JP4312104B2 - Magnetic body internal structure measuring method and apparatus - Google Patents

Description

  The present invention relates to a method and an apparatus for measuring a magnetic material, and specifically, a method for nondestructively measuring the internal structure of a magnetic material such as a steel plate, a welded state, an internal defect, or a strain or stress distribution, and the like. Relates to the device.

  Conventionally, the internal state of a magnetic steel sheet has been measured and inspected in a nondestructive manner. One example is the inspection of welds on steel plates. Spot welding is generally used for assembling various sheet metal products including the automobile industry. Spot welding is a resistance welding process in which stacked metal base materials are sandwiched between the tips of electrodes that have been appropriately shaped at the tip, and current is concentrated and heated locally in a relatively small part, while simultaneously pressing with the electrode. It is.

  Usually, the spot welded portion has a cross-sectional structure shown in FIG. The surface of the welded portion is recessed as compared with the outside of the welded portion due to pressurization (indentation portion), and the size of the recess is referred to as an indentation diameter. The inside of the welded portion is formed by a nugget portion (welded portion) that is the center of the welded portion and a peripheral crimp portion. The nugget portion is a portion where the metal has once melted and solidified. On the other hand, a crimping | compression-bonding part is a part crimped | bonded by metal surfaces. The dimension of the nugget part is called the nugget diameter, and the sum of the nugget part and the crimping part (the part that is actually joined) is called the joint diameter. In spot welding, superposed metal base materials are welded with dots, but in many cases, it is inspected whether the welding strength is sufficient in order not to increase the number of unnecessary welding points.

  As a method of nondestructively measuring the welding strength, conventionally, an AC magnetic field generated by a coil through which a high-frequency current is applied is applied to the spot weld, and the resulting change in the coil inductance results in a change in the spot weld. A method for determining pass / fail is known. This conventional method utilizes the property that the magnetic permeability changes between the nugget portion and the outside of the nugget portion, and detects the change in the magnetic permeability as a change in inductance, thereby determining the quality of the spot welded portion. (See Patent Document 1).

  Recently, a method for nondestructively measuring various internal structures of a magnetic body including an nugget diameter as well as a nugget diameter by using a static magnetic field has been proposed (see Patent Document 2). In this method, a static magnetic field is applied to the object to be measured to magnetize the object to be measured, and after the static magnetic field is cut off, the time variation of local magnetic flux at multiple positions near the object to be measured is measured. In addition, there is no problem of variations in measured values due to frequency fluctuations associated with the method using an alternating magnetic field.

  The magnitude of the magnetic flux generated when a static magnetic field is applied and the time constant with which the magnetic flux decays with time when the static magnetic field is interrupted are the magnetic resistance (or inductance) inside the magnetic material through which the magnetic flux passes. It is affected by the difficulty of magnetic flux return (hysteresis) and the electrical resistance of the magnetic material. Further, how much signal is induced by the time change of the magnetic flux with respect to the means for measuring the time change of the magnetic flux depends on the spatial arrangement of the magnetic substance to be measured and the measurement means. For this reason, the strength of the local magnetic flux generated when a static magnetic field is applied and what kind of signal the local magnetic flux that decays in time induces the measurement means It has a distribution that reflects the internal structure.

  According to this method, the change of the magnetic flux is caused by the change corresponding to the attenuation of the first magnetic flux caused by the static magnetic field and the second magnetic flux caused by the eddy current induced by the attenuation of the first magnetic flux. Considering the combination with the change in the second magnetic flux corresponding to the attenuation, the internal structure can be predicted using the time constants of these two attenuations.

JP-A-5-149923

Japanese Patent No. 3098193

  Although various internal structures of the magnetic material can be predicted by the method disclosed in Patent Document 1, it is a problem to further increase the S / N ratio and enable more accurate measurement. In view of such problems, the present invention provides a method and apparatus for measuring the internal structure of a magnetic body by measuring temporal changes in local magnetic flux at a plurality of positions in the vicinity of an object to be measured that is a magnetic body. A mechanism and an apparatus for accurately measuring the internal structure of a magnetic material are provided by reexamining the generation mechanism of the time change in detail. The “internal structure” used in the present invention includes not only a mechanical structure but also a broad meaning including magnetic properties and chemical composition.

  The magnetic substance internal structure measuring method of the present invention occurs in a step of applying a static magnetic field to a non-measurement part of a magnetic substance, a step of cutting off the static magnetic field, and the whole part to be measured after the static magnetic field is cut off. A step of measuring a change in magnetic flux, a step of measuring a change in local magnetic flux generated at a plurality of positions in the vicinity of the measured portion after the static magnetic field is interrupted, and a change in the magnetic flux generated in the entire measured portion. Regression representing the local magnetic flux change occurring at a plurality of positions in the vicinity of the measured part A step of obtaining a function, and a step of obtaining a characteristic value relating to the internal structure of the magnetic body based on at least one of a coefficient of the regression function and an attenuation constant.

