JP2011122897A - Shape determination method of metal object - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a shape determination method of a metal object discriminating and detecting the shape of the metal object. <P>SOLUTION: A metal object intermingled in a detection object 2 is detected by using sensor coils 5, 6. A detection signal outputted from the sensor coils 5, 6 is subjected to signal processing analysis to determine the shape of the metal object. In analysis, a signal from an input timing of the detection object 2 into the sensor coils 5, 6 until an outgoing timing thereof after passing the sensor coils 5, 6, is standardized into a standardized signal. When there are three or more peak values in the standardized signal, it is determined that a needle exists, and when there are only two peak values in the standardized signal, it is determined that iron powder exists. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a metal object shape determination method for detecting metal. More particularly, the present invention relates to a method for determining the shape of a metal object that detects a metal shape such as a magnetic needle or powder according to the shape.

  Metal detection devices are used in various fields. A typical example is detection of metallic foreign matter mixed in a product in the product manufacturing process. Specifically, foreign substances such as metal fragments may be mixed in the inspection object in the manufacturing process of the inspection object such as food and medicine. In order to guarantee the safety of the inspection object, since it is necessary to remove the foreign object from the inspection object mixed with the metal foreign object, it is essential to detect the metal foreign object. In addition, there are a number of counterfeit counterfeiting measures that cause needles to be mixed into foods at retail stores such as supermarkets and convenience stores as well as the manufacturing process of products. These need to be detected and eliminated as well.

  The applicants of the present invention proposed the inventions described in Patent Documents 1 and 2. The apparatuses described in Patent Documents 1 and 2 detect a metallic foreign object in an object to be detected. Using the proposed sensor coil, we succeeded in detecting minute magnetic materials. Specifically, Patent Document 1 discloses a metal foreign object detection method and a metal foreign object detection device that detect a metal foreign object mixed in an object to be detected. According to the present invention, not only the metal foreign matter such as stainless steel mixed in the detection object is detected, but also the metal mixed in the detection object such as food, medicine, industrial material wrapped in the conductive packaging material. Foreign objects can also be detected.

  Japanese Patent No. 3857271 discloses a metal foreign object detection method, in which a foreign object mixed in a process of manufacturing a detected object in a package is detected in a conveyance path, and the conveyance A metal comprising a metal foreign object detecting step for detecting a metal foreign object mixed in the detected object by generating a magnetic field by a detection unit provided in the middle of the road and having a coil with a structure in which a conductor is wound around a core. In the foreign matter detection method, a minute magnetic field is generated by applying a voltage or supplying current to one of the coils, and a magnetic field detected from the metal foreign matter in response to the minute magnetic field is used as a detection voltage or a detection current of the coil. A detection signal output step of detecting and outputting a detection signal; and a signal analysis step of analyzing the detection signal to analyze the signal in order to identify the metal foreign object, The voltage applied to the coil or the supplied current is very small, and among the magnetization (BH) characteristics of the core constituting the coil, a magnetic flux density (B) and a magnetic field ( H) is a non-linear part having a minute value near 0, the current or voltage is a frequency of several hundred Hz to several tens kHz, and when the metal foreign matter passes near the coil, The frequency changes.

Japanese Patent No. 3857271 No. 3857271 is a metal foreign object detection device comprising: a conveying means having a conveyance path for conveying the detected object in order to detect a metal foreign object mixed in the process of manufacturing the detected object in the package; A metallic foreign matter detecting means for detecting a metallic foreign matter mixed in the detected object by generating a magnetic field by a detecting portion provided in the middle of the transport path and having a coil with a configuration in which a conductor is wound around a core. In the metal foreign object detection device, a voltage is applied to one coil or a current is supplied to generate a minute magnetic field, and a detected magnetic field from the metal foreign object in response to the minute magnetic field is detected as a detection voltage or a detection current of the coil. Detection signal output means for detecting and detecting a detection signal, and signal analysis means by signal analysis for analyzing the detection signal to identify the metal foreign matter, The voltage applied to the coil or the supplied current is very small, and among the magnetization (BH) characteristics of the core constituting the coil, a magnetic flux density (B) and a magnetic field ( H) uses a non-linear part having a minute value near 0, and the current or voltage flowing through the coil has a frequency of several hundred Hz to several tens kHz, and the metal foreign matter passes through the vicinity of the coil. In this case, the frequency changes.
The detection unit of the metal foreign object detection device may be configured such that the detected magnetic field from the metal foreign object affects the minute magnetic field and the voltage applied to the coil or the frequency of the current flowing through the coil. Changes to output the detection signal.

  This metallic foreign object detection device of Patent Document 1 detects a magnetic field detected from a metallic foreign object in response to a minute magnetic field by generating a minute magnetic field by a detection unit having a single coil wound around a core. It detects as a voltage or a detection current and outputs a detection signal. A very small magnetic field is a voltage that is applied to a coil at a frequency of several hundreds of Hz to several tens of kHz, or a current that is supplied to the coil. Is a non-linear portion in which the magnetic flux density (B) and the magnetic field (H) representing a small value near 0 are used.

  When the metal foreign object passes in the vicinity of the coil, the formation of magnetic force lines interlinking with the coil is disturbed, and the amplitude, phase, and frequency of the signal voltage are changed, thereby detecting the metal foreign object. This apparatus has an excellent sensitivity capable of detecting metallic foreign objects of 1 mm or less. Further, it is possible to detect an elongated metal object such as a needle in an aluminum package and an antioxidant made of metal powder. Patent Document 3 filed and filed by the applicant of the present invention discloses "a metal detector sensor and a metal detector". The metal detector sensor disclosed in Patent Document 3 is a metal detector sensor for detecting a metal object in an object to be detected, and is disposed in contact with the core, and generates a magnetic field line due to a static magnetic field. And a magnet for magnetizing the metal object.

  This metal detector disclosed in Patent Document 3 includes a transport unit having a transport path for transporting an object to be detected, and a coil of a conductive wire that is provided in the middle of the transport path and generates a magnetic field by passing an electric current. In a metal detector comprising a sensor coil wound around a metal core, the magnet is arranged in a range covered by a magnetic field generated by the sensor coil, generates a magnetic line of force due to a static magnetic field, and magnetizes the metal object. The current is an alternating current for generating an alternating magnetic field, and the magnet is a magnet disposed in contact with the core.

