JP5442532B2 - Method and apparatus for measuring misalignment of laminated iron core and computer program - Google Patents

Description

  The present invention relates to a technical field of a measuring method, a measuring apparatus, and a computer program for measuring a stacking deviation on a laminated surface of laminated cores such as a motor core and a transformer core, which are formed by laminating a plurality of plate-like iron cores.

  This type of measuring method and measuring apparatus is intended for measurement, for example, a laminated iron core in which a plurality of plate-like iron cores formed by punching a strip-like thin plate material using a predetermined die is laminated. Such a laminated iron core is used as, for example, a motor core or a transformer core. Since the shape and dimensions of the laminated iron core used for such applications are important factors for determining the performance of the motor and the transformer, it is required that the shape and dimensions be accurately formed. Therefore, in the manufacturing process of the laminated core, evaluation measurement is performed on the shape and dimensions of the laminated core in order to determine whether or not the manufactured laminated core satisfies such requirements.

  For example, in Patent Document 1, for each of a plurality of iron cores constituting a laminated iron core, by measuring the surface shape of the cut surface over the entire circumference in the circumferential direction, noise caused by a fractured portion formed at the time of punching is detected. Techniques for forming data that is not included are disclosed. Patent Document 2 discloses a technique for measuring the stacking misalignment of the laminated iron core by transferring the surface shape of the laminated surface of the laminated iron core to silicon rubber and measuring the irregularities transferred onto the silicon rubber. Yes.

JP 2006-194838 A JP 2003-65751 A

  One of the important measurement items in the evaluation of the laminated core is stacking deviation (that is, the degree of variation in the degree of overhang of each core when viewed from the lamination direction). As a method for measuring the stacking deviation, a method of acquiring and analyzing unevenness data (hereinafter referred to as “profile” as appropriate) in the surface state of the laminated surface by scanning the laminated surface with a sensor probe is considered. It is done. In this method, it is possible to evaluate the stacking deviation by detecting the degree of overhang of the end of each iron core and calculating the difference between them, but the profile obtained by the sensor probe has a concave shape existing between the laminated cores. A lot of information other than the overhang of the end, such as a gap, is also included. For this reason, it is difficult to determine the information required for evaluation from a profile that contains a lot of such unnecessary information, so the analysis of the profile has to rely exclusively on human work. There is a problem. When the evaluation is performed based on human work, there are various demerits that the analysis results increase in dispersion due to factors such as human error and that the analysis requires a lot of time, labor and cost.

  In addition, in the above-mentioned patent document 1, only the evaluation method about the shape and dimension of each iron core which comprises a laminated iron core is disclosed, and it cannot apply about the stacking gap regarding the lamination surface of a laminated iron core. Moreover, in patent document 2, although it can be evaluated indirectly from the uneven | corrugated shape of the silicon rubber to which the lamination | stacking surface of the laminated iron core was transcribe | transferred, even in the said method, it acquired from the unevenness | corrugation of the silicon rubber surface after all. It is necessary to analyze the profile, and the above technical problems at the time of analysis are not solved at all. In Patent Document 2, since the evaluation is performed indirectly by transferring the laminated surface, there is a possibility that a new error may occur at the time of the transfer, and the evaluation accuracy of stacking deviation may be deteriorated. There are some problems.

  The present invention has been made in view of the above-mentioned problems, for example, and provides a stacking misalignment measuring method and a measuring apparatus for a stacking core capable of quickly and accurately evaluating the stacking misalignment on the stacking surface of the stacked core. And

  In order to solve the above problems, a method for measuring a stacking deviation of a laminated core of the present invention is a method for measuring a stacking misalignment of a laminated core in which a plurality of plate-like iron cores are stacked, and is along the stacking direction of the cores. The profile acquisition step of acquiring the profile of the laminated surface of the laminated core, and the first region in which the absolute value of the inclination exceeds a predetermined threshold among the acquired profiles is a valley between the adjacent cores. A valley identification step that identifies the profile, a profile division step that divides the profile into sub-profiles corresponding to each of the iron cores with the first region as a boundary, and a maximum value of the waveform of the sub-profile for each sub-profile. And the difference between the maximum and minimum values calculated for each sub-profile and the maximum value calculation step for each sub-profile. Characterized in that it comprises a loading shift calculating step calculates Te.

