JP4491596B2 - Method and apparatus for measuring characteristics of sensor for detecting acceleration - Google Patents

Description

  The present invention relates to a method and apparatus for measuring characteristics of a sensor for detecting acceleration, and the technical field to which the present invention belongs is a field in which measurement of motion is indispensable, for example, vehicle collision safety, automobile suspension control, robots, transportation. This is a field for measuring acceleration in equipment, nuclear power generation related equipment, ships, aerospace equipment, information equipment, measurement of response to vibration of human body, and environmental vibration.

  Known sensors that detect acceleration include acceleration sensors and inertial sensors. As accelerometers (acceleration sensors), AC accelerometers (acceleration sensors) and DC accelerometers (acceleration sensors) are known, and in the frequency characteristics, input accelerometers in an AC frequency band that does not include a DC component are used. An accelerometer (acceleration sensor) that responds is called an AC accelerometer (AC acceleration sensor), and an accelerometer (acceleration sensor) that responds to input acceleration in the AC frequency band including the DC component in frequency characteristics. DC accelerometer (DC acceleration sensor).

  Conventionally, a method in which an acceleration sensor is installed on a one-dimensional shaking table and the movement of the shaking table or the other mounting surface of the installed acceleration sensor is measured with a laser interferometer is regarded as the most reliable method, and is the primary standard. Has been used. In this method, the sensitivity axis direction of the acceleration sensor is matched with the direction of motion of the shaking table. This technology was also used in the key comparision for MRA (Mutual Recognition Arrangement) conducted by the International Bureau of Weights and Measures. As a method of generating impact acceleration for calibration by impact acceleration, a metal projectile was launched from a simple launch tube and collided with the end face of the rod, and an elastic wave pulse generated inside the rod attached an acceleration sensor. A technique for evaluating the frequency characteristics of an acceleration sensor based on the impact acceleration generated at the end face when the light is reflected by the other end face has already been put into practical use (see, for example, Patent Document 1).

  Further, as described in Non-Patent Document 1, a metal sphere is brought into contact with and fixed to one end face of a round bar, and another metal sphere is caused to fly and collide with the metal sphere. When the elastic wave pulse generated inside is reflected by the other end face of the rod to which the acceleration sensor is attached, a method for evaluating the frequency characteristic of the acceleration sensor by the impact acceleration generated at the end face has been internationally standardized.

  In these methods, since the duration of impact acceleration cannot be controlled, the duration of impact acceleration to be generated is too short. That is, there is a drawback that the frequency band is too wide. Furthermore, neither current calibration nor vibration calibration can function as an acceleration measurement standard with the current technology. The reason is that since the motion is applied in the direction of the sensitivity axis of the acceleration sensor, an important element of acceleration, which is originally a vector quantity, is discarded from the beginning. Measuring acceleration must mean measuring the physical quantity that has directionality at the same time as the magnitude of the vector of acceleration amplitude, and therefore the acceleration measurement standard must be the basis for measuring both. . In that sense, the current acceleration measurement standard, whether vibration or impact, cannot be a standard for measuring acceleration in its original meaning as a vector. The key measure parison of the International Bureau of Weights and Measures can be the standard for acceleration, but not the standard for acceleration.

  Further, in such a situation, in the measurement of the most important peak sensitivity in the measurement of impact acceleration, the influence of the lateral sensitivity cannot be considered. This is extremely important and it must be said that the reliability of measurement of fractures and collisions is significantly impaired.

Japanese Patent Publication No. 6-52270 ISO / FDIS 16063-13, Part 13: Primary shock calibration using laser interferometry

  As described above, the acceleration is a vector. A signal applied to the acceleration sensor is one element of the acceleration set, and a set of output signals of the acceleration sensor is a set of vectors, that is, a vector space. That is, if it is recognized that the acceleration is a vector, the acceleration sensor has a function of converting the input acceleration vector in the real space into the output acceleration vector in the signal space. Mathematically, acceleration is a numeric vector and transforms the input vector space into the output vector space, so the sensitivity of the acceleration sensor must be in a matrix, so calibration is all elements of matrix sensitivity. , Is very natural to define.

  The situation in which a large number of multi-axis acceleration sensors are produced as semiconductor acceleration sensors can be said to require the development of calibration technology that can meet the original natural requirements.

  An object of the present invention is to provide a sensor characteristic measuring method and apparatus for detecting acceleration that solve the above-described problems.

  A more specific object of the present invention is to provide a method of constructing an apparatus that can easily generate impact acceleration by controlling duration, peak value, waveform, waveform spectrum, etc. Is a method and apparatus for calibrating from the standpoint of being a vector, that is, a method and apparatus for defining matrix sensitivity.

According to one aspect of the method for measuring the characteristics of a sensor for detecting acceleration according to the present invention, a plurality of flying bodies collide with one end face of a round bar, whereby the translational acceleration generated in a direction perpendicular to the other end face is Measured by a sensor having a single-axis or multi-axis sensitivity axis attached to the other end surface of the first and second axes, and when this is repeated a plurality of times, the jigs have different three-dimensional shapes . By using a plurality of jigs , the angle formed between the mounting plane of the round bar on the other end surface and the mounting plane of the sensor is made different from each other for each of the plurality of measurements, whereby the translational acceleration is calculated . vector whose components direction cosine to the coordinate system of the sensor, after as a linearly independent states, by solving a system of linear equations based on the previous SL plurality of measurement results, the translational acceleration vector Against and obtains the sensitivity matrix of the sensor.

Another aspect of the present invention is a sensor having a single-axis or multi-axis sensitivity axis for detecting acceleration, and a jig for mounting the sensor, wherein a plurality of flying objects collide with one end surface of a round bar. Thus , when the measurement for applying the translational acceleration to the other end face of the round bar is repeated a plurality of times , a vector whose component is a direction cosine of the translational acceleration determined for each measurement to the coordinate system of the sensor is obtained. , so that the linearly independent condition, so that the angle of the mounting plane to the other side of the round bar of the sensor is different, and a plurality pieces of jigs having different three-dimensional shapes, the measurement for each of the jig And calculating means for finding a sensitivity matrix of the sensor with respect to the translational acceleration vector by solving simultaneous linear equations based on each measurement result obtained by executing the above.

According to another aspect of the present invention, an impact is applied to one end face of a round bar for each measurement, and an elastic wave pulse generated inside the round bar due to the impact is reflected by the other end face of the round bar. When measuring the movement of the end face by a sensor having a single-axis or multi-axis sensitivity axis attached to the other end face of the round bar via a jig and detecting acceleration, and an input acceleration measuring device. The vector whose component is the direction cosine of the translational impact acceleration generated in the direction perpendicular to the other end face of the round bar to the coordinate system of the sensor is linearly independent between the measurements. As a tool , a plurality of jigs having different three-dimensional shapes are used, and the angle between the jig mounting plane on the other end surface of the round bar and the sensor mounting plane of the jig is different for each measurement. is not running, the said obtained by each measurement Capacitors and by solving the simultaneous linear equations based on the measurement result of the input acceleration measuring device, you and obtains the sensitivity matrix of the sensor.

Still another aspect of the present invention includes means for applying an impact to one end face of a round bar for each measurement, an input acceleration measuring device for measuring movement of the other end face of the round bar, and the other end of the round bar. A plurality of jigs for attaching the sensor to an end face, the other of the round bars of an elastic wave pulse generated inside the round bar by an impact applied to one end face of the round bar by the means for applying the impact the movement of the end surfaces generated by the reflection at the end face, when measured by a sensor for detecting acceleration having a sensitive axis of the single or multi-axis and the input acceleration measuring device, repeated a plurality of times this, together A plurality of jigs having different three-dimensional shapes, and a vector whose component is a direction cosine of the translational acceleration generated on the other end face of the round bar to the coordinate system of the sensor is linearly independent between each measurement. so as, the other of the round bar Solving the mounting plane of the end face and the angle is different jig with mounting plane of the sensor, the system of linear equations based on the measurement result of each measurement of the sensor and the input acceleration measuring device obtained by executing And a sensor characteristic measuring device for detecting acceleration, comprising an arithmetic means for obtaining a sensitivity matrix of the sensor.

Here, the input acceleration measuring device is at least one strain gauge affixed to a side surface of the round bar, and the computing means is an elastic wave incident on the other end surface of the round bar from the strain gauge output signal. when obtaining the transient distortion signal pulse, it may be to have use of at least first-order terms of the solution of Sukaraku is expanded analytical solution to the series.

