WO2020138339A1 - 製品管理方法 - Google Patents
製品管理方法 Download PDFInfo
- Publication number
- WO2020138339A1 WO2020138339A1 PCT/JP2019/051245 JP2019051245W WO2020138339A1 WO 2020138339 A1 WO2020138339 A1 WO 2020138339A1 JP 2019051245 W JP2019051245 W JP 2019051245W WO 2020138339 A1 WO2020138339 A1 WO 2020138339A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- gas diffusion
- primary
- storage medium
- electrode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M4/00—Electrodes
- H01M4/86—Inert electrodes with catalytic activity, e.g. for fuel cells
- H01M4/88—Processes of manufacture
- H01M4/8803—Supports for the deposition of the catalytic active composition
- H01M4/8807—Gas diffusion layers
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
- H01M8/00—Fuel cells; Manufacture thereof
- H01M8/10—Fuel cells with solid electrolytes
- H01M8/1004—Fuel cells with solid electrolytes characterised by membrane-electrode assemblies [MEA]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30164—Workpiece; Machine component
Definitions
- the present invention relates to a product management method. More specifically, the present invention relates to a product management method for managing a product made of carbon paper by a computer.
- a fuel cell is formed by sandwiching a membrane electrode structure (so-called MEA) formed by disposing an electrolyte layer containing a solid polymer between an anode electrode body and a cathode electrode body with a pair of separators. .. Further, the fuel cell stack is configured by stacking a plurality of such fuel cells, and is mounted in a vehicle as a power source that generates electric power to be supplied to a drive motor, for example.
- MEA membrane electrode structure
- the fuel cell stack is configured by stacking a plurality of such fuel cells, and is mounted in a vehicle as a power source that generates electric power to be supplied to a drive motor, for example.
- carbon paper with conductivity and acid resistance is often used as the base material of the anode electrode body and the cathode electrode body that make up each fuel cell. Therefore, a plurality of carbon papers are used for one fuel cell stack. Further, the performance of the fuel cell stack is greatly influenced by the performance of each carbon paper. Therefore, in the manufacturing process for manufacturing the fuel cell stack, what kind of fuel cell stack is formed by stacking the cathode electrode body, the anode electrode body, the fuel cell unit including these electrode structure bodies, or these fuel cell units There is a demand for a management system that manages whether carbon paper is used.
- Patent Document 1 discloses a management system that manages a fuel cell stack by attaching an identifier on which information about the inspection result is printed to the MEA. Therefore, it is conceivable to attach an identifier to the carbon paper in the process of manufacturing the fuel cell stack using such a technique. However, in this case, since it is necessary to directly attach the identifier to the carbon paper, there is a possibility that the portion of the carbon paper to which the identifier is attached cannot be used for power generation or the carbon paper may be damaged. is there. Further, in order to obtain information from the identifier, it is necessary to use a dedicated reader, and management may be complicated.
- the object of the present invention is to provide a product management method for managing parts and products based on carbon paper, which can be managed without damaging the carbon paper.
- a material for example, carbon paper CA, CC and a first gas diffusion layer 25b and a second gas diffusion layer 26b, which will be described later
- a base material for example, A structure (for example, an anode electrode body 25 to be described later) manufactured through a material forming step (for example, a bonding step S8 to be described later) for bonding or coating a first electrode catalyst layer 25a and a second electrode catalyst layer 26a described below.
- a cathode electrode body 26, and a membrane electrode structure 21) or a product (for example, a fuel cell 2 or a fuel cell stack 1 described later) having the structure as a component is managed by a computer (for example, a computer 5). Therefore, the primary work data (for example, the specific work data (for example, a specific position P, PA, PC described later) in a portion where the material is not bonded or coated on each of the plurality of works is taken.
- a registration step (for example, anode-side initial registration described later) of acquiring primary feature vector data in FIG. 6 described later) and storing the acquired primary work data of each work in a storage medium (for example, storage medium 55 described later).
- Reference work data is acquired by capturing an image of the specific portion of the work after the registration process and the step S2 and the cathode side initial registration process S5), and the reference work data is stored in the storage medium. And a collating process (for example, a cell managing process S9, a cell inspecting process S11, and a stack assembling/stack managing process S12, which will be described later).
- a collating process for example, a cell managing process S9, a cell inspecting process S11, and a stack assembling/stack managing process S12, which will be described later).
- the specific portion is a part of a surface (for example, a second surface CAb or CCb described later) of the workpiece to which the material is not bonded or coated.
- the product management method uses the secondary work data (for example, the secondary work data in FIG. 6 to be described later) by capturing an image of the specific portion of the work after the registration process and the material forming process.
- the secondary work data is collated with a plurality of primary work data stored in the storage medium, and the primary work data having the highest degree of coincidence and the secondary work data are associated with each other.