  The step of regressing the measurement value of the change in the magnetic flux generated in the entire measured part is caused by the excitation current and the corresponding eddy current in the entire measured part corresponding to the change in the magnetic flux generated in the entire measured part. A step of regressing may be included.

  Further, the step of obtaining a regression function representing local magnetic flux changes occurring at a plurality of positions in the vicinity of the measured part includes local magnetic flux changes induced by the static magnetic field cutoff and the static magnetic field cutoff. The magnetic flux change caused by the eddy current induced in the entire measured part by the magnetic field, the magnetic flux change caused by the local eddy current induced at the measurement position by the interruption of the static magnetic field, and the local You may make it include the process of synthesize | combining the transient response of the means which measures the change of magnetic flux, and calculating | requiring the said regression function.

  Further, the step of obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the part to be measured includes the magnitude of the local magnetic flux induced by the static magnetic field cutoff at the time of the static magnetic field cutoff. The damping constant after interruption, the magnitude of the magnetic flux generated by the local eddy current at the time of interruption of the static magnetic field, the time constant of the change after interruption, and the transient response of the means for measuring the change of the local magnetic flux A step of obtaining at least one of the magnitude at the time of the static magnetic field interruption and the attenuation constant after the interruption may be included.

  Further, the step of measuring the change in the excitation current that generates the static magnetic field may include the step of generating the static magnetic field by an excitation coil and directly measuring the excitation coil current after the static magnetic field is interrupted.

  The magnetic substance internal structure measuring device of the present invention includes means for applying a static magnetic field to a non-measurement part of a magnetic substance, means for measuring a change in magnetic flux generated in the whole part to be measured, and a plurality of parts near the part to be measured. Means for measuring a change in local magnetic flux generated at a position, and a measured value of magnetic flux generated in the entire measured part after the static magnetic field is interrupted and a local value generated at a plurality of positions near the measured part. Means for obtaining a regression function representing local magnetic flux changes occurring at a plurality of positions in the vicinity of the portion to be measured based on a measured value of magnetic flux, and a characteristic value relating to the internal structure of the magnetic material based on the regression function A means for obtaining.

  Here, the static magnetic field may be applied by an exciting coil. Further, the exciting coil current may be measured to measure a change in magnetic flux that occurs in the entire part to be measured. Furthermore, in order to measure a local magnetic flux change occurring at a plurality of positions, it can be measured using a coil in which an inductor is connected in series.

  In this way, the time change of the magnetic flux applied to the entire non-measurement part is obtained, and the change in the pickup coil signal of a plurality of minute parts is obtained reflecting this, so that it is more accurate than the conventional one. It was possible to construct a simple model and obtain a measurement signal with a large S / N ratio.

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing a schematic configuration and operation of a measuring apparatus according to the present embodiment. This measuring apparatus includes an exciting coil 4 having n turns including an iron core 1 made of ferrite, and an array sensor 3 that detects a change in magnetic flux density in the vicinity of a steel plate 2 that is a sample. The array sensor 3 includes a plurality of magnetic sensors arranged in a line, and is held at a predetermined interval so that a change in magnetic flux can be detected. As the magnetic sensor, various sensors such as a Hall element can be used. In this embodiment, a pickup coil is used. In the figure, a resistance r ₃ ⁱ corresponding to the i-th magnetic sensor is shown. An exciting coil driving circuit 5 that drives the exciting coil 4 includes a voltage source V, a switch SW, and a resistor R. When applying a static magnetic field, the switch SW is turned on, and to interrupt the static magnetic field, the switch SW is turned off. The current of the magnetic sensor induced by the change in the magnetic field near the object to be measured due to the static magnetic field interruption is converted into a voltage by the resistor r ₃ ⁱ and output. As for the outputs of the magnetic sensors constituting the array sensor 3, only one output is selected by a switching selection circuit (not shown) and sequentially output. As the load resistance r ₃ ⁱ of the pickup coil, one common r ₃ can be used for each coil.

  In this embodiment, a static magnetic field is applied to the welded portion of the steel plate 2 by this apparatus, and then the static magnetic field is interrupted, and the change in magnetic flux density after the interruption is measured by the array sensor 3. The part is determined. In the present invention, a model described below is assumed as a model for giving a change in magnetic flux after the static magnetic field is interrupted.