  According to the present invention, the profile related to the laminated surface of the laminated iron core is divided into sub-profiles corresponding to each of the laminated iron cores, and for each sub-profile, the maximum value is obtained from the waveform corresponding to the uneven shape on the surface of the iron core. By calculating, it is possible to extract information necessary for calculating the stacking deviation, that is, the vertex positions of each of the laminated cores. As a result, unnecessary information included in the profile, for example, information on concave gaps existing between the laminated iron cores can be accurately determined, and the stacking deviation can be easily calculated without relying on human work. It becomes possible.

  In one aspect of the method for measuring the stacking deviation of the laminated core according to the present invention, a regression line calculating step for calculating a regression line based on the maximum value calculated for each sub-profile, and the slope of the calculated regression line include: A correction step of correcting the profile so as to be zero, and a load shift recalculation step of re-executing the profile dividing step, the maximum value calculation step, and the load shift calculation step based on the corrected profile It is characterized by providing.

  According to this aspect, when the stacking misalignment is measured for the laminated cores installed in an inclined state, the tilt is included in the stacking misalignment, and thus it may be difficult to obtain an accurate measurement result. Even in such a case, in this embodiment, the measurement accuracy can be effectively improved by correcting the measurement result using the regression line.

  In another aspect of the method for measuring a misalignment of a laminated core according to the present invention, the valley identification step includes a second region having an absolute value of an inclination equal to or less than the predetermined threshold in the acquired profile at the apex of the core. A noise removing step that specifies that the second region is a valley when the first region occupies more than a predetermined ratio within a predetermined range including the second region in the profile. It is further provided with the feature.

  As described above, in the present invention, the valley of the laminated core is specified based on the absolute value of the profile inclination. For this reason, the absolute value of the slope approaches zero near the bottom of the valley, and there is a risk that inaccuracy in specifying whether or not it is a valley due to the influence of noise may increase. In this aspect, such inaccuracies can be effectively improved, and the vertices can be identified with higher accuracy for a plurality of stacked iron cores.

  In another aspect of the method for measuring misalignment of a laminated core according to the present invention, the profile obtaining step obtains the profile along a plurality of the laminating directions.

  According to this aspect, by acquiring a profile along the laminated surface, the stacking deviation is measured two-dimensionally when the laminated surface is a plane, and three-dimensionally when the laminated surface is a curved surface. Can be very practical.

  In order to solve the above-mentioned problem, the stacking misalignment measuring device for a laminated core according to the present invention is a device for measuring the stacking misalignment of a laminated core in which a plurality of plate-like cores are stacked, and the stacking direction of the cores. A profile acquisition means for acquiring a profile of the laminated surface of the laminated core, and a first region in which the absolute value of the inclination exceeds a predetermined threshold among the acquired profiles, a valley between adjacent iron cores A valley specifying means for specifying the profile, a profile dividing means for dividing the profile into sub-profiles corresponding to each of the iron cores with the first region as a boundary, and a maximum value of a waveform of the sub-profile. The maximum value calculation means for calculating each profile and the difference between the maximum value and the minimum value among the maximum values calculated for each sub-profile are accumulated. Characterized in that it comprises a loading shift calculating means for calculating as a record.

  The laminated core stacking deviation measuring apparatus according to the present invention can be suitably realized by the above-described method for manufacturing an electro-optical device (including various aspects).

  Incidentally, the computer program of the present invention can also adopt various aspects in response to the various aspects of the laminated core stacking deviation measuring apparatus of the present invention described above. Specifically, the computer program product may be configured by a computer-readable code (or computer-readable instruction) that functions as the above-described laminated core stacking deviation measuring apparatus of the present invention.

