JP4585709B2 - Simple calibration method for electromagnetic molten steel level detector - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an electromagnetic molten steel level detector that measures the distance to a detected object by using eddy current generated in the detected object, and particularly to a molten metal level in a casting mold during continuous casting of the molten metal. The present invention relates to a simple signal calibration method.
[0002]
[Prior art]
FIG. 3 shows a general configuration of an electromagnetic molten steel level meter 10 provided with a transmission / reception coil. In FIG. 3, B is a mold, M is molten steel, and S is a detector. The detector S is composed of a transmission coil T and a reception coil R, 11 is an AC power source, 12 is an AC amplifier, 13 is a synchronous detector that converts AC voltage into DC voltage, 14 is a DC amplifier, and 15 is offset adjustment. 16 is a linear processor. The linear processor 16 is a device that outputs a non-linear DC output voltage into a linear form, that is, converts the voltage into a voltage corresponding to the measurement distance.
[0003]
In the configuration of FIG. 3, the detector S is installed at a position away from the mold B by a certain distance dm, the transmission coil T is energized from the AC power supply 11, and an AC magnetic field is generated by the transmission coil T. Since an induced voltage is generated in the receiving coil R in this magnetic field, this induced voltage is amplified by the AC amplifier 12, detected by the synchronous detector 13, and converted into a DC voltage. The output value of the DC voltage is mainly determined by the distance dm between the detector S and the mold B and the distance between the detector S and the molten steel M. A DC voltage is input to the DC amplifier 14 and a negative voltage corresponding to a voltage value determined by the distance S between the detector S and the mold is added from the offset adjuster 15. That is, the signal component due to the surrounding hardware is removed by the voltage added from the offset adjuster 15, and the DC voltage obtained by this is obtained only for the effective signal due to the molten metal surface level change by the molten steel M, and the linearity of the next stage The distance to the hot water level can be measured via the processor 16.
[0004]
FIG. 2 shows an example of the level characteristic result using the detector S and the distance dm between the molds as parameters, and the vertical axis shows the DC voltage output from the DC amplifier 14. As shown in FIG. 3, the detector S is provided at a position away from the mold wall at the upper part of the mold by a gap dm, and is used for molten steel level measurement. The detector S is housed in an insulator case (not shown) and can be adapted to a high temperature environment.
As apparent from FIG. 2, as the distance dm between the detector S and the mold decreases, the eddy current loss in the mold increases and the output voltage decreases. Here, if the data of dm = 15 is used as the reference data for linearization, there is a problem that a linearization error occurs in other data as is obvious.
As a method for solving this problem, for example, a molten metal surface detecting device described in Japanese Patent Publication No. 3-48448 is known, and a pair of non-magnetic and heat resistant materials close to the molten metal. An electromagnetic field generated by applying an alternating voltage to one of the pair of coils, and a detection unit in which the coils are arranged in parallel at a symmetric position about the perpendicular of the molten metal surface with a certain interval. A detecting / smoothing unit that uses the induced voltage of the other coil provided therein as a DC voltage signal, and a storage unit that stores in advance a voltage value that is determined by the action of the molten metal container and does not correspond to the distance between the detecting unit and the molten metal And a correction function unit that corrects the output voltage of the detection / smoothing unit using the stored voltage.
[0005]
[Problems to be solved by the invention]
In the molten steel level meter based on the detection principle of electromagnetic induction according to the above-described conventional example, the detection voltage of the secondary coil changes when the gap dm with the non-detection object around the detector S, for example, the mold, fluctuates. Measurement error occurs.
Further, the reference data of the linear processor 16 is usually obtained by moving the unit length (for example, 10 mm) using a simulated hot water surface (lead plate, stainless plate, etc.). On the other hand, the molten steel that is the actual detection target and the simulated molten metal surface have different material properties (conductivity), so that a conversion error cannot be avoided by linear calculation using the reference data.
Conventionally, in order to reduce these problems, (1) Reproducibility of detector installation position, (2) Selection of material close to the material characteristics of actual molten steel, (3) Arbitrary fixed point of reference data and actual molten steel level When the values match (determine a fixed point with an electrode-type level meter, etc.), the difference value with each voltage is obtained and stored, and thereafter the detected voltage obtained successively and the difference value are added to perform data correction (offset adjustment) However, there was a limit to error reduction.