  Specifically, a magnet is fixed to the sensor in the vicinity of the coil sensor or directly, and an unequal static magnetic field linked to the sensor coil is formed. When the object to be detected crosses this unequal magnetic field, the metal foreign object is temporarily magnetized, and at the same time, the magnetic field generated from the magnetized object to be detected disturbs the magnetic field linked to the sensor coil. The sensor coil sends out the magnetic field disturbance as a detection signal.

Japanese Patent No. 3857271 JP 2005-188985 A Japanese Patent No. 3875161

  However, metal detectors that are widely used in the food manufacturing industry and pharmaceutical manufacturing industry recognize magnetic fields due to eddy currents when metals enter the vicinity of the sensor, but recognize the size and shape of the metal. not going. Furthermore, the conventional detector erroneously detects a metal foreign object when a part of the product is made of metal or a metal object such as an oxygen scavenger made of metal powder is intentionally enclosed. For example, when placing a product in a container with a metal lid, the contents must be inspected before the lid is closed. The inspection cannot be carried out. For this reason, many restrictions are imposed on the reliability of the inspection process and foreign object inspection at the time of product manufacture.

  Also, when a part of a product is wrapped with a metal wire or sealed and packaged, the metal wire is detected by the metal foreign object detection device and cannot be distinguished from the metal foreign object. Moreover, a metal plate may be attached in the product or on the packaging surface. Therefore, in the metal foreign object detection device, it is desired that the metal object intentionally included in the product can be distinguished from the metal foreign object.

The present invention has been made based on the technical background as described above, and achieves the following objects.
An object of the present invention is to provide a metal object shape determination method capable of determining and detecting a metal object shape.
Another object of the present invention is to provide a metal object shape determination method capable of distinguishing and detecting metal objects attached to pharmaceutical products, cosmetics, foodstuffs, etc., or enclosed together with their packages, and metal foreign objects. To do.

In order to achieve the above object, the present invention employs the following means.
The metal object shape determination method according to the first aspect of the present invention includes a transport step of transporting an object to be detected on a transport path, and a magnet booster provided in the middle of the transport path and disposed in the vicinity of the sensor coil. While magnetizing or strengthening the magnetic substance in the object to be detected, a change in the magnetic field near the sensor coil is detected by the sensor coil and a detection signal is output to detect a metal object in the object to be detected. A metal object detection method comprising: a metal object detection step, wherein when the metal object passes in the vicinity of the sensor coil, (I) an object formed by the magnet booster is linked to the sensor coil. The metal object crosses the unequal magnetic field, (II) the magnetic object is disturbed by the metal object crossing the magnetic field, and (III) the magnetic flux linked to the sensor coil changes, IV) a detection signal output step in which the rate of change of the magnetic flux interlinking the sensor coil becomes a detection voltage or a detection current of the sensor coil to output the detection signal; and the detected object A signal analysis step of analyzing the detection signal from the timing of input to the sensor coil to the timing of passing through the sensor coil to identify the shape of the metal object, and the signal waveform of the detection signal Changes depending on the shape of one metal selected from linear (one-dimensional), planar (two-dimensional), spherical (three-dimensional), and massive (multi-dimensional), and the signal The analyzing step is characterized in that the shape of the metal object is specified by comparing the signal waveform of the detection signal with a value measured in advance and a value indicating the shape of the metal.

The metal object shape determination method according to the second aspect of the present invention is the method according to the first aspect, wherein the (unequal) magnetic field is formed by the magnets constituting the magnet booster with the same magnetic pole directed toward the sensor coil (long direction). The magnetic field is formed by being fixed to the sensor coil.
The metal object shape determining method according to the third aspect of the present invention is a rod-shaped metal according to the first or second aspect, wherein there are three or more peak values in the waveform of the detection signal in the signal analyzing step. When there are only two peak values in the waveform of the detection signal, it is determined that it is a metal powder.

  According to a fourth aspect of the present invention, there is provided the metal object shape determining method according to the first or second aspect, wherein the detection signal from the timing when the detected object is input to the sensor coil to the timing when the detected object passes through the sensor coil. Is normalized and converted into a standardized signal, and the standardized signal is divided into five regions from the first region to the fifth region from the beginning of the time axis, and the standardized signal in the first region (A ′) is predetermined. If there is a waveform that is higher than the intensity, it is determined that the object to be detected includes a one-dimensional metal object, the first area (A ′) does not have a waveform that exceeds a predetermined intensity, and the third area (B ) And the fourth region (C) have a waveform with a predetermined intensity or higher, and when there is a waveform with a predetermined intensity or higher in one of the second region (A) and the fifth region (D), It is determined that the object to be detected contains a one-dimensional metal object, and the first The region (A ′), the second region (A), and the fifth region (D) have no waveform with a predetermined intensity or higher, and the third region (B) and the fourth region (C) have a predetermined strength. When there is the above waveform, it is determined that the object to be detected includes a two-dimensional metal object.

  According to a fifth aspect of the present invention, there is provided the metal object shape determining method according to the first or second aspect, wherein the first sensor coil is disposed on a surface parallel to the transport path, and is parallel to the first sensor coil. The second sensor coil is disposed on a surface parallel to the transport path across the transport path, and a third surface is parallel to the transport path and orthogonal to the first and second sensor coils. The sensor coil is arranged, and the detection signal output step outputs the three detection signals of the detection signals of the first to third sensor coils, respectively, When the three-dimensional shape of the metal object included in the detected object is specified by combining the three determinations that are analyzed and output in the analysis step, and the three-dimensional shape is specified, Or third, and When the detection signals detected by the two sensor coils are determined to be one-dimensional, the metal object has a one-dimensional shape and is detected by the first and / or third and second sensor coils. When each of the detection signals is determined to be two-dimensional, the metal object has a three-dimensional shape, the detection signal detected by the first or third sensor coil is determined to be one-dimensional, and When the detection signal detected by the two sensor coils is determined to be two-dimensional, the metal object has a two-dimensional shape, and the detection signal detected by the first or third sensor coil is two-dimensional. When the determination and the detection signal detected by the second sensor coil are determined to be one-dimensional, the metal object has a two-dimensional shape.