FIG. 2 shows a state in which the switch SW is turned on and a static magnetic field is applied to the steel plate 2. By turning on the switch SW, the voltage from the voltage source V is applied to the exciting coil 4. Most of the magnetic flux φ generated by the exciting current I ₁ flowing through the exciting coil 4 passes through the exciting coil 4 because the magnetic permeability of the steel plate 2 is high. In the figure, the i-th pickup coil 3 ⁱ constituting the array sensor 3 is shown. When the static magnetic field is interrupted, a current flows through the pickup coil 3 ⁱ in response to the attenuation of the static magnetic field and is converted into a voltage by the load resistance r ₃ ⁱ to detect a change in the magnetic flux flowing through the pickup coil 3 ^i. It will be.

At this time, as shown in FIG. 3, a minute region in which the magnetic flux density can be regarded as substantially uniform is considered. Here, it is assumed that the magnetic flux density is almost uniform in a magnetic circuit having a minute area surrounded by the pickup coil. Figure showed i-th magnetic circuit 6 ⁱ surrounded by the pickup coil 3 ^i. As shown in the figure, the magnetic flux entering the steel plate in this region flows through without interruption. The magnetomotive force acting on the magnetic circuit 6 ⁱ is only due to the current I ₁ flowing through the exciting coil having n turns, and thus becomes nI ₁ . On the other hand, the magnetic resistance of the magnetic circuit having a small area is obtained as the sum R ⁱ of the magnetic resistances of the iron core, the sample (steel plate), and the air gap. Therefore, the magnetic flux N ⁱ flowing through the local magnetic circuit 6 ⁱ is obtained by dividing the magnetomotive force nI ₁ by the magnetic resistance R ⁱ , that is, nI ₁ / R _i . Flux N ⁱ of the static magnetic field generated by the switch SW is turned on, those reflecting the size of the local air gap to increase or decrease depending on the indentation of the local permeability mu _2, the steel sheet surface of the steel sheet become. The magnetic circuit of the magnetic flux flowing through the entire excitation coil, because regarded as composite circuit local magnetic circuit in parallel, the magnetic resistance is determined as the parallel resistance of the magnetoresistive R ⁱ of local magnetic circuit .

FIG. 4 shows a state in which the switch SW is turned off to block the static magnetic field. The exciting coil current I ₁ is consumed by the load resistance r ₁ and decreases. At this time, since the magnetomotive force also decreases, the magnetic flux passing through the magnetic path 6 ⁱ also starts to decrease. Then, an eddy current I ₂ ⁱ flows against the decrease in the magnetic flux inside the steel plate 2 which is a conductive material. This eddy current is consumed and reduced by the electric resistance r ₂ ⁱ of the steel plate 2. On the other hand, since the magnetic flux surrounded by the coil begins to decrease, the pickup coil 3 ⁱ responds transiently and generates an electromotive force. By this electromotive force, a current I ₃ ⁱ defined by the resistance r ₃ ⁱ flows, and I ₃ ⁱ × r ₃ ⁱ can be measured as a voltage.

Thus, when the static magnetic field is interrupted, unlike the state in which the static magnetic field is constantly applied, as the magnetomotive force related to the magnetic circuit 6 ⁱ , in addition to the excitation current I ₁ flowing through the excitation coil, a vortex There is a current I ₂ ⁱ and a current I ₃ ⁱ flowing through the pickup coil.

In the following, an equation that accurately describes the state of change in magnetic flux when the static magnetic field is interrupted is established. First, considering an excitation circuit using current I ₁ flowing through the excitation coil 4, almost all magnetic flux flows through the excitation coil 4, and an electromotive force is generated by multiplying the time derivative of the total magnetic flux N ^tot by the number of turns n. Also, according to Kirchhoff's law, the algebraic sum of the electromotive force of the closed circuit is equal to the algebraic sum of the voltage drop in the closed circuit,

It becomes. Here, regarding the current flowing through the pickup coil, only the current I ₃ ^j flows while only the j-th pickup coil is closed, and the other coils do not flow when opened.

In the equivalent circuit of the sample through which the eddy current I ₂ ⁱ flows, the i-th magnetic circuit may be considered, and the generated eddy current is consumed by the resistance r ₂ ⁱ ,

It becomes.
In the pickup circuit using the current I ₃ ⁱ flowing through the pickup coil, the pickup is formed in one turn in this example, and the current I ₃ ⁱ is consumed by the load resistance r ₃ of the pickup coil.

It becomes.

  When the equations (1), (3), and (5) are modified by introducing the self-induction coefficient L and the mutual induction coefficient M, the following equations (6) to (8) are obtained for the excitation circuit, the sample equivalent circuit, and the pickup circuit. ) Is guided.

Here, in the excitation circuit equation (6), since the entire current is considered, the sum Σ of the magnetic circuit is also considered for the mutual inductance M ₁₂ with the eddy current (sample equivalent circuit). As described above, regarding the current flowing through the pickup coil, the current I ₃ ⁱ flows only in the closed i-th pickup coil.

