Detection value correction system, coefficient calculation method and detection value correction method
Through the detection value correction system and coefficient calculation method, the detection accuracy reduction caused by reducing the magnetic impedance sensor spacing is solved. Through correction processing and coefficient calculation, the accuracy of magnetic field detection is improved.
Patent Information
- Application Number
- CN202080019333.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-03-13
- Filing Date
- 2020-03-05
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2040-03-05
AI Technical Summary
Reducing the adjacency distance of the magnetic impedance sensor will cause a decrease in detection accuracy, because the magnetic flux in the rotation direction will expand to the peripheral sensor, affecting the detection accuracy.
The detection value correction system and coefficient calculation method are adopted, and the correction processing and coefficient calculation processing are used to correct the detection value of the adjacent sensor, and the first coefficient k1 and the second coefficient k2 are calculated to reduce the impact of other sensors on the sensor of concern.
It effectively improves the detection accuracy of the sensor array, reduces errors due to small adjacency spacing, and ensures high-precision magnetic field detection.
Smart Images

Figure CN113544528B_ABST
Abstract
Description
[0001] This application is based on Japanese patent application No. 2019-045780 filed on March 13, 2019, the contents of which are incorporated herein by reference. Technical Field
[0002] The present invention relates to a detection value correction system for correcting a detection value of a sensor, a detection value correction method, and a coefficient calculation method for calculating a coefficient used for correction of the detection value. Background Art
[0003] Conventionally, a magneto-impedance (MI) sensor is known that includes an amorphous wire and a coil wound around the amorphous wire. Also, it is known that a magnetic sensor array in which MI sensors are arranged in a row can be used to detect magnetic distribution (see, for example, Patent Document 1).
[0004] Prior art literature
[0005] Patent Literature
[0006] Patent Document 1: Japanese Patent Application Publication No. 2002-169614 Summary of the invention
[0007] In addition, when the magnetic distribution is detected by the MI sensor, the resolution of the distribution position is determined by the adjacent spacing of the MI sensor. Therefore, in order to improve the resolution of the distribution position, it is considered necessary to reduce the adjacent spacing of the MI sensor. However, if the adjacent spacing of the MI sensor is reduced, the following problems will occur due to the operating principle of the MI sensor.
[0008] When a pulse current is passed through the amorphous wire of the MI sensor, the direction of the spin in the amorphous wire is consistent with the direction orthogonal to the amorphous wire due to the magnetic field generated by the current. If the current becomes zero in this state and the magnetic field generated by the current disappears, the direction of each spin will be toward the direction corresponding to the external magnetic field. In this way, the rotation changes from the direction orthogonal to the amorphous wire to the direction corresponding to the external magnetic field, thereby generating a magnetic flux with a strength corresponding to the external magnetic field. In accordance with the change in the magnetic flux, an electromotive force is generated in the coil. Therefore, the MI sensor detects the external magnetic field by generating an electromotive force corresponding to the external magnetic field in the coil.
[0009] However, the magnetic flux generated by the change in the direction of rotation will spread to the periphery to a certain extent. Therefore, if the adjacent spacing of the MI sensors is small, the magnetic flux generated by the change in the direction of rotation of a certain MI sensor will reach the MI sensors around it and be detected by the MI sensors. As a result, the detection accuracy of the magnetic field in the adjacent MI sensors may decrease.
[0010] Similar to the MI sensor, in sensors other than the MI sensor, when a plurality of sensors are arranged and each sensor affects the detection values of other sensors around it, the detection accuracy of the sensor may decrease if the adjacent pitch of the sensors is reduced.
[0011] An object of the present invention is to provide a detection value correction system and a detection value correction method for easily correcting detection values of sensors arranged in a row, and a coefficient calculation method for calculating coefficients used for correction of detection values.
[0012] A detection value correction system of one example of the present invention corrects detection values of a plurality of sensors arranged in a row for detecting physical quantities, the detection value correction system comprising: a correction processing unit that performs correction processing on the detection value of a focus sensor that is a sensor to be corrected among the plurality of sensors, based at least on the detection value of a first sensor adjacent to the focus sensor.
[0013] Moreover, a detection value correction method of an example of the present invention corrects the detection values of multiple sensors arranged in a row and detecting physical quantities, wherein a correction process is performed on the detection value of a focus sensor that is a sensor to be corrected among the multiple sensors, at least based on the detection value of a first sensor adjacent to the focus sensor.
[0014] Moreover, a coefficient calculation method of an example of the present invention is to calculate a first coefficient k1 and a second coefficient k2 for one of a plurality of sensors arranged in a row and detecting a physical quantity, wherein the first coefficient k1 represents the influence of a first sensor adjacent to the sensor on the sensor, and the second coefficient k2 represents the influence of a second sensor adjacent to the one sensor in the opposite direction relative to the first sensor on the sensor, and the coefficient calculation method includes the following steps: (a) acquiring detection values of the sensors in a state where the physical quantity is unevenly assigned to the plurality of sensors arranged in a row; (b) temporarily assigning different values to the plurality of first coefficients k1, corresponding to the first coefficients k1 assigned the temporary values, respectively; (b1) labeling the plurality of sensors with consecutive numbers and setting them as sensors 1 to Q, and the detection value of any Xth sensor X among sensors 1 to Q is B X , the detection value of the X+1th sensor (X+1) adjacent to one side of the sensor X is B X+1 The detection value of the X-1th sensor (X-1) adjacent to the other side of the sensor X is B X-1, the detection value of the X+2th sensor (X+2) adjacent to the sensor X in the opposite direction relative to the sensor (X+1) is B X+2 The detection value of the X-2th sensor (X-2) adjacent to the sensor X in the opposite direction relative to the sensor (X-1) is B X-2 , the second coefficient is k2, and X is 3 to (Q-2), the value D is calculated based on the following formula (2): X (b2) For n, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(n) and value D(n-1) at which the square of the difference between the value D(n) and the value D(n-1) of the adjacent sensor n and sensor (n-1) reaches the maximum, and calculate the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ], (b3) for m, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(m) and value D(m-1) at which the square of the difference between the value D(m) and the value D(m-1) of the adjacent sensor m and sensor (m-1) reaches the minimum, and calculate the square of the difference between the minimum value D(m) and the value D(m-1) as MIN[{D(m)-D(m-1)} 2 ], (b4) based on the following formula (3), calculate a plurality of evaluation values U(k1) corresponding to the values of the plurality of coefficients k1; and (c) explore the maximum evaluation value U(k1) among the plurality of evaluation values U(k1), determine a first coefficient k1 corresponding to the maximum evaluation value U(k1) and a second coefficient k2 obtained based on the first coefficient k1, as the first coefficient and the second coefficient used in the correction process,
[0015] D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 …(2)
[0016] U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]…(3). BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 This is an explanatory diagram conceptually showing an example of the configuration of a magnetism measuring device using a detection value correction method according to an embodiment of the present invention.
[0018] Figure 2 Yes means Figure 1 A flowchart of an example of the operation of the magnetism measuring device shown.
[0019] Figure 3 This is an explanatory diagram showing the coefficients k2, -k1, 1, -k1, and k2 in a graph.
[0020] Figure 4 This is an explanatory diagram showing the coefficients k2, -k1, 1, -k1, and k2 in a graph.
[0021] Figure 5 This is an explanatory diagram showing the coefficients k2, -k1, 1, -k1, and k2 in a graph.
[0022] Figure 6 This is an explanatory diagram showing the coefficients k2, -k1, 1, -k1, and k2 in a graph.
[0023] Figure 7 This is a flowchart showing an example of a coefficient calculation method according to one embodiment of the present invention.
[0024] Figure 8 This is a flowchart showing an example of a coefficient calculation method according to one embodiment of the present invention.
[0025] Fig. 9 This is an explanatory diagram conceptually showing an example of the configuration of an X-ray measuring device using a detection value correction method according to an embodiment of the present invention.
[0026] Fig.10 This is an explanatory diagram conceptually showing an example of the structure of a sensor to which the detection value correction method according to one embodiment of the present invention can be applied.
