when an external force is virtually applied to the structure under test C2 such as a bridge, which exists in the virtual space B, for example. The virtual model control device 16 is provided.

  As shown in FIG. 1, the structure under test C1 of the actual model test apparatus 12 is a part of the test piece (shown by an ellipse) of the structure under test C2 existing in the virtual space B of the virtual model control apparatus 16. Corresponds to the D portion).

  Further, the test apparatus 10 includes a data control apparatus 18 that controls data exchange between the real model test control apparatus 14 and the virtual model control apparatus 16. In addition, as shown in FIG. 2, data is exchanged between the virtual model control device 16 and the data control device 18 via, for example, a LAN network 22 that exchanges data compliant with the TCP / IP protocol. It is like that.

  Further, as shown in FIGS. 1 and 2, the real model test apparatus 12, the real model test control apparatus 14, the virtual model control apparatus 16, and the data control apparatus 18 are collectively controlled. The control device 20 is provided.

  Meanwhile, in the conventional test apparatus 100, a delay occurs due to sampling of the digital controller in the data control apparatus 108 or mechanical factors. The control delay is regarded as “negative attenuation”, and the analysis result of the virtual model may diverge. For this reason, it is necessary to prevent the control delay from affecting the analysis result as much as possible.

  In the test apparatus 10 of the present invention, for example, in the virtual model control apparatus 16, the behavior when an external force is virtually applied to the structure under test C <b> 2 existing in the virtual space B of the virtual model control apparatus 16 is previously determined. Calculate to create virtual model test prediction data.

Further, for example, in the real model test control device 14, the real test device 12 actually exists in the physical space A corresponding to a part of the test piece of the structure under test C 2 existing in the virtual space B in the real model test device 12. The actual model test prediction data is created by calculating in advance the behavior when an external force is actually applied to the structure C1 based on the virtual model test prediction data .

Based on the virtual model test prediction data and the real model test prediction data, the real model test prediction data is stored as a control prediction value in , for example, the virtual model control device 16.

Based on this control prediction value, in the real model test apparatus 12, the structure under test C1 actually existing in the physical space A corresponding to a part of the test piece of the structure under test C2 existing in the virtual space B On the other hand, the test is started by actually applying an external force.

  Then, the behavior data during the actual test in the real model test device 12 and the behavior data of the structure under test C2 existing in the virtual space B of the virtual model control device 16 are compared, and based on the comparison result (difference). Thus, the test conditions in the actual model test apparatus 12 are controlled.

  That is, the behavior data of the structure under test C2 existing in the virtual space B of the virtual model control device 16 is calculated based on the behavior data during the actual test step in the actual model test device 12, and based on the calculation result. Thus, the test condition of the next test step in the actual model test apparatus 12 is controlled.

  Specifically, such control is performed as follows as shown in FIG.

  First, as shown in FIG. 3, at the time of t = 0, the virtual model control device 16 inputs force such as ground acceleration. At the same time, that is, at the time t = 0, as shown by the arrow P1 in FIG. 3, the control predicted value stored in advance in the virtual model control device 16 from the virtual model control device 16 is Input to the model test control unit 14.

  Thus, at the time t = 0, the actual model test apparatus 12 actually applies an external force to the structure under test C1 that actually exists in the physical space A in the test step S1.

  Then, at the time point t = δ, in the real model test apparatus 12, the measured value such as, for example, the restoring force measured by the real model test control apparatus 14 is the virtual model as indicated by the arrow P2 in FIG. It is fed back to the control device 16 and becomes ground motion acceleration data at the time point t = δ in the virtual model control device 16.

  At the time point t = δ, the actual model test control device 14 uses the actual model test device 12 in the physical space A in the test step S2 based on the control prediction value stored in advance. Actually, external force is continuously applied to the test structure C1.

  On the other hand, in the virtual model control device 16, for example, a measurement value such as a restoring force measured by the real model test control device 14 fed back from the real model test device 12 and a virtual model at a time point t = δ. The behavior data of the structure under test C2 existing in the virtual space B of the control device 16 is compared, and the test condition in the test step S2 in the actual model test device 12 is controlled based on the comparison result (difference).