As described above, in the model of the present invention, attention is paid to the current I ₁ flowing through the exciting coil, the eddy current I ₂ ⁱ and the current I ₃ ⁱ flowing through the pickup coil as currents affecting the change in magnetic flux. Since the eddy current I ₂ ⁱ cannot be measured, it is necessary to measure the current I ₁ flowing through the exciting coil and the current I ₃ ⁱ flowing through the pickup coil.

FIG. 5 shows a conceptual diagram of a measurement circuit according to an embodiment of the present invention. A static magnetic field is applied to the sample by the exciting coil 4 provided with the magnetic shield SH. An exciting coil drive circuit 5 that supplies an exciting current for exciting the exciting coil 4 is formed of a voltage source V, a resistor r1, and a semiconductor switch SW. When the semiconductor switch SW is on, an exciting coil current flows through the circuit of the voltage source V, the exciting coil 4 and the semiconductor switch SW, and a static magnetic field is formed by the coil 4. When the semiconductor switch SW is turned off, a current flows through the resistor r _{i due} to the electromotive force generated in the exciting coil 4. A change in magnetic flux at that time is detected by a pickup coil constituting the array sensor 3, and a detection signal is output. In the figure, a signal from the i-th (for example, 1 ≦ i ≦ 15) pickup coil 3 ⁱ is input from the i-th input changeover switch S _i . Furthermore, in the present embodiment, the change in the exciting coil current is also input from the input selector switch S _0. The input signal is amplified by the amplifier Amp and output to the signal processing circuit.

When such a circuit is used, the current I ₁ flowing through the exciting coil and the current I ₃ ⁱ flowing through the pickup coil are detected and output as signals by the changeover switches S ₀ and S _i . In particular, since the change of the entire static magnetic field can be directly input to the signal processing circuit by the change of the exciting coil current, it is suitable for carrying out the present invention.

It is considered that the current I ₃ ⁱ flowing through the i-th pickup coil is almost completely described by the simultaneous equations (6) to (8). Therefore, by solving the simultaneous equations (6) to (8), the change in the magnetic flux density after the static magnetic field interruption can be completely understood. However, in order to solve this, it is necessary to make it continuous for every fine area where the magnetic flux density is uniform over the entire magnetic path, for example, every 20 to 30 areas. Therefore, if the simultaneous equations (6) to (8) remain, I ₃ ⁱ cannot be obtained other than numerical calculation.

Therefore, in the present invention, an approximation that substantially describes the pickup coil current I ₃ ⁱ is employed, and if necessary, a correction term is inserted to obtain a solution. As will be described in detail below, the change in the magnetic flux as a whole, that is, the exciting coil current I ₁ that attenuates and the corresponding eddy current total are obtained by an appropriate approximation and then obtained from the equations (6) and (7). Based on the exciting coil current I ₁ and eddy current, the local magnetic flux change, that is, the pickup coil current I ₃ ⁱ is approximately obtained from the equations (7) and (8).

First, in equation (6) describing the excitation coil current (excitation circuit), when measuring the excitation coil current, the circuit of the pickup coil is off and the pickup coil I ₃ ⁱ can be ignored. Further, the eddy current I ₂ ⁱ can be considered to be coupled to the exciting coil current as a whole, and can be handled as the combined current I ₂ ^eff .

On the other hand, the equation (7) describing the eddy current (sample equivalent circuit) also deals with the sum of the minute magnetic circuit, that is, the combined current I ₂ ^eff . The current flowing through the pickup coil is negligible, and the exciting coil current I ₁ and the eddy current I ₂ ^eff as a whole are expressed as the following two simultaneous equations (9) and (10).

The simultaneous equations (9) and (10) can be easily solved, and initial conditions I ₁ = 0, I ₂ ^eff = 0, L ₁ dI ₁ / dt + M ₁₂ ^eff dI ₂ ^eff / dt = 0, L ₂ ^eff dI ₂ ^{If eff} / dt + M ₁₂ ^eff dI ₁ / dt = 0, the result is as follows.

Thus, the exciting coil current I ₁ is described as the sum of two exponential functions. Here, I ₂ ^eff cannot be directly measured by an eddy current generated in the sample. Only the excitation current I ₁ is a measurement target.

Here, in order to evaluate the exciting current I ₁ , consider a case where the sample is a magnetic material such as ferrite and the electrical conductivity is extremely low. In this case, since eddy current does not flow (I ₂ ^eff = 0), the equation (9) becomes an equation representing simple attenuation by the attenuation constant α1 = r1 / L ₁ , and the solution is a single unit as follows: Expressed with an exponential function.