[0027] [Explanation of Symbols]
[0028] 1: Magnetic measurement device (detection value correction system)
[0029] 1a: X-ray measuring device (detection value correction system)
[0030] 1b: Flow measurement system (detection value correction system)
[0031] 2: Magnetic sensor array
[0032] 2a: X-ray sensor array
[0033] 3: Power supply
[0034] 4: Voltage measurement unit
[0035] 5: Calculation processing unit
[0036] 21: Amorphous wire
[0037] 51: Detection and processing department
[0038] 52: Correction Processing Department
[0039] 53: Coefficient calculation processing unit
[0040] 100: X-ray
[0041] A, B, B1~B10, Bn, BQ: detection value
[0042] C1~C10、Cn:correction value
[0043] d: interval
[0044] E1~E10, F, F1~FQ, M, M1~M10: Sensor
[0045] k1: coefficient (first coefficient)
[0046] k2: coefficient (second coefficient)
[0047] L: Coil
[0048] PD: Photodiode
[0049] R: River
[0050] S1~S6, S101~S112, S121~S123: Steps
[0051] SC: Scintillator
[0052] U: Evaluation value
[0053] X: sensor number DETAILED DESCRIPTION
[0054] Hereinafter, embodiments of the present invention will be described based on the drawings. In addition, the same reference numerals in the drawings represent the same structures, and their descriptions are omitted. Figure 1 The magnetism measuring device 1 shown corresponds to an example of the detection value correction system of the present invention.
[0055] Figure 1 The magnetic measuring device 1 shown includes a magnetic sensor array 2, a power supply 3, a plurality of voltage measuring units 4, and an operation processing unit 5. The magnetic sensor array 2 includes an amorphous wire 21 and a plurality of sensors M1 to M10 (sensors) labeled with sensor numbers 1 to 10. Hereinafter, the sensors M1 to M10 are collectively referred to as sensors M. The sensor M is a so-called MI sensor (magnetoresistive sensor) that detects the strength of a magnetic field.
[0056] The amorphous wire 21 is, for example, a linear magnetic conductor such as CoFeSiB. The diameter of the amorphous wire 21 is, for example, set to be less than 100 μm. An insulating layer of, for example, acrylic resin is formed on the periphery of the amorphous wire 21. In addition, instead of the amorphous wire 21, a wire having a magnetic anisotropic film coated on a linear body, or a wire such as permalloy which is a Ni-Fe alloy, may be used.
[0057] A plurality of coils L are wound in a row at predetermined intervals d from each other on the amorphous wire 21. One sensor M is formed by the amorphous wire 21 and one coil L wound around the outer circumference thereof. Figure 1 2 shows an example in which ten coils L are wound around one amorphous wire 21 to form ten sensors M1 to M10. The number of sensors M may be less than ten or may be more than ten.
[0058] The power source 3 is connected to both ends of the amorphous wire 21 . The power source 3 outputs a current having a preset current value in a pulse form to the amorphous wire 21 in accordance with a control signal from the calculation processing unit 5 .
[0059] In addition, the present invention is not necessarily limited to the example of winding a plurality of coils L around a single amorphous wire 21 to form a plurality of sensors M. The amorphous wire 21 may be separated for each sensor M. Furthermore, current may be supplied from a power source to each separated amorphous wire 21. However, by winding a plurality of coils L around a single amorphous wire 21, the current flowing to each sensor M is equal. As a result, the deviation of the detection value of each sensor M is reduced, so a structure in which a plurality of coils L are wound around a single amorphous wire 21 is more preferable.
[0060] The voltage measuring unit 4 is configured using, for example, an amplifier, a peak hold circuit, and an analog-to-digital converter. Each voltage measuring unit 4 measures the voltage induced in each coil L of the sensors M1 to M10, and generates a signal indicating the measurement result as the detection values B1 to B10 of the sensors M1 to M10. 10 And sent to the calculation processing unit 5.
[0061] The calculation processing unit 5 is composed of, for example, a central processing unit (CPU) that performs a predetermined calculation process, a random access memory (RAM) that temporarily stores data, a hard disk drive (HDD) and / or a flash memory that stores a predetermined control program, and their peripheral circuits, etc. The calculation processing unit 5 functions as a detection processing unit 51, a correction processing unit 52, and a coefficient calculation processing unit 53 by executing a predetermined control program, etc.
[0062] The detection processing unit 51 obtains the outputs of the voltage measuring units 4 as the detection values B1 to B2 of the sensors M1 to M10. 10 .
[0063] The correction processing unit 52 performs correction processing on the detection value B of the sensor Mn of interest that is the sensor to be corrected among the sensors M1 to M10. n The correction processing unit 52 performs correction processing on the sensors M1 to M10 in sequence as the sensors Mn of interest, thereby correcting the detection values B1 to B2 of the sensors M1 to M10. 10 The correction processing unit 52 corrects the detected values B1 to B 10 The correction values C1 to C 10 .
[0064] The coefficient calculation processing unit 53 calculates the coefficient based on the detection values B1 to B 10 , to calculate the coefficients k1 and k2 described later.
[0065] In addition, the detection value correction system is not necessarily limited to the example in which a plurality of sensors, the correction processing unit 52, and the coefficient calculation processing unit 53 are constituted as a single device. For example, the magnetic measuring device may not include the correction processing unit 52 and the coefficient calculation processing unit 53. In addition, a correction device including the correction processing unit 52 and the coefficient calculation processing unit 53 may be constituted by a computer such as a personal computer that is independent of the magnetic measuring device. In addition, the detection values B1 to B1 detected by the magnetic measuring device may be corrected by the correction device. 10 In this case, the magnetic measuring device and the correction device constitute a detection value correction system.
[0066] Furthermore, the detection value correction system does not necessarily need to include a plurality of sensors. The detection value correction system may also be a correction device including the correction processing unit 52 and the coefficient calculation processing unit 53. Furthermore, the detection value correction system may also include a correction processing device including the correction processing unit 52 and a coefficient calculation processing device including the coefficient calculation processing unit 53.
[0067] Furthermore, the magnetism measuring device 1 may be configured as a testing device further including a testing unit that detects the correction values C1 to C2 generated by the correction processing unit 52. 10 For example, the magnetic measuring device 1 may be an inspection device as described below, that is, the inspection unit compares the reference value and the correction value C1 to C2 stored in advance in the storage unit. 10The comparison is performed to determine the quality of the inspection object or to detect the mixing of magnetic foreign matter. More specifically, the magnetic measuring device 1 can also be configured as an inspection device as follows, that is, by using the magnetic sensor array 2 to detect the magnetic field on the battery surface, thereby detecting metallic foreign matter mixed into the battery.
[0068] Next, the operation of the magnetism measuring device 1 configured as described above will be described. First, the user brings the magnetic sensor array 2 close to an object whose magnetic field is to be measured, or places the magnetic sensor array 2 at a location where the magnetic field is to be measured. Then, the magnetism measuring device 1 starts operating.
[0069] Reference Figure 2 First, the detection processing unit 51 causes a current of a preset current value to flow to the amorphous wire 21 in a pulsed state through the power supply 3 (step S1 ).
[0070] The magnetic flux generated by the current is generated in a concentric circle around the amorphous wire 21. Since the coil L is wound around the outer periphery of the amorphous wire 21, the magnetic flux generated by the current is in substantially the same direction as the direction of the winding of the coil L. In the coil L, when the magnetic flux intersecting the winding changes, a voltage is generated due to electromagnetic induction. However, the magnetic flux generated by the current does not intersect the winding of the coil L, so no voltage is generated in the coil L relative to the current flowing through the amorphous wire 21.
[0071] On the other hand, when current flows through the amorphous wire 21, the direction of rotation in the amorphous wire 21 is consistent with the direction orthogonal to the amorphous wire 21 by the magnetic field generated by the current during the flow of the current. Subsequently, when the pulse drops and the current becomes zero, and the magnetic field generated by the current disappears, the direction of each rotation of the amorphous wire 21 is oriented toward the direction corresponding to the external magnetic field. In this way, the rotation changes from the direction orthogonal to the amorphous wire 21 to the direction corresponding to the external magnetic field, thereby generating a magnetic flux of a strength corresponding to the external magnetic field. The magnetic flux accompanying the change in direction of the rotation includes a component in a direction intersecting with the winding of the coil L, and therefore a voltage is generated in the coil L in accordance with the change in the magnetic flux accompanying the change in direction of the rotation.