- an anode electrode material for example, a first electrode catalyst layer 25a described below
- a first gas diffusion layer for example, a first gas diffusion layer 25b described below
- the cathode electrode material for example, the second electrode catalyst layer 26a described later
- the second gas diffusion layer for example, the second gas diffusion layer 26b described later
- the formed anode electrode body for example, the anode electrode body 25 described later.
- a cathode electrode body for example, a cathode electrode body 26 described later
- an ion exchange layer for example, an electrolyte membrane 24 described later
- Primary anode electrode data (eg, specific anode PA data) is obtained by capturing an image of a first specific location (eg, a specific location PA described later) in a portion where the anode electrode material is not bonded or coated on each of the first gas diffusion layers.
- Primary feature vector data within a broken line 6a of FIG. 6 described below) and the obtained primary anode electrode data of each first gas diffusion layer is stored in a storage medium (for example, a storage medium 55 described later).
- a side registration step for example, an anode side initial registration step S2 described later
- a second specific portion for example, a later-described later-described portion
- Primary cathode electrode data is acquired by capturing an image of a specific portion PC), and the acquired primary cathode of each second gas diffusion layer.
- a cathode side registration step of storing electrode data in the storage medium for example, a cathode side initial registration step S5 described later
- the product management method is such that the primary anode electrode data of the first gas diffusion layer included in the integrated electrode structure and the primary cathode electrode of the second gas diffusion layer included in the integrated electrode structure. It is preferable to further include a cell management step (for example, a cell management step S9 described later) of storing the data in the storage medium in association with the data.
- a cell management step for example, a cell management step S9 described later
- the product management method may include a plurality of first gas diffusion layers included in a laminated body (for example, a fuel cell stack 1 described later) configured by laminating a plurality of the integrated electrode structures.
- /Stack management step S12) is preferably further provided.
- the pattern of the surface of the work made of carbon paper as a base material is different.
- primary work data is acquired by capturing an image of a predetermined specific portion for each of a plurality of works, and the acquired primary work data of each work is stored in a storage medium.
- reference work data is acquired by capturing an image of a specific portion of the work after the registration step, and the reference work data is collated with the data stored in the storage medium.
- the product management method of the present invention when performing the registration process, the collation process, etc., it is only necessary to capture an image of a specific portion of the work, so that the work may be scratched or the performance of the parts or products may be degraded.
- the specific portion for obtaining the primary work data and the reference work data is set to the portion of the work where the material is not joined or coated.
- the pattern on the surface of the work changes due to this processing.
- the change in the pattern due to this processing is smaller on the surface on which the material is not bonded or coated than on the surface on which the material is bonded or coated. Therefore, in the product management method of the present invention, the specific portion is set as a part of the surface of the work on which the material considered to have a small change in the pattern due to the processing is not joined or coated. Therefore, according to the product management method of the present invention, by determining the specific portion at such a position, it is possible to reduce the change in the pattern of the specific portion before and after the material forming process, and thus it is possible to improve the accuracy in the matching process.
- the pattern at a specific location may change before and after the material forming process. Therefore, in the updating process, secondary work data is acquired by capturing an image of a specific portion of the work after the registration process and the material forming process, and the secondary work data and a plurality of primary works stored in the storage medium are acquired. The work data is collated, and the primary work data having the highest degree of coincidence and the secondary work data are associated and stored in the storage medium.
- the secondary work data acquired after the material forming process in this way in the storage medium in association with the primary work data acquired before the material forming process, the reference work can be stored in the subsequent collating process. Since the data can be collated with the latest secondary work data, the accuracy in this collation process can be further improved.
- the primary cathode electrode data is obtained by capturing an image of the first specific portion in the portion where the cathode electrode material is not joined or coated on each of the plurality of first gas diffusion layers.
- the obtained primary cathode electrode data of each first gas diffusion layer is stored in a storage medium, and the anode electrode material is bonded or coated to each of the plurality of second gas diffusion layers in the anode side registration step.
- the primary anode electrode data is acquired by capturing an image of the second specific portion in the unprocessed portion, and the acquired primary anode electrode data of each second gas diffusion layer is stored in the storage medium.
- reference electrode data is acquired by capturing an image of the first specific portion of the first gas diffusion layer that has undergone the cathode side registration step or the second specific portion of the second gas diffusion layer that has undergone the anode side registration step. Then, the reference electrode data is collated with the data stored in the storage medium.
- what kind of gas diffusion does an integrated electrode structure including a pair of cathode electrode body and anode electrode body, or a laminated body formed by laminating the integrated electrode structure It can be specified which layer is being used.