FIG. 6 is a diagram showing a state of attenuation of the exciting coil current when the same ferrite as the iron core of the exciting coil is used as a sample. The figure shows a result 200 obtained by fitting the measured value 100 in the range of elapsed time 15 to 52 using the exponential function of Expression (13). The regression residual 300 is shown at the bottom of the figure. As can be seen from the figure, when trying to fit over a long period of time, the displacement of the fitting increases at a point where the elapsed time is small. This is because the attenuation constant λ _A determined by the self-inductance L1 of the exciting coil and the load resistance r ₁ is constant and the self-inductance L ₁ is constant without depending on time when the exciting coil current decays in a relatively long time. Indicates that it cannot be done. That is, it is necessary to capture the change in the magnetic permeability μ that defines the self-inductance L ₁ as the excitation current decreases and the magnetic field decreases. Considering the time derivative of self-inductance in this way, Equation 9 is as follows.

This equation can be integrated by variable separation, but since the solution does not express the exciting coil current I ₁ explicitly, the component proportional to the square of time t is considered as an approximate function.

From the above, it was found that the excitation coil current I ₁ is well expressed. FIG. 7 is a diagram in which the example of the previous ferrite is fitted with the elapsed time of 20 to 52 according to Equation 15. Also in FIG. 7, the measured value 100, the regression function 200, and the residual 300 are shown. Compared with the previous example, it can be seen that it matches very well.

  In this way, it has been found that the following may be used as a function expressing the exciting coil current and the eddy current of the entire sample.

Compared to the previous equation, λ _A (t) was λ _A · t in the previous equation, but in this equation, a correction term proportional to λ _A · t + λ _A2 · t ² and t ² is added. ing. Since the eddy current rapidly decreases, the term due to the eddy current is expressed only by the term λ _B · t proportional to time.

  Here, FIG. 8 shows an example measured for the welded portion of an actual steel plate. This is a result of fitting a state of attenuation of the exciting coil current by Equation 16 by measuring the spot weld. In the regression function 200, an eddy current term 500 appears as a transient term in addition to the current decay term 400 which is a steady term. This term can be interpreted as appearing as a result of the entire region in which the magnetic flux flows during the measurement responding by generating an eddy current to the change in the magnetic flux, which is combined with the excitation circuit.

Next, an equation describing the pickup coil current I ₃ ⁱ given as a solution to the simultaneous equations (7) and (8) is obtained. The pickup coil current is driven by the exciting coil current combined with the eddy current of the entire sample, and at the same time, it is interpreted as being combined only with the eddy current immediately below the pickup coil (on the same minute magnetic circuit). Accordingly, the pickup coil current I ₃ ⁱ is described by the following simultaneous equations while ignoring the time dependence of the inductance.

Hereinafter, the derivation process of the equation will be omitted and the outline will be described. First, the pickup coil current I ₃ ⁱ that satisfies the above equation responds independently to two exponential function inputs of a transient response term that is a general solution when I ₁ is not present and a special solution I _i (t). Expressed as a sum of terms. That is, it is expressed as a linear sum of a term due to the current of the pickup coil itself having the damping constant λc ⁱ , a term due to the exciting coil current having the damping constant λ _A (t) and a term due to the entire eddy current having the damping constant λ _B. Suppose that it is given by the following equation.

Then, the general solution and the special solution are obtained, and the attenuation coefficient λ and the coefficients A to C are determined. Here, it should be noted that the coefficient A ₃ ⁱ is differentiated in consideration of the dependence on the time t. Furthermore, the initial conditions are I ₂ ⁱ = I ₃ ⁱ = 0, L ₂ ⁱ dI ₂ ⁱ / dt + M ₂₃ ⁱ dI ₃ ⁱ / dt + M ₁₂ ⁱ dI ₁ / dt = 0, L ₃ ⁱ dI ₃ ⁱ / dt + M ₂₃ ⁱ dI _When calculation is performed assuming ₂ ⁱ / dt + M ₁₃ ⁱ dI ₁ / dt = 0, the pickup coil current I ₃ ⁱ is expressed as follows.

Where each coefficient is

And each attenuation constant is

It becomes.
As described above, the quantities with the subscript i reflect the local magnetic resistance penetrating the pickup coil, so the attenuation constant λc and the coefficients A to C with the subscript i are used for nondestructive inspection. be able to. For example, in the non-destructive inspection of a welded portion, the shape and dimensions of the nugget portion can be obtained by obtaining a sudden change point of the distribution of λc ₁ ⁱ as described in Japanese Patent No. 3098193. By determining the sudden change point of the distribution of ₂ ⁱ , the shape and dimensions of the joint can be determined.