  According to the present invention, the profile related to the laminated surface of the laminated iron core is divided into sub-profiles corresponding to each of the laminated iron cores, and for each sub-profile, the maximum value is obtained from the waveform corresponding to the uneven shape on the surface of the iron core. By calculating, it is possible to extract information necessary for calculating the stacking deviation, that is, the vertex positions of each of the laminated cores. As a result, unnecessary information included in the profile, for example, information on concave gaps existing between the laminated iron cores can be accurately determined, and the stacking deviation can be easily calculated without relying on human work. It becomes possible.

It is a schematic diagram which shows the typical structure of the laminated iron core which is a measuring object. It is a block diagram which shows the whole structure of the stacking deviation measuring apparatus which concerns on a present Example. It is a graph which shows an example of the profile acquired by scanning the laminated surface of a laminated iron core with a scanning probe. It is a flowchart figure of the valley specific process which concerns on a present Example. It is a conceptual diagram which shows typically the process of calculating inclination Q (i) from the waveform of the profile P. FIG. It is a flowchart figure of the noise correction process which concerns on a present Example. It is a conceptual diagram which shows typically the correction process performed in the noise correction process which concerns on a present Example. It is a flowchart figure of the profile division | segmentation process which concerns on a present Example. It is a flowchart figure of the misalignment calculation process which concerns on a present Example. It is a flowchart figure of the correction process which concerns on a present Example. It is a conceptual diagram which shows typically the process performed with respect to a profile in a correction process. It is a flowchart figure of the valley determination process which concerns on this modification.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, but are merely illustrative examples. Not too much.

  First, with reference to FIG. 1, the structure of the laminated core which is the measuring object of the stacked misalignment measuring apparatus and the stack misalignment measuring apparatus capable of executing the method of the present invention will be described. FIG. 1 is a schematic diagram showing a typical structure of a laminated iron core to be measured.

  In the laminated core 1 according to this embodiment, a plurality of disk-shaped iron cores 1a shown in FIG. 1A are laminated along a predetermined direction (hereinafter referred to as X direction) as shown in FIG. 1B. It has a crimped structure. As shown in FIG. 1A, each iron core 1a includes a first opening 11 that is concentrically opened in the center in a plan view, and a first opening 11 that is A plurality of second openings 12 opened in a rectangular shape along the outer periphery are provided. The “lamination surface” according to the present invention means a surface formed by the outer periphery of the main body 10, the end of the iron core 1 a defining the first opening 11 and the second opening 12 being continuous in the lamination direction. To do. It is needless to say that the laminated core 1 shown in FIG. 1 is merely an example of a typical shape and is not limited to this.

  The laminated iron core 1 is manufactured, for example, by laminating a plurality of iron cores 1a formed by punching a strip-shaped thin plate material made of an electromagnetic steel plate using a predetermined die and caulking and bonding them. Specifically, in the press mold, it may be crimped via a protrusion formed on the adjacent iron core and a recess into which the protrusion is press-fitted, or stacked in the press mold. An adhesive that melts by heating and pressing may be used between adjacent iron cores.

  Then, with reference to FIG. 2, the whole structure of the stacking deviation measuring apparatus which can perform the stacking deviation measuring method of the laminated core which concerns on this invention is demonstrated. FIG. 2 is a block diagram illustrating the overall configuration of the misalignment measuring apparatus according to the present embodiment.

  The misalignment measuring device 100 according to the present invention (hereinafter simply referred to as “measuring device 100”) includes a main body 200 including a control unit 210 and a storage unit 220, and a measuring unit 300 connected to the measuring device main body 200. Consists of. The control unit 210 includes an arithmetic processing unit such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit) and a buffer memory such as a RAM (Random Access Memory), and can control the operation of the measurement unit 300. It is a configured control unit. The control unit 210 executes an application program (hereinafter referred to as “application” as appropriate) stored in the storage unit 220, so that the measuring apparatus 100 is an example of the “stacked core stacking displacement measuring apparatus” according to the present invention. Configured to work. The application is an example of the “computer program” according to the present invention.

  An input device 400 and a display device 500 are connected to the main body 200. The input device 400 includes a keyboard, an input pen, and a pointing device (not shown) such as a mouse as appropriate, and is configured to allow an appropriate input operation by the user. The display device 500 is various display devices such as a plasma display device and a liquid crystal display device, and is configured to be able to display a screen related to an application executed by the measurement device 100.