Further, the input acceleration measuring device is a plurality of strain gauges attached to the side surface of the round bar in the axial direction, and the waveform due to the difference in the attachment position in the axial direction with respect to the plurality of strain gauge outputs. The difference can be converted into a waveform at a single point by calculating an addition average of each strain gauge output, and signal processing for reducing noise can be performed.

Still another aspect of the present invention is that two or more flying objects that are independent from each other for each measurement, and the flying objects having different distances between the tip and one end surface of the round bar are simultaneously driven by high-pressure gas. The vertical impact is applied to one end face of the round bar for each measurement by causing it to collide with one end face of the round bar in order with a slight time difference, and the impact is generated inside the round bar. The movement of the end face generated by reflection of the elastic wave pulse on the other end face of the round bar has a single-axis or multi-axis sensitivity axis attached to the other end face via a jig, and accelerates the acceleration. When measuring with the sensor to be detected and the input acceleration measuring device, a vector whose component is a direction cosine of the translational acceleration generated on the other end face of the round bar to the coordinate system of the sensor is linearly independent between each measurement. To be in the state of As the jig, using a plurality of different fixture sterically shapes, perform the angle of the mounting plane of the mounting plane the sensor to the other end surface of the rod by different for each measurement In addition, it is characterized in that a sensitivity matrix of the sensor is obtained by solving simultaneous linear equations based on measurement results of the sensor and the input acceleration measuring device obtained by each measurement.

Still another aspect of the present invention is two or more flying bodies independent from each other, wherein the flying bodies having different distances between the tip and one end face of the round bar are simultaneously driven by the high-pressure gas to fly, Means for applying an impact to one end face of the round bar by sequentially colliding with one end face of the round bar with a slight time difference; an input acceleration measuring device for measuring the movement of the other end face of the round bar; and the impact The movement of the end face generated by reflection of the elastic wave pulse generated inside the round bar by the impact applied to one end face of the round bar by means of A direction cosine of the translational acceleration generated on the other end face of the round bar to the coordinate system of the sensor when the sensor having the sensitivity axis of the axis and detecting the acceleration by the input acceleration measuring device is measured a plurality of times. component Vector, so that the linearly independent condition between each measurement, a plurality of jig mounting the sensor on the other end surface of the rod, and the mounting plane to the other end face of the rod that Solving simultaneous linear equations based on the measurement results of the sensor and the input acceleration measuring device obtained by executing the respective measurements with jigs having different three-dimensional shapes so that the angle formed with the mounting plane of the sensor is different. And a sensor characteristic measuring device for detecting acceleration, comprising an arithmetic means for obtaining a sensitivity matrix of the sensor.

  Here, the two or more flying bodies are supported so as to be separable into a pressure receiving body that can directly receive the pressure of the high-pressure gas and move within the launch tube, and the flying direction in the supported state. The top tip position can be different.

  Further, the two or more flying bodies are a first flying body located on a central axis of the pressure receiving body, and are located outside the first flying body and coaxially with the first flying body. And at least one second flying object.

  Furthermore, any one of the first and second flying bodies may be integral with the pressure receiving body.

  Furthermore, at least one of the two or more flying objects may have a material of the collision end portion different from that of the other portions.

  Further, each of the two or more flying objects may have at least a part of a material different from that of the other flying objects.

  Furthermore, a metal sphere can be contacted and fixed to one end surface of the round bar, and the impact can be applied to the metal sphere.

  Furthermore, the input acceleration measuring device may be at least one strain gauge attached to a side surface of the round bar.

Further, a plurality pasted the strain gauge in the axial direction of the rod, with respect to each strain gauge outputs, the differences in the axial direction of the pasted wave due to a difference in position, the averaging of the strain gauge output By obtaining this, it is possible to perform signal processing for converting into a waveform at one point and adding it to reduce noise.

Further, the computing means is based on the strain gauge output signal and the strain of the incident acoustic wave pulse, which is a measurement result of a laser interferometer that directly measures the movement of the other end face of the round bar . Obtaining a correction function by dividing the Laplace transform of strain at the representative position by Laplace transform of strain of the incident acoustic wave pulse on the sensor mounting end face, and obtaining the correction function by the strain of the strain gauge output, By multiplying the frequency characteristic of the sensor that detects acceleration, the transfer function of the acceleration detection sensor is obtained, and the elasticity that is incident on the end face of the round bar to which the sensor is attached, which is necessary to obtain the input acceleration applied to the sensor. can and Turkey asked the strain of the wave pulse.

Further, when obtaining the strain of the elastic wave pulse, the output signal of the strain gauge is divided by a correction function derived using the measurement result of a laser interferometer that directly measures the movement of the other end face of the round bar. Both the measurement result of another laser interferometer that directly measures the movement of the mounting surface of the sensor in the jig and the measurement result of the laser interferometer that directly measures the movement of the other end face of the round bar with, the direction cosine can and corrected child.

Further, when the calculation means obtains a transient strain signal of an elastic wave pulse incident on the other end face of the round bar from the strain gauge output signal, the calculation means is at least a first order of a sparse, which is an analytical solution developed in a series . Terms can be used.

  Furthermore, the input acceleration measuring device may be a laser interferometer that directly measures the movement of the other end face of the round bar.

Further, when the acceleration sensor is calibrated, the calculation means controls the frequency bandwidth of the impact acceleration signal applied to the sensor, thereby obtaining the sensor's frequency obtained as a function of the frequency band of the impact acceleration signal. after raising the peak sensitivity values, actually it is the this to determine the signal frequency bandwidth and the peak sensitivity of the input impact acceleration can be used in the acceleration measurements.

  The present invention relates to a dynamic displacement in a direction perpendicular to the end face generated in the process in which an elastic wave generated by applying an impact to the end face of a round bar propagates inside the bar to reach the other end face and reflects. Speed and acceleration are input signals to the sensor that detects the acceleration attached to the end face, and these input signals that are functions of time are measured by the input acceleration measuring device for calibration, and the output signal and calibration of the sensor that detects the acceleration The dynamic displacement detection function, speed detection function, and acceleration detection function of the sensor that detects acceleration by performing signal processing operations such as Fourier transform, Laplace transform, and filter operation on the output signal of the input acceleration measurement device Enables measurement of gain-frequency characteristics, phase-frequency characteristics, and peak sensitivity of sensors that detect acceleration for each function as a matrix

  Furthermore, the present invention can be used for calibration of an acceleration sensor used for controlling an automobile occupant protection airbag.

  Furthermore, the present invention can be used for calibration of inertial sensors used for robot motion control.

  Furthermore, the present invention can be used for calibration of an inertial sensor used for measurement such as human body motion or vibration received by a human body or animal behavior monitor.

  Furthermore, the present invention can be used for calibration of inertial sensors related to inertial navigation devices mounted on automobiles, submarines, torpedoes, missiles, aircraft, and guided bombs.

  Furthermore, the present invention can be used for calibration of inertial sensors used for preventing vibration of image equipment and video equipment.

  Furthermore, the present invention can be used for calibrating an acceleration sensor used for controlling an automobile suspension.

