JPH07113625B2 - Steel plate flaw detection device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a flaw detection device for a steel sheet, and more particularly to improvement in detection accuracy of micro flaws.

[Prior Art] In order to detect a flaw generated in, for example, a steel plate or a steel pipe, an ultrasonic flaw detection method, an eddy current flaw detection method, a leakage magnetic flux flaw detection method or the like has been conventionally used.

Among these flaw detection methods, the magnetic flux leakage flaw detection method is conventionally used in every place because it can measure even in a bad environment compared to other methods.

FIG. 11 shows the structure of a conventional flaw detection device for steel plates using the magnetic flux leakage flaw detection method. As shown in the figure, the flaw detection device includes a non-magnetic hollow roll 2 that rotates as the steel plate 1 moves,
The hollow roll 2 is provided with a coil 3 wound around a magnetized core 4 for magnetizing the steel plate 1, and a sensor 5 provided between the magnetized cores 4 and configured to detect a leakage magnetic flux, such as a Hall element. .

When detecting a flaw in the steel sheet 1, a direct current is passed through the coil 3 to generate a magnetic field. Then, the moving steel plate 1 is magnetized by the magnetizing core 4 via the roll 2. When the magnetized steel plate 1 has a flaw, a leakage magnetic flux is generated in the flaw portion. This leakage magnetic flux is detected by the sensor 5, and the flaw existing on the steel plate 1 is detected by the magnitude of the output signal from the sensor 5.

In this case, since the output signal of the sensor 5 contains a noise signal, in order to remove the influence of this noise, a threshold value of a certain level is set in advance, and the output signal of the sensor 5 exceeding this threshold value causes a flaw. Is detected to prevent false detection of defects.

[Problems to be Solved by the Invention] However, the above noise is caused by differences in steel types, rolling conditions, etc.
Alternatively, it has various noise levels due to the influence of other magnetic substances at the measurement location. In order to remove these various noises and prevent erroneous detection of flaws, conventionally, the highest level of noise has been set as the threshold value. For this reason, although major flaws can be detected, minute flaws indicated by a signal below the threshold value output from the sensor 5 cannot be detected, and there is a disadvantage that the flaw measurement accuracy decreases.

On the other hand, the higher the sensitivity of the sensor 5, such as the magnetic sensor based on the principle of Japanese Patent Application No. 1-110108, the higher the
Although smaller flaws (defects) can be detected, the tendency to pick up various noises is increased, so that demands for noise removal and highly accurate flaw detection are increased.

The present invention has been made to solve the above disadvantages, and an object of the present invention is to obtain a steel plate flaw detection apparatus capable of detecting minute flaws with high sensitivity and improving the flaw measurement accuracy.

[Means for Solving the Problems] A steel sheet flaw detection device according to the present invention is a first embodiment that receives a magnetic sensor and an output of the magnetic sensor to detect major flaws in the steel sheet.
Of the second threshold value calculating means and the second threshold value calculating means, and the second comparing means for receiving the output of the magnetic sensor and detecting a small flaw in the steel sheet.

The magnetic sensor is a coil installed around the surface of a steel plate that is wound around a ferromagnetic core and moves continuously, and an AC power supply that supplies AC power of a constant frequency and a constant voltage to this coil through a fixed impedance. Means for detecting the positive side voltage and the negative side voltage of the voltage output from both ends of the coil respectively, and the difference between the positive side voltage and the negative side voltage detected by the DC voltage detecting means for calculating the magnetic field strength. And a magnetic detection means for detecting the height.

The first comparing means compares the detection signal from the magnetic sensor with a predetermined first threshold value, and compares the detection signal exceeding the first threshold value and the detection signal not reaching the first threshold value. Determine.

The second threshold value calculating means receives the detection signal from the first comparing means, which does not reach the first threshold value, and detects the peak value of the detection signal for each predetermined moving distance of the steel sheet. Means, a representative value selecting means for selecting a representative value from a plurality of peak values detected by the peak value detecting means, and an average value of representative values sequentially sent from the representative value selecting means to calculate a second threshold value. And an average value calculating means.

The second comparing means is the second threshold value calculated by the second threshold value calculating means.
The threshold value and the detection signal of the magnetic sensor are compared.