The measurement unit 300 is provided so as to extend from the support unit 330 provided on a pedestal (not shown), a holding unit 330 attached to be movable in the vertical direction along the support column, and the holding unit 330. Scanning probe 340. The scanning probe 340 is further movable along the arrow direction in the figure, and can scan the laminated surface of the laminated core 1 along the X direction. In FIG. 2, in order to clearly show the positional relationship of the scanning probe 340 with respect to the laminated surface, a part of the laminated core 1 is omitted to show a sectional view, and the vicinity of the tip of the sensor probe 340 is enlarged. It is shown.
The scanning probe 340 may be a contact method or a non-contact method. At the time of measurement, scanning is performed while pressing the tip of the scanning probe 340 along the laminated surface of the laminated core 1, thereby transmitting an electrical signal from a sensor attached to the tip of the scanning probe 340 connected to the holding unit 330. It can be taken into the main body 200 via the line 22. Thereby, the surface shape of the laminated surface of the laminated iron core 1 is acquired as an electrical signal, and can be appropriately analyzed by an application stored in the storage unit 220. The analysis process in the control unit 210 will be described in detail later.

  The electronic signal thus captured from the scanning probe 340 is stored in the storage unit 220 as the profile P. Here, FIG. 3 is a graph showing an example of the profile P acquired by scanning the laminated surface of the laminated core 1 with the scanning probe 340. 3 illustrates a part of the waveform of the profile P acquired when the scanning surface of the laminated core 1 is scanned with the scanning probe 340 along the X direction. The profile P is composed of a total of L data points P (i), and each of the data points P (i) has a coordinate value (X coordinate) in the X direction and a laminated iron core corresponding to the X coordinate. The surface height Y of the laminated surface 1 is a pair. In the following description, the i-th data among the data included in the profile P is particularly expressed as P (i).

As shown in FIG. 3, the waveform of the profile P has a shape in which a convex region 5 corresponding to the apex of the iron core 1a and a concave region 6 corresponding to a gap between adjacent iron cores 1a are repeated. The stacking deviation of the laminated core 1 is obtained by extracting L local maximum values from a plurality of convex regions 5 having such a waveform of the profile P, and then extracting the maximum and minimum values of the extracted L local maximum values. Is obtained by calculating the difference between In the measuring apparatus 100 according to the present embodiment, such calculation of the stacking deviation is realized by the operation described below.
<Operation example>

An example of the operation of the measuring apparatus 100 will be described in detail. In the operation example in the present embodiment, the analysis process in the main body 200 of the measurement apparatus 100 mainly includes a valley identification process, a noise correction process, a profile division process, a stacking deviation calculation process, and a correction process. Hereinafter, these processes will be sequentially described.
<Valley specific processing>

In the valley specifying process, the profile P stored in the storage unit 220 is read, and the position of the valley between adjacent iron cores 1a in the waveform of the profile P is specified.

  Here, with reference to FIG. 4, the process of the valley identification process will be specifically described. FIG. 4 is a flowchart of the valley identification process according to the present embodiment.

  First, the control unit 210 reads the profile P stored in the storage unit 220 by accessing the storage unit 220 (step S101). Further, the control unit 210 reads the inclination determination sample number a1 input from the input device 400 by the user (step S102). The inclination determination sample number a1 is a constant used in step S104 described later, and can be appropriately defined by the user.

  Subsequently, the control unit 210 sets the increment variable i to “1”, that is, the initial state (step S103).

  Subsequently, the control unit 210 uses the a1 data points located around the data points P (i) and P (i) (typically adjacent to P (i)) to generate the waveform of the profile P. A slope Q (i) at the data point P (i) is calculated (step S104). Particularly in this embodiment, the slope Q (i) is calculated using the least square method. In this embodiment, the inclination Q (i) is calculated for all data points P (i). However, some data may be used to prevent a delay in the processing speed of the control unit 210 and to improve the processing speed. The slope Q (i) may be calculated only for the point P (i).