[0072] Thus, the voltage generated in each coil L represents the strength of the external magnetic field at each position of the sensors M1 to M10. Each voltage measuring unit 4 measures the peak voltage generated in the coil L of the sensors M1 to M10. Each voltage measuring unit 4 uses the signal representing the measured voltage as the detection value B1 to B2 of the sensors M1 to M10. 10 And output to the calculation processing unit 5.
[0073] The detection processing unit 51 acquires the voltages measured by the voltage measuring units 4 as detection values B1 to B2. 10 (Step S2).
[0074] Here, the detection values B1 to B2 of sensors M1 to M10 are 10 The error caused by the influence of other sensors M is included. For example, in addition to the magnetic flux generated by the rotation of the amorphous wire 21 at the position of sensor M5, the coil L of sensor M5 also detects the magnetic flux generated by the rotation of the positions of sensors M4 and M6 adjacent to sensor M5. Moreover, the coil L of sensor M5 may further detect the magnetic flux generated by the rotation of the positions of sensors M3 and M7 adjacent to the sensors M4 and M6. Therefore, the detection values B1 to B10 of sensors M1 to M10 are 10 The error caused by the influence of other sensors M is included.
[0075] Therefore, the correction processing unit 52 performs correction on the detection values B1 to B1 obtained by the detection processing unit 51. 10 The correction process is performed to reduce the error caused by the influence of other sensors M. Steps S3 to S6 described below correspond to an example of the correction process.
[0076] First, the correction processing unit 52 initializes a variable n to 1 (step S3 ).
[0077] Next, the correction processing unit 52 sets the detection value of the target sensor Mn to be corrected to B. n , which is equivalent to the detection value of the first sensor adjacent to one side of the sensor Mn of interest, sensor M(n+1), being B n+1 , which is equivalent to the detection value of the first sensor M(n-1) adjacent to the other side of the sensor Mn of interest being B n-1 , which is equivalent to the detection value of sensor M(n+2) being B, which is the second sensor adjacent to the sensor Mn of interest in the opposite direction relative to sensor M(n+1). n+2 , which corresponds to the detection value B of the second sensor M(n-2) adjacent to the sensor Mn of interest in the opposite direction relative to the sensor M(n-1). n-2 When , the detection value B of the sensor Mn is calculated based on the following formula (1): n Correction value C n (Step S4).
[0078] C n =k2B n-2 -k1B n-1 +B n -k1B n+1 +k2B n+2 …(1)
[0079] The coefficient k1 (first coefficient) and the coefficient k2 (second coefficient) are calculated in advance by, for example, a coefficient calculation method described later, and are stored in the storage unit.
[0080] When n=1, the correction processing unit 52 sets B n-1 =B n-2 =0, when n=2, set to B n-2 =0, when n=9, set to B n+2 =0, when n=10, set it to B n+1 =B n+2 =0, to perform the calculation of formula (1).
[0081] According to formula (1), the detection value of the sensor of interest is added with the value obtained by multiplying the detection value of the first sensor by the first coefficient, and the value obtained by multiplying the detection value of the second sensor by the second coefficient which is smaller than the first coefficient is subtracted. In this way, the detection value of sensor Mn which is the sensor of interest can be corrected.
[0082] Next, the correction processing unit 52 compares the number 10 of sensors M with the variable n (step S5). If the variable n is not 10 (No in step S5), the correction processing unit 52 adds 1 to the variable n (step S6), and repeats steps S4 and S5 again.
[0083] On the other hand, if the variable n is 10 (Yes in step S5), the correction values C1 to C 10 Since all the detection values have been corrected, the correction processing unit 52 ends the correction processing.
[0084] Next, the correction of the detection values B1 to B1 by the above formula (1) is described. 10 The reason is to reduce the error caused by the influence of other sensors M.
[0085] When the correct detection value of sensor Mn, that is, the detection value of sensor Mn is not affected by other sensors M, it is A n The coefficient representing the degree of influence that sensor Mn receives from adjacent sensors M(n-1) and M(n+1) is numerated as k1, and the coefficient representing the degree of influence that sensor Mn receives from sensors M(n-2) and M(n+2) further adjacent to sensors M(n-1) and M(n+1) is numerated as k2=k1 according to equation (4) described later. 2 When the sensor Mn detects the value B n It is represented by the following formula (A).
[0086] B n =k1 2 A n-2 -k1A n-1 +A n-k1A n+1 +k1 2 A n+2 …(A)
[0087] Therefore, if the correction value C calculated by the above formula (1) is n Correct, when substituting formula (A) into formula (1), the corrected value C should be n =A n .
[0088] Therefore, if equation (A) is substituted into equation (1), the following is obtained. 2 , 0.5<k1<1.
[0089] C n =k1 2 B n-2 -k1B n-1 +B n -k1B n+1 +k1 2 B n+2
[0090] =k1 2 (k1 2 A n-4 +k1A n-3 +A n-2 +k1A n-1 +k1 2 A n )
[0091] -k1(k1 2 A n-3 +k1A n-2 +A n-1 +k1A n +k1 2 A n+1 )
[0092] +k1 2 A n-2 +k1A n-1 +A n +k1A n+1 +k1 2 A n+2
[0093] -k1(k1 2 A n-1 +k1A n +A n+1 +k1A n+2 +k1 2 A n+3 )
[0094] +k1 2(k1 2 A n +k1A n+1 +A n+2 +k1A n+3 +k1 2 A n+4 )
[0095] Here, since 0.5<k1<1, k1 3 ≒k1 4 Very small.
[0096] Therefore, if k1 is approximated 3 =k1 4 =0, then
[0097] C n =k1 2 A n-2
[0098] -k1 2 (A n-2 +A n )-k1A n-1
[0099] +k1 2 A n-2 +k1A n-1 +A n +k1A n+1 +k1 2 A n+2
[0100] -k1 2 (A n +A n+2 )-k1A n+1
[0101] +k1 2 A n+2
[0102] =k1 2 (A n-2 -2A n +A n+2 )
[0103] +k1(-A n-1 +A n-1 +A n+1 -A n+1 )
[0104] +A n
[0105] Here, (-A n-1 +A n-1 +A n+1 -An+1 ) is zero, so
[0106] C n =k1 2 (A n-2 -2A n +A n+2 )+A n …(B).
[0107] Furthermore, the detection value correction system and detection value correction method of the present invention are intended to solve the following problem: if the adjacent spacing of sensors is reduced, the detection accuracy of the sensors may decrease. Therefore, the adjacent spacing of sensors is sufficiently small relative to the spatial extension range of the physical quantity of the detection object such as the magnetic field. As a result, the adjacent spacing of sensors is set so that the difference in the physical quantity of the detection object at the configuration position of adjacent sensors Mn, M(n+1) or sensors Mn, M(n-1) becomes the physical quantity A at the position of sensor Mn. n For example, it is 1% or less.
[0108] In this way, A n+2 and A n-2 is the magnetic field strength (physical quantity) at the adjacent sensor positions of sensor Mn, so the difference in magnetic field strength (physical quantity) at the position of sensor Mn is about 1%+1%=2%. Therefore, it can be approximated as A n ≒A n-2 ≒A n+2 , so it can be approximated as (A n-2 -2A n +A n+2 )≒0. In addition, since 0.5<k1<1, k1 2 (A n-2 -2A n +A n+2 ) is closer to zero, and can be approximated as k1 2 (A n-2 -2A n +A n+2 )≒0, so the formula (B) can be approximated as C n ≒A n .
[0109] Therefore, it can be seen that the correction value C calculated by the above formula (1) is n It can be approximated as the correct detection value A when sensor Mn is not affected by other sensors M n .
[0110] Therefore, the correction value C calculated by equation (1) is n It can be approximated as the correct detection value A when sensor Mn is not affected by other sensors Mn Therefore, through the processing of steps S1 to S6, all the detection values B1 to B 10 Correction is performed to obtain correction values C1 to C 10 Therefore, it is easy to correct the detection values of the sensors M arranged in a row.