According to the metal object shape determination method of the present invention, the metal object shape can be determined and detected.
According to the metal object shape determination method of the present invention, a metal object attached to a pharmaceutical product, cosmetics, food, etc. In addition, it is now possible to distinguish and detect metal foreign objects that are accidentally mixed by shape discrimination.

FIG. 1 illustrates an outline of a metal detector 1 according to an embodiment of the present invention. FIG. 2 is a diagram showing a configuration example of the sensor coil, FIG. 2A is a front view of the sensor coil 5, FIG. 2B is a plan view of FIG. 2A, and FIG. FIG. 3 is a cross-sectional view taken along line AA in FIG. FIG. 3 is a diagram illustrating an arrangement example of the magnet booster 8. FIG. 4 is a diagram illustrating an arrangement example of the magnet booster 8. FIG. 5 is a block diagram illustrating a configuration example of the control unit 3 of the metal detector 1. FIG. 6 is a conceptual diagram illustrating a waveform example of a detection signal. FIG. 7 is a diagram for explaining a state in which an object to be detected is detected by the metal detector 1 according to the embodiment of the present invention. FIG. 8 is a diagram showing a pattern indicating the shape of a metal object. FIG. 9 is a flowchart illustrating an operation example when the shape of the metal object is determined by the control unit 3 of the metal detector 1. FIG. 10 is a diagram illustrating a result of analyzing a magnetic field when a rod-shaped metal object passes in the vicinity of the sensor coil by FEMM analysis. FIG. 11 is a diagram illustrating a result of analyzing a magnetic field when a rectangular metal object passes in the vicinity of the sensor coil by FEMM analysis. FIG. 12 is a diagram illustrating a result of analyzing a magnetic field when a circular metal object passes in the vicinity of the sensor coil by the FEMM analysis. FIG. 13 is a diagram illustrating a state where a bar-like metal object such as a needle is stabbed into the detected object 2 in the first embodiment.

  FIG. 1 illustrates an outline of a metal detector 1 according to an embodiment of the present invention. The metal detector 1 detects or detects a metal object in the detection object 2 and notifies it. The detected object 2 is a product such as food or medicine. These objects to be detected 2 are generally packaged with a packaging material or the like or contained in a container. The detected object 2 is an aluminum packaging bag that is a container for wrapping food, and food may be packaged. Further, the metal detector 1 detects metal foreign matters mixed in these detection objects 2.

The detected object 2 can have a metal structure such as an oxygen scavenger made of iron powder, a metal lid, and a metal name tag. The metal detector 1 detects the metal structure and the metal foreign object by dividing them into shapes. Hereinafter, the determination rule for determining the shape of the metal object from the detection principle will be described in detail.
[Outline of metal detector]
The metal detector 1 includes a slide 4, a control unit 3, and sensor coils 5 and 6, which are transport means for transporting the object 2 to be detected.

  The slide 4 slides the detection object 2 on a plurality of rollers (not shown). The conveying means is not limited to the slide 4. The conveying means may be a belt conveyor that rotationally drives an endless belt by a driving motor provided on one side, and carries the object 2 to be detected on the belt. The transport means may transport the detected object 2 by a human hand. In FIG. 1, the conveyance direction for conveying the detection object 2 is indicated by an arrow. The control unit 3 and the sensor coils 5 and 6 are disposed around the center of the slide 4 in the length direction in the transport path in which the detection object 2 is transported.

  The sensor coils 5 and 6 are for detecting a change in magnetic field disturbance generated from a magnetic metal such as a metal object or a metal foreign object, and output a voltage or current detection signal corresponding to the change in the magnetic field. The control unit 3 performs signal processing on the detection signals received from the sensor coils 5 and 6 and determines whether or not there is a metal foreign object in the detected object 2. In this embodiment, the sensor coil 5 and the sensor coil 6 have the same structure, and a detailed description thereof will be described later.

  The control unit 3 receives the detection signal from the sensor coil 5 and / or the sensor coil 6, processes the signal, determines whether or not there is a metal foreign object in the detected object 2, and to that effect (Warning sound, display, etc.). The control unit 3 can be installed at any location as long as it receives a signal from the sensor coil 5 and / or the sensor coil 6. The metal detector 1 includes an optical sensor 7 for detecting the presence or absence of the object 2 to be detected. The optical sensor 7 is installed on the front side of the sensor coil 5 and the sensor coil 6.

  The optical sensor 7 detects the timing when the detection object 2 is conveyed and enters the sensor coil 5 and the sensor coil 6, and transmits this detection to the control unit 3. In other words, when viewed in the transport direction of the detected object 2, the detected object 2 passes through the optical sensor 7 and passes through the space between the sensor coil 5 and the sensor coil 6. The optical sensor 7 can detect the timing when the detection object 2 enters between the sensor coil 5 and the sensor coil 6 and can transmit the detected signal to the control unit 3. The structure can be of any operation scheme.

  The control unit 3 receives the signal from the optical sensor 7 and can calculate the timing when the detected object 2 passes through the sensor coil 5 and the sensor coil 6, and uses it to detect from the sensor coils 5 and 6. The signal is analyzed, and it is determined whether or not the detected object 2 contains a metal foreign object. When the metal object is contained in the detected object 2, the control unit 3 outputs an output signal to that effect. This output signal is sent to another device such as an electric arm (not shown) provided in or connected to the metal detector 1 to remove the detected object 2 including the metal foreign matter. Processing is performed.

  The control unit 3 needs to be able to grasp the conveyance speed of the slide 4 being driven. The grasping and calculation of the conveyance speed is a well-known technique and is not the gist of the present invention, and therefore its explanation is omitted. A magnet booster 8 corresponding to the length of the sensor coils 5 and 6 is attached to the sensor coil 5 and the sensor coil 6 in order to magnetize and reinforce the magnetic metal contained in the detected object 2. The magnet booster 8 will be described later.

[Sensor coil]
The sensor coil 5 and the sensor coil 6 are arranged at a predetermined distance from each other. The sensor coil 5 and the sensor coil 6 are arranged on two parallel surfaces. The detected object 2 passes through the space between the sensor coil 5 and the sensor coil 6. In the example of FIG. 1, the sensor coil 5 and the sensor coil 6 are disposed vertically with the slide 4 interposed therebetween. The sensor coil 5 and the sensor coil 6 are arranged on a plane parallel to the slide 4. The sensor coil 5 and the sensor coil 6 have an elongated shape and are arranged so as to be parallel to each other.