  Here, FIG. 5 is a conceptual diagram schematically showing a process of calculating Q (i) from the waveform of the profile P. The waveform of the profile P is specified by L data points P (i) constituting the profile P, and the control unit 210 determines P (i based on a1 data points adjacent to the data point P (i). ) To calculate the slope Q (i). Typically, a total of (a1 + 1) data points P (i) from P (i−a1 / 2) to P (i + a1 / 2) are used to calculate the slope Q (i).

  Returning to FIG. 4 again, subsequently, the control unit 210 determines whether or not the absolute value of the slope Q (i) calculated in step S104 is smaller than a predetermined threshold value Q1 (step S105). As a result, when the absolute value of Q (i) is smaller than Q1 (step S105: YES), the control unit 210 determines that the attribute of the data point P (i) corresponding to the Q (i) is “vertex”. (Step S106). On the other hand, when the absolute value of Q (i) is not smaller than Q1 (step S105: NO), the control unit 210 determines that the attribute of the data point P (i) corresponding to the Q (i) is “valley”. (Step S107).

After determining the attribute of the data point P (i) in this way, the control unit 210 assigns i + 1 to the increment variable i (step S108), and the increment variable i + 1 is from L, which is the total number of the data points P (i). It is determined whether it is larger (step S109). As a result, when the increment number i + 1 is equal to or smaller than L (step S109: NO), the control unit 210 returns the process to step S104 and repeats the above steps. On the other hand, when the increment variable i + 1 is larger than L (step S109: YES), the control unit 210 ends the process (END).
<Noise correction processing>

  In the noise correction process, correction is performed to improve the identification accuracy of the attribute (that is, “vertex” or “valley”) of each data point P (i) identified in the above-described valley identification process. Is done. Note that it is not always necessary to execute this processing when it is desired to improve the processing speed preferentially or when sufficient specific accuracy can be obtained without executing the processing.

  Here, the processing process of the noise correction processing will be specifically described with reference to FIG. FIG. 6 is a flowchart of the noise correction process according to the present embodiment.

  First, the control unit 210 reads the noise determination sample number a2 input from the input device 400 by the user (step S201). Here, the noise determination sample number a2 is a constant referred to in step S203 described later.

  Subsequently, the control unit 210 sets the increment variable i to “1”, that is, the initial state (step S202).

  Subsequently, the control unit 210 sets a2 data points adjacent to the data point P (i) (typically, (a2 + 1) data points from P (i−a2 / 2) to P (i + a2 / 2)). ) And the number of data points P (i) whose attribute is “vertex” is included in the referenced data points (step S203). Here, the number of data points having the attribute “vertex” included in the series of data points referred to around the data point P (i) in step S203 is referred to as C (i).

When C (i) is equal to or smaller than the predetermined threshold C1 (step S204: YES), the control unit 210 changes the attribute of the data point P (i) corresponding to C (i) to “valley” (step S205). .
On the other hand, when C (i) is larger than the predetermined threshold C1 (step S204: NO), the controller 210 advances the process to step S206.

  FIG. 7 is a conceptual diagram schematically showing the correction process performed in step S205. In FIG. 7, as a result of the valley specifying process, the data point P (i) determined to have the attribute “vertex” is indicated by a white symbol, and the data point P () determined to have the attribute “valley”. i) is indicated by black symbols. In step S204, a2 data points around the specific data point P (i) are referred to, and the data point P (i) whose attribute is determined to be “vertex” within the range is outlined. It is determined whether there are more symbols than C1.

  In particular, as shown in FIG. 7, the absolute value of the slope Q (i) approaches zero at the bottom of the waveform of the profile P. Is likely to be erroneously determined to be “vertex”. Therefore, in step S204, if there are extremely few data points having the attribute “vertex” in the vicinity thereof, the determination is made as an erroneous determination due to noise, and the determination in the valley identification process is corrected. In other words, if the attribute is truly “vertex”, as shown in FIG. 7, a considerable number of data points P (i) around it also have the “vertex” attribute. Therefore, when only a specific data point P (i) has the “vertex” attribute even though the number of data points having the “vertex” attribute is extremely small, an unnatural determination is made due to noise. As a result, the attribute of the data point P (i) is corrected to “valley”.