Moreover, the molten metal surface detecting device described in Japanese Patent Publication No. 3-48448 has a problem that the apparatus becomes complicated and the measurement takes time.
The present invention has been made in view of such circumstances, and provides a simple calibration method for an electromagnetic molten steel level detector that is simple and can substantially eliminate detection errors without using new information.
[0006]
[Means for Solving the Problems]
A simple calibration method for an electromagnetic molten steel level detector according to the first invention that meets the above-mentioned object is a detector comprising a transmitting coil and a receiving coil arranged above and below, an AC power source that applies a current to the transmitting coil, Using an electromagnetic molten steel level detector having a synchronous detector that converts an AC voltage induced in the receiving coil into a DC signal, and an offset adjuster that extracts only an effective signal output from the synchronous detector Before starting casting , move the simulated molten metal for each unit level to obtain data, obtain a data table that approximates the broken line between the data, and in actual use, it is detected by the electrode level meter actual values in the known molten metal surface level (Vm) and data values in該既Chiyumen levels in the data table for the (Vd), the data value (Vd) / measured value (Vm) Ranaru obtains a normalization factor (K1), by multiplying the normalization factor with the actual measured value (K1), to measure the depth to the molten steel surface from said data table.
[0007]
In addition, a simple calibration method for an electromagnetic molten steel level detector according to a second aspect of the present invention includes a detector comprising a transmission coil and a reception coil arranged above and below, an AC power source for applying a current to the transmission coil, and the reception Casting using an electromagnetic molten steel level detector having a synchronous detector that converts an AC voltage induced in a coil into a DC signal and an offset adjuster that extracts only an effective signal output from the synchronous detector. Before starting , move the simulated hot water surface for each unit level to obtain data, obtain a data table that approximates the broken line between the data, and in actual use, known data detected by the electrode level meter Found in molten metal surface level (Vm) and data values in該既Chiyumen levels in the data table and the seeking (Vd), consisting of the measured value (Vm) / data value (Vd) A normalization factor (K2) is obtained, the data table is multiplied by the normalization factor (K2), a correction data table is created, and the measured value of the electromagnetic molten steel level detector is compared with the correction data table. And measure the depth to the actual molten steel surface.
In the simple calibration method for the electromagnetic molten steel level detector according to the first and second inventions, the detected voltage may be linearly processed.
[0008]
As described above, in the simple calibration method of the electromagnetic molten steel level detector according to the first and second inventions, the detected voltage of the electromagnetic molten steel level detector is mainly the distance dm between the detector S and the mold B, and Although determined by the distance between the detector S and the molten steel M, the influence of surrounding hardware on the output voltage has a certain proportional relationship if the detector S is specified. Therefore, a preliminary experiment is performed in advance for an arbitrary detector S to obtain a data table that approximates a polygonal line including data for each unit level serving as a reference. If the ratio between the data table value (Vd) and the actual measurement value (Vm) is obtained, this ratio relationship is maintained even for an arbitrary hot water level.
In the first invention, based on the data table, the data value sought (Vd) / measured value normalization factor consisting of (Vm) (K1), by multiplying the measured value of the normalization factor (K1) The actual measurement value is corrected, and the molten metal level is measured by comparing the correction value with the data table and output.
In the second invention, the data table is corrected in accordance with the measurement state, and the actual measured value is collated with the corrected data table to measure the distance.
In the first and second inventions, it is preferable that the final output is converted into a voltage proportional to the measurement distance, that is, it is output by performing linear processing.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Next, embodiments of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a block diagram of an apparatus to which a simple calibration method for an electromagnetic molten steel level detector according to an embodiment of the present invention is applied. In FIG. 1, the same components as those used in FIG. 3 are denoted by the same reference numerals, and detailed description thereof is omitted.
[0010]
Here, in FIG. 1, 20 is a device to which a simple calibration method of an electromagnetic molten steel level detector according to an embodiment of the present invention is applied, 21 is a correction computing unit that processes the output from the DC amplifier 14, 22 is an interrupt signal generator, and for example, a contact signal that conducts when the molten metal level reaches a predetermined level may be input. The device 20 is configured by adding a new correction calculator to the conventional electromagnetic molten steel level detector.