[0111] Next, the coefficients k1 and k2 are described. The coefficients k1 and k2 vary depending on the difference in sensitivity caused by manufacturing deviations of the sensors M, the intervals between the sensors M, the distance from the magnetic sensor array 2 to the object to be measured, and the like. Therefore, for example, it is preferable that the coefficients k1 and k2 of the magnetic sensor array 2 are calculated and stored in the storage unit of the operation processing unit 5 after the magnetic sensor array 2 is manufactured, or the coefficients k1 and k2 of the magnetic sensor array 2 are calculated and stored in the storage unit of the operation processing unit 5 after the magnetic sensor array 2 is installed in the magnetic measurement device 1.
[0112] In the formula (1), B n-2 , B n-1 , B n , B n+1 , B n+2 The coefficients of are k2, -k1, 1, -k1, k2 including the sign, and the coefficients function as a filter. Since the sensor Mn is affected by the sensors M on its left and right to the same extent, the coefficients are based on the detection value B. n The arrangement is symmetrical with respect to the center.
[0113] Figure 3 Indicates the case where k1 = 0.6 and k2 = 0.1, Figure 4 This shows the case where k1=0.7 and k2=0.2.
[0114] The coefficient k1 is a numerical expression of the degree of influence of the sensors M(n+1) and M(n-1) as the first sensor on the sensor Mn of interest. The coefficient k2 is a numerical expression of the degree of influence of the sensors M(n+2) and M(n-2) as the second sensor on the sensor Mn of interest. The sensors M(n+2) and M(n-2) are farther from the sensor Mn of interest than the sensors M(n+1) and M(n-1), so the influence of the sensors M(n+2) and M(n-2) on the sensor Mn of interest is smaller than the influence of the sensors M(n+1) and M(n-1) on the sensor Mn of interest. Therefore, the influence represented by the coefficient k2 is smaller than the influence represented by the coefficient k1, so k1>k2.
[0115] The so-called influence is nothing but energy. For example, when a sensor M has an influence on another sensor M, energy will be transferred from the sensor that exerted the influence to the sensor that was affected.
[0116] thus, Figure 3 , Figure 4 The bars representing the coefficients can be considered as representing the exchange of energy by setting the energy received from others as positive and the energy given to others as negative. Although there is such an exchange of energy, the total amount of energy does not change, so the sum of k2, -k1, 1, -k1, and k2 is approximately zero. In other words, the value obtained by subtracting the sum of coefficients k1 from the sum of coefficients k2 plus 1 is approximately zero.
[0117] as a result, Figure 3 , Figure 4 The total of positive values in is equal to the total of negative values. Therefore, from the positive coefficients k2, 1, k2 and the negative coefficients k1, k1, the following equation (5) is obtained.
[0118] k2+1+k2=k1+k1=2k1
[0119] 2k2+1=2k1
[0120] 2k2=2k1-1
[0121] k2=k1-1 / 2…(5)
[0122] Figure 5 Indicates the case where coefficient k1 is 1.
[0123] When the coefficient k1 exceeds 1, the energy received by the adjacent sensors M on both sides is greater than the energy emitted from the sensor Mn of interest, resulting in a contradiction. Therefore, k1<1.
[0124] Figure 6 The coefficient k1 is 0.5. When the coefficient k1 is 0.5, the coefficient k2 becomes 0 according to equation (5), and the coefficient k2 disappears. Therefore, the range of the coefficient k1 is 0.5<k1<1.
[0125] In addition, the relationship between coefficient k1 and coefficient k2 is not necessarily limited to the example represented by equation (5). The relationship between coefficient k1 and coefficient k2 is appropriately set according to the characteristics of the sensor used. For example, in the case of a sensor having the following characteristics, the relationship of equation (4) below is applicable, wherein the influence of the distant sensor on the sensor of interest decreases rapidly as the distance from the sensor of interest increases.
[0126] k2=k1 2 …(4)
[0127] Furthermore, for example, in the case of a sensor having the characteristic that the influence decreases in inverse proportion to the square of the distance from the sensor of interest, the relationship of the following equation (6) is applicable.
[0128] k2=k1 / 4…(6)
[0129] Next, refer to Figure 7 , Figure 8 The coefficient calculation method according to one embodiment of the present invention will be described. First, for example, the user places a magnet near a portion of the Q sensors M1 to MQ, for example, the sensor M5. Thus, an inhomogeneous magnetic field is applied to the sensors M1 to MQ (step S101). Figure 1 In the example shown, Q=10. It is preferable that the magnetic field applied to sensors M1 to MQ changes greatly with respect to each position of sensors M1 to MQ so that the difference in detection value detected by each sensor M is as large as possible. Sensors M1 to MQ correspond to sensors 1 to Q.
[0130] In a state where a non-uniform magnetic field is applied to the sensors M1 to MQ, the coefficient calculation processing unit 53 acquires the detection values B1 to B2 of the sensors M1 to MQ. Q (Step S102: process (a)).
[0131] Next, the coefficient calculation processing unit 53 temporarily sets the coefficient k1 to 0.51 (step S103). Next, the coefficient calculation processing unit 53 sets the sensor number X to 3 as an initial value (step S104). Next, the coefficient calculation processing unit 53 sets the detection value of the sensor MX to B. X , the detection value of sensor M(X+1) adjacent to one side of sensor MX is B X+1 , the detection value of the sensor M(X-1) adjacent to the other side of the sensor MX is B X-1 , the detection value of sensor M(X+2) adjacent to sensor MX in the opposite direction relative to sensor M(X+1) is B X+2 The detection value of sensor M(X-2) adjacent to sensor MX in the opposite direction relative to sensor M(X-1) is B X-2 , the value D(X) is calculated based on the following formula (2) and the above-mentioned formula (5) (step S105: step (b1)).
[0132] D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 …(2)
[0133] In step S105, in order to obtain coefficients k1 and k2, equation (4) may be used instead of equation (5), or equation (6) may be used. Alternatively, for example, a relational expression other than equations (4) to (6) obtained by experiment may be used to obtain coefficients k1 and k2.
[0134] Next, the coefficient calculation processing unit 53 compares the sensor number X with Q-2 (step S106). If the sensor number X is not equal to Q-2 (No in step S106), the coefficient calculation processing unit 53 adds 1 to the sensor number X (step S107) and repeats steps S105 and S106 again.
[0135] On the other hand, if the sensor number X is equal to Q-2 (YES in step S106 ), since the values D(3) to D(Q-2) have been calculated, the coefficient calculation processing unit 53 shifts the process to step S108 .
[0136] In step S108, the coefficient calculation processing unit 53 searches for the value D(n) and the value D(n-1) of the adjacent sensor Mn and sensor M(n-1) among the values D(3) to D(Q-2) at which the square of the difference between the value D(n) and the value D(n-1) reaches the maximum, and calculates the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ](Step S108: step (b2)). n is an integer from 4 to (Q-2).
[0137] According to step S108, the sensor Mn and the sensor M(n-1) whose detection values B of the adjacent sensors M have the largest difference will be found.
[0138] Next, the coefficient calculation processing unit 53 searches for the value D(m) and value D(m-1) at which the square of the difference between the value D(m) and the value D(m-1) of the adjacent sensor Mm and sensor M(m-1) in the values D(3) to D(Q-2) reaches the minimum, and calculates the square of the difference between the value D(m) and the value D(m-1) reaching the minimum as MIN[{D(m)-D(m-1)} 2 ](Step S109: step (b3)). m is an integer from 4 to (Q-2).
[0139] According to step S109, the sensor Mm and the sensor M(m-1) whose detection values B difference between the adjacent sensors M is the smallest are found.
[0140] Next, the coefficient calculation processing unit 53 calculates the evaluation value U(k1) based on the following equation (3) (step S110: process (b4)).
[0141] U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]…(3)
[0142] When the evaluation value U(k1) is the coefficient k1 set in step S103 or S112, the square of the difference between the values obtained by correcting the detection values B of the adjacent sensors M by using the formula (1) is MAX[{D(n)-D(n-1)} 2 ], and the square of the difference between the values obtained by correcting the detection values B of the adjacent sensors M by using the formula (1) for the sensor Mm and the sensor M(m-1) whose detection values B are the smallest, that is, MIN[{D(m)-D(m-1)} 2 ] difference.