Returning to FIG. 6 again, subsequently, the control unit 210 substitutes i + 1 for the increment variable i (step S206), and determines whether or not the increment variable i + 1 is larger than L which is the total number of data points P (i). (Step S207). As a result, when the increment variable i + 1 is equal to or smaller than L (step S207: NO), the control unit 210 returns the process to step S203 and repeats the above steps. On the other hand, when the increment variable i + 1 is larger than L (step S207: YES), the control unit 210 ends the process (END).
<Profile division processing>

  In the profile division process, the profile P for which the valley is determined in the above process is divided into M sub-profiles SP (j).

  Here, the processing process of the profile division process will be specifically described with reference to FIG. FIG. 8 is a flowchart of profile division processing according to the present embodiment.

  First, the control unit 210 reads the division determination sample number a3 input from the input device 400 by the user (step S301). Here, the division determination sample number a3 is a constant referred to in step S303 to be described later.

  Subsequently, the control unit 210 sets the increment variable i to “1”, that is, an initial state (step S302).

  Subsequently, the control unit 210 refers to the a3 + 1 data points including the data point P (i) (typically, data points P (i−a3 / 2) to P (i + a3 / 2)) It is determined whether or not N or more data points P (i) having the “valley” attribute exist continuously in the data points (step S303). As a result, when N or more data points P (i) having the “valley” attribute exist continuously (step S303: YES), it is specified that the data point P (i) is in the valley region. (Step 304).

  Subsequently, the control unit 210 substitutes i + 1 for the increment variable i (step S305), and determines whether or not the increment variable i + 1 is larger than L, which is the total number of data points P (i) (step S306). As a result, when the increment variable i + 1 is equal to or smaller than L (step S306: NO), the control unit 210 returns the process to step S303 and repeats the above steps. On the other hand, when increment variable i + 1 is larger than L (step S306: YES), control unit 210 advances the process to step S307.

As described above, after specifying the valley between the adjacent iron cores 1a in the profile P, the control unit 210 sets the profile P to the sub-profile SP (j) corresponding to each iron core 1a with the specified valley as a boundary. The process is divided (step S307), and the process ends (END). In the present embodiment, in particular, the laminated iron core 1 is formed by laminating M iron cores 1a, and the profile P is divided into M sub-profiles SP (j) corresponding to each iron core 1a.
<Load misalignment calculation processing>

  In the stacking deviation calculation process, the stacking deviation of the laminated core 1 can be calculated by specifying the maximum value of the waveform for each of the divided sub-profiles SP (j).

  Here, with reference to FIG. 9, the processing process of the misalignment calculation process will be specifically described. FIG. 9 is a flowchart of the stacking deviation calculation process according to the present embodiment.

  First, the control unit 210 sets the increment variable j to “1”, that is, the initial state (step S401).

  Subsequently, the control unit 210 calculates a maximum value SPmax (j) in the waveform included in the sub-profile SP (j) (Step S402).

  Subsequently, the control unit 210 substitutes j + 1 for the increment variable j (step S403), and determines whether or not the increment variable j + 1 is larger than M, which is the total number of sub-profiles SP (j) (step S404). As a result, when the increment variable j + 1 is equal to or smaller than M (step S404: NO), the control unit 210 returns the process to step S402 and repeats the above steps. On the other hand, when increment variable j + 1 is larger than M (step S404: YES), control unit 210 advances the process to step S405.

Subsequently, the control unit 210 refers to the total M maximum values SPmax (j) calculated above, and specifies the maximum value SPmax1 and the minimum value SPmax2 from among them (step S405). Then, the difference between the maximum value SPmax1 and the minimum value SPmax2 is calculated by the following formula, and the stacking deviation Δ of the laminated core is calculated (step S406).
Δ = SPmax1-SPmax2 (1)
<Correction process>

  Here, if the laminated iron core 1 itself, which is a measurement target, is inclined, it is conceivable that the measurement error increases due to the stacking deviation calculated as described above. Therefore, by performing correction in the correction process, it is possible to effectively eliminate such a measurement error and improve the measurement accuracy of stacking deviation.