A data table in which the detector S described above is disposed in the correction calculator 21 with a predetermined distance dm from the mold B 1 and records a DC voltage obtained by subtracting an offset voltage from the output voltage for each level. Is remembered. Since this data table is data for every 10 mm (unit level) as shown in the graph of FIG. 2, for example, the data table is regarded as a straight line and approximated by a broken line. Therefore, in the data table of the polygonal line approximation, the actual measurement value up to the molten metal surface and the output voltage correspond one-to-one.
[0011]
Next, the detector S is attached to an arbitrary mold B1 that is to be measured. In this case, it is not necessary to measure the distance between the mold B1 and the detector S, but the detector S needs to be set at a fixed position with respect to the mold B1.
When the molten metal surface in the mold B1 rises, the electrode sensor G comes into contact with the molten metal surface, and is output from the interrupt signal generator 22 to the correction calculator 21. Since the level of the molten metal when the electrode sensor G is in contact with the molten metal surface is known, the output from the DC amplifier 14 at this time, that is, the measured value (Vm) of the known molten metal surface level is stored at the same time as the data table. The data value (Vd) at the known hot water level is output and stored, and the following equation (1) is executed by an arithmetic unit.
Data value (Vd) / actual value (Vm) = normalization coefficient (K1) (1)
Then, when the normalization coefficient (K1) is stored and the apparatus 20 using the detector S is used to measure an arbitrary molten metal surface of the mold B1, the actual value from the DC amplifier 14 is measured. The correction value (Vam) is calculated by multiplying the output value (Vmx), which is a measured value, by the normalization coefficient (K1), and this correction value (Vam) is compared with the data table to determine the level of the molten metal surface. The calculated hot water level is subjected to linear processing, converted into a voltage proportional to the measurement distance, and output.
[0012]
In the above embodiment, the actual measurement value (Vm) output from the DC amplifier 14 is corrected with the normalization coefficient (K1) based on the voltage generated in the receiving coil R of the detector S. As a method according to the embodiment, the present invention can also be implemented by the following method.
That is, based on the signal generated from the interrupt signal generator 22, the output from the DC amplifier 14 at this time, that is, the actual measured value (Vm) of the known molten metal level is stored, and at the same time, the known molten metal level in the data table is stored. The data value (Vd) at the level is output and stored, and the arithmetic unit calculates and stores the normalization coefficient (K2) composed of the actual measurement value (Vm) / data value (Vd). Then, by multiplying the normalization factor (K2) to the voltage data of each unit for each level of the data table, to create the correction data table. Naturally, linear interpolation, that is, polygonal line approximation is performed between the unit levels of the correction data table.
By using this correction data table and comparing it with the actual measured voltage, that is, the output of the DC amplifier 14, it is possible to output the molten metal level.
[0013]
In the above embodiment, a plurality of reference data for polygonal line approximation is not required. The detector based on the detection principle of electromagnetic induction has non-linear characteristics. However, unlike the present embodiment, the point where the molten steel M does not exist ( absolute base point ) and the electrode level meter for calibration (ie, the electrode sensor G). It is possible to theoretically convert and reproduce the reference data by determining a total of two other points to be measured by the above method and correcting the gain. Since this is made possible only by four arithmetic operations, a simple and practical correction method is realized without causing any problem of processing time.
[0014]
For reference, the adjustment procedure of the apparatus to which the embodiment of the present invention that is the basis of the present invention is applied will be described more specifically as follows (1) to (4).
(1) First, the detector S is installed in the mold B1 to be measured.
(2) Creation of reference data table: The offset adjuster 15 is adjusted except for the simulated molten metal surface, and the output voltage of the DC amplifier 14 is set to zero. Thereby, the influence of adjacent conductors such as the mold B1 is removed.
(3) Next, a simulated hot water surface is set, and the hot water surface is moved every 10 mm to create a reference data table.