Further, it is preferable to supply a pulse current added with a DC bias to the coil of the magnetic sensor.

Further, the flaw detection performance is further improved by variably controlling the DC bias added to the pulse current according to the magnetic detection signal from the magnetic detection means.

[Operation] According to the present invention, a coil of a ferromagnetic material core wound around a ferromagnetic material core and having a constant frequency and a constant voltage is supplied with alternating current power, and exhibits a hysteresis characteristic by a magnetic field generated by an alternating current flowing through the coil. The impedance of the coil is changed according to the magnetic susceptibility, and the output voltage across the coil is changed to a positive / negative symmetrical rectangular waveform. The coil is arranged close to the steel plate, and the leakage magnetic flux from the steel plate changes the positive and negative voltage values of the output voltage across the coil. The changing positive side voltage and negative side voltage are respectively detected, the difference between the detected positive side voltage and negative side voltage is calculated, and the leakage magnetic flux of the steel sheet is detected as an electric signal. As a result, finer flaws can be detected with extremely high sensitivity.

The detection signal from the magnetic sensor is compared with a predetermined first threshold value by the first comparing means, and the detection signal exceeding the first threshold value is output as a main flaw detection.

Further, the second threshold value calculating means calculates the peak value of the detection signal which does not reach the first threshold value determined by the first comparing means,
The second threshold value is calculated by detecting for each predetermined moving distance of the steel sheet, selecting a representative value from the plurality of detected peak values, and calculating an average value of the sequentially obtained representative values.

A small flaw in the steel plate is detected from the detection signal of the magnetic sensor based on the second threshold value.

Further, by supplying a pulse current to which a DC bias is added to the coil of the magnetic sensor, the measurement span of the leakage magnetic flux can be expanded.

Further, the operating point of the magnetic sensor can be automatically compensated by variably controlling the DC bias added to the pulse current according to the magnetic detection signal from the magnetic detection means.

[Embodiment] FIG. 1 is a block diagram showing an embodiment of the present invention.
As shown in the figure, the flaw detection device compares the magnetic sensor 5 with the first threshold value set in the first threshold value setting means 6 by receiving the magnetic detection signal from the magnetic sensor 5, The first comparing means 7 for detecting flaws, the second threshold calculating means 8 for calculating the second threshold, and the signals from the second threshold calculating means and the first comparing means 7 are received. The second comparison means 9 and the display means 10 for detecting a small flaw in the steel plate are provided.

The magnetic sensor 5 includes a rod-shaped ferromagnetic core 11 and a ferromagnetic core.
A coil 12 wound around 11, an AC power supply means 13 for supplying a constant frequency and a constant voltage AC power to the coil 12 via a fixed impedance 14, a DC voltage detection means 15 connected to both ends of the coil 12, and a magnetic field. And a detection means 16. The DC voltage detecting means 15 has a positive polarity detector and a negative polarity detector, and detects the positive side voltage and the negative side voltage of the output voltage of the coil 12, respectively. The magnetic detection means 16 receives the output of the DC voltage detection means 15 and obtains the difference between the positive side voltage and the negative side voltage of the coil 12,
Detects magnetic field strength.

The second threshold value calculation means 8 is the first threshold value from the first comparison means 7.
Receiving a detection signal that does not reach the threshold value, a peak value detecting means 17 for detecting the peak value of the detection signal for each predetermined moving distance of the steel plate, and a plurality of peak values detected by the peak value detecting means 17. Representative value selecting means 18 for selecting a representative value from the average value, and average value calculating means 19 for calculating an average value of the representative values sequentially sent from the representative value selecting means 18 to calculate a second threshold value.
Consists of.

When the flaw detection device configured as described above is used to detect a flaw in the steel sheet 1, for example, as shown in FIG.
The coil 12 wound around the ferromagnetic core 11 of the magnetic sensor 5.
A non-magnetic hollow roll 2 that rotates as the steel plate 1 moves
The coil 3 for magnetizing the steel plate 1 provided inside is installed between the wound magnetic cores 4. The coil 12 wound around the ferromagnetic core 11 of the magnetic sensor 5 is shown in FIG.
As shown in FIG. 3, the coil 3 for magnetizing the steel plate 1 is installed in a hollow roll 2a having the same diameter as the hollow roll 2 and the hollow roll 2 having a magnetized core 4 around which the steel plate 1 is wound. Good.