  Since the sensor coil 5 and the sensor coil 6 have the same structure in this example, the structure will be described by taking the sensor coil 5 as an example. FIG. 2 is a diagram illustrating a structure example of the sensor coil 5. 2 (a) is a front view of the sensor coil 5, FIG. 2 (b) is a plan view of FIG. 2 (a), and FIG. 2 (c) is a cross-sectional view taken along line AA of FIG. 2 (a). is there. The sensor coil 5 has an elongated bar shape as shown in FIG. The sensor coil 5 has a configuration in which a coil 11 is wound along a groove portion of an iron core 10 of a conductive material having an elongated rod shape and an E-shaped cross-sectional structure.

  The iron core 10 is formed by laminating silicon steel plates or amorphous plates. Alternatively, a ferrite core or a permanent magnet may be used. A minute voltage is applied to the sensor coil 5. The design is such that almost no current flows. Of the magnetization (BH) characteristics of the sensor coil 5, a metal part is detected using a non-linear portion where the magnetic flux density (B) representing the magnetization characteristics and the magnetic field (H) are very small values near zero. . The technical idea about this magnetization characteristic and its utilization method is described in detail in Patent Document 1. All or part of the invention of Patent Document 1 constitutes the present invention.

  However, the voltages, currents, and frequencies applied to the sensor coils 5 and 6 may be all or partly different from those in Patent Document 1 in the embodiment of the present invention. The sensor coils 5 and 6 perform a detection operation even when no current is supplied and no voltage is applied. For example, the bias voltage of the amplifier is applied to the sensor coils 5 and 6 in advance. As a result, the input bias current of the amplifier flows through the coil 11 of the sensor coils 5 and 6. In addition, when the magnetic flux across the sensor coils 5 and 6 changes with the change of the peripheral magnetic field, the current and / or voltage flowing through the sensor coils 5 and 6 change to become an output signal.

[Magnet booster]
The movement of the magnet booster 8 directly attached to the sensor coils 5 and 6 is to magnetize the magnetic metal object, and immediately thereafter, the magnetic metal object can be sensed by the sensor coils 5 and 6. The magnet booster 8 of the metal detector 1 according to the embodiment of the present invention is for magnetizing and strengthening a magnetic metal. The magnet booster 8 is not necessary when the metal object or metal foreign matter contained in the detection object 2 is magnetized with sufficient strength. The sensor coils 5 and 6 of the present invention are mainly used for detecting minute metals of 1 mm or less.

  Therefore, the magnet booster 8 is useful for strengthening the magnetization of a metal object or metal foreign substance contained in the detection object 2 as much as possible to increase the detection sensitivity. FIGS. 3 and 4 show examples of the magnet booster 8. In the present embodiment, the magnet booster 8 is made of a permanent magnet. As shown in FIG. 3, the magnet booster 8 is composed of a plurality of magnet elements 8 a, and the magnet elements 8 a are arranged so as to face the same magnetic pole toward the sensor coil 5 and the sensor coil 6. In the example of FIG. 3, the magnet element 8 a is arranged with the south pole facing the sensor coil 5. When the magnet booster 8 is used for the metal detector 1, the metal detector 1 can detect the metal object while magnetizing or strengthening the magnetization of the two magnetic bodies in the object to be detected.

  The sensor coils 5 and 6 detect a change in the magnetic field near the sensor coils 5 and 6 and output a detection signal. The magnetic field in the vicinity of the sensor coils 5 and 6 is due to the static magnetic field formed by the magnet booster 8 and the magnetic field formed by the sensor coils 5 and 6. Since the magnet booster 8 is fixed to the sensor coils 5 and 6, a symmetric magnetic field is not formed, but an unequal magnetic field is formed. The detected object 2 passes through the unequal magnetic field. This is illustrated in FIG. 10, FIG. 11, and FIG. 12, which will be described later.

  As is apparent from FIGS. 10, 11, and 12, the magnetic field is strong and unequal near the magnet booster 8. The formation of the unequal lines of magnetic force is due to the interaction between the magnetic field of the magnet booster 8 and the sensor coils 5 and 6. Magnetic field lines generated from the magnet booster 8 link the sensor coils 5 and 6. When the metal object crosses the magnetic field, the magnetic field is disturbed, and the magnetic flux linked to the sensor coils 5 and 6 changes. The ratio of the temporal change of the magnetic flux interlinking the sensor coils 5 and 6 becomes the detection voltage or detection current of the sensor coils 5 and 6 and becomes a detection signal.

  As shown in FIG. 4, the magnet booster 8 is arranged so that the south pole faces the sensor coil 5 and the sensor coil 6. The magnet booster 8 is disposed so as to be parallel to the longitudinal axes of the sensor coil 5 and the sensor coil 6. When a small metal piece approaches the vicinity of the sensor coil 5, for example, austenitic stainless steel (SUS304, etc.) causes martensite-induced transformation due to plastic deformation such as fracture, crushing, bending, and chipping, and has weak magnetism. The part is magnetized by the action of the static magnetic field of the magnet booster 8 to generate a magnetic pole.

  The change in the magnetic pole affects the magnetic field of the sensor coil 5, and the sensor coil 5 outputs an output corresponding to this influence as a detection signal. This detection signal is signal-analyzed by a subsequent circuit to determine whether or not it is a metal foreign object. The length of the magnet booster 8 is ideally the same length as the sensor coil 5 or the sensor coil 6. When a metal object is magnetized by the action of the static magnetic field of the magnet booster 8 to generate a magnetic pole, the detection sensitivity is remarkably improved as compared with the prior art. In particular, for a ferromagnetic material having a large magnetization enhancement effect, the detection sensitivity is greatly improved by the influence of the magnet booster 8.

  Examples of the ferromagnetic material include iron, nickel, cobalt contaminants, alloys thereof, martensitic stainless steel, and the like. A magnetic booster 8 provided on the side surfaces of the sensor coils 5 and 6 forms a spatially unequal magnetic field. That is, the magnetic field shape is three-dimensionally unequal. When metal objects having different shapes pass through the magnetic field, the magnetic field changes temporally and spatially depending on the conveyance speed and shape, and therefore the flux linkage of the sensor coils 5 and 6 depends on the shape of the metal object. Change. In response to the change of the interlinkage magnetic flux, finally, an output signal output from the sensor coils 5 and 6 is uniquely generated for each shape of the metal object.