[0143] In other words, the evaluation value U(k1) represents the difference between the correction value difference at which the difference in correction value of the detection value B between adjacent sensors M reaches the maximum and the correction value difference at which the difference in correction value of the detection value B between adjacent sensors M reaches the minimum relative to the uneven magnetic field applied to the magnetic sensor array 2 at the temporary coefficient k1.
[0144] That is, the larger the evaluation value U(k1), the greater the difference between the portion with large position change and the portion with small position change in the correction value obtained at the coefficient k1.
[0145] For the evaluation value U(k1), the more accurately the coefficient k1 represents the influence of the sensor M on other adjacent sensors M, the more accurately the inhomogeneity of the magnetic field applied to the magnetic sensor array 2 is reflected in the evaluation value U(k1), and as a result, the evaluation value U(k1) becomes a larger value. On the other hand, the more the coefficient k1 deviates from the influence of the sensor M on other adjacent sensors M, the more the inhomogeneity of the magnetic field applied to the magnetic sensor array 2 is diffused, and the evaluation value U(k1) becomes a smaller value.
[0146] Therefore, the processing recorded in steps S103 to S123 is: while changing the coefficient k1, the evaluation value U(k1) is calculated, and the coefficient k1 when the maximum evaluation value U(k1) is obtained is obtained as the coefficient k1 that most accurately represents the influence of the sensor M on other adjacent sensors M, that is, the correct coefficient k1.
[0147] Next, the coefficient calculation processing unit 53 compares the coefficient k1 with 0.99. If the coefficient k1 is not equal to 0.99 (No in step S111), the coefficient calculation processing unit 53 adds 0.01 to the coefficient k1 and repeats steps S104 to S111 again. On the other hand, if the coefficient k1 is equal to 0.99 (Yes in step S111), since the calculation of the plurality of evaluation values U(0.51), U(0.52), U(0.53), ..., U(0.99) has been completed, the process proceeds to step S121.
[0148] In addition, in step S112, as long as different values can be set for the plurality of coefficients k1, it is not necessary to set them to values that differ by 0.01. The smaller the value added in step S112, i.e., the difference, the higher the accuracy of finding the coefficient k1, but on the other hand, the greater the amount of computational processing. Therefore, the value added in step S112 can be appropriately set in consideration of accuracy and processing volume.
[0149] In step S121, the coefficient calculation processing unit 53 searches for the maximum evaluation value U(k1) from the evaluation values U(0.51) to U(0.99) (step S121). Next, the coefficient calculation processing unit 53 obtains the coefficient k1 corresponding to the maximum evaluation value U(k1) as the coefficient k1 for correction processing (step S122). For example, if the evaluation value U(0.77) is the maximum, it can be estimated that 0.77 is the correct coefficient k1, so the coefficient calculation processing unit 53 sets the coefficient k1 for correction processing to 0.77. The coefficient k1 is used in the above step S4.
[0150] Next, the coefficient calculation processing unit 53 uses the same relational expression as that used in step S105, for example, equation (5), to calculate the coefficient k2 for correction processing from the coefficient k1 corresponding to the maximum evaluation value U(k1) (step S123). For example, in the case of coefficient k1=0.77, if coefficient k1=0.77 is substituted into equation (5), coefficient k2=0.27.
[0151] As described above, through the processing of steps S101 to S123, the coefficients k1 and k2 can be calculated.
[0152] In addition, the coefficients k1 and k2 are not necessarily limited to the examples calculated by the coefficient calculation method described above. Since the coefficients k1 and k2 are values indicating the influence of other sensors on the sensor of interest, for example, the detection value A of the sensor of interest alone without being influenced by other sensors can be measured by experiment, and the detection value B of the sensor of interest without being influenced by the second sensor and the detection value C of the first sensor can be measured only by the sensor of interest and the first sensor, and then the detection values D, E, and F detected by the sensor of interest, the first sensor, and the second sensor can be measured in the same manner as the structure actually used, and the coefficients k1 and k2 can be calculated based on the detection values A to F.
[0153] That is, the detection value D can also be used as the detection value B n , take the detection value E as the detection value B n+1 , B n-1 , take the detection value F as the detection value B n+2 , B n-2 Substituting into equation (1), we can obtain the correction value C which is roughly equal to the detection value A. n The coefficients k1 and k2 can be found by using the method
[0154] In order to perform measurement without being affected by other sensors, for example, amorphous wires separated for each sensor may be used. Figure 1 The sensors M1 to M10 shown in the figure can be used to perform the experimental measurement described above. Fig. 9 In the case of the sensors E1 to E10 shown in the figure, it is sufficient to separate the scintillator SC for each sensor and to arrange only the necessary sensors to perform the experimental measurement. Fig.10 In the case of the sensors F1 to FQ shown, only necessary sensors need only be arranged in the fluid to perform the above-described experimental measurement.
[0155] However, the coefficients k1 and k2 are affected by the distance between the sensors, the characteristic deviation of the sensors, etc. Therefore, due to the distance change caused by the assembly accuracy of the sensors actually used for measurement or the deviation of the sensor characteristics, when the coefficients k1 and k2 are obtained by experimental measurement in a state where the sensors are dispersed as described above, the accuracy of the coefficients may decrease.
[0156] On the other hand, according to the coefficient calculation method, the coefficients k1 and k2 can be calculated in a state where the sensor used for actual measurement is already assembled, so it is easy to improve the accuracy of the coefficients k1 and k2. Therefore, it is more preferable to calculate the coefficients k1 and k2 using the coefficient calculation method.
[0157] Fig. 9 The X-ray measuring device 1a shown in FIG. Figure 1 The magnetism measuring device 1 shown is different in that it includes sensors E1 to E10 instead of sensors M1 to M10. The X-ray measuring device 1a is configured similarly to the magnetism measuring device 1 in other respects, so the description thereof is omitted and the characteristic features of the X-ray measuring device 1a are described below.
[0158] The X-ray measuring device 1a includes: a plurality of photodiodes PD arranged in a row; and a scintillator SC as a fluorescent plate that emits light by X-rays. The plurality of photodiodes PD are arranged to face the scintillator SC. Furthermore, sensors E1 to E10 are formed by each photodiode PD and a portion of the scintillator SC facing each photodiode PD. The sensors E1 to E10 arranged in a row in this way form an X-ray sensor array 2a.
[0159] When the X-rays to be measured are irradiated to the scintillator SC, the scintillator SC emits light in an amount corresponding to the X-ray intensity. The light emitted by the scintillator SC is detected by a photodiode PD arranged at a position corresponding to the light emission position, and a voltage corresponding to the light emission amount is output from the photodiode PD.
[0160] For example, when Fig. 9 When the X-ray 100 shown is irradiated to the scintillator SC of the sensor E5 and the scintillator SC emits light, the light is detected not only by the photodiode PD of the sensor E5 but also by the surrounding sensors E4, E6, etc. That is, the sensors E1 to E10 are equivalent to an example of a sensor that affects the detection value B of other sensors around the sensors E1 to E10 when detecting X-rays (physical quantity).
[0161] Each voltage measuring unit 4 measures the output voltage of each photodiode PD in the sensors E1 to E10, and generates a signal indicating the measurement result as the detection value B1 to B2 of the sensors E1 to E10. 10 And sent to the calculation processing unit 5.
[0162] The calculation processing unit 5 detects the values B1 to B10 based on the values B1 to B10 detected by the sensors E1 to E10. 10 And Figure 1 The operation processing unit 5 shown in FIG. 1 operates in the same manner. Specifically, except that the sensors E1 to E10 are irradiated with uneven X-rays in step S101, the same coefficient calculation method as described above is executed. In addition, the detection value correction method other than step S1 is executed. Thus, it is easy to correct the detection values B1 to B2 of each sensor E1 to E10. 10 Make corrections.