(4) Calibration of the detection signal based on the measured molten metal level: When molten steel is put into the mold and the cast molten metal level rises, an interrupt signal from, for example, an electrode type level meter for detecting a fixed point provided separately is received. The output voltage is taken in, and at the same time, the calculation shown in the above equation (1) is performed to obtain and store the normalization coefficient (K1). Next, in order to obtain correction data of the detection signal, the detection signal sequentially obtained, that is, the output voltage (Vmx) of the DC amplifier 14 and the normalization coefficient (K1) are multiplied.
(5) The correction data that has been gain-corrected is output by performing a linear conversion by a known broken line approximation (linear approximation between two points) using the reference data.
By the correction calculation process as described above, a simple method for calibrating a molten steel level meter that can compensate for the influence of the material of the object to be detected and the influence on the characteristics of the gap between the detector and the mold has been realized.
[0015]
【The invention's effect】
As apparent from the above description, the simple calibration method of the electromagnetic molten steel level detector according to the first invention is based on the measured value (Vm) at the known molten metal level and the data value (Vd) in the data table. Obtain the normalization coefficient (K1) consisting of the data value (Vd) / actual measurement value (Vm), multiply the actual measurement value by the normalization coefficient (K1), and calculate the depth from the data table to the molten steel surface. Measuring. And in the simple calibration method of the electromagnetic molten steel level detector which concerns on 2nd invention, the measured value (Vm) in a known molten metal surface level and the data value (Vd) in a data table are calculated | required, and measured value (Vm) / Obtain the normalization coefficient (K2) consisting of the data value (Vd), multiply the data table by the normalization coefficient (K2), create a correction data table, and the depth to the actual molten steel surface, The depth to the actual molten steel surface is measured by comparing the measured value of the molten steel level detector with the correction data table. Accordingly, it is possible to provide a simple calibration method for an electromagnetic molten steel level detector that has a simple apparatus configuration and can substantially eliminate detection errors without using new information. Can do.
[Brief description of the drawings]
FIG. 1 is a block diagram showing an apparatus configuration of a simple calibration method for an electromagnetic molten steel level detector according to an embodiment of the present invention.
FIG. 2 is a graph showing the relationship between the output of a DC amplifier and the surface level.
FIG. 3 is a block diagram of an electromagnetic molten steel level detector according to a conventional example.
[Explanation of symbols]
11: AC power supply, 12: AC amplifier, 13: synchronous detector, 14: DC amplifier, 15: offset adjuster, 16: linear processor, 20: device, 21: correction calculator, 22: interrupt signal generator, S: detector, B1: mold, M: molten steel, T: transmitter coil, R: receiver coil, G: electrode sensor

Claims (2)

A detector comprising a transmitter coil and a receiver coil disposed above and below; an AC power source for applying a current to the transmitter coil; a synchronous detector for converting an AC voltage induced in the receiver coil into a DC signal; Using an electromagnetic molten steel level detector having an offset adjuster that extracts only the effective signal of the signal output by the detector , the simulated molten metal surface is moved for each unit level before starting casting , and data is acquired. Find a data table that approximates the line between the data,
In the actual use, seeking data values in該既Chiyumen level measured value in the detected known bath level level by electrode level meter and (Vm) in the data table and (Vd), the data value ( Vd) / normalized value (K1) consisting of actual measurement value (Vm) is obtained,
A simple calibration method for an electromagnetic molten steel level detector, wherein an actual measured value is multiplied by the normalization coefficient (K1) to measure a depth from the data table to the molten steel surface. A detector comprising a transmitter coil and a receiver coil disposed above and below; an AC power source for applying a current to the transmitter coil; a synchronous detector for converting an AC voltage induced in the receiver coil into a DC signal; Using an electromagnetic molten steel level detector having an offset adjuster that extracts only the effective signal of the signal output by the detector , the simulated molten metal surface is moved for each unit level before starting casting , and data is acquired. Find a data table that approximates the line between the data,
In the actual use, seeking data values in該既Chiyumen level measured value in the detected known bath level level by electrode level meter and (Vm) in the data table and (Vd), Found ( Vm) / data value (Vd) is obtained as a normalization coefficient (K2), the data table is multiplied by the normalization coefficient (K2), and a correction data table is created.
A simple method for calibrating an electromagnetic molten steel level detector, wherein the measured value of the electromagnetic molten steel level detector is compared with the correction data table to measure the depth to the actual molten steel surface.

Images