  Here, the processing process of the correction process will be specifically described with reference to FIG. FIG. 10 is a flowchart of the correction process according to the present embodiment.

  First, the controller 210 reads M maximum values SPmax (j) calculated in step S402 (step S501). Then, the control unit 210 calculates a regression line from the read M maximum values SPmax (j) using the least square method (step S502).

Subsequently, the control unit 210 specifies the slope of the regression line calculated in step S502, and corrects the entire profile P so that the slope becomes zero (step S503). Specifically, the correction amount is calculated so that the slope of the regression line becomes zero, and the correction amount is applied to each of the data points P (i) constituting the profile P.
In addition, as in the case of a modification example to be described later, when calculating the stacking deviation in a plane by calculating the stacking deviation on a plurality of lines along the laminated surface, the regression plane is calculated instead of the regression line. The profile P may be corrected so that the slope of the regression plane becomes zero.

  Thereafter, the control unit 210 re-executes the above-described profile division process and stacking deviation calculation process based on the corrected data point P (i) (steps S504 and S505).

  Here, FIG. 11 is a conceptual diagram schematically showing processing performed on the profile P in steps S502 and S503. In step S502, as shown in FIG. 11A, a regression line is calculated based on the profile P using the least square method. The regression line has a certain slope. On the other hand, in step S503, the regression line shown in FIG. 11A is changed so that the slope of the regression line becomes zero (that is, parallel to the horizontal axis), and the profile is corrected in accordance with the change. As a result, as shown in FIG. 11B, the magnitude of the corrected stacking deviation Δ can be reduced as compared with the stacking deviation Δ before correction shown in FIG. As described above, by performing the correction, it is possible to effectively eliminate the error and calculate the stacking deviation Δ with high accuracy.

In this embodiment, each data point P (i) constituting the profile P is corrected based on the slope of the regression line. However, the maximum value SPmax (j directly used for calculating the stacking deviation Δ is used. ) And only the misalignment calculation process may be re-executed in step S504 and step S505.
<Modification>

  The laminated core stacking deviation measuring method and measuring apparatus according to the present invention can not only determine the stacking deviation on a single line along the laminated surface as described above, but also exemplify the following modifications. Thus, it is also possible to calculate the stacking deviation in a plane. That is, in the stacked core stacking deviation measuring method and measuring apparatus according to the modified example, it is possible to measure a planar stacking offset by acquiring profiles on a plurality of lines with respect to the stacking surface.

  Here, with reference to FIG. 12, the process of the valley determination process according to the present modification will be specifically described. FIG. 12 is a flowchart of a valley determination process according to the modification. In FIG. 12, the steps common to the above-described embodiment are denoted by the same reference numerals, and the description thereof is omitted as appropriate.

  First, the control unit 210 reads the number of scanning lines a4 input from the input device 400 by the user (step S601). Here, the scanning line number a4 is the number of scanning lines of 2 or more necessary for measuring the stacking deviation in a plane. Note that the user side may individually specify which line on the stacked surface is actually scanned by the scanning probe 340.

Subsequently, the control unit 210 sets the increment variable k corresponding to the number of scanning lines to “1” (step S602) , that is, the initial state, and acquires the profile P along the first scanning line (note that the single scanning is performed). The method for obtaining the profile P on the line is as described in the above embodiment). After the acquisition of the profile P on the scan line is completed, the control unit 210 substitutes k + 1 for the increment variable ( step S603 ), and determines whether the increment variable k + 1 is larger than the total scan line number a4 ( step S604). ). As a result, when the increment number k + 1 is equal to or smaller than a4 ( step S604 : NO), the control unit 210 returns the process to step S101 and repeats the above-described process, thereby obtaining the profile P on the next scanning line. I do. On the other hand, when the increment number k + 1 is larger than a4 ( step S604 : YES), the control unit 210 ends the process on the assumption that the profile P is acquired for all the a4 scanning lines.

  By executing the above-described profile dividing process, stacking shift calculation process, and correction process based on the profile P acquired for the a4 scanning lines in this manner, a planar stacking shift can be measured.