In explaining the operation of the flaw detection device configured as described above, first, the principle of the magnetic sensor 5 will be described with reference to the voltage waveform diagram of FIG. 3 and the magnetization characteristic diagram of the ferromagnetic core of FIG. To do.

When an alternating voltage E having a constant frequency and a constant voltage is supplied to the coil 12 as shown in the voltage waveform diagram of FIG.
The voltage E ₀ generated at both ends of is determined by the following equation based on the resistance value R of the fixed impedance 14 and the impedance Z _S of the coil 12.

E ₀ = represented by _{E · Z S / (R +} Z S).

The impedance Z _{S of the} coil 12 changes in proportion to the magnetic permeability of the ferromagnetic core 11. That is, when an alternating current is passed through the coil 12 without applying an external magnetic field, the magnetic flux of the coil 12 magnetizes the ferromagnetic core 11. The magnetic permeability of the ferromagnetic core 11 is not constant but changes depending on the strength of the magnetic field, and the magnetization curve exhibits a hysteresis characteristic as shown in FIG. In FIG. 4, B is the magnetic flux density, n is the number of turns of the coil 2, and i is the coil current.

Therefore, the output voltage E ₀ generated at both ends of the coil 12 has a positive and negative symmetrical rectangular waveform, as shown in FIG. 3 (b). And in the state where no external magnetic field is applied, the positive voltage V ₁
And the negative side voltage V ₂ become equal.

When an external magnetic field is applied to the coil 12 in this state, the ferromagnetic core
The magnetic flux crossing 11 is a composite magnetic flux of the magnetic flux generated in the coil 12 and the magnetic flux of the external magnetic field. As a result, the waveform generated at both ends of the coil 12 has a difference between the positive voltage V ₁ and the negative voltage V ₂ as shown in FIG. 3 (c). The external magnetic field can be indirectly measured by comparing the positive side voltage V ₁ and the negative side voltage V ₂ of the output voltage generated at both ends of the coil 12 and obtaining the difference.

Then, when the magnetic flux density of the external magnetic field is measured by the magnetic sensor 5, as shown in FIG. 5, a high output voltage of 0 to 500 mV can be obtained for a minute magnetic flux density of 0 to 10 Gauss. The magnetic flux density can be measured with a much higher detection sensitivity than the output voltage of the Hall element, which is several millivolts.

Next, the operation of the flaw detection device using the magnetic sensor 5 based on the above principle will be described.

For example, as shown in FIG. 2 (a), a coil wound around a ferromagnetic core 11 between magnetized cores 4 of a hollow roll 2 provided close to the surface of a continuously moving steel plate 1. Install 12. In this state, AC power having a constant frequency and a constant voltage as shown in the waveform diagram of FIG. 3A is supplied to the coil 12 from the AC power supply means 13 via the fixed impedance 14. The coil 12 generates a magnetic field by the supplied alternating current, and its magnetic flux causes the ferromagnetic core 11 to have a magnetic field as shown in FIG.
Magnetize until the magnetic flux density B becomes saturated. The strength of the magnetic field of the ferromagnetic core 11 repeats hysteresis characteristics due to the alternating current flowing in the coil 12, and the impedance of the coil 12 is changed at the same frequency as the frequency of the alternating current. On the other hand, a leakage magnetic flux of the magnetic flux generated in the magnetized core 4 provided in the hollow roll 2 and passing through the steel plate 1 is applied to the coil 12. The leakage magnetic flux from the steel plate 1 changes depending on the flaw and the size of the flaw of the steel sheet 1. Therefore, as shown in FIG. 3C, a difference occurs between the positive side voltage V ₁ and the negative side voltage V ₂ of the output voltage on both sides of the coil 12. Therefore, the DC voltage detecting means 15 detects the positive side voltage V ₁ and the negative side voltage V ₂ of the coil 12 to detect the magnetic force.
Send to 16. The magnetic detection means 16 calculates the voltage difference between the positive side voltage V ₁ and the negative side voltage V ₂ sent to obtain a magnetic detection signal proportional to the leakage magnetic flux of the steel plate 1.