(1) In the conventional calibration technique, since the accelerometer was calibrated only with the acceleration amplitude, the acceleration could not be measured with the accelerometer. The reason is that the acceleration is a vector quantity having a magnitude and direction. Since it is possible to calibrate with acceleration as a vector according to the present invention, the first thing that must be clarified is that this method means that the accelerometer is calibrated with acceleration, and the accelerometer can measure acceleration. That's what it means.
If a multi-dimensional multi-axis shaking table is used, it is possible to calibrate with acceleration as a similar vector (for example, Japanese Patent Application No. 2003-123417), but it is difficult to generate a large acceleration with the shaking table, The present invention solves this problem.
(2) Furthermore, in a multi-dimensional multi-axis shaking table, it takes time to set the control, and the frequency must be scanned, so that much time is required for the entire data collection. The present invention using impact ameliorates this problem.
(3) It is widely recognized that it is necessary to consider the influence of lateral sensitivity in peak acceleration measurement.
(4) The metrology law stipulates that the Minister of Economy, Trade and Industry designates a specified standard device as a primary national standard, and states that currently calibrators that can only be calibrated using acceleration values are being used as specified standards for acceleration standards. Has been. It is necessary to change the name to a specific standard device for acceleration values and to specify a device that can be calibrated by acceleration as a vector as a national standard for acceleration. In other words, the recognition that the measurement law needs to be revised is spread.
(5) The CCAUV / International Bureau of Weights and Measures performed key comparision based on the calibration technology for acceleration values. However, when the 3D impact acceleration generator spreads throughout the world, the impact acceleration key comparision for mutual recognition. It is necessary to carry out. In this key comparison, both peak sensitivity and frequency response must be experimented in terms of matrix sensitivity.
(6) An acceleration sensor capable of vector synthesis is developed. This improves the accuracy of impact acceleration measurement when the direction of motion is unknown. It is important for measuring phenomena that are not well understood. The acceleration sensor for controlling the operation of the airbag for protecting the vehicle occupant must be able to measure the impact acceleration as a vector. At present, when trying to guarantee safety from the legal aspect, it is extremely unrealistic that the driver must turn the steering wheel to make a frontal collision when trying to collide. In the case of a collision that cannot be avoided, the preventive safety device that is used needs to be unrealistic that it automatically steers to make a frontal collision.
(7) A reference acceleration sensor for an acceleration sensor capable of vector synthesis must be developed and will be developed.
(8) In an application where it is meaningless if acceleration as a vector cannot be measured, such as an acceleration sensor installed in a vehicle for the safety of collision of an automobile occupant, the use is widened and the safety of the vehicle is increased.
(9) Like the seismometers used for measuring earthquakes with pitch and roll, the reliability of the measurement of accelerometers used in application fields that are meaningless unless they can be measured in three dimensions.
(10) Utilization technology of a multi-axis acceleration sensor used for automobile suspension control and the like is improved. The industry is aware that they are not using multi-axis acceleration sensors. In the future, such a consciousness will disappear if the calibration technique described in this patent spreads.
(11) The lateral sensitivity is not given in the piezoelectric uniaxial acceleration widely used as the reference acceleration sensor at present, but the lateral sensitivity is given by the present invention. Since the piezoelectric reference acceleration sensor output constitutes a one-dimensional output space, the current three-axis acceleration sensor that assembles such acceleration sensors in a three-orthogonal coordinate system is used to synthesize the vector as an accurate vector. Can be measured.
(12) A new type of accelerometer that incorporates a numerical calculator having a vector composition function appears on the market as a product. As a result, acceleration measurement in real time becomes possible.
(13) The business of calibrating accelerometers will become more sophisticated, and at the same time, Japan's international position in the world for vibration and impact measurement standards will improve.
(14) It is possible to measure an error caused by rotation of the acceleration sensor, which is conventionally difficult to measure, called a scallling effect.
(15) Performance evaluation of a multi-degree-of-freedom inertial sensor (particularly made of semiconductor) having a translational motion acceleration detection function, a rotational angular velocity detection function, and a rotational angular acceleration detection function at the same time is possible, thereby improving the performance of the product.
(16) The EU Directive issued in June 2002, which requires the numerical targets to be clearly documented in law in 2005, can also be handled. This standard is based on the human body vibration standards specified in ISO2631 and ISO8041, and corresponds to human body vibration standards that require measurement with acceleration as a vector if measured by an acceleration sensor calibrated according to the present invention. It becomes possible.
(17) A vibration generator for calibrating an impact accelerometer that enables vector calibration becomes widespread.
(18) Since the acceleration sensor, such as a motion generation device, can measure vector acceleration in the sense of 眞, the accuracy of motion control is improved.
(19) Multi-dimensional motion generators for inertial sensor calibration are widely used.
(20) The recognition of the importance of matrix expression for peak sensitivity spreads.
(21) The concept of lateral peak sensitivity and main axis peak sensitivity is widespread.

(Explanation of the entire calibration device)
FIG. 1A shows the configuration of a sensor calibration apparatus that detects acceleration due to impact acceleration using multiple flying objects. Reference numeral 1 denotes a metal rod that is sufficiently long compared to its diameter, and is horizontally supported by bearings placed in two grooves (not shown) at a plurality of locations, and is not constrained in the axial direction. Reference numeral 2 denotes a launch tube as a means for applying an impact to one end face of the round bar 1 and simultaneously drives three flying bodies 3A, 3B, 3C (see FIG. 1B) simultaneously with high-pressure air. It is positioned and fixed by appropriate fixing means so as to fly and collide sequentially with one end face of the round bar 1 with a minute time difference. The structure of the flying object will be described later.

  Reference numeral 4 denotes a sensor for detecting acceleration, which is attached to the other end face of the round bar 1 through each of three jigs (details will be described later) (in FIG. 1, one of the jigs 5 is attached). Show).

  The elastic wave pulse generated by applying an impact to one end face of the round bar 1 as described above propagates inside the round bar, reaches the other end face, and is reflected. The impact acceleration in the direction perpendicular to the end face (in the axial direction of the round bar 1) generated in the reflection process becomes an input to the sensor 4 attached to the end face via a jig. Each of the three jigs has two mounting planes, that is, a mounting plane to the other end surface of the round bar 1 and a mounting plane of the sensor 4, and the angle formed by these two mounting planes is three. The jigs are three-dimensionally different from each other. The sensor 4 is attached to the other end face of the round bar 1 in a state as will be described later (FIG. 2) via the jig, so that the sensitivity of the sensor 4 is matrixed. Can be obtained as

  The sensor for detecting the acceleration is attached to the other end face (end face to which the jig is attached) of the round bar 1 in three different modes. This will be described with reference to FIG.

2, an orthogonal coordinate system fixed to the other end face of the round bar 1 and X _b Y _b Z _b, axial round bar 1 to Z _b-axis, i.e., the other round bar 1 by reflection of the acoustic wave pulse Match the direction of movement of the end face of. Three impact accelerations are applied to one end face of the round bar 1, and the first to third different jigs are used for each application.

An orthogonal coordinate system X ₁ Y ₁ Z ₁ is an orthogonal coordinate system fixed to the sensor 4 when the sensor 4 is attached to the other end face of the round bar 1 via a first jig, and the direction of impact acceleration application Is the Z ₁ axis.

Similarly, the Cartesian coordinate system X ₂ Y ₂ Z ₂ is a Cartesian coordinate system fixed to the sensor 4 when the sensor 4 is attached to the other end face of the round bar 1 via the second jig, acceleration application direction is Z ₂ axes.

Similarly, an orthogonal coordinate system X ₃ Y ₃ Z ₃ is an orthogonal coordinate system fixed to the sensor 4 when the sensor 4 is attached to the other end face of the round bar 1 via a third jig, acceleration application direction is Z ₃ axes.

In this way, by attaching to the other end face of the round bar 1 through each of the three jigs, the direction vector cosine (cosα ₁ , cosβ to the orthogonal coordinate system fixed to the sensor 4 of the acceleration vector applied to the sensor 4 is obtained. ₁ , cosγ ₁ ), (cosα ₂ , cosβ ₂ , cosγ ₂ ), (cosα ₃ , cosβ ₃ , cosγ ₃ ) can be set to be linearly independent. Therefore, the acceleration generated on the other end face of the round bar 1 by applying the impact acceleration three times is represented by the orthogonal coordinate system X _b Y _b Z _b fixed to the other end face of the round bar 1 as shown in Table 1. And the orthogonal coordinate system (X ₁ Y ₁ Z ₁ , X ₂ Y ₂ Z ₂ , X ₃ Y ₃ Z ₃ ) fixed to the sensor 4 can also be expressed as shown in Table 1. The meanings of the symbols in Table 1 are shown in Table 2.

  Thereby, although details will be described later, simultaneous linear equations relating to all elements of the matrix sensitivity can be derived as a function of frequency. Since the coefficient determinant of simultaneous linear equations can be proved to be non-zero, it can always be solved. Therefore, solving the simultaneous linear equations is nothing but calibration. The fact that simultaneous linear equations can be solved is proved mathematically at the end of this specification.

  Reference numeral 6 denotes a plurality of strain gauges as calibration input acceleration measuring devices for measuring the movement of the other end face of the round bar 1, and in this embodiment, the side face of the round bar 1 (both ends in the axial direction of the round bar 1). A plurality of surfaces are affixed in the circumferential direction and the axial direction on the surface (excluding the surface).

  Reference numeral 7 denotes a laser beam applied to the other end face of the round bar 1 and is emitted from a laser interferometer (not shown). Reference numeral 8 denotes another laser beam irradiated on the mounting plane of the sensor 4 of the jig 5 and is emitted from another laser interferometer (not shown). The detection output signal of the sensor 4, the detection output signal of the strain gauge 6, and the detection output signal of the laser interferometer are supplied to an arithmetic device (usually configured by a computer), and an acceleration as will be described later by the arithmetic device. Various calculations are performed for the calibration of the sensor that detects.