  That is, the virtual model control device 16 calculates the behavior data of the structure under test C2 existing in the virtual space B of the virtual model control device 16 based on the behavior data of the actual test step S1 in the real model test device 12. . Then, based on the calculation result, as indicated by the arrow P3 in FIG. 3, the corrected control prediction value is input to the real model test control device 14, and thereby the next test in the real model test device 12 is performed. The test conditions of the step (test step S2) are controlled by the actual model test control device 14.

  In this case, for example, a measured value such as a restoring force measured by the real model test control device 14 fed back from the real model test device 12 and the test target existing in the virtual space B of the virtual model control device 16. When the behavior data of the structure C2 is compared and there is no difference in the comparison result, the actual model test control device 14 uses the actual model in the test step S2 based on the control prediction value stored in advance. In the test apparatus 12, an external force is actually continuously applied to the structure under test C1 that actually exists in the physical space A.

  Similarly, at the time point t = 2δ, the actual model test apparatus 12 measures the measured value such as the restoring force measured by the actual model test control apparatus 14 as indicated by the arrow P4 in FIG. This is fed back to the virtual model control device 16 and is converted into ground motion acceleration data at time t = 2δ in the virtual model control device 16.

  At the time of t = 2δ, the actual model test control device 14 uses the actual model test device 12 in the physical space A in the test step S3 based on the control prediction value stored in advance. Actually, external force is continuously applied to the test structure C1.

  On the other hand, in the virtual model control device 16, for example, a measurement value such as a restoring force measured by the real model test control device 14 fed back from the real model test device 12 and a virtual model at the time of t = 2δ. The behavior data of the structure under test C2 existing in the virtual space B of the control device 16 is compared, and the test conditions in the test step S3 in the actual model test device 12 are controlled based on the comparison result (difference).

  That is, the virtual model control device 16 calculates the behavior data of the structure under test C2 existing in the virtual space B of the virtual model control device 16 based on the behavior data of the actual test step S2 in the real model test device 12. . Then, based on the calculation result, as indicated by the arrow P5 in FIG. 3, the corrected control prediction value is input to the real model test control device 14, and thereby the next test in the real model test device 12 is performed. The test conditions of the step (test step S3) are controlled by the actual model test control device 14.

  Such a cycle is repeated as shown in FIG. As a result, as shown in FIG. 3, in the real model test apparatus 12, for example, a measured value such as a restoring force measured by the real model test control apparatus 14 and ground acceleration data in the virtual model control apparatus 16 are obtained. Can be acquired at the same time with almost no control delay.

  That is, as shown in the graph of FIG. 5, in the conventional test apparatus 100, a control delay occurs in the command (dotted line), but in the test apparatus 10 of the present invention in FIG. I understand that it is not.

  As shown in the graph of FIG. 7, when a model test was performed on the ramen viaduct assuming actual seismic waves, the maximum error displacement was 3 mm as shown in the E part of FIG. 7. Yes, it can be seen that the control value (dotted line) of the virtual model control device 16 almost follows the movement of the actual model test device 12 (solid line).

  In addition, as shown in the block diagram of FIG. 4, such control is performed by inputting to a normal PID controller 30, performing feedforward control 32, and controlling amount compensation feedback control for each control sampling. This is done by correcting by 34. That is, the feedback from the controlled object is monitored by the variable weight 36 and the input value is corrected, thereby performing control with improved followability compared to normal PID control.

  According to the test apparatus 10 of the present invention configured as described above, a control predicted value is created and stored based on the virtual model test predicted data and the actual model test predicted data, and the actual model is determined based on the control predicted value. In the test apparatus 12, the test is started by actually applying an external force to the structure under test C1 that actually exists in the physical space A.

  Then, the behavior data during the actual test in the actual model test device 12 is compared with the behavior data of the structure under test C2 existing in the virtual space B of the virtual model control device 16, and based on the comparison result, the actual data Test conditions in the model test apparatus 12 are controlled.