In the present embodiment, with the attenuation of the exciting coil current I _1, then since to obtain a pick-up coil signal I ₃ ^i, can be determined more accurately pickup coil signal I ₃ ^i. Further, if the exciting coil current I ₁ is directly measured, it is convenient for processing the pickup coil signal I ₃ ⁱ . That is, the attenuation constants λ _A and λ _B can be determined by regression of the exciting coil current I ₁ . Also, the attenuation constants λ _A and λ _B reflect the total magnetoresistance surrounded by the sample and the excitation coil. It can also be used to determine whether the sensor is correctly in contact with the sample.

  In addition, for example, the rise of the BH curve is different between the welded portion and the base material, that is, a dent is formed at the welded portion, and the magnetic resistance increases with the melting of the welded portion. Therefore, if the welding part and the base material are measured under different excitation conditions, the amount of change will be different. Therefore, it is possible to distinguish between the welded portion and the base material portion by performing the measurement twice while changing the excitation condition and looking at the difference in the coefficient or the attenuation constant. Thus, the internal structure of the sample can also be estimated by performing measurement a plurality of times under different excitation conditions and comparing the obtained coefficient and / or attenuation constant.

Further, when an inductor is inserted in series with the pickup coil, and the attenuation constant of the pickup coil current itself is fixed to about 10 times the attenuation constant of the excitation coil, _one of the attenuation constants λc ₁ and λc ₂ is thereby obtained. Can be forcibly fixed. This facilitates the process of determining the regression equation.

Finally, the measurement method of the present embodiment can be summarized as follows.
(1) Measure exciting coil current.
(2) Fit the attenuation of the exciting coil current I _1A (t) using the following equation as a regression function.

(3) Fit the eddy current component I _1B (t) from the residual of the attenuation of the exciting coil current using the following equation as a regression function.

(4) The following equation is fitted as a regression function as a portion I ₁ ⁱ (t) due to the change of the exciting coil current (and eddy current component) in the i-th pickup signal I ₃ ⁱ (t), and the coefficient A ₃ ⁱ and B ₃ ⁱ are determined.

(5) Fit the i-th pickup signal I ₃ ⁱ (t) using the following equation as a regression function.

(6) The internal structure is estimated from the coefficient of the pickup signal I ₃ ⁱ (t) determined as described above and the attenuation constant.

Here, this embodiment is an example for carrying out the present invention, and the form of the regression function, the coefficient, and the time dependence of the attenuation constant are appropriately selected. For example, in this embodiment, the term of the square of time t is inserted in order to fit the exciting coil current. However, in consideration of necessity such as when using an air-core coil, a normal term excluding this term is used. An exponential function may be used. Further, the coefficient A ⁱ may not be dependent on the time t. Further, for convenience, it is also effective to fit C ₃₁ ⁱ xp (−λ _c1 t) and B ₃ ⁱ exp (−λ _B t) having close attenuation constants as one function.

Example 1
For spot lap welding of mild steel (plate thickness 1.2mm, TS270MPa class) plates, we produce welded joints with a sufficiently large nugget diameter (4√t, t: plate thickness mm) and joints in a pressure-welded state with almost no melting Then, non-destructive inspection was attempted by applying the present invention.

About 0.33 A is applied to the exciting coil current, and the voltage of the exciting coil resistance r ₁ (FIG. 1) after being cut off is measured. The voltage of the load resistance r ₃ ⁱ is measured for each pickup coil, and the pickup coil is measured. Got the signal. FIG. 9 shows an example of the result of measuring the pickup coil signal.

  The signal based on the exciting coil current was regressed according to equation (16), and the pick-up coil signal for each pickup coil (in this case, coil Nos. 0 to 15) was regressed according to equation (23).

10 (a) and 10 (b) show the coefficient A ₃ ⁱ for the joint in pressure contact and the welded nugget diameter 4√t, and the relative values of the values at both ends (coils No. 0 and 15). It is shown as a value. The horizontal axis in the figure is the coil number, and the vertical axis is the relative value of A ₃ ⁱ . Compared with the dent indicated by the coefficient A ₃ ⁱ of the pressure contact state shown in FIG. 10A, the dent indicated by the coefficient A ₃ ⁱ of the welded portion having a nugget diameter of 4√t shown in FIG. 10B becomes deeper. Yes. Thereby, it turns out that a press-contact state and a molten state can be distinguished. Based on this figure, for example, the depth or area of the dent in the figure can be obtained and used as a parameter to be a measure of the molten state.

In this case, it is not necessary to fit all four functions according to the equation (23). If λ _A (t) is determined from the regression of the excitation coil current measurement value, the latter half of the measurement data, the case of FIG. Then, the coefficient A ₃ ⁱ can be determined from the ratio of the measured value after time 25 and exp (−λ _A (t)). Furthermore, such a measurement can be performed a plurality of times under different excitation conditions, and the melting state can be determined from their mutual values (such as the depth and area of the dent).