- the product management method of the present invention when performing the cathode side registration step, the anode side registration step, the collation step, etc., it is only necessary to take an image of a specific portion of the gas diffusion layer, so that the gas diffusion layer is damaged or laminated It does not reduce the performance of the body. Further, in the product management method of the present invention, the specific portion for acquiring the primary cathode electrode data, the primary anode electrode data, and the reference electrode data is set to the portion of the gas diffusion layer where the electrode material is not bonded or coated.
- an image of the specific portion can be easily captured even on the integrated electrode structure or the laminated body, and the obtained reference electrode data is used as the primary cathode electrode data or the primary anode data.
- the origin and characteristics of this gas diffusion layer can be specified by collating with the electrode data.
- the data is stored in the storage medium in association with the primary anode electrode data of the layer. This makes it possible to specify what kind of origin, characteristics, and the like the gas diffusion layer is used in in the fuel cell unit including the integrated electrode structure.
- the primary cathode electrode data of each of the plurality of first gas diffusion layers included in the stacked body configured by stacking the integrated electrode structures are included in the same stacked body.
- the data is stored in the storage medium in association with the primary anode electrode data of each of the second gas diffusion layers. This makes it possible to identify what kind of origin, characteristics, and the like the gas diffusion layer is used in in the fuel cell stack including the stack.
- FIG. 1 It is a figure which shows the structure of the fuel cell stack which is the management target of the product management method which concerns on one Embodiment of this invention. It is a figure which shows the structure of the management system used in the product management method. It is a functional block diagram implement
- a product management method is executed in a manufacturing process for manufacturing the fuel cell stack 1 as shown in FIG. 1, and the plurality of fuel cell stacks 1 manufactured in this manufacturing process and the plurality of fuels constituting the fuel cell stack 1 are manufactured.
- Battery cell 2 a plurality of membrane electrode structures 21 constituting these plurality of fuel battery cells 2, a plurality of electrode bodies 25, 26 constituting these plurality of membrane electrode structures 21, and these electrode bodies 25, 26
- the plurality of gas diffusion layers 25b, 26b, etc. constituting the above are managed as a work made of carbon paper, which will be described later, and their manufacturing numbers, origins, etc. are managed.
- FIG. 1 is a diagram showing a configuration of a fuel cell stack 1 which is a management target of the product management method according to the present embodiment. In FIG. 1, only a part of the plurality of fuel battery cells 2 forming the fuel battery stack 1 is shown.
- the fuel cell 2 includes a membrane electrode structure 21 (hereinafter referred to as “MEA 21”), and a first separator 22 and a second separator 23 that sandwich the MEA 21.
- the MEA 21 includes, for example, an electrolyte membrane 24 as an ion exchange layer containing a solid polymer such as a perfluorosulfonic acid thin film, an anode electrode body 25 provided on one surface of the electrolyte membrane 24, and the other of the electrolyte membrane 24. And a cathode electrode body 26 provided on the surface.
- the anode electrode body 25 is a porous body including a first electrode catalyst layer 25a facing one surface of the electrolyte membrane 24 and a first gas diffusion layer 25b laminated on the first electrode catalyst layer 25a.
- the cathode electrode body 26 is a porous body including a second electrode catalyst layer 26a facing the other surface of the electrolyte membrane 24 and a second gas diffusion layer 26b laminated on the second electrode catalyst layer 26a.
- the first electrode catalyst layer 25a and the second electrode catalyst layer 26a include, for example, catalyst particles (electrode catalyst) configured by supporting a catalyst metal such as platinum on a carbon catalyst carrier such as carbon black (electrode catalyst), and high ion conductivity. And a molecular binder.
- the electrode catalyst may be made of only a catalyst metal such as platinum black and may not include a catalyst carrier.
- the first gas diffusion layer 25b and the second gas diffusion layer 26b are sheet materials using carbon paper as a base material, for example.
- the first gas diffusion layer 25b is arranged so as to face the first separator 22, and the second gas diffusion layer 26b is arranged so as to face the second separator 23.
- carbon separators are used as the first separator 22 and the second separator 23, for example, metal separators may be used instead.
- An anode side gas inlet communication hole (not shown) for supplying the anode side gas and an anode side gas for discharging the anode side gas are formed on the surface of the first separator 22 facing the first gas diffusion layer 25b.
- An anode-side gas flow channel 27 that communicates with the outlet communication hole (not shown) is formed.
- a cathode-side gas inlet communication hole (not shown) for supplying the cathode-side gas and a cathode-side gas for discharging the cathode-side gas are formed on the surface of the second separator 23 facing the second gas diffusion layer 26b.
- a cathode-side gas flow channel 28 that communicates with the outlet communication hole (not shown) is formed.
- the fuel cell stack 1 is configured by stacking a plurality of fuel cell cells 2 as described above.
- FIG. 2 is a diagram showing the configuration of the management system S used in the product management method.