Therefore, it can be said that each sensor coil 5 and 6 outputs a different signal for each metal shape.
[Block diagram of metal detector 1]
FIG. 5 is a block diagram showing an outline of signal processing in the metal detector 1. Detection signals are output from the sensor coils 5 and 6. Since this detection signal is a weak signal, it is amplified by the amplifier 20 and a high frequency component is cut from the amplified detection signal by the LPF 21. Further, the signal output from the LPF 21 is amplified by the amplifier 22, and the component due to the DC power source is cut from the amplified signal by the filter 23.

  The signal output from the filter 23 is waveform-shaped by the waveform shaper 24 and input to the digital computer processing unit 25. The digital computer processing unit 25 analyzes the signal, determines whether or not it is a metallic foreign object, and outputs the result of this determination as a signal. The waveform shaper 24 shapes the waveform of the input signal, converts it into a digital signal, adjusts the reception sensitivity, and outputs the digital signal to the digital computer processing unit 25. The waveform shaper 24 includes an amplitude comparison / signal extraction circuit, a conveyance speed timing circuit, and the like.

  Further, the digital computer processing unit 25 receives a signal of the inspection object 2 from the optical sensor 7 and obtains information that the detection object 2 passes through the sensor units 5 and 6. Therefore, the digital computer processing unit 25 receives the digital signal from the waveform shaper 24, and combines the metal contained in the detected object 2 with the transport speed of the slide 4, the passing signal of the detected object 2, and the like. Determine foreign matter. When the output signals of the sensor coils 5 and 6 have a strength necessary for signal processing of the subsequent circuit and have a sufficient strength, the amplifier 20 is not inevitable. When the signal output from the LPF 21 has a strength necessary for signal processing of a circuit at a subsequent stage and has a sufficient strength, the amplifier 22 is not inevitable.

[Determination rules based on foreign object shape]
FIG. 6 is a diagram illustrating a waveform example of a detection signal in which the detection object 2 is detected by the sensor coil 5 or the sensor coil 6. FIG. 6 is a diagram illustrating a waveform example of a detection signal obtained by detecting the detected object 2 with the single sensor coil 5 or the sensor coil 6. The detection signal consists of a plurality of waveforms. The horizontal axes of FIGS. 6A and 6B are time axes indicating the time when the detected object 2 passes near the sensor coils 5 and 6. The vertical axis in FIG. 6A and FIG. 6B indicates the voltage amplitude.

  Each symbol a ', a, b, c, and d indicates the maximum intensity of each waveform of the detection signal. The sensor coils 5 and 6 output detection signals indicating the waveforms of a ′, a, b, c, and d during the period from when the detected object 2 approaches the vicinity of the sensor coils 5 and 6 until it passes. . The detection signals output from the sensor coils 5 and 6 are analog signal waveforms, and generally include noise signals. For this purpose, the detection signal is A / D converted to convert the analog detection signal into a digital signal (discrete value).

  A plurality of sampling discrete values subjected to A / D conversion are stored in a memory, and a moving average of a plurality of sampling values before and after the sampling value is taken and smoothed. In this way, the detection signal becomes smooth and signal processing becomes easy. In other words, when the waveform detected by the sensor coils 5 and 6 is normalized between the time when the detected object 2 approaches the vicinity of the sensor coils 5 and 6 and passes, the metal objects included in the detected object 2 Depending on the shape, all or part of the waveforms a ′, a, b, c, and d are output from the sensor coils 5 and 6.

  FIG. 6A shows detection signals detected by the sensor coils 5 and 6. FIG. 6B shows the detection signal converted into its absolute value intensity by signal processing. The waveforms at this time are A ', A, B, C, and D corresponding to a', a, b, c, and d. Waveforms A ′ to D are absolute value intensities of the waveforms a ′ to d. When the metal object passes through the space between the sensor coils 5 and 6 and is detected by the sensor coil 5 and the sensor coil 6, the position from the metal object to the sensor coil 5 and the position from the metal object to the sensor coil 6 are Different.

  Therefore, the sensor coil 5 and the sensor coil 6 output detection signals having different intensity waveforms. If the metal object is completely in the middle between the sensor coils 5 and 6, the sensor coil 5 and the sensor coil 6 detect signals of the same magnitude. When a metal object passes between the sensor coils 5 and 6 and is detected by the sensor coil 5 and the sensor coil 6, the magnitudes of the respective detection signals are large, but the waveforms of both detection signals are correlated, and the waveform The shape is almost the same. In particular, since this effect appears greatly in the vicinity of the sensor coils 5 and 6, it is possible to reliably detect minute debris with the sensor coil 5.

  FIG. 7 is a diagram for explaining a state in which the detection object 2 is detected by the metal detector 1 according to the embodiment of the present invention. In this example, the detected object 2 has a three-dimensional structure, and each surface thereof is defined as follows. First, let the surface of the detected object 2 facing the sensor coil 5 installed on the upper side of the slide 4 be the A surface. The surface of the detected object 2 facing the sensor coil 6 installed on the lower side of the slide 4 is defined as a C surface. In this drawing, the surface of the detected object 2 facing the front side of the paper surface is a B surface. In this drawing, the surface of the detected object 2 facing the back of the paper surface is defined as B 'surface.

  Surfaces perpendicular to the conveyance direction of the detection object 2 are a D surface and a D ′ surface. Of the D plane and D ′ plane, the plane that passes through the first sensor coil 5 is the D plane. FIG. 8 is a table showing detection waveform patterns of metal objects detected by the metal detector 1. In the case where the metal object has an elongated shape such as a needle and a volume such as iron powder, the detected waveform is shown. The first column of this table indicates the type of metal object. The second column of this table indicates the position where the metal object of the first column is arranged in the detected object 2. For example, the needle 1 is disposed at 90 degrees with respect to the C plane.