  As shown in FIG. 3, the three flying bodies 3A, 3B, and 3C constitute a triple-structured multiple flying body, of which the innermost flying body 3A has an axial shape and its base end. Is fixed by being screwed into the center of a flange 3D as a pressure receiving body provided in the firing tube 2 so as to be airtightly movable in the axial direction.

The intermediate tubular flying body 3B is slidably disposed on the outside of the flying body 3A, and further, the tubular flying body 3C is disposed on the outside of the flying body 3B so as to be slidable with respect to each other. In this state, the flying object 3B has a length that does not protrude further downstream than the flying object 3A in the flying direction. body 3C, from the projectile 3B is the total length of the flying direction of which has (projectile 3A and the flange 3D has a length which does not protrude on the downstream side of the flight direction and L _1, 2 single projectile 3B, L ₁ > L ₂ > L ₃ ) where L ₂ and L ₃ are the lengths in the flight direction of 3C, respectively. That is, the three flying bodies 3A, 3B, and 3C are supported by the flange 3D and have different tip positions in the flying direction.

  With the flange 3D not projecting from the opening end of the launch tube 2, the tip of the innermost projecting body 3A touches one end surface of the round bar 1, and the center of the flange 3D, that is, the projecting object. The distance and the positional relationship between the launch tube 2 and one end face of the round bar 1 are adjusted so that the axis of 3A coincides with the axis of the round bar 1.

  The end of the firing tube 2 on the closing side is connected to a high-pressure air source 8 via an opening / closing valve 7, and the opening / closing of the opening / closing valve 7 is controlled by a valve opening / closing controller 9. When the flange 3D supporting the two flying bodies 3B, 3C is on the closed end side of the launch tube 2 and the opening / closing valve 7 is opened from the closed state by the valve opening / closing control device 9, high-pressure air passes through the opening / closing valve 7. The flange 3D is instantaneously supplied into the launch tube 2, and the flange 3D is driven by the supplied high-pressure air. As a result, the flange 3D moves at high speed toward one end face of the round bar 1, and the two projectiles are moved by the flange 3D. 3B and 3C are simultaneously driven and flew (the flying object 3A integrated with the flange 3D also flies at the same time). First, the innermost flying object 3A first collides with one end surface of the round bar 1, and the flange The two flying bodies 3B and 3C supported by 3D are separated from the flange 3D, the intermediate flying body 3B next collides with one end face of the round bar 1, and the outermost flying body 3C is the round bar 1 Collides with one end face last

The three flying bodies 3A, 3B, 3C can also be configured as shown in FIG. That is, the center portion of the flange 3D is protruded from the outer portion, the innermost flying body 3A is screwed and fixed to the center of the protruding portion, and the intermediate tubular flying body 3B is supported by the protruding portion. Are arranged slidably on the outside of the flying body 3A, and further, the tubular flying body 3C is slidably arranged on the outside of the flying body 3B so as to be supported by the outer portion of the protruding portion of the flange 3D. and, the two projectile 3B, 3C and is obtained by so as to detachably supporting the flange 3D, and in its supported state, the sum of the flying object 3A and the flange 3D flight direction of the length of the L _1, When the lengths of the two flying bodies 3B and 3C in the flying direction are L ₂ and L ₃ , respectively, L ₁ > L ₃ > L ₂ is satisfied. In such a configuration, it is not necessary to make the length of the outermost flying object 3C shorter than the length of the intermediate flying object 3B. As described above, the frequency bandwidth of the impact acceleration generated on the end face of the round bar can be more easily controlled by changing the relationship between the lengths of the three flying objects as compared with that of FIG. In both the configuration of FIG. 3 and the configuration of FIG. 4, the innermost flying body 3A can be separated from the flange 3D in the same manner as the other two flying bodies, without being structured to be screwed into the flange 3D. It may be. Also in the configuration of FIG. 4, the three flying bodies 3A, 3B, 3C are different from each other in the tip position in the flying direction while being supported by the flange 3D.

Further, the three flying bodies 3A, 3B, 3C can be configured as shown in FIG. In this example, the outermost flying body 3C is screwed and fixed to the flange 3D, and an intermediate flying body 3B is slidably disposed inside the outermost flying body 3C, and the inner side of the intermediate flying body 3B. The innermost flying body 3A is slidably arranged on the other side, and these two flying bodies 3A and 3B are detachably supported by the flange 3D. In this supported state, the flying body 3C and the flange are supported. the sum of the flying direction of the length of the 3D and _{L 1,} 2 single projectile 3A, the flying direction of 3B length respectively When _{L 3, L 2,} is obtained by the _{L 1>} L _2> L ₃ . Also in this configuration, the three flying bodies 3A, 3B, and 3C are different from each other in the tip position in the flying direction while being supported by the flange 3D. According to this configuration, the three flying bodies 3A, 3B, 3C are simultaneously driven to fly and sequentially collide with one end face of the round bar 1 in the order of the flying body 3C, the flying body 3B, and the flying body 3A.

FIG. 6 shows an example of four flying bodies 3A, 3B, 3C, 3E. In this example, an intermediate flying body 3B is slidably disposed inside the outermost flying body 3C, and another intermediate flying body 3E is slidably disposed inside the intermediate flying body 3B. Then, the innermost flying body 3A is slidably disposed inside the other intermediate flying body 3E, and these four flying bodies 3A, 3B, 3C, 3E are detachably supported by the flange 3D. If the lengths of the four flying bodies 3A, 3B, 3C, and 3E in the flying direction are L ₄ , L ₁ , L ₂ , and L ₃ in the supporting state, respectively, L ₁ > L ₂ > L ₃ > L ₄ Also in this configuration, the four flying bodies 3A, 3B, 3C, and 3E are supported by the flange 3D, and the tip positions in the flying direction are different from each other. According to this configuration, the four flying bodies 3A, 3B, 3C, and 3E are simultaneously driven to fly, and one of the round bars 1 is in the order of the flying body 3B, the flying body 3C, the flying body 3E, and the flying body 3A. It collides sequentially with the end face.

  Further, the plurality of flying bodies may be a plurality of spheres having different diameters. In this case, an appropriate support structure is provided for the flange 3D (for example, a plurality of pipes having an inner diameter in which the respective spheres are inserted are fixed to the flanges, and the spheres are inserted therein). When inserted into the launch tube 2 and driven with high-pressure air and the flange 3D is stopped at the open end of the launch tube 2, each sphere jumps out of the respective tube and collides with one end face of the round bar 1.

  Reference numeral 12 denotes a laser light source, and laser light emitted from the laser light source is reflected by two mirrors 13 (one mirror is semi-transmissive and transmits a part of the laser light) and The light passes through the vicinity of one of the end faces perpendicularly and enters the two light receiving elements 14. The light reception outputs of the two light receiving elements 14 are input to a counter 15 that measures the speed when the flying object collides with the round bar 1. Immediately before the flying object is driven to fly and collide with one end face of the round bar 1, the tip of the innermost flying object 3A first shields the laser beam on the launch tube 2 side, and then the round bar 1 The side laser beam is shielded. When the laser beam is blocked, the light receiving output of the light receiving element is turned off. Therefore, when the flying object 3A collides, the counter 15 counts the timing between the two light receiving output signals from the two light receiving elements 14 between OFF and OFF, and the flying object collision speed is based on the count result. Can be measured. The same configuration as described above is provided for each of the intermediate flying body 3B and the outermost flying body 3C. That is, the laser beams from the laser light sources other than the laser light source are shielded immediately before the tips of the intermediate projecting body 3B and the outermost projecting body 3C collide with one end surface of the round bar 1. Two laser light sources, two mirrors for each of the two laser light sources and two light receiving elements (not shown) are provided, and the light reception outputs of the two (four total) light receiving elements are input to the counter 15. Then, the same measurement as described above is performed. The measurement result of the counter 15 is supplied to the above-described arithmetic device and used for the calibration process of the sensor 4 that detects the acceleration.

  According to the above configuration, the flying bodies 3 </ b> A, 3 </ b> B, and 3 </ b> C are caused to collide with one end face of the round bar 1 and an impact is applied to generate an elastic wave pulse inside the round bar 1. At that time, by shifting the timing of the collision of each flying object with one end face of the round bar 1, the duration of the elastic wave pulse as a whole can be extended according to the principle of superposition. Details will be described later with reference to FIG.

Timing of the collision of the projectile has a difference in length of the projectile depends on the impact velocity V _1. Assuming that the speed drop due to friction can be ignored, the entire flying object collides with the end face of the round bar at the same collision speed, but the length of the innermost flying object 3A is L _p1 and the length of the intermediate flying object 3B is L _p2 . Then, the time difference of collision becomes (L _p1 -L _p2 ) / V ₁ .