  Note that, in the above-described modification, it is possible to measure planar stacking deviation by executing profile division processing, stacking shift calculation processing, and correction processing using data acquired for a total of a4 scanning lines as one profile. However, the data acquired for a total of a4 scanning lines are set as different profiles (that is, a4 profiles corresponding to a4 scanning lines), and profile division processing, stacking deviation calculation processing, and correction processing are individually performed for each profile. You may measure a planar stacking deviation by performing.

  As described above, according to the stacking misalignment measuring method and apparatus of the laminated core according to the present embodiment, the profile relating to the laminated surface of the laminated core is divided into sub-profiles corresponding to each of a plurality of laminated cores. By calculating the maximum value from the waveform corresponding to the uneven shape on the surface of the iron core for each profile, it is possible to extract information necessary for calculating the stacking deviation, that is, the vertex positions of each of the laminated iron cores. it can. As a result, unnecessary information included in the profile, for example, information on concave gaps existing between the laminated iron cores can be accurately determined, and the stacking deviation can be easily calculated without relying on human work. It becomes possible.

  INDUSTRIAL APPLICABILITY The present invention can be used, for example, as a stacking misalignment measuring method and measuring apparatus for stacking cores capable of evaluating stacking misalignment on the stacking surface of the stacked cores.

DESCRIPTION OF SYMBOLS 1 Laminated iron core, 1a Iron core, 100 Measuring device, 200 Main body part, 210 Control part, 220 Storage part, 300 Measuring part, 340 Scanning probe , 400 Input device, 500 Display device

Claims (6)

A method for measuring a stacking deviation of a laminated core formed by laminating a plurality of plate-like iron cores,
A profile obtaining step for obtaining a profile of the laminated surface of the laminated core along the lamination direction of the iron core;
A valley specifying step for specifying the first region in which the absolute value of the inclination exceeds a predetermined threshold in the acquired profile as a valley between the adjacent iron cores,
A profile dividing step of dividing the profile into sub-profiles corresponding to each of the iron cores with the first region as a boundary;
A maximum value calculating step for calculating the maximum value of the waveform of the sub-profile for each of the sub-profiles;
A stacking misalignment measuring method for a laminated core, comprising: a stack misalignment calculating step of calculating a difference between a maximum value and a minimum one of the maximum values calculated for each sub-profile as a stack misalignment. A regression line calculating step for calculating a regression line based on the maximum value calculated for each sub-profile;
A correction step of correcting the profile so that the slope of the calculated regression line becomes zero;
The laminated core according to claim 1, further comprising: a stacking shift recalculation step that re-executes the profile dividing step, the maximum value calculation step, and the stacking shift calculation step based on the corrected profile. How to measure misalignment. The valley identification step identifies the second region of which the absolute value of the slope is equal to or less than the predetermined threshold in the acquired profile as the apex of the iron core,
In the predetermined range including the second region in the profile, the method further comprises a noise removing step of specifying the second region as a valley when the first region occupies more than a predetermined ratio. The method for measuring the misalignment of the laminated core according to claim 1 or 2.   4. The method for measuring misalignment of a laminated core according to claim 1, wherein the profile obtaining step obtains the profile along a plurality of the laminating directions. 5. An apparatus for measuring a stacking deviation of a laminated core in which a plurality of plate-like iron cores are laminated,
Profile acquisition means for acquiring a profile of the laminated surface of the laminated core along the lamination direction of the iron core;
A valley identification unit that identifies the first region in which the absolute value of the slope of the acquired profile exceeds a predetermined threshold as being a valley between the adjacent iron cores,
Profile dividing means for dividing the profile into sub-profiles corresponding to each of the iron cores with the first region as a boundary;
A maximum value calculating means for calculating the maximum value of the waveform of the sub-profile for each sub-profile;
A stack misalignment measuring device for a laminated core, comprising: stack misalignment calculating means for calculating a difference between a maximum value and a minimum one of the maximum values calculated for each sub-profile as a stack misalignment.   A computer program for causing a computer system to function as the stacking misalignment measuring apparatus for laminated cores according to claim 5.

Images