  Therefore, in the real model test apparatus 12, the control delay is extremely small, the analysis result of the virtual model is not likely to diverge, the analysis is simple, a large and high-speed control apparatus is unnecessary, and an accurate analysis can be performed.

  Moreover, for example, for a structure having speed dependency such as a seismic isolation device with a damper, a test piece (structure under test C1) in the actual physical space A and a computer (virtual model control device 16) It is possible to fuse the behavior results of the test piece model (the structure under test C2) in the virtual space B, and to analyze the test results closer to the actual phenomenon.

  Thereby, it can apply to the simulation test which performs arbitrary input in the middle of a test, and feeds back the reaction to the test object which can calculate behavior to some extent beforehand.

  In the above-described embodiment, the control device includes the real model test control device 14, the virtual model control device 16, the data control device 18, and the control device 20. However, the control device includes a single control device, It is also possible to construct a plurality of control units that exchange data and commands.

  The preferred embodiment of the present invention has been described above. However, the present invention is not limited to this, and the test apparatus 10 according to the present invention can be used as a test apparatus, for example, an automobile component (such as a drive system or undercarriage). Metal parts, rubber parts, shock absorbers, etc.), finished products such as finished automobiles, and civil engineering-related structures (bridge girder, bridges, seismic isolation rubber for buildings, etc.) It can be applied to various testing devices such as a material testing device, a vibration testing device, a fatigue testing device for performing a test, a vibration test, a fatigue test, a characteristic test, etc. Can be changed.

The present invention is applied to a test apparatus for performing various tests such as a loading test by applying an external force to a structure to be tested such as a building such as a bridge, a building, and a house. Can do.

DESCRIPTION OF SYMBOLS 10 Test apparatus 12 Real model test apparatus 14 Real model test control apparatus 16 Virtual model control apparatus 18 Data control apparatus 20 Control apparatus 30 PID controller 32 Feedforward control 34 Control amount compensation feedback control 36 Variable weight 100 Test apparatus 102 Real model test Device 104 Real model test control device 106 Virtual model control device 108 Data control device 110 Network A Physical space B Virtual space C1 Structure under test C2 Structure under test

Claims (4)

A test apparatus for performing various tests by applying an external force to a structure under test,
An actual model test device that performs various tests by actually applying external force to the structure under test that actually exists in physical space,
A control device for controlling test conditions in the real model test device;
A virtual space computing device for computing the behavior of the virtual model in the virtual space;
With
By the virtual space arithmetic device, the behavior when the external force is virtually applied to the structure under test existing in the virtual space is calculated in advance, and virtual model test prediction data is created,
Virtual by said controller, corresponds to a part of the specimen under test structures existing in the virtual space, relative to the test structure actually present in the physical space, actually the behavior when loaded with external force Pre-calculate based on model test prediction data, create actual model test prediction data,
Based on the virtual model test prediction data and the actual model test prediction data, the control device stores the actual model test prediction data as a control prediction value,
Based on the control predicted value by the control device, with respect to the structure under test actually existing in the physical space corresponding to a part of the test piece of the structure under test existing in the virtual space in the real model test apparatus. Actually start the test with external force applied,
The control device compares the behavior data during the actual test in the actual model test device with the behavior data of the structure under test existing in the virtual space, and based on the comparison result, the test in the actual model test device. A test apparatus characterized by being configured to control conditions.   The test apparatus according to claim 1, wherein the control of the test conditions in the control apparatus is performed by PID control, feedforward control, and control amount compensation feedback control.   Based on the calculation result, the control of the test conditions in the control device calculates the behavior data of the structure under test existing in the virtual space based on the behavior data during the actual test step in the real model test device. The test apparatus according to claim 1, wherein the test apparatus is configured to control a test condition of a next test step in the actual model test apparatus. The control device is
An actual model test control device for controlling test conditions in the actual model test device;
A virtual model control device for calculating the behavior when an external force is virtually applied to the structure under test existing in the virtual space;
A data control device for controlling the exchange of data between the real model test control device and the virtual model control device;
The test apparatus according to claim 1, further comprising:

Images