  The elastic wave pulse generated by the collision of the flying object with one end face of the round bar 1 propagates through the inside of the round bar 1 and reaches the other end face of the round bar 1 to be reflected. The impact acceleration in the direction perpendicular to the end face generated during the reflection process becomes an input to the sensor 4 that detects the acceleration attached to the end face via the jig 5.

  The impact acceleration generated on the other end face of the round bar 1 is detected by the laser interferometer provided at a position where the other end face can be irradiated with the laser light from the laser interferometer or a plurality of positions on the side face of the round bar 1. This is performed with the strain gauge 6 affixed to (including the case of one place in the axial direction).

Here, the impact acceleration generated on the other end face of the round bar due to the single collision of each flying object on one end face of the round bar 1 and the three flying bodies 3A, 3B, 3C are rounded at different collision timings. The relationship with the impact acceleration that occurs when the end surface of the rod is continuously collided will be described with reference to FIG. As shown in FIG. 7, 21 is a waveform of impact acceleration measured by the strain gauge only generated by the collision of the innermost flying object 3 </ b> A, and 22 is a distortion gauge generated only by the collision of the intermediate flying object 3 </ b> B. The waveform of the measured impact acceleration, 23 is the waveform of the impact acceleration measured by the strain gauge generated only by the collision of the outermost flying object 3C. From these, the impact acceleration generated by the collision of one flying object It can be seen that the duration of is short. 24 three projectile 3A, 3B, and controls the collision timing of 3C (in FIG. 7, the innermost projectile 3B collision from the middle after a time lapse of alpha ₁ of the projectile 3A collide, likewise, alpha ₂ (the outermost flying object 3C collided after the time of α ₁ <α ₂ )), showing the waveform of the impact acceleration measured by the strain gauge as a whole caused by superposition, and the impact acceleration as a whole duration alpha ₃ of, it can be seen that longer than the duration of the individual impact acceleration above. Reference numeral 25 denotes a distortion waveform of the superimposed incident pulse obtained when the three flying bodies 3A, 3B, and 3C collide. Reference numeral 26 denotes a combined strain waveform (not contributing to the generation of impact acceleration) of the elastic wave pulse reflected from the other end face of the round bar 1.

  When the movement of the other end face of the round bar 1 is measured with a laser interferometer, the other end face of the round bar 1 to which a sensor for detecting acceleration to be calibrated is attached is directly irradiated with laser. In the case of measuring with a strain gauge, noise can be suppressed by converting the output results of a plurality of gauges into values of strain gauges at a representative position subjected to signal processing. At the same time, the correction function previously obtained by directly measuring the movement of the other end face of the round bar 1 with the above-mentioned laser interferometer is applied to the frequency characteristics of the gauge output at the representative position, and further necessary. Accordingly, the movement of the sensor mounting plane of the jig 5 is directly measured by another laser interferometer, and another correction function obtained in advance is applied to obtain a result equivalent to the result measured by the laser interferometer. It can also be done.

  The contact surface between the outermost flying object 3C and the launch tube 2 is lubricated. In order to control (for example, narrow) the frequency band of the elastic wave pulse generated inside the rod by the launch of each flying object, a polymer material, plastic, wood, or the like can be attached to the tip of the flying object. Alternatively, a flying body having a structure in which the flying body main body portion has a laminated structure with different materials such as metal, polymer material, plastics, and wood can be used.

(Contents of arithmetic processing performed in the arithmetic unit)
<Motion of the end face of the round bar>

Actually, since it is impossible to attach a strain gauge to the boundary between the other end face of the round bar 1 and the side face of the round bar 1, L _n (n = n = 1... N) is assumed to be pasted at a position (see FIG. 2) separated from each other. Further, the representative position of the gauge was affixed to a plurality of positions in the axial direction of the rod and L _1. In this case, at each position of L _n (n = 1... N), the incident wave and the reflected wave on the end face to which the sensor for detecting acceleration is attached must be observed separately.

By calculating the average of output signals using a plurality of strain gauges, it is possible to reduce the influence of noise and measure minute dynamic strain, that is, low peak acceleration.

For that purpose, the following equation is used.

Since the motion acceleration of the end face is obtained from the equation (11), the frequency response of the sensor that detects the acceleration is obtained according to the following equation.

  When a plurality of strain gauges are attached in the circumferential direction of the round bar 1 and more than one strain gauge is attached in the axial direction, when only one strain gauge is attached in the axial direction. (That is, a plurality of strain gauges are attached in the circumferential direction) converts the output from each of a plurality of strain gauge attachment positions in the axial direction into an output at one representative position in the axial direction. No calculation is necessary. The strain gauge output pasted at one position in the axial direction is merely regarded as the gauge output pasted at the representative point position.

<When the sensitivity axis for translation acceleration is a single axis sensor>
For each measurement, the axial acceleration (translational acceleration) of the other end face of the round bar 1 to which a sensor 4 (inertia sensor or acceleration sensor) for detecting acceleration is attached via a jig is measured as described above. The detection signal from the sensor 4 that detects acceleration is processed as follows. The measurement is performed three times by changing the jig for each measurement and adjusting the jig so that a state as described below is obtained (details will be described later).

One sensitivity axis of the sensor for detecting acceleration is defined as the Z axis, and its output signal is _defined as (0, 0, a _oz (t)). Of course, the X-axis component and the Y-axis component are zero. On the other hand, if the input acceleration signal of the sensor is a _ix (t), a _iy (t), a _iz (t), the following equation is established. At this time, a coordinate system fixed to a sensor for detecting acceleration using acceleration a _i1 , a _i2 , a _i3 for each measurement generated on the end face by an elastic wave pulse reflected on the other end face of the round bar 1 as a vector. Adjust the jig so that it is linearly independent when viewed from above.

See FIG. 2 for symbols α ₁ , α ₂ , α ₃ , β ₁ , β ₂ , β ₃ , γ ₁ , γ ₂ , γ ₃ . It is not assumed that the input acceleration is on the sensitivity axis of the sensor that detects the acceleration. Assuming that it is on the sensitivity axis, the input acceleration is not considered as a vector. At this time, the matrix sensitivity is represented by a 1 × 3 matrix (S _zx , S _zy , S _zz ). S _zx is cross sensitivity because cross sensitivity since showing the relationship between the Z axis output signal to the X-axis component of the input acceleration, the S _zy represent the relation between the Z-axis output signal to the Y-axis component of the input acceleration, S _zz is Z Since the relationship with the output signal with respect to the Z-axis component of the axis input acceleration is expressed, the spindle sensitivity is expressed. At this time, the relationship between the output signal and the input signal is expressed by the following equation. L represents Laplace transform.

On the left side of equation (17), if the calculation result for the k-th frequency component of each input signal is G _ik (ω _k ), the following simultaneous linear equations for sensitivity at ω _k are derived.

The coefficient determinant of the equation (18) is zero because it is assumed that the input acceleration vector in the axial direction of the round bar is linearly independent when viewed from the coordinate system fixed to the sensor for detecting acceleration. Since it cannot be, it can always be solved regardless of the value of ω _k . In this way, since the main axis sensitivity and the lateral sensitivity are obtained for the kth frequency component, the main axis sensitivity and the lateral sensitivity are obtained as a function of frequency by performing the same calculation for the necessary frequency components. Can do.

When it is not obtained as a matrix, only S _zz is obtained. In this case, a sensor for detecting acceleration is directly attached to the other end face of the round bar 1 without using a jig. The Z axis of the coordinate system fixed to the end face of the rod matches the sensitivity axis of the sensor that detects acceleration.

<In the case of a sensor with a biaxial sensitivity axis for translational acceleration>
For each measurement, the axial acceleration (translational acceleration) of the other end face of the round bar 1 to which a sensor 4 (inertia sensor or acceleration sensor) for detecting acceleration is attached via a jig is measured as described above. The detection signal from the sensor 4 that detects acceleration is processed as follows. The measurement is performed three times by changing the jig for each measurement and adjusting the jig so that a state as described below is obtained (details will be described later).

The two sensitivity axes of the sensor that detects acceleration are the Y-axis and the Z-axis, and the output signals are (0, a _oy (t), a _oz (t)). ω is the angular frequency. Of course, the X-axis component is zero. On the other hand, the input acceleration signal of the sensor is (a _ix (t), a _iy (t), a _iz (t)). At this time, a coordinate system fixed to a sensor for detecting acceleration using acceleration a _i1 , a _i2 , a _i3 for each measurement generated on the end face by an elastic wave pulse reflected on the other end face of the round bar 1 as a vector. When the jig is adjusted so as to be linearly independent when viewed from the above, the following relationship is established.