(Example 2)
In addition to the coefficient A ₃ ⁱ , other regression results can be used for measurement. Here, a method using the coefficient C ₃₁ ⁱ and the exponent λ _c1 ⁱ will be described. FIG. 11 shows the result of regression using one of the pickup coil signals of the first embodiment using the coefficient A ₃ ⁱ and, for simplicity, a simple exponential function A ₃ ⁱ exp (−λ _A t). Indicates. Here, the area of the shaded portion in the figure corresponds to the eddy current component and reflects the value of the coefficient C ₃₁ ⁱ .

12 (a) and 12 (b) show the area of the hatched portion in FIG. 11 for each pickup coil in the pressure-welded state and the welded state, respectively, and the relative values to the measured values at both ends (coils No. 0 and 15) It is shown as a value. The horizontal axis of the figure is the coil number, and the vertical axis is the area relative value of the shaded portion in FIG. 11 corresponding to the eddy current component. In this case, the excitation current is about 0.48A. Compared to the pressure contact state shown in FIG. 12 (a), the weld portion with a nugget diameter of 4√t shown in FIG. 12 (b) has a large convex shape, and distinguishes between the pressure contact state and the molten state. You can see that The area of this convex portion can be used as an auxiliary parameter of A ₃ ⁱ , unlike A ₃ ^{i in} behavior with respect to the excitation condition.

It is a figure which shows the outline of the measuring apparatus of this invention. It is a figure explaining the magnetic circuit of the measuring apparatus of this invention. It is a figure explaining the magnetic circuit enclosed by the pick-up coil of the measuring apparatus of this invention. It is a figure explaining the change of the magnetic flux after interrupting the exciting current of the measuring apparatus of this invention. It is a figure which shows an example of the measuring circuit of this invention. It is a figure which shows an example which regressed the exciting coil current which applied the static magnetic field to the ferrite sample according to this invention. It is a figure which shows the other example which regressed the exciting coil current which applied the static magnetic field to the ferrite sample according to this invention. It is a figure which shows an example which regressed the exciting coil current which applied the static magnetic field to the welding part of the steel plate according to this invention. It is a figure which shows an example of a pickup coil signal. (A), (b) are diagrams showing the relative values of the coefficients A ₃ ⁱ of pressure contact with the welding state. It is a figure which shows an example of the regression result of a pickup coil current. (A), (b) is a figure which shows the relative value of the area of the shaded part of FIG. 11 of a press-contact state and a welding state, respectively. It is a figure which shows the cross-section of a general spot weld part.

Explanation of symbols

1 ... Iron core 2 ... Sample (steel plate)
3 ... the array sensor 4 ... exciting coil 5 ... resistance of resistor r ₃ ⁱ ... pickup coils of the load resistor r ₂ ⁱ ... sample of the exciting coil drive circuit r ₁ ... exciting coil (steel plate)

Claims (17)