The magnetic detection signal S shown in FIG. 6 obtained by continuously measuring the leakage magnetic flux of the steel plate 1 moving by the magnetic sensor 5 is sent to the first comparing means 7. The first comparison means 7 compares the sent detection signal S with the first threshold value TH ₁ set in advance in the first threshold value setting means 6, and the first threshold value T of the detection signal S is compared.
Determine the detection signal S ₂ to the detection signals S ₁ beyond the H ₁ does not exceed the first threshold value TH _1, the detection signal of the main flaw detection signals S ₁ exceeding the first threshold value TH ₁ Is sent to the display means 10 and displayed.

On the other hand, the first threshold value TH ₁ determined by the first comparison means 7
The detection signal S _{2 which} does not exceed the threshold value is sent to the peak value calculating means 17 of the second threshold value calculating means 8. As shown in FIG. 7, the peak value detecting means 17 detects the peak value Pi of the detection signal S ₂ for each moving distance L sent from the moving distance detecting means 20 of the steel plate 1, for example, L = 100 mm. This peak value Pi is sent to the representative value selecting means 18, and a typical value is determined from the sent peak values Pi (i = 1 to n) every predetermined number of times n. As shown in FIG. 12, the representative value has a distribution A of detection signals due to noise and a distribution B of detection signals due to flaws.
When the distribution is such that there is a region C where noise and flaws cannot be distinguished, and Pm at which the distribution of the detection signal S _{2 that} does not exceed the first threshold TH ₁ is the minimum is the representative value, noise and flaws are Area where noise is detected as a flaw in area C where
And a region C ₂ to detect the C ₁ and flaws as a noise to a minimum each can increase the reliability of flaw detection. Thus the first and the process defining a representative value of the threshold TH ₁ and not exceeding the detection signal peak value from the distribution of S ₂ Pi (i = 1~n) is complicated. In order to simplify this process, for example, the peak value Pa at the third height is selected from the sent peak values Pi (i = 1 to n), and a plurality of peak values Pi (i = 1 ~
It may be set as the representative value of n). By determining the peak value Pa at the third height from the peak values Pi (i = 1 to n) sent in this way, as shown in FIG. 6, the peak value Pa rapidly increases due to noise. It is possible to prevent the influence of the detection signal Sn. Further, if the representative value Pa is determined in this way, the probability of detecting noise that is indistinguishable from minute flaws as flaws or detecting minute flaws that cannot be distinguished from noise as noise is somewhat high, but with simple processing It is possible to detect almost all flaws except for the flaw corresponding to the lower limit of the distribution density of flaws.

The representative value Pa of this peak value is sent to the average value calculation means 19. The average value calculating means 19 averages the representative values Paj (j = 1 to m) of the peak values sequentially sent according to the movement of the steel sheet 1 every predetermined m times, and the average value is the second threshold value TH. ₂ is sent to the second comparison means 9.

The second comparing means 9 receives the second threshold value TH ₂ sent from the second threshold value calculating means 8 and the detection signal S ₂ not exceeding the first threshold value TH ₁ sent from the first comparing means 7. Compare the second
The detection signal S ₂ exceeding the threshold value TH ₂ is sent to the display means 10 and displayed as a detection signal of a minute flaw in the steel plate 1.

The second threshold value TH ₂ obtained by the average value calculation means 19 of the second threshold value calculation means 8 may be multiplied by the safety factor α.

In this way, the steel plate 1 detected with high detection sensitivity by the magnetic sensor 5
The magnetic flux leakage detection signal S from the first threshold value TH ₁ is compared with the second threshold value TH ₂ which changes depending on the compression conditions and the like, so that small flaws as well as major flaws are detected. It is possible to greatly improve the accuracy of flaw detection.

Furthermore, when measuring the magnetic flux leakage, the temperature fluctuates and the coil 12
Even if the winding resistance and the magnetic permeability of the ferromagnetic core 11 change, the impedance of the coil 12 due to this change changes equally with the positive and negative polarities of the magnetizing current, so they compensate each other.
Drift does not occur in the difference between the positive side voltage and the negative side voltage of the coil 12 due to temperature change, and the leakage magnetic flux can be accurately measured without the influence of temperature fluctuation.