  The third column and the fourth column of this table show the first determination rule and the second determination rule, which are conditions for specifying the metal object. The fifth column of this table shows the shape of the detected waveform. When the metal object is the needle 1, the needle 1 is disposed at 90 degrees on the C surface of the object 2 to be detected, the first determination rule is h1 ≠ 0, and the second determination rule is h2 ≠ 1. Waveforms A, B, and C are detected. However, the waveform D is not detected.

This h1 is calculated | required by the maximum intensity | strength of waveform A, B, C, and D, and becomes following Formula.
[First determination rule] h1 = (A + D) / (B + C)
Similarly, h2 of the second determination rule is obtained from the maximum intensity of the waveforms B and C, and is expressed by the following equation.
[Second determination rule] h2 = B / C
As apparent from this table, like the needles 1 to 4, the rod-shaped metal object has the waveforms B and C and the waveform A or the waveform D. When the metal object is iron powder, only waveforms B and C are detected, and waveform A and waveform D are not detected.

Although not shown in the fifth column of the table of FIG. 8, there is a third determination rule. The third determination rule determines whether h3 is greater than zero. It is determined by the waveform A ′. A rod-shaped metal object always has waveforms a ′ and A ′.
[Third determination rule] h3 = A ′
Further, when attention is paid to the waveform D, the waveform D becomes a rising signal in the case of a rod-shaped metal object.

[Determination flowchart]
FIG. 9 is a flowchart showing an operation example in which the metal detector 1 determines the shape of the metal object. The control unit 3 of the metal detector 1 starts determination (step 1). First, the control unit 3 makes a determination based on the third determination rule (step 2). In the third determination rule, it is determined whether h3 is greater than 0 (step 2). When h3 is larger than 0, it is determined as a one-dimensional material (step 9). This third determination rule is based on the waveform a ′. In other words, the one-dimensional material is determined based on whether or not the waveform A ′ is present.

  If the third determination rule is not satisfied, it is confirmed whether or not the first determination rule and the second determination rule are satisfied (steps 3 to 5). The first determination rule confirms whether or not the value of h1 is 0 (step 3). In the second determination rule, it is confirmed whether or not the value of h2 is 1 (steps 4 and 5). When the determinations of both the first determination rule and the second determination rule are “Yes” and both the first determination rule and the second determination rule are satisfied, it is determined that the material is a two-dimensional material (Step 3 → 5). → 8).

  When the determinations of both the first determination rule and the second determination rule are “No” and both the first determination rule and the second determination rule are not satisfied, it is determined that the material is a one-dimensional material (Step 3 → 4). → 6). When one of the first determination rule and the second determination rule is satisfied, it is determined that the determination is impossible (step 3 → 4 or 5 → 7).

  A ', A, B, C, and D rarely become zero due to the characteristics on the circuit. For this reason, basically, when the value is near 0 or much smaller than other waveforms, it is regarded as 0 and each calculation is performed. Similarly, when two values of A, B, C, and D are within a predetermined range, the two values are set to the same value. This predetermined range is determined, for example, as 5%, 10%, 20%, etc., and becomes the determination accuracy of the metal detector 1. If this predetermined range is increased, false detection is reduced. A3 of h3 is rarely zero at all because of circuit characteristics.

  Therefore, basically, when the value is near 0 or much smaller than other waveforms, h3 is regarded as 0 and each calculation is performed. Similarly, in the case of h1, it is rare that the values of A and D become 0 at the same time or one of them. Therefore, basically, when the value is near 0 or much smaller than other waveforms, A and D are regarded as 0 and each calculation is performed. Similarly, in the case of h2, it is rare that B and C have the same value. Basically, when B2 and C are almost the same value within a predetermined range, B and C are the same value. Each calculation.

  In the present embodiment, the metal detector 1 uses a digital computer processing unit 25 as means for determining the shape of a metal object. Although not shown, the digital computer processing unit 25 includes at least a central processing unit (CPU) and a memory (RAM, ROM, etc.). The program stored in the memory is executed by the central processing unit, and the shape of the metal object Make a decision. In particular, this program causes the central processing unit to execute each step of the pro chart shown in FIG. Further, the digital computer processing unit 25 can realize the same operation by a circuit means including an analog circuit or a digital circuit in addition to the microcomputer including the central processing unit and the memory. At this time, each step of the flowchart of FIG. 9 is realized as a circuit means.

[FEMM analysis]
Changes in the magnetic field when a rod-shaped metal object passes near the sensor coils 5 and 6 can be analyzed by computer simulation. FIG. 10 is a diagram illustrating an analysis result when a rod-shaped metal object passes in the vicinity of the sensor coils 5 and 6. This analysis was performed using FEMM (Finite Element Method Magnetics) analysis, which is one of magnetic field analysis methods. FEMM analysis is an analysis method for analyzing a magnetic field using a finite element method, and is widely used for analysis of a low-frequency alternating magnetic field and a static magnetic field. In the drawing, magnetic lines showing the same potential are indicated by solid lines.

  First, FIG. 10A is a diagram showing a magnetic field when a rod-shaped metal object approaches the sensor coils 5 and 6. In this simulation, the sensor coils 5 and 6 and the magnet booster 8 illustrated in FIGS. 1, 2, 3, and 4 were assumed. FIG. 10B shows a state in which the magnetic field starts to change as the rod-shaped metal object approaches the sensor coil 6. FIG. 10C shows that a rod-shaped metal object is next to the sensor coil 6 and the magnetic field is greatly disturbed.

  FIG. 10 (d) shows that the rod-shaped metal object has come to a position between the sensor coil 5 and the sensor coil 6. FIG. 10E shows the state of the magnetic field when the rod-shaped metal object passes through the sensor coils 5 and 6. At this time, the disturbance of the magnetic field remains large. From here, the disturbance of the magnetic field is reduced, and as shown in FIG. 10 (f), when the rod-shaped metal object passes the sensor coils 5 and 6, the magnetic field returns to the state shown in FIG. 10 (a). . As can be seen from this analysis, it is understood that the magnetic field generated from the sensor coils 5 and 6 is disturbed when the magnetic metal passes through the vicinity of the sensor coils 5 and 6.