Applying a static magnetic field to the part to be measured of the magnetic material;
Cutting off the static magnetic field;
Measuring a change in magnetic flux generated in the entire measured part after the static magnetic field is interrupted;
Measuring a change in local magnetic flux generated at a plurality of positions near the portion to be measured after the static magnetic field is interrupted;
The measured value of the change in magnetic flux generated in the entire measured part is regressed, and then the local change in magnetic flux generated in a plurality of positions in the vicinity of the measured part is regressed to the plurality of positions in the vicinity of the measured part. Obtaining a regression function representing the local magnetic flux change that occurs,
A method for measuring the internal structure of a magnetic body, comprising: obtaining a characteristic value related to the internal structure of the magnetic body based on at least one of a coefficient of the regression function and an attenuation constant.   2. The magnetic body according to claim 1, wherein the step of measuring a change in magnetic flux generated in the entire portion to be measured after the static magnetic field is interrupted includes a step of measuring the magnitude of the excitation current that generates the static magnetic field and the change thereof. Internal structure measurement method.   The step of regressing the measured value of the change in the magnetic flux generated in the entire measured part is caused by the excitation current and the corresponding eddy current in the entire measured part corresponding to the change in the magnetic flux generated in the entire measured part. The method for measuring the internal structure of a magnetic body according to claim 1, comprising a step of returning as if it were. A step of obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the measured part,
A change in local magnetic flux induced by the interruption of the static magnetic field;
A change in magnetic flux caused by an eddy current induced in the entire measured part by the interruption of the static magnetic field;
A change in magnetic flux caused by a local eddy current induced in the measurement position by the interruption of the static magnetic field;
A transient response of the means for measuring the local flux change;
The method for measuring the internal structure of a magnetic body according to claim 1, further comprising the step of obtaining the regression function by combining The step of obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the measured part further includes
The magnitude of the local magnetic flux induced by the interruption of the static magnetic field when the static magnetic field is interrupted and the damping constant after the interruption, and the magnitude and interruption of the magnetic flux generated by the local eddy current when the static magnetic field is interrupted 5. The method according to claim 4, further comprising: determining at least one of a time constant of a later change, a magnitude of the transient response of the means for measuring the change of the local magnetic flux when the static magnetic field is interrupted, and an attenuation constant after the interrupt. The magnetic body internal structure measuring method as described.   3. The magnetic body according to claim 2, wherein the step of measuring the change in the excitation current that generates the static magnetic field includes the step of generating the static magnetic field by an excitation coil and directly measuring the excitation coil current after the static magnetic field is cut off. Internal structure measurement method.   The step of measuring the change in the excitation current that generates the static magnetic field is caused by the change in the excitation coil current after the static magnetic field is interrupted by the steady response of the excitation coil current and the eddy current induced in the entire measured part. Recognizing as a composite of transient response of exciting coil current, regression is performed with two exponential functions to obtain respective attenuation constants, and coefficients of the two exponential functions are obtained from local magnetic flux changes at a plurality of positions in the vicinity of the measured part. The method for measuring the internal structure of a magnetic body according to claim 6, further comprising a step of obtaining.   The step of measuring a change in local magnetic flux that occurs at a plurality of positions near the portion to be measured after the static magnetic field is cut off uses a coil in which an inductor is connected in series as a means for measuring the local magnetic flux, The magnetic body internal structure measuring method according to any one of claims 1 to 7, further comprising a step of forcibly determining and measuring a decay constant of a transient response of the coil itself.   The method for measuring the internal structure of a magnetic body according to any one of claims 1 to 8, further comprising a step of performing measurement twice while changing the conditions of the static magnetic field and obtaining a difference between measurement values of each time. Means for applying a static magnetic field to the part to be measured of the magnetic material;
Means for measuring a change in magnetic flux generated in the entire measured part;
Means for measuring a local magnetic flux change occurring at a plurality of positions in the vicinity of the portion to be measured,
A plurality of positions in the vicinity of the measured part based on a measured value of the magnetic flux generated in the entire measured part after the static magnetic field is interrupted and a measured value of a local magnetic flux generated in a plurality of positions in the vicinity of the measured part. A means for obtaining a regression function representing a local magnetic flux change occurring in
Means for determining a characteristic value related to the internal structure of the magnetic body based on the regression function.   The magnetic body internal structure measuring device according to claim 10, wherein the means for applying a static magnetic field is an exciting coil.   The magnetic body internal structure measuring device according to claim 11, wherein the means for measuring a change in magnetic flux generated in the entire measured part includes a means for measuring an exciting coil current.   Means for measuring changes in local magnetic flux generated at a plurality of positions near the portion to be measured after interruption of the static magnetic field is a coil in which an inductor is connected in series, and the attenuation constant of the transient response of the coil itself is set. The apparatus for measuring an internal structure of a magnetic body according to any one of claims 10 to 12, wherein the measurement is forcibly determined.   Means for obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the part to be measured includes a local magnetic flux magnitude at the plurality of positions immediately after the static magnetic field is interrupted and the plurality of positions after the interruption. 14. The magnetic body internal structure measuring apparatus according to claim 10, wherein a local magnetic flux attenuation constant is obtained. Means for obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the measured part,
A change in local magnetic flux induced by the interruption of the static magnetic field;
A change in magnetic flux caused by an eddy current induced in the entire measured part by the interruption of the static magnetic field;
A change in magnetic flux caused by a local eddy current induced in the measurement position by the interruption of the static magnetic field;
A transient response of the means for measuring the local flux change;
The apparatus for measuring the internal structure of a magnetic body according to claim 14, wherein the regression function is obtained by combining. Means for obtaining a regression function representing a local magnetic flux change occurring at a plurality of positions in the vicinity of the portion to be measured,
The magnitude of the local magnetic flux induced by the interruption of the static magnetic field when the static magnetic field is interrupted and the damping constant after the interruption, and the magnitude and interruption of the magnetic flux generated by the local eddy current when the static magnetic field is interrupted The at least one of a time constant of a subsequent change, a magnitude of a transient response of the means for measuring the local magnetic flux change when the static magnetic field is interrupted, and an attenuation constant after the interruption is obtained. 15. The magnetic body internal structure measuring device according to 15.   The means for measuring a change in the excitation current that generates the static magnetic field is caused by a change in the excitation coil current after the static magnetic field is cut off by a steady response of the excitation coil current and an eddy current induced in the entire measured part. Recognizing the combination of the transient responses of the exciting coil currents, regression is performed with two exponential functions to obtain respective attenuation constants, and the coefficients of the two exponential functions are determined from changes in local magnetic flux at a plurality of positions near the measured portion. The apparatus for measuring an internal structure of a magnetic body according to claim 16, wherein:

Images