In addition, by magnetizing the ferromagnetic core 11 until it is sufficiently saturated with the current flowing through the coil 12, the output voltage generated at both ends of the coil 12 is clipped to a constant value, and the output voltage at both ends of the coil depends only on the strength of the external magnetic field. Since the amplitude and phase of the positive electrode and the negative electrode are changed, the AC power supply means 13
Even if the output voltage and the resistance value of the fixed impedance 14 change a little, the detection sensitivity does not change, and the leakage magnetic flux can be measured with high sensitivity.

In the above embodiment, the case where continuous AC power as shown in FIG. 3 (a) is supplied from the AC power supply means 13 to the coil 12 wound around the ferromagnetic core 11 of the magnetic sensor 1. As described above, by outputting a pulse current from the AC power supply means 13 and adding a DC bias to this pulse current, the reference point of the magnetization curve of the ferromagnetic core 11 is changed, and the measurement span of the magnetic flux density detection is changed. By shifting and moving, it is possible to further improve the defect detection accuracy.

FIG. 8 shows a magnetic sensor 5 of an embodiment in the case of supplying a pulse current with a DC bias applied to the coil 12. As shown in the figure, a high-frequency voltage pulse current is supplied from the AC power supply means 13 to the adder 21, and a DC bias is supplied from the DC power supply 22 to the adder 21. In the adder 21, a DC bias is added to the high frequency voltage to generate a combined voltage, and the combined voltage is amplified by the power amplifier 23 and supplied to the coil 12 via the fixed impedance 14 composed of a resistor.

In this embodiment, the value of the DC bias current from the DC power source 22 is changed to 0 mA, 50 mA, 100 mA, 150 mA, 200 mA,
As a result of measuring the output voltage of the coil 12 with respect to the change of the magnetic flux density of the external magnetic field, the detection sensitivity characteristic shown in FIG. 9 was obtained.

As can be seen from FIG. 9, when the DC bias current is added to the pulse current, the linear characteristic of the output voltage does not change greatly even if the value of the DC bias current is changed. The obtained range is about 0 to 40 gauss, and the measurement span of the magnetic flux density can be moved compared to about 0 to 20 gauss when the DC bias current is 0 mA, and the measurement area can be changed easily. .

Further, the DC bias current from the DC power supply 22 may be made variable according to the detection signal S output from the magnetic detection means 16 so that a good measurement span can always be obtained even if the measurement conditions change.

FIG. 10 is a block diagram showing an embodiment in which the DC bias current is varied. As shown in the figure, the magnetic detection means 16
Detection signal S of the differential amplifier through the low-pass filter 24
Send to 25. The differential amplifier 25 is a detection signal S and a reference voltage generator.
The reference voltage from 26 is compared and the difference voltage is compared with the DC power supply 22
Send to. The DC power supply 22 changes the DC bias current by this difference voltage and sends it to the adder 21.

In this way, the operating point of the magnetic sensor 5 is detected by the detection signal S of the magnetic detection means 16, the differential voltage from the reference voltage of the reference voltage generator 26 is obtained by the differential amplifier 26, and the DC bias from the DC power source 22 is obtained. By the control, the detection signal S of the magnetic detection means 16 in the state where there is no defect is automatically compensated so that it is always 0V.

By automatically compensating the operating point in the center of the measurement span of the magnetic sensor 5 as described above, a good measurement span can always be ensured even if the measurement conditions change, and the flaw detection performance can be further improved.

[Effects of the Invention] As described above, the present invention provides a coil wound around a ferromagnetic core with AC power of a constant frequency and a constant voltage, and exhibits a hysteresis characteristic by a magnetic field generated by an AC current flowing through the coil. The impedance of the coil is changed according to the magnetic permeability of the ferromagnetic core, and the output voltage across the coil is changed into a positive and negative symmetrical rectangular wave. The coil is arranged close to the steel plate, and the leakage magnetic flux from the steel plate changes the positive and negative voltage values of the output voltage across the coil. The changing positive side voltage and the negative side voltage are respectively detected, the difference between the detected positive side voltage and negative side voltage is calculated, and the leakage flux of the steel sheet is detected as an electric signal. Can be measured with high accuracy.