See FIG. 2 for symbols α ₁ , α ₂ , α ₃ , β ₁ , β ₂ , β ₃ , γ ₁ , γ ₂ , γ ₃ . It is not assumed that the input acceleration is on a plane (sensitivity plane) determined by the two sensitivity axes of the sensor that detects the acceleration. Assuming that it is on a plane determined by the sensitivity axis, the input acceleration is not regarded as a vector. At this time, the matrix sensitivity is represented by a 2 × 3 matrix shown below.

When the equation (20) is written down for the accelerations a _i1 , a _i2 , a _i3 generated by the elastic wave pulse reflected from the end face of the round bar, the following three equations are obtained.

In formulas (21), (22), and (23), when the left side is Laplace transformed and the transfer functions of the respective axis outputs and a _i1 , a _i2 , and a _i3 are obtained, formula (24) is obtained.

Since the determinant of the coefficient matrix of equation (24) can be proved to be non-zero and can be solved regardless of the frequency, the main axis sensitivity and the lateral sensitivity can be obtained as a function of the frequency.

When not obtained as a matrix, only S _yy and S _zz are obtained. In this case, a sensor for detecting acceleration is directly attached to the other end face of the round bar 1 without using a jig. Since the Z axis of the coordinate system fixed to the end face of the rod needs to coincide with the sensitivity axis of the sensor that detects acceleration, the data of the sensor that detects acceleration is taken in the Z axis direction, and then the Y of the sensor that detects acceleration Axis data needs to be taken. It is important that the mounting conditions are the same.

<When the sensitivity axis for translational acceleration is a triaxial sensor>
For each measurement, the axial acceleration (translational acceleration) of the other end face of the round bar 1 to which a sensor 4 (inertia sensor or acceleration sensor) for detecting acceleration is attached via a jig is measured as described above. The detection signal from the sensor 4 that detects acceleration is processed as follows. The measurement is performed three times by changing the jig for each measurement and adjusting the jig so that a state as described below is obtained (details will be described later).

The three sensitivity axes of the sensor for detecting acceleration are the X axis, the Y axis, and the Z axis, and the output signals are (a _ox (t), a _oy (t), and a _oz (t)). On the other hand, the X-axis component, Y-axis component, and Z-axis component of the input acceleration signal of the sensor are set to (a _ix (t), a _iy (t), a _iz (t)). At this time, a coordinate system fixed to a sensor for detecting acceleration using acceleration a _i1 , a _i2 , a _i3 for each measurement generated on the end face by an elastic wave pulse reflected on the other end face of the round bar 1 as a vector. When the jig is adjusted so as to be linearly independent when viewed from the above, the following relationship is established.

It is not assumed that the input acceleration is on a space (sensitivity space) determined by the three sensitivity axes of the sensor that detects the acceleration. Assuming that it is in the sensitivity space determined by the sensitivity axis, the input acceleration is not regarded as a vector. The sensor that detects acceleration projects the input acceleration space onto the sensitivity vector space.
At this time, the matrix sensitivity is represented by a 3 × 3 matrix shown below.

When the equation (26) is written down for the accelerations a _i1 , a _i2 , a _i3 for each measurement generated on the end surface by the elastic wave pulse reflected on the other end surface of the round bar 1, the following three equations are obtained. become.

In equations (27), (28), and (29), when the left side is subjected to Laplace transform and the transfer functions of the respective axis outputs and a _i1 , a _i2 , and a _i3 are obtained, equation (30) is obtained.

Since the determinant of the coefficient matrix of Equation (30) can be proved to be non-zero and can be solved regardless of the frequency, the main axis sensitivity and the lateral sensitivity can be obtained as a function of the frequency. When not obtained as a matrix, only S _xx , S _yy , and S _zz are obtained. In this case, a sensor for detecting acceleration is directly attached to the other end face of the round bar 1 without using a jig. Since the Z axis of the coordinate system fixed to the end face of the rod needs to coincide with the sensitivity axis of the sensor that detects acceleration, the data of the sensor that detects acceleration is taken in the Z axis direction, and then the Y of the sensor that detects acceleration It is necessary to take data in the axial direction and then take data in the X-axis direction of a sensor that detects acceleration. The order is not important. It is important that the mounting conditions are the same.

<Measurement method of peak sensitivity>
Conventionally, when the peak value of impact acceleration is obtained by a sensor that detects acceleration, the influence of lateral sensitivity cannot be considered. The reason is that there was no basic idea of defining the sensitivity of the sensor for detecting acceleration as a matrix and obtaining the element as a function of frequency. The peak sensitivity in the sensitivity axis direction is written as PS _xx , PS _yy , PS _zz . The peak sensitivity in the direction orthogonal to the sensitivity axis is written as PS _xy , PS _xz , PS _yx , PS _yz , PS _zx , PS _zy . When the sensitivity axis is one sensor, in the case of two sensors, the Z axis is used as the sensitivity axis common to the case of three sensors, and therefore the peak sensitivity calculation method will be described for the Z axis. The same applies to the X axis and the Y axis.

  (a) For sensors that detect acceleration with a single sensitivity axis

Here, the peak sensitivity determined by the maximum value of the + side input signal and the maximum value of the + side output signal is PS ⁺ , and the peak sensitivity determined by the maximum value of the − side input signal and the maximum value of the − side output signal is PS ^- to define. As a result, the following formula is derived from the above table.

Here, all peak sensitivities PS ⁺ _zx , PS ^- _zx , PS ⁺ _zy , PS ^- _zy , PS ⁺ _zz , and PS ^- _zz are measured in a state where the internal seismo system of the sensor that detects acceleration is in resonance. must not. Peak sensitivity increases when resonance occurs. Therefore, as shown in FIG. 8, if the peak sensitivity is expressed as a function of the frequency band of the impact acceleration applied to the sensor for detecting the acceleration, the usable frequency band and the peak sensitivity can be determined as the value starts to increase. It becomes possible to ask. Control of the frequency band of the impact acceleration applied to the sensor for detecting the acceleration becomes possible by multiplexing the flying objects (multiple flying objects).

  (b) In the case of a sensor that detects acceleration with two sensitivity axes

  Since the peak sensitivity with respect to the Z axis is the same as the result already described in (a), the result with respect to the Y axis is written as follows.

Here all peak sensitivities PS ⁺ _zx , PS ^- _zx , PS ⁺ _zy , PS ^- _zy , PS ⁺ _zz , PS ^- _zz , PS ⁺ _yx , PS ^- _yx , PS ⁺ _yy , PS ^- _yy , PS ⁺ _yz , PS ^- _yz should not be measured in a state where the internal seismogram of the sensor that detects acceleration is causing resonance. Peak sensitivity increases when resonance occurs. Therefore, as shown in FIG. 8, if the peak sensitivity is expressed as a function of the frequency band of the impact acceleration applied to the sensor for detecting the acceleration, the usable frequency band and the peak sensitivity can be determined as the value starts to increase. It becomes possible to ask. Control of the frequency band of the impact acceleration applied to the sensor for detecting the acceleration becomes possible by multiplexing the flying objects (multiple flying objects).

  (c) For a sensor that detects acceleration with three sensitivity axes

The peak sensitivities for the Y and Z axes are the same as those already described in (b) and (a), so the results for the X axis are as follows.

Here all peak sensitivities PS ⁺ _zx , PS ^- _zx , PS ⁺ _zy , PS ^- _zy , PS ⁺ _zz , PS ^- _zz , PS ⁺ _yx , PS ^- _yx , PS ⁺ _yy , PS ^- _yy , PS ⁺ _yz , PS ^- _yz , PS ⁺ _xx , PS ^- _xx , PS ⁺ _xy , PS ^- _xy , PS ⁺ _xz , PS ^- _xz are measured when the internal seismo system of the sensor that detects acceleration is in resonance. Don't be. Peak sensitivity increases when resonance occurs. Therefore, as shown in FIG. 7, if the peak sensitivity is expressed as a function of the frequency band of the impact acceleration applied to the sensor for detecting acceleration, the usable frequency band and the peak sensitivity are determined by the point where the value starts to rise. It becomes possible to ask. Control of the frequency band of the impact acceleration applied to the sensor for detecting the acceleration becomes possible by multiplexing the flying objects (multiple flying objects).