  FIG. 11 is a diagram illustrating a result of analyzing a magnetic field when a rectangular metal object passes in the vicinity of the sensor coils 5 and 6 by FEMM analysis. FIG. 12 is a diagram showing a result of analyzing the magnetic field when a circular metal object passes in the vicinity of the sensor coils 5 and 6 by the FEMM analysis. 11 and 12 show the state of the magnetic field when the metal object passes in the vicinity of the sensor coils 5 and 6, similarly to FIG. FIG. 11A and FIG. 12A are diagrams showing a magnetic field when a metal object approaches the sensor coils 5 and 6.

  FIG. 11B and FIG. 12B show how the magnetic field starts to change as the metal object approaches the sensor coil 6. FIG. 11C and FIG. 12C show that the metal object is next to the sensor coil 6 and the magnetic field disturbance is increased. FIG. 11D and FIG. 12D show that the metal object is at a position between the sensor coil 5 and the sensor coil 6. FIG. 11E and FIG. 12E show the state of the magnetic field when the metal object passes through the sensor coils 5 and 6. FIGS. 11 (f) and 12 (f) show the magnetic field when the metal object passes through the sensor coils 5 and 6. This FEMM analysis illustrated in FIG. 10, FIG. 11, and FIG. 12 was first performed by the inventors of the present invention in the metal object detection industry and in academic societies.

[Example]
Here, an example will be described. In this example, metal objects such as needles and iron powder were measured using a metal foreign object detector. The foreign metal detector used in this example is a part of the sensor of Toku Engineering Co., Ltd. (Tachikawa City, Tokyo, Japan) with a foreign object detector “Observation (registered trademark)” NIP-IIID. It is. That is, the length of the magnet and the location of the magnet of the conventional metal foreign object detector were changed.

  The state after the change is illustrated in FIG. Further, the belt conveyor of the metal foreign object detector was changed to the slide 4. Specifically, the pulley of the belt conveyor of the metal foreign object detector was replaced, the conveying speed of the slide 4 was set to 40 to 150 m / min, and the detected object 2 at that speed was detected to enable data collection. Signal determination is possible at this speed. It was assumed that the speed at which a person can move the detection object 2 by holding the detection object 2 by hand is 80 to 90 m / min.

  No special voltage is applied to the sensor coil of the metal foreign object detector. The bias voltage of the OP amplifier was applied to the sensor coil. The sensor coil has a structure in which a coil is wound around an E-shaped core. The sensor coil is connected to the input of the OP amplifier through a resistor. Accordingly, the input bias current of the OP amplifier flows through the coil. The power supply for supplying current was ± 15 V as the power supply voltage of the OP amplifier.

  As the material to be detected, a sewing needle was used as a one-dimensional material, and AGELESS (registered trademark) was used as a two-dimensional material. This sewing needle is a general needle having a length of 42 mm, a maximum diameter of 0.8 mm, and a material of steel. This ageless is 30.5 mm in length, 20.5 mm in width, 3 mm in thickness, and the content is iron powder. FIG. 13 is a diagram for explaining the object to be detected 2, and is a diagram for explaining the state of sticking a bar-like metal object such as a needle. As shown in FIG. 13, the detection object 2 is a regular cube having a side length of 115 cm.

  As shown in FIGS. 7 and 13, the material to be detected arranged in the object to be detected defined each side of the object 2 to be detected. As shown in FIG. 13, the material to be detected was arranged at an angle with respect to each of the surfaces A to D. The one-dimensional material and the two-dimensional material were measured by changing the inclination with respect to the sensor coil. The interval between the sensor coil 5 and the sensor coil 6 was 155 mm. This interval is based on the slide 4. The conveyance speed of the detection object 2 was 80 m / min and 90 m / min.

  The distance between the upper surface of the sensor coil 6 and the C surface of the lower surface of the detection object 2 was 5 mm. In the test of the two-dimensional material, the measurement of the B surface and the D surface was performed at three heights of 10 mm, 57.5 mm, and 105 mm from the lower sensor coil 6. Furthermore, as an additional measurement, the measurement was also performed at a height of 75.5 mm which is the middle between the upper and lower sensor coils 5 and 6. The detection signal was obtained from the second stage of the signal amplification amplifier. An oscilloscope (manufactured by Tektronix Co., Ltd.) was used for acquiring the detection signal. The detection signal was sampled every 1 msec, for a total of n = 10000 times.

  Then, only the detection signal from the lower sensor coil was evaluated. Data collected with an oscilloscope was shaped according to the following procedure. First, 1024 pieces of data were extracted from the detection signal. Since the above signal frequency was 20 Hz or less as a result of the FFT analysis, this was subjected to Fourier transform, and data having a frequency of 50 Hz or more was cut with a low-pass filter. Further, inverse Fourier transform was performed. The inverse Fourier transformed data was converted to absolute value data.

  Finally, this data was determined according to the first to third determination rules described above to determine the shape of the metal object. A needle or ageless was placed on each surface A to D of the object 2 to be detected and detected by the sensor coil 6. The following Tables 1 and 2 show the determination rates at which the needle or ageless can be accurately determined by the lower sensor alone or the upper sensor alone. This determination rate is a summary of the first to third determination rules. Therefore, when the upper and lower sensor coils 5 and 6 are combined, the determination rate is almost 100%.

  A needle was placed at a predetermined height from each surface C of the object 2 to be detected and detected by the sensor coil 6. Table 3 below shows the determination rate at which the needle could be accurately determined at this time.

[Three-dimensional shape recognition]
A linear object such as a needle is a one-dimensional object to be detected. A planar object such as iron powder is a two-dimensional object to be detected. As described above, the shape detection of the one-dimensional and two-dimensional objects to be detected can be realized by signal processing of the detection signals using the three E-core sensor coils. However, three or more E-core sensor coils are required to detect the shape of a three-dimensional object that is a three-dimensional object.

  A three-dimensional recognition process can be performed by combining the individual signals of the four E-type core sensor coils. A third sensor coil can be used in a plane perpendicular to the sensor coil 5 and the sensor coil 6 and parallel to the transport direction. If it demonstrates in FIG. 7, the sensor coil 5 and the sensor coil 6 are parallel to A surface and C surface, and a 3rd sensor coil is parallel to B surface, and can also be said to be a side sensor coil. Therefore, unlike the upper and lower sensor coils 5 and 6, the waveform of the detection signal detected by the side sensor coil is the same as the signal when the needles are parallel in the upper and lower sensor coils 5 and 6.