[0163] Fig.10This is an explanatory diagram conceptually showing an example of the structure of a sensor to which the detection value correction method according to one embodiment of the present invention can be applied. Fig.10 The Q sensors F1 to FQ shown are flow sensors for measuring the flow rate or flow velocity of the fluid. The sensors F1 to FQ may be various known flow meters such as electromagnetic flow meters, Karman vortex flow meters, impeller flow meters, float flow meters, thermal flow meters, diaphragm flow meters, ultrasonic flow meters, and Coriolis flow meters.
[0164] Fig.10 In the example shown, sensors F1 to FQ are arranged in a row so as to extend in a direction intersecting the flow direction of water as a fluid in the river R. Hereinafter, sensors F1 to FQ are collectively referred to as sensors F.
[0165] The processing unit 5 is housed and installed in a housing (not shown) at the river bank of the river R. The sensors F1 to FQ measure the flow rate or flow velocity at the installation location of each sensor, and use a wired or wireless communication circuit (not shown) to communicate the detected values B1 to B 10 Send to the calculation processing unit 5.
[0166] The calculation processing unit 5 detects the values B1 to B1 of the sensors F1 to FQ based on the values B1 to B1 of the sensors F1 to FQ. Q And Figure 1 The operation processing unit 5 shown in FIG. 1 operates in the same manner. Specifically, except that an uneven amount of water is passed through the sensors F1 to FQ in step S101, the same coefficient calculation method as described above is performed. In addition, the detection value correction method other than step S1 is performed. As a result, it is easy to correct the detection values B1 to B2 of each sensor F1 to FQ. Q The sensors F1 to FQ and the calculation processing unit 5 constitute a flow rate measurement system 1 b as an example of a detection value correction system.
[0167] When the flow of river R hits sensors F1~FQ, Fig.10 As shown by the arrow in , the flow is disturbed. The disturbance of the river flow also affects the flow measured by the adjacent sensor F. That is, when the sensors F1 to FQ detect the flow rate or flow velocity (physical quantity) of the fluid, they affect the detection value B of other sensors F around each sensor F. The calculation processing unit 5 can easily correct the detection value B of such a sensor F.
[0168] Furthermore, although an example in which the detection value of the sensor Mn as the sensor of interest is corrected based on the equation (1) is shown, the present invention is not necessarily limited to the example in which the equation (1) is used.
[0169] Moreover, the correction processing only needs to correct the detection value of the sensor of interest based on the detection values of the first and second sensors and the first and second coefficients according to the required correction accuracy. For example, the correction can also be performed by further considering parameters other than the detection values of the first and second sensors and the first and second coefficients.
[0170] The correction processing only needs to be based on the detection value of at least the first sensor. It can be based only on, for example, the detection value of the first sensor according to the required correction accuracy, or it can be based only on the detection value of the first sensor and the first coefficient, or it can be based only on the detection values of the first and second sensors and the first coefficient, or it can be based only on the detection values of the first and second sensors and the second coefficient.
[0171] Furthermore, an example is shown in which two first sensors are provided on both sides of the sensor of interest, and two second sensors are provided on both sides outside the sensor of interest, but a configuration in which each of the first and second sensors is provided on one side of the sensor of interest may also be adopted. Even in a configuration in which each of the first and second sensors is provided on one side of the sensor of interest, a certain degree of correction effect can be obtained.
[0172] That is, a detection value correction system of an example of the present invention corrects the detection values of multiple sensors arranged in a row and detecting physical quantities, and the detection value correction system includes: a correction processing unit that performs correction processing on the detection value of a focus sensor that is a sensor to be corrected among the multiple sensors, based on the detection value of a first sensor adjacent to the focus sensor.
[0173] Moreover, a detection value correction method of an example of the present invention corrects the detection values of multiple sensors arranged in a row and detecting physical quantities, wherein a correction process is performed on the detection value of a focus sensor that is a sensor to be corrected among the multiple sensors, at least based on the detection value of a first sensor adjacent to the focus sensor.
[0174] According to this structure, when a plurality of sensors are arranged in a row, the detection value of the sensor of interest that is the sensor to be corrected is corrected based on the detection value of the first sensor adjacent to the sensor of interest, so that correction can be made to reduce the influence of the first sensor on the sensor of interest. Therefore, it is easy to correct the detection values of the sensors arranged in a row.
[0175] Furthermore, it is preferable that the correction process further corrects the detection value of the sensor of interest based on a first coefficient indicating the influence of the first sensor on the sensor of interest.
[0176] According to this configuration, the detection value of the sensor of interest is corrected based on the detection value of the first sensor and the first coefficient indicating the influence of the first sensor on the sensor of interest, so that the correction accuracy can be easily improved.
[0177] Furthermore, it is preferable that the correction process further corrects the detection value of the sensor of interest based on the detection value of a second sensor that is adjacent to the sensor of interest in the opposite direction relative to the first sensor.
[0178] According to this configuration, the detection value of the sensor of interest can be corrected so that not only the influence of the first sensor on the sensor of interest but also the influence of the second sensor on the sensor of interest is reduced.
[0179] Furthermore, it is desirable that the correction process further corrects the detection value of the sensor of interest based on a second coefficient, the second coefficient indicating the influence of the second sensor on the sensor of interest, and the second coefficient is smaller than the first coefficient.
[0180] According to this configuration, the detection value of the sensor of interest is corrected in consideration of the second coefficient indicating the influence of the second sensor on the sensor of interest, so that the correction accuracy can be easily improved.
[0181] Furthermore, preferably, when the detection value of the sensor of interest is B n , the detection value of the first sensor adjacent to one side of the sensor of interest is B n+1 , the detection value of the first sensor adjacent to the other side of the sensor of interest is B n-1 , the detection value of the second sensor adjacent to the sensor of interest on the opposite side is B n+2 , the detection value of the second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor adjacent to the other side of the sensor of interest is B n-2 , when the first coefficient is k1 and the second coefficient is k2, the correction processing is based on the following formula (1) to calculate the correction value C of the detection value of the sensor of interest: n processing.
[0182] C n =k2B n-2 -k1B n-1 +B n -k1B n+1 +k2B n+2 …(1)
[0183] According to this configuration, the detection value of the sensor of interest can be corrected by using the numerical calculation of equation (1).
[0184] Furthermore, it is preferred that the method further comprises a coefficient calculation processing unit, wherein the coefficient calculation processing unit performs the following steps: (a) obtaining the detection value of each sensor in a state where the physical quantity is assigned to the plurality of sensors arranged in a row; (b) temporarily assigning different values to the plurality of first coefficients k1, and corresponding to the first coefficients k1 assigned the temporary values, respectively, (b1) labeling the plurality of sensors with consecutive numbers and setting them as sensors 1 to Q, and the detection value of any X-th sensor X among the sensors 1 to Q is B X , the detection value of the X+1th sensor (X+1) adjacent to one side of the sensor X is B X+1 The detection value of the X-1th sensor (X-1) adjacent to the other side of the sensor X is B X-1 , the detection value of the X+2th sensor (X+2) adjacent to the sensor X in the opposite direction relative to the sensor (X+1) is B X+2 The detection value of the X-2th sensor (X-2) adjacent to the sensor X in the opposite direction relative to the sensor (X-1) is B X-2 , the second coefficient is k2, and X is 3 to (Q-2), the value D is calculated based on the following formula (2): X (b2) For n, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(n) and value D(n-1) at which the square of the difference between the value D(n) and the value D(n-1) of the adjacent sensor n and sensor (n-1) reaches the maximum, and calculate the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ], (b3) for m, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(m) and value D(m-1) at which the square of the difference between the value D(m) and the value D(m-1) of the adjacent sensor m and sensor (m-1) reaches the minimum, and calculate the square of the difference between the minimum value D(m) and the value D(m-1) as MIN[{D(m)-D(m-1)} 2 ], (b4) based on the following formula (3), calculate multiple evaluation values U(k1) corresponding to the values of the multiple first coefficients k1; and (c) explore the maximum evaluation value U(k1) among the multiple evaluation values U(k1), determine the first coefficient k1 corresponding to the maximum evaluation value U(k1) and the second coefficient k2 obtained based on the first coefficient k1, as the first coefficient and the second coefficient used in the correction process.
[0185] D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 …(2)
[0186] U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]…(3)
[0187] According to this configuration, the first coefficient k1 and the second coefficient k2 suitable for the correction process can be calculated.