<Proof that simultaneous linear equations can be solved without the coefficient determinant becoming zero>
Appendix Proof that simultaneous linear equations can be solved without zero coefficient determinant

The fact that the determinant shown in (101) is not zero can be proved by using the determinant property that the value of the determinant does not change even if the rows are replaced. That is, if the current third row of (101) is exchanged with the current second row, the fifth row and the third row at that stage are exchanged, and the fourth and fifth rows at that stage are exchanged, It becomes the following matrix.

The value of this determinant is not zero due to the following assumptions derived from the assumption that the input acceleration seen from the coordinate system fixed to the acceleration sensor is linearly independent.

Similarly, regarding the matrix of equation (102), the following is proved when considered in the same manner as equation (101).

(A) is a figure which shows the structure of the acceleration sensor calibration apparatus by the impact acceleration using the multiple flying body in embodiment of this invention, (b) shows the structure of three flying bodies (one part shows a cross section) ). It is a figure which shows the relationship between the coordinate system fixed to the end surface of a round bar, and the coordinate system fixed to the sensor which detects an acceleration. It is a figure which shows an example of the structure of a multiple flying body. It is a figure which shows another example of the structure of a multiple flying body. It is a figure which shows another example of the structure of a multiple flying body. It is a figure which shows another example of the structure of a multiple flying body. The impact acceleration generated on the other end face of the round bar due to the single collision of each flying object on one end face of the round bar 1 and the three flying bodies 3A, 3B, 3C are shifted by the collision timing and one of the round bars It is a figure which shows the relationship with the impact acceleration which generate | occur | produces when it is made to collide with the end surface of this. It is a figure which shows the relationship between the peak sensitivity of an acceleration sensor, and the frequency band of the impact acceleration applied to an acceleration sensor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Metal round bar 2 Launch tube 3A, 3B, 3C Flying object 3D Flange 4 Acceleration sensor 5 Jig 6 Strain gauge 7, 8 Laser beam
9 Valve 10 High pressure air source
11 Valve open / close control device
12 Laser light source 13 Mirror 14 Light receiving element 15 Counter

Claims (36)