  By combining a total of three judgments obtained by analyzing and outputting each of the three detection signals from the sensor coil 5, the sensor coil 6, and the side sensor coil, the three-dimensional of the metal object included in the detected object 2 is obtained. The shape can be specified. For example, if the detection signals detected by three sensor coils all have a one-dimensional shape, the metal object can be determined to have a one-dimensional shape. If the detection signals detected by the three sensor coils all have a two-dimensional shape, the metal object can be determined to have a three-dimensional shape.

  Similarly, when two detection signals of the three sensor coils indicate a two-dimensional shape and the detection signals of the remaining sensor coils indicate a one-dimensional shape, it can be determined that the metal object has a two-dimensional shape. When the detection signals of the two sensor coils that are not arranged in parallel to each other indicate a two-dimensional shape, the metal object can be determined to have a two-dimensional shape. When the detection signals of two sensor coils that are not arranged in parallel to each other indicate a one-dimensional shape, the metal object can be determined to have a one-dimensional shape.

  In this way, using the waveforms from the three sensor coils, it is possible to create an algorithm that analyzes not only the shape of the object but also the position, size, and inclination. Of course, the number of sensor coils can be increased, and for example, four or five can be used to further improve the detection accuracy. However, although it is theoretically possible, it is not possible to recognize the shapes of all metal objects, but to the extent that powder, straight lines, round objects (lumps), and cube corners can be determined. In this example, the difference between the ageless powder aggregate and the needle could be identified.

  The present invention may be used in a field where metal objects are detected and used according to their shape, and in particular, in the field of manufacturing pharmaceuticals, cosmetics, and foodstuffs, it may be used for detection of metallic foreign matter mixed in products. The present invention detects a magnetic metallic substance mixed in an object to be detected 2 wrapped in a conductive packaging material such as aluminum vapor deposition or aluminum foil. The detection object 2 can be used in the field of detecting magnetic metallic foreign matter mixed in frozen foods, food materials such as grains, pharmaceuticals, industrial materials and the like.

DESCRIPTION OF SYMBOLS 1 ... Metal detector 2 ... Detected object 3 ... Control part 4 ... Slide 5,5 ... Sensor coil 7 ... Optical sensor 8 ... Magnet booster

Claims (5)

A transport process for transporting the object to be detected on the transport path;
A magnet booster provided in the middle of the transport path and disposed in the vicinity of the sensor coil, while magnetizing or strengthening the magnetic substance in the detected object, changes the magnetic field in the vicinity of the sensor coil. A metal object detection method comprising: a metal object detection step for detecting a metal object in the object to be detected by detecting a sensor coil and outputting a detection signal;
When the metal object passes in the vicinity of the sensor coil, (I) the metal object crosses the unequal magnetic field formed by the magnet booster and linked to the sensor coil, and (II) the metal object The magnetic field is disturbed by crossing the magnetic field, (III) the magnetic flux linked to the sensor coil is changed, and (IV) the rate of temporal change of the magnetic flux linking the sensor coil is the sensor coil. A detection signal output step of outputting the detection signal as a detection voltage or a detection current, and
A signal analysis step of analyzing the detection signal from the timing at which the detection object is input to the sensor coil to the timing at which the detection object passes through the sensor coil to identify the shape of the metal object, and
The signal waveform of the detection signal depends on the shape of one metal selected from linear (one-dimensional), planar (two-dimensional), spherical (three-dimensional), and massive (multi-dimensional). Change,
In the signal analysis step, the signal waveform of the detection signal is measured in advance and is compared with a value indicating the shape of the metal to identify the shape of the metal object. Shape determination method. In the metal object shape determination method according to claim 1,
The (unequal) magnetic field is a magnetic field formed by fixing the magnets constituting the magnet booster to the sensor coil with the same magnetic pole facing the sensor coil (in the long direction). An object shape determination method. In the metal object shape determination method according to claim 1 or 2,
In the signal analysis step,
When there are three or more peak values in the waveform of the detection signal, it is determined that the metal is a rod-shaped metal,
When there are only two peak values in the waveform of the detection signal, it is determined that the metal powder is a metal powder. In the metal object shape determination method according to claim 1 or 2,
The detection signal from the timing when the detected object is input to the sensor coil to the timing when the detection object passes through the sensor coil is normalized and converted into a normalized signal,
The standardized signal is divided into five regions from the first region to the fifth region from the beginning of the time axis,
If the normalized signal in the first region (A ′) has a waveform with a predetermined intensity or more, it is determined that the detected object contains a one-dimensional metal object,
The first region (A ′) has no waveform of a predetermined intensity or higher, the third region (B) and the fourth region (C) have a waveform of a predetermined intensity or higher, and the second region (A). And when there is a waveform having a predetermined intensity or more in one region of the fifth region (D), it is determined that the detected object contains a one-dimensional metal object,
The first region (A ′), the second region (A), and the fifth region (D) have no waveform of a predetermined intensity or more, and the third region (B) and the fourth region (C). If there is a waveform having a predetermined intensity or more, the object to be detected is determined to contain a two-dimensional metal object. In the metal object shape determination method according to claim 1 or 2,
Placing the first sensor coil on a surface parallel to the transport path;
The second sensor coil is arranged on a surface parallel to the first sensor coil and across the transport path and parallel to the transport path,
Arranging the third sensor coil on a surface parallel to the transport path and orthogonal to the first and second sensor coils;
The detection signal output step outputs the three detection signals of the detection signals of the first to third sensor coils,
By combining a total of the three determinations output by analyzing and outputting each of the three detection signals, the three-dimensional shape of the metal object included in the detected object is specified,
When specifying the three-dimensional shape,
When the detection signals detected by the first and / or third and second sensor coils are each determined to be one-dimensional, the metal object has a one-dimensional shape,
When the detection signals detected by the first and / or third and second sensor coils are each determined to be two-dimensional, the metal object has a three-dimensional shape,
When the detection signal detected by the first or third sensor coil is determined to be one-dimensional and the detection signal detected by the second sensor coil is determined to be two-dimensional, the metal object is Has a two-dimensional shape,
When the detection signal detected by the first or third sensor coil is determined to be two-dimensional and the detection signal detected by the second sensor coil is determined to be one-dimensional, the metal object is 2 A method for determining the shape of a metal object, characterized by having a dimensional shape.