[0188] Furthermore, it is preferred that the second coefficient k2 is obtained based on the following equation (4).
[0189] k2=k1 2 …(4)
[0190] According to this configuration, it is possible to obtain the second coefficient k2 suitable for a case where the influence of a distant sensor on the sensor of interest rapidly decreases as the distance from the distant sensor increases.
[0191] Furthermore, the second coefficient k2 can also be obtained based on the following formula (5).
[0192] k2=k1-1 / 2…(5)
[0193] The first coefficient k1 and the second coefficient k2 represent the influence between sensors, and the so-called influence is nothing more than the exchange of energy. According to this structure, the second coefficient k2 suitable for the following situation can be obtained, that is, the energy is exchanged between the sensor of interest, the first sensor and the second sensor, and the energy of these sensors as a whole is fixed.
[0194] Furthermore, the second coefficient k2 can also be obtained based on the following formula (6).
[0195] k2=k1 / 4…(6)
[0196] According to this structure, it is possible to obtain the second coefficient k2 which is suitable for the case where the influence becomes smaller in inverse proportion to the square of the distance from the sensor of interest.
[0197] Moreover, preferably, the first coefficient k1 is greater than 0.5 and less than 1.
[0198] According to this configuration, it is easy to set the first coefficient k1 to an appropriate value.
[0199] Furthermore, each of the sensors may affect detection values of other sensors around the sensor when detecting the physical quantity.
[0200] According to the correction processing described above, it is easy to correct the detection value of such a sensor.
[0201] Furthermore, the physical quantity may be a magnetic field, and each of the sensors may be an MI sensor.
[0202] According to the correction processing described above, it is easy to correct the detection value of such an MI sensor.
[0203] Furthermore, the physical quantity may be X-rays, and each of the sensors may be an X-ray sensor including a fluorescent plate and a photosensor, wherein the fluorescent plate emits light due to the X-rays, and the photosensor detects the light emitted by the fluorescent plate.
[0204] According to the correction processing described above, it is easy to correct the detection value of such an X-ray sensor.
[0205] Furthermore, the physical quantity may be a flow rate or a flow velocity of a fluid, and the sensors may be arranged such that the row extends in a direction intersecting with a flow direction of the fluid.
[0206] According to the correction processing described above, it is easy to correct the detection value of such a sensor.
[0207] Furthermore, it is preferable that the plurality of sensors are further included.
[0208] According to this structure, the detected value correction system can detect and correct the physical quantity.
[0209] Moreover, a coefficient calculation method of an example of the present invention calculates a first coefficient k1 and a second coefficient k2 for one of a plurality of sensors arranged in a row and detecting a physical quantity, wherein the first coefficient k1 represents the influence of a first sensor adjacent to the sensor on the sensor, and the second coefficient k2 represents the influence of a second sensor adjacent to the one sensor in the opposite direction relative to the first sensor on the sensor, and the coefficient calculation method includes the following steps: (a) acquiring detection values of the sensors in a state where the physical quantity is unevenly assigned to the plurality of sensors arranged in a row; (b) temporarily assigning different values to the plurality of first coefficients k1, corresponding to the first coefficients k1 assigned the temporary values, respectively; (b1) labeling the plurality of sensors with consecutive numbers and setting them as sensors 1 to Q, and the detection value of any Xth sensor X among sensors 1 to Q is B X , the detection value of the X+1th sensor (X+1) adjacent to one side of the sensor X is B X+1The detection value of the X-1th sensor (X-1) adjacent to the other side of the sensor X is B X-1 , the detection value of the X+2th sensor (X+2) adjacent to the sensor X in the opposite direction relative to the sensor (X+1) is B X+2 The detection value of the X-2th sensor (X-2) adjacent to the sensor X in the opposite direction relative to the sensor (X-1) is B X-2 , the second coefficient is k2, and X is 3 to (Q-2), the value D is calculated based on the following formula (2): X (b2) For n, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(n) and value D(n-1) at which the square of the difference between the value D(n) and the value D(n-1) of the adjacent sensor n and sensor (n-1) reaches the maximum, and calculate the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ], (b3) for m, which is an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in step (b1), search for the value D(m) and value D(m-1) at which the square of the difference between the value D(m) and the value D(m-1) of the adjacent sensor m and sensor (m-1) reaches the minimum, and calculate the square of the difference between the minimum value D(m) and the value D(m-1) as MIN[{D(m)-D(m-1)} 2 ], (b4) based on the following formula (3), calculate multiple evaluation values U(k1) corresponding to the values of the multiple coefficients k1; and (c) explore the maximum evaluation value U(k1) among the multiple evaluation values U(k1), determine the first coefficient k1 corresponding to the maximum evaluation value U(k1) and the second coefficient k2 obtained based on the first coefficient k1, as the first coefficient and the second coefficient used in the correction process.
[0210] D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 …(2)
[0211] U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]…(3)
[0212] According to this configuration, the first coefficient k1 and the second coefficient k2 suitable for the correction process can be calculated.
[0213] Furthermore, in the detection value correction method, it is preferred that the correction process is further based on the detection value of a second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor and a second coefficient representing the influence of the second sensor on the sensor of interest, and when the detection value of the sensor of interest is B n , the detection value of the first sensor adjacent to one side of the sensor of interest is B n+1 , the detection value of the first sensor adjacent to the other side of the sensor of interest is B n-1 , the detection value of the second sensor adjacent to the sensor of interest on the opposite side is B n+2 , the detection value of the second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor adjacent to the other side of the sensor of interest is B n-2 , when the first coefficient is k1 and the second coefficient is k2, the correction value C of the detection value of the sensor of interest is calculated based on the following formula (1): n .
[0214] C n =k2B n-2 -k1B n-1 +B n -k1B n+1 +k2B n+2 …(1)
[0215] According to this configuration, the detection value of the sensor of interest can be corrected by numerical calculation using equation (1).
[0216] Such a detection value correction system and detection value correction method can easily correct the detection value of each sensor even when a plurality of sensors are arranged in a row. Moreover, such a coefficient calculation method can calculate coefficients suitable for correction processing in the detection value correction system and detection value correction method.
[0217] In addition, the specific implementation or example described in the specific implementation is only for clarifying the technical content of the present invention, and the present invention should not be limited to such specific examples and interpreted narrowly.
Claims
1. A detection value correction system for correcting detection values of a plurality of sensors arranged in a row for detecting physical quantities, the detection value correction system comprising: a correction processing unit that performs correction processing for a detection value of a sensor of interest that is a sensor to be corrected among the plurality of sensors, based on a detection value of a first sensor adjacent to the sensor of interest, The correction processing further corrects the detection value of the sensor of interest in a manner that reduces the detection value of the sensor of interest based on a first coefficient, wherein the first coefficient represents the influence of the first sensor on the sensor of interest, The correction processing further corrects the detection value of the sensor of interest based on the detection value of a second sensor, the second sensor being adjacent to the sensor of interest in the reverse direction relative to the first sensor, and the arrangement order of the second sensor adjacent to the sensor of interest in the reverse direction is the sensor of interest, the first sensor, and the second sensor, The correction processing further corrects the detection value of the sensor of interest in a manner that increases the detection value of the sensor of interest based on a second coefficient, wherein the second coefficient represents the influence of the second sensor on the sensor of interest, And the second coefficient is smaller than the first coefficient.