In addition, the high-accuracy detection signal from the magnetic sensor is compared with a predetermined first threshold value to detect a major flaw with the detection signal exceeding the first threshold value and to calculate the second threshold value. Since the small flaw of the steel sheet is detected from the detection signal of the magnetic sensor with the second threshold value obtained by the means as a reference, the flaw of the steel sheet can be accurately detected.

Further, by supplying a pulse current added with a DC bias to the coil of the magnetic sensor, the measurement span of the leakage magnetic flux can be expanded, and the flaw accuracy can be further improved.

In addition, the DC bias added to the pulse current is variably controlled according to the magnetic detection signal from the magnetic detection means, so that the operating point of the magnetic sensor is automatically compensated, and even if the measurement condition changes, it is always good. The measurement span can be secured, and the flaw detection performance can be further improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIGS. 2 (a) and 2 (b) are layout diagrams of the coils of the above embodiment, and FIGS. 3 and 4 are views of the magnetic sensor of the above embodiment. Showing the principle,
3 (a), (b), and (c) are voltage waveform diagrams,
FIG. 4 is a magnetization characteristic diagram of the ferromagnetic core, FIG. 5 is an output characteristic diagram of the magnetic sensor of the above embodiment, FIG. 6 is a detection signal distribution diagram showing detection signals of the magnetic sensor with respect to movement of the steel plate, and FIG.
FIG. 8 is an explanatory diagram showing the operation of the second threshold value calculating means, FIG. 8 is a block diagram showing another embodiment, FIG. 9 is an output characteristic diagram of the embodiment shown in FIG. 8, and FIG. FIG. 11 is a block diagram showing a third embodiment, FIG. 11 is a sectional view showing a conventional example, and FIG. 12 is a peak value distribution chart showing the operation of the representative value selecting means. 1 ... Steel plate, 5 ... Magnetic sensor, 7 ... First comparing means, 8 ... Second threshold value calculating means, 9 ... Second comparing means, 11 ... Ferromagnetic core, 12 ... ... coil, 13 ... AC power supply means, 14 ... fixed impedance, 15 ... DC voltage detection means, 16 ... magnetic detection means, 17 ... peak value detection means, 18 ... representative value selection means, 19 ... ... mean value calculation means,
21 …… Adder, 22 …… DC power supply.

Claims (3)

[Claims] 1. A coil wound around a ferromagnetic core and installed near the surface of a continuously moving steel plate, and an alternating current supplying a constant frequency and a constant voltage to the coil via a fixed impedance. A power supply means, a DC voltage detection means for detecting a positive side voltage and a negative side voltage of the voltage output from both ends of the coil, and a difference between the positive side voltage and the negative side voltage detected by the DC voltage detection means. A magnetic sensor including a magnetic detection means for detecting the strength of a magnetic field and a detection signal from the magnetic sensor are compared with a predetermined first threshold value, and the detection signal exceeds the first threshold value. And a detection signal that does not reach the first threshold value to detect a major flaw in the steel plate, and a predetermined comparison method that receives the detection signal that does not reach the first threshold value from the comparison means. Distance of the rolled steel plate A peak value detecting means for detecting the peak value of the detection signal; a representative value selecting means for selecting a representative value from a distribution of a plurality of peak values detected by the peak value detecting means; A second threshold value calculating means including an average value calculating means for calculating an average value of the representative values to calculate a second threshold value; and a second threshold value calculated by the second threshold value calculating means. A flaw detecting device for a steel sheet, comprising: a second comparing means for comparing a detection signal of the magnetic sensor to detect a minute flaw in the steel sheet. 2. The flaw detection apparatus for a steel sheet according to claim 1, wherein the AC power supply means of the magnetic sensor outputs a pulse current, and a DC bias adding means for adding a DC bias to the pulse current is provided. 3. A flaw detection device for a steel sheet according to claim 2, further comprising control means for variably controlling a DC bias added by said DC bias adding means in accordance with a magnetic detection signal from said magnetic detection means.