By causing a plurality of flying objects to collide with one end face of a round bar, translational acceleration generated in a direction perpendicular to the other end face is attached to the other end face of the round bar via a jig. measured by a sensor for detecting an acceleration sensitive axis of the shaft, when repeating this process several times, as the jig, using a plurality of jigs having different three-dimensional shapes, the other end face of the rod By making the angle between the mounting plane to the sensor and the mounting plane of the sensor different from each other for each of the plurality of measurements , a vector whose component is a direction cosine of the translational acceleration to the coordinate system of the sensor is linear. on which was formed to be independent of the state, before Symbol solving the multiple simultaneous linear equations based on the measurement results of, and obtains the sensitivity matrix of the sensor relative to the translational acceleration vector Characteristic measurement method of a sensor for detecting the speed. An impact is applied to one end face of the round bar for each measurement, and the movement of the end face caused by reflection of an elastic wave pulse generated inside the round bar by the impact at the other end face of the round bar The other end face of the round bar when measured by a sensor that has a single-axis or multi-axis sensitivity axis attached to the other end face of the bar with a jig and detects acceleration and an input acceleration measuring device As the jig , the three-dimensional shapes are different from each other so that a vector whose component is a direction cosine of the translational impact acceleration generated in the direction perpendicular to the sensor is in a linearly independent state between each measurement. Using a plurality of jigs, the angle formed between the mounting plane of the jig on the other end face of the round bar and the mounting plane of the sensor of the jig is different for each measurement, and each measurement is performed. The sensor obtained by By solving the simultaneous linear equations based on the measurement result of the degree measuring apparatus, characteristic measurement method of a sensor for detecting an acceleration and obtains the sensitivity matrix of the sensor. In claim 2,
The input acceleration measuring device is at least one strain gauge attached to a side surface of the round bar,
When obtaining a transient strain signal of an elastic wave pulse incident on the other end face of the round bar from the strain gauge output signal, using at least a first-order term of a Scalak solution that is an analytical solution developed in a series. A characteristic measurement method for a sensor that detects a characteristic acceleration. In claim 2,
The input acceleration measuring device is a plurality of strain gauges attached to the side surface of the round bar in the axial direction,
For the multiple strain gauge outputs, the difference in waveform due to the difference in the pasting position in the axial direction is converted to a single waveform by calculating the average of each strain gauge output and added to reduce noise. A method for measuring characteristics of a sensor for detecting acceleration, characterized by performing signal processing. For each measurement, two or more flying objects that are independent from each other, the flying objects having different distances between the tip and one end surface of the round bar are simultaneously driven by the high-pressure gas to fly, and the round bar By sequentially colliding with one end face with a slight time difference, an impact in the vertical direction is applied to one end face of the round bar for each measurement, and an elastic wave pulse generated inside the round bar due to the impact is applied to the round bar. A sensor having a single-axis or multi-axis sensitivity axis that detects the movement of the end face generated by reflection on the other end face via a jig, and detects the acceleration, and an input acceleration measuring device when measured by the, as a vector to a direction cosine components of the coordinate system of the sensor translational acceleration generated in the other end face of the rod is linearly independent states between each measurement, the jig As a tool, each other Using different a plurality of jigs of the body shape, the other angle of the mounting plane of the mounting plane the sensor to the end face of the rod and run by different for each measurement, obtained by the respective measurement An acceleration detection sensor characteristic measurement method comprising: solving a simultaneous linear equation based on a measurement result of the sensor and the input acceleration measurement device to obtain a sensitivity matrix of the sensor. In claim 5,
The two or more flying bodies are separably supported by a pressure receiving body that can directly receive the pressure of a high-pressure gas and move within the launch tube, and the tip positions in the flying direction are different in the supported state. A characteristic measurement method for a sensor that detects acceleration. In claim 6,
The two or more flying bodies include a first flying body located on a central axis of the pressure receiving body and at least an outer side of the first flying body and coaxial with the first flying body. A sensor characteristic measuring method for detecting acceleration, comprising: a second flying object. In claim 6 or 7,
One of the said 1st and 2nd flying bodies is integral with the said pressure receiving body, The characteristic measuring method of the sensor which detects the acceleration characterized by the above-mentioned. In any of claims 5 to 8,
At least one of the two or more flying objects is characterized in that the material of the collision end portion is different from the material of the other portions, and a method for measuring the characteristics of a sensor for detecting acceleration. In any one of Claims 5 thru | or 9,
Each of the two or more projectiles is characterized in that at least a part of the material is different from the material of other projectiles, and a sensor characteristic measuring method for detecting acceleration. In any of claims 5 to 10,
A method for measuring characteristics of a sensor for detecting acceleration, characterized in that a metal sphere is brought into contact with and fixed to one end face of the round bar and the impact is applied to the metal sphere. In any of claims 5 to 11,
The input acceleration measuring device is at least one strain gauge affixed to a side surface of the round bar, and is a sensor characteristic measuring method for detecting acceleration. In claim 12,
A plurality of strain gauges are affixed in the axial direction of the round bar, and for each strain gage output, the difference in waveform due to the difference in the affixing position of the strain gage in the axial direction is expressed by elastic wave propagation theory, A method for measuring the characteristics of a sensor for detecting acceleration, comprising: converting to an elastic waveform at a representative position based on a solution, and performing signal processing to reduce noise by adding and averaging the converted strain gauge output waveforms . In claim 12 or 13,
Based on the strain gauge output signal and the strain of the incident elastic wave pulse, which is the measurement result of the laser interferometer that directly measures the movement of the other end face of the round bar, the Laplace transform of the strain at the representative position of the strain gauge is performed. , Rutotomoni obtain a correction function by dividing the Laplace transform of strain of the incident acoustic wave pulses to the sensor mounting end surface, the correction function was determined by strain of the strain gauge output, the frequency of the sensor for detecting the acceleration The transfer function of the sensor is obtained by multiplying the characteristics, and the distortion of the elastic wave pulse incident on the end face of the round bar to which the sensor is attached, which is necessary for obtaining the input acceleration applied to the sensor, is obtained. A sensor characteristic measurement method for detecting acceleration. In claim 14,
When determining the strain of the elastic wave pulse, the output signal of the strain gauge is divided by a correction function derived using the measurement result of a laser interferometer that directly measures the movement of the other end face of the round bar. Correct and use both the measurement result of another laser interferometer that directly measures the movement of the mounting surface of the sensor in the jig and the measurement result of the laser interferometer that directly measures the movement of the other end face of the round bar A method for measuring characteristics of a sensor for detecting acceleration, wherein the direction cosine is corrected. In any one of Claims 12 and 13,
When obtaining a transient strain signal of an elastic wave pulse incident on the other end face of the round bar from the strain gauge output signal, using at least a first-order term of a Scalak solution that is an analytical solution developed in a series. A characteristic measurement method for a sensor that detects a characteristic acceleration. In any of claims 5 to 11,
The input acceleration measuring device is a laser interferometer that directly measures the movement of the other end face of the round bar, and is a sensor characteristic measuring method for detecting acceleration. In any of claims 5 to 17,
When determining the characteristics of the sensor, by controlling the frequency bandwidth of the impact acceleration signal applied to the sensor, the peak sensitivity value of the sensor obtained as a function of the frequency bandwidth of the impact acceleration signal is increased. A method for measuring the characteristics of a sensor for detecting acceleration, characterized by obtaining a signal frequency bandwidth and peak sensitivity of an input impact acceleration that can be actually used for acceleration measurement. A sensor having a single-axis or multi-axis sensitivity axis to detect acceleration;
A jig for attaching the sensor, when a plurality of flying objects collide with one end face of the round bar, and repeatedly performing the measurement of applying translational acceleration to the other end face of the round bar , The translation plane determined for each measurement is formed of a plane that is attached to the other surface of the round bar of the sensor so that a vector whose component is a direction cosine of the translation acceleration to the coordinate system of the sensor is linearly independent. angles different, and a plurality pieces of jigs having different three-dimensional shapes,
Computation means for solving a simultaneous linear equation based on each measurement result obtained by executing the measurement for each jig and obtaining a sensitivity matrix of the sensor with respect to the translational acceleration vector is provided. Sensor characteristic measurement device that detects acceleration. Means for applying an impact to one end face of the round bar for each measurement; an input acceleration measuring device for measuring the movement of the other end face of the round bar; and a plurality of sensors for attaching the sensor to the other end face of the round bar. A jig generated by reflection of an elastic wave pulse generated inside the round bar by the impact applied to one end face of the round bar by the means for applying the impact on the other end face of the round bar; the movement of the end face, measured by the sensor and the input acceleration measuring device for detecting acceleration having a sensitive axis of the single or multi-axis, when repeating this process several times, a plurality of jigs having different three-dimensional shapes a is, as the vector to the direction cosine components of the coordinate system of the sensor translational acceleration generated in the other end face of the rod is linearly independent states between each measurement, the other of said rod Mounting plane to the end face And the angle is different jig with mounting plane of the serial sensor, wherein solving the system of linear equations based on the measurement result of each measurement of the sensor and the input acceleration measuring device is obtained by executing, the sensitivity matrix of the sensor An apparatus for measuring characteristics of a sensor for detecting acceleration, characterized by comprising a calculating means for obtaining. In claim 20,
The input acceleration measuring device is at least one strain gauge attached to a side surface of the round bar,
The calculation means obtains a transient strain signal of an elastic wave pulse incident on the other end face of the round bar from the strain gauge output signal, and at least a first-order solution of the Scalak solution that is an analytical solution developed in a series. A sensor characteristic measuring apparatus for detecting acceleration, characterized by using a term. In claim 20,
The input acceleration measuring device is a plurality of strain gauges attached to the side surface of the round bar in the axial direction,
The arithmetic means converts the difference in waveform due to the difference in the axial attachment position with respect to the plurality of strain gauge outputs into a waveform at a single point by calculating an average of each strain gauge output, and noise. A device for measuring characteristics of a sensor for detecting acceleration, characterized by performing signal processing for reducing noise. Two or more flying objects that are independent of each other and having different distances between the tip and one end face of the round bar are simultaneously driven by the high-pressure gas to fly, and the one end face of the round bar is minutely moved. Means for applying an impact to one end face of the round bar by sequentially colliding with a time difference; an input acceleration measuring device for measuring the movement of the other end face of the round bar; and one of the round bars by the means for applying the impact. The movement of the end face caused by the reflection of the elastic wave pulse generated inside the round bar by the impact applied to the end face of the round bar on the other end face of the round bar When measuring multiple times by the sensor to be detected and the input acceleration measuring device, a vector whose component is a direction cosine of the translational acceleration generated on the other end face of the round bar to the coordinate system of the sensor is So that the linearly independent states, a plurality of jig mounting the sensor on the other end surface of the rod, formed by the attachment plane and the mounting plane of the sensor to the other end surface of the rod The sensitivity matrix of the sensor is obtained by solving simultaneous linear equations based on the measurement results of the jig and the input acceleration measuring device obtained by executing the respective measurements with jigs having different three-dimensional shapes so that the angles are different from each other. An apparatus for measuring characteristics of a sensor for detecting acceleration, characterized by comprising an arithmetic means. In claim 23,
The two or more flying bodies are separably supported by a pressure receiving body that can directly receive the pressure of a high-pressure gas and move within the launch tube, and the tip positions in the flying direction are different in the supported state. Sensor characteristic measuring device that detects acceleration. In claim 24,
The two or more flying bodies include a first flying body located on a central axis of the pressure receiving body and at least an outer side of the first flying body and coaxial with the first flying body. A sensor characteristic measuring apparatus for detecting acceleration, characterized by having one second flying object. In claim 24 or 25,
A sensor characteristic measuring apparatus for detecting acceleration, wherein any one of the first and second flying bodies is integral with the pressure receiving body. A device according to any of claims 23 to 26.
At least one of the two or more flying objects is a sensor characteristic measuring device for detecting acceleration, characterized in that the material of the collision end is different from the material of the other parts. A device according to any of claims 23 to 27.
Each of the two or more flying objects is a sensor characteristic measuring device for detecting acceleration, characterized in that at least a part of the material is different from the material of another flying object. A device according to any one of claims 23 to 28.
A sensor characteristic measuring device for detecting acceleration, further comprising a metal sphere fixed in contact with one end face of the round bar and applying the impact to the metal sphere. 30.
The input acceleration measuring device is at least one strain gauge affixed to a side surface of the round bar, and is a sensor characteristic measuring device for detecting acceleration. In claim 30,
A plurality of strain gauges are affixed in the axial direction of the round bar, and the arithmetic means elastically changes a waveform difference due to a difference in affixing position of the strain gauge in the axial direction with respect to each strain gauge output. Based on wave propagation theory, that is, Scalak's solution, it is converted to an elastic waveform at the representative position, and the converted strain gauge output waveform is added and averaged to perform signal processing that reduces noise and detects acceleration Sensor characteristic measurement device. In claim 30 or 31,
The calculation means is based on the strain gauge output signal and the strain of the incident elastic wave pulse, which is a measurement result of a laser interferometer that directly measures the movement of the other end face of the round bar. Is obtained by dividing the Laplace transform of strain in the Laplace transform of the strain of the incident elastic wave pulse on the sensor mounting end surface, and the correction function is obtained by the strain of the strain gauge output. An elastic wave pulse incident on the end face of the round bar to which the sensor is attached, which is necessary for obtaining the transfer function of the acceleration detection sensor by multiplying the frequency characteristic of the sensor to be detected and obtaining the input acceleration applied to the sensor. An apparatus for measuring characteristics of a sensor for detecting acceleration, characterized by obtaining a strain of the sensor. In claim 32,
When determining the strain of the elastic wave pulse, the output signal of the strain gauge is divided by a correction function derived using the measurement result of a laser interferometer that directly measures the movement of the other end face of the round bar. Correct and use both the measurement result of another laser interferometer that directly measures the movement of the mounting surface of the sensor in the jig and the measurement result of the laser interferometer that directly measures the movement of the other end face of the round bar A sensor characteristic measuring device for detecting acceleration, wherein the direction cosine is corrected. In any one of Claim 30, 31.
The arithmetic means obtains at least a first-order term of scrack as an analytical solution developed in a series when obtaining a transient strain signal of an elastic wave pulse incident on the other end face of the round bar from the strain gauge output signal. A sensor characteristic measuring apparatus for detecting an acceleration characterized by being used. 30.
The input acceleration measuring apparatus is a laser interferometer that directly measures the movement of the other end face of the round bar, and is a sensor characteristic measuring apparatus for detecting acceleration. In any of claims 23 to 35
When the acceleration sensor is calibrated, the arithmetic means controls the frequency bandwidth of the impact acceleration signal applied to the sensor, thereby obtaining the peak sensitivity of the sensor as a function of the frequency bandwidth of the impact acceleration signal. A sensor characteristic measuring device for detecting acceleration, characterized by obtaining a signal frequency bandwidth and peak sensitivity of an input impact acceleration that can be actually used for acceleration measurement after increasing the value.

Images