2. The detection value correction system according to claim 1, characterized in that: When the detection value of the sensor of interest is B n , the detection value of the first sensor adjacent to one side of the sensor of interest is B n+1 , the detection value of the first sensor adjacent to the other side of the sensor of interest is B n-1 , the detection value of the second sensor adjacent to the sensor of interest on the opposite side is B n+2 , the detection value of the second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor adjacent to the other side of the sensor of interest is B n-2 , when the first coefficient is k1 and the second coefficient is k2, the correction processing is to calculate the correction value C of the detection value of the sensor of interest based on the following formula (1): n processing, C n =k2B n-2 -k1B n-1 +B n -k1B n+1 +k2B n+2 ...(1)。 3. The detection value correction system according to claim 2, characterized in that: The invention also includes a coefficient calculation processing unit, which performs the following steps: (a) acquiring detection values of each sensor in a state where the physical quantity is applied to a plurality of sensors arranged in a row; (b) temporarily assigning different values to a plurality of the first coefficients k1, and corresponding to each of the first coefficients k1 assigned the temporary value, (b1) The plurality of sensors are assigned consecutive numbers and are referred to as sensors 1 to Q. The detection value of any X-th sensor X among the sensors 1 to Q is B. X , the detection value of the X+1th sensor (X+1) adjacent to one side of the sensor X is B X+1 , the detection value of the X-1th sensor (X-1) adjacent to the other side of the sensor X is B X-1 , the detection value of the X+2th sensor (X+2) adjacent to the sensor X in the opposite direction relative to the sensor (X+1) is B X+2 , the detection value of the X-2th sensor (X-2) adjacent to the sensor X in the opposite direction relative to the sensor (X-1) is B X-2 , when the second coefficient is k2 and X is 3 to (Q-2), the value D(X) is calculated based on the following formula (2): D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 ...(2) (b2) For n being an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in the step (b1), search for the value D(n) and the value D(n-1) of the adjacent sensor n and the sensor (n-1) at which the square of the difference between the value D(n) and the value D(n-1) reaches the maximum, and calculate the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ], (b3) For m being an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in the step (b1), search for the value D(m) and value D(m-1) of the adjacent sensor m and sensor (m-1) at which the square of the difference between the value D(m) and the value D(m-1) reaches the minimum, and calculate the square of the difference between the minimum value D(m) and the value D(m-1) as MIN[{D(m)-D(m-1)} 2 ], (b4) calculating a plurality of evaluation values U(k1) corresponding to a plurality of values of the first coefficient k1 based on the following formula (3); U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]…(3) as well as (c) searching for a maximum evaluation value U(k1) among the plurality of evaluation values U(k1), and determining a first coefficient k1 corresponding to the maximum evaluation value U(k1) and a second coefficient k2 obtained based on the first coefficient k1 as the first coefficient and the second coefficient used in the correction process.
4. The detection value correction system according to claim 2, characterized in that: The second coefficient k2 is obtained based on the following formula (4): k2=k1 2 …(4)。 5. The detection value correction system according to claim 2, characterized in that: The second coefficient k2 is obtained based on the following formula (5): k2=k1-1 / 2…(5).
6. The detection value correction system according to claim 2, characterized in that: The second coefficient k2 is obtained based on the following formula (6): k2=k1 / 4…(6).
7. The detection value correction system according to claim 2, characterized in that: The first coefficient k1 is greater than 0.5 and less than 1.
8. The detection value correction system according to any one of claims 1 to 7, characterized in that: When detecting the physical quantity, each of the sensors affects detection values of other sensors around the sensor.
9. The detection value correction system according to any one of claims 1 to 7, characterized in that: The physical quantity is the magnetic field, Each of the sensors is a magneto-impedance sensor.
10. The detection value correction system according to any one of claims 1 to 7, characterized in that: The physical quantity is X-rays, Each of the sensors is an X-ray sensor including a fluorescent plate that emits light by the X-rays and a photosensor that detects the light emitted by the fluorescent plate.
11. The detection value correction system according to any one of claims 1 to 7, characterized in that: The physical quantity is the flow rate or flow velocity of the fluid, The sensors are arranged such that the row extends in a direction intersecting with the flow direction of the fluid.
12. The detection value correction system according to any one of claims 1 to 7, characterized in that: Also included are the plurality of sensors.
13. A coefficient calculation method, for one of a plurality of sensors arranged in a row and detecting a physical quantity, calculating a first coefficient k1 and a second coefficient k2, wherein the first coefficient k1 represents an influence of a first sensor adjacent to the sensor on the sensor, and the second coefficient k2 represents an influence of a second sensor adjacent to the one sensor in the opposite direction relative to the first sensor on the sensor, the coefficient calculation method comprising the following steps: (a) acquiring detection values of the sensors in a state where the physical quantity is unevenly applied to a plurality of sensors arranged in a row; (b) temporarily assigning different values to a plurality of the first coefficients k1, and corresponding to each of the first coefficients k1 assigned the temporary value, (b1) The plurality of sensors are assigned consecutive numbers and are referred to as sensors 1 to Q. The detection value of any X-th sensor X among the sensors 1 to Q is B. X , the detection value of the X+1th sensor (X+1) adjacent to one side of the sensor X is B X+1 , the detection value of the X-1th sensor (X-1) adjacent to the other side of the sensor X is B X-1 , the detection value of the X+2th sensor (X+2) adjacent to the sensor X in the opposite direction relative to the sensor (X+1) is B X+2 , the detection value of the X-2th sensor (X-2) adjacent to the sensor X in the opposite direction relative to the sensor (X-1) is B X-2 , when the second coefficient is k2 and X is 3 to (Q-2), the value D(X) is calculated based on the following formula (2): D(X)=k2B X-2 -k1B X-1 +B X -k1B X+1 +k2B X+2 … (2) (b2) For n being an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in the step (b1), search for the value D(n) and the value D(n-1) of the adjacent sensor n and the sensor (n-1) at which the square of the difference between the value D(n) and the value D(n-1) reaches the maximum, and calculate the square of the difference between the maximum value D(n) and the value D(n-1) as MAX[{D(n)-D(n-1)} 2 ], (b3) For m being an integer from 4 to (Q-2), among the values D(3) to D(Q-2) calculated in the step (b1), search for the value D(m) and value D(m-1) of the adjacent sensor m and sensor (m-1) at which the square of the difference between the value D(m) and the value D(m-1) reaches the minimum, and calculate the square of the difference between the minimum value D(m) and the value D(m-1) as MIN[{D(m)-D(m-1)} 2 ], (b4) calculating a plurality of evaluation values U(k1) corresponding to a plurality of values of the first coefficient k1 based on the following formula (3); U(k1)=MAX[{D(n)-D(n-1)} 2 ]-MIN[{D(m)-D(m-1)} 2 ]… (3) as well as (c) searching for the largest evaluation value U(k1) among the plurality of evaluation values U(k1), and determining a first coefficient k1 corresponding to the largest evaluation value U(k1) and a second coefficient k2 obtained based on the first coefficient k1 as the first coefficient and the second coefficient used in the correction process.
14. A detection value correction method for correcting detection values of a plurality of sensors arranged in a row for detecting physical quantities, characterized in that: performing a correction process of correcting a detection value of a sensor of interest that is a sensor to be corrected among the plurality of sensors based on a detection value of a first sensor adjacent to the sensor of interest, The correction processing further corrects the detection value of the sensor of interest in a manner that reduces the detection value of the sensor of interest based on a first coefficient, wherein the first coefficient represents the influence of the first sensor on the sensor of interest, The correction processing further corrects the detection value of the sensor of interest based on the detection value of a second sensor, the second sensor being adjacent to the sensor of interest in the reverse direction relative to the first sensor, and the arrangement order of the second sensor adjacent to the sensor of interest in the reverse direction is the sensor of interest, the first sensor, and the second sensor, The correction processing further corrects the detection value of the sensor of interest in a manner that increases the detection value of the sensor of interest based on a second coefficient, wherein the second coefficient represents the influence of the second sensor on the sensor of interest, And the second coefficient is smaller than the first coefficient.
15. The detection value correction method according to claim 14, characterized in that: The correction process is further based on a detection value of a second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor and a second coefficient representing an influence of the second sensor on the sensor of interest, When the detection value of the sensor of interest is B n , the detection value of the first sensor adjacent to one side of the sensor of interest is B n+1 , the detection value of the first sensor adjacent to the other side of the sensor of interest is B n-1 , the detection value of the second sensor adjacent to the sensor of interest on the opposite side is B n+2 , the detection value of the second sensor adjacent to the sensor of interest in the opposite direction relative to the first sensor adjacent to the other side of the sensor of interest is B n-2 , when the first coefficient is k1 and the second coefficient is k2, the correction processing is based on the following formula (1) to calculate the correction value C of the detection value of the sensor of interest: n processing, C n =k2B n-2 -k1B n-1 +B n -k1B n+1 +k2B n+2 ...(1)。