- the management system S is incorporated in a part of the manufacturing line 3 in which the manufacturing process of the fuel cell stack (see FIG. 5 described later) is performed.
- the management system S includes a camera C that captures images of n pieces (n is an integer of 2 or more) of works W1,..., Wn that are management targets that flow through the manufacturing line 3, and an image obtained by the camera C.
- a computer 5 in which a program for processing data is installed.
- the work pieces W1,..., Wn flowing through the production line 3 are the fuel cell stack 1, the fuel cell 2, the membrane electrode structure 21, the anode electrode body 25, the cathode electrode body 26, the first gas diffusion layer 25b, and The second gas diffusion layer 26b and the like.
- the workpieces W1,..., Wn include gas diffusion layers 25b and 26b having carbon paper as a base material.
- the pattern formed on the surface of carbon paper differs depending on each individual. Therefore, the management system S manages these works W1,..., Wn by using the pattern of carbon paper at a specific location P predetermined for these works W1,.
- the camera C takes an image of a specific portion P of the works W1,..., Wn flowing through the manufacturing line 3, and sends the obtained image data to the computer 5.
- FIG. 3 is a functional block diagram realized in the computer 5 by a program installed in the computer 5.
- the feature vector calculation unit 51 that calculates the feature vector data that characterizes the work W
- the feature vector calculation unit 51 calculates the feature vector data.
- An initial registration unit 52 for newly constructing a management database by associating the characteristic vector data and various information (such as a carbon sheet manufacturing number and unique information described later) that characterize this work W in the storage medium 55
- An update registration unit 53 that collates the feature vector data calculated by the vector calculation unit 51 with a management database stored in the storage medium 55 and updates the management database.
- FIG. 4 is a diagram showing a configuration of the feature vector calculation unit 51.
- the feature vector calculation unit 51 includes, for example, 256 types of filters from the first filter F1 to the 256th filter F256, and based on the image data of the specific location P obtained by the camera C by using these, Feature vector data, which is 256-dimensional vector data having 256 scalar amounts up to the 256th feature amount as components, is calculated.
- Each of the filters F1 to F256 outputs the first to the 256th feature amount that is an integer within a predetermined range (for example, 0 to 99) when the pixel value that characterizes the image data of the specific portion P is input.
- Each of the filters F1 to F256 is composed of, for example, a four-layer neural network including first, second, third, and fourth layers.
- the neural network that constitutes each of the filters F1 to F256 prepares a plurality of carbon papers as samples, and learns in advance so that an appropriate sample can be identified by the feature amount vector calculated from the image data of these samples.
- the old one is used. More specifically, by using a camera, images of a predetermined portion of a plurality of carbon papers prepared as samples are imaged under different times and environments, and learning image data and verification image data are sampled for each sample.
- the neural network forming each of the filters F1 to F256 includes feature vector data calculated using the learning image data prepared for each sample, and a verification image obtained by imaging each sample under different times and environments.
- the degree of coincidence with the feature vector data calculated using the data is evaluated, and learning is performed so that the degree of coincidence between the two feature vector data obtained from the same sample is the highest.
- the degree of coincidence between the two feature vector data can be quantified by calculating the norm value of the difference vector data whose component is the difference between the respective components of the two feature vector data. That is, the smaller the norm value of the difference vector data, the higher the degree of coincidence between the two feature vector data can be evaluated.
- FIG. 5 is a diagram showing a procedure of a manufacturing process of a fuel cell stack in which the product management method according to the present embodiment is incorporated.
- the manufacturing process of the fuel cell stack includes the anode side trim step S1, the anode side initial registration step S2, the anode side cleaning step S3, the cathode side trim step S4, the cathode side initial registration step S5, and the cathode side cleaning step S6.
- the base paper roll RA1 is cut by a trim device (not shown), thereby making a plurality of Cut out a rectangular carbon paper CA.
- FIG. 5 illustrates a case where four carbon papers CA are cut out in a row along the width direction of the raw paper roll RA1, but the present invention is not limited to this.
- the plurality of carbon papers CA cut out from the base paper roll RA1 by the anode side trim step S1 become the first gas diffusion layer 25b.
- the camera C captures an image of a specific location PA predetermined for each carbon paper CA as shown in FIG.
- the specific portion PA is defined as a portion of the carbon paper CA where the anode electrode material is not joined in the joining step S8 described later. More specifically, the specific portion PA is set to a part of the second surface CAb of the carbon paper CA opposite to the first surface CAa to which the anode electrode material is bonded in the bonding step S8.
- the feature vector calculation unit 51 calculates the feature vector data based on the image data of the specific portion PA imaged by the camera C.
- the initial registration unit 52 sets the feature vector data calculated by the feature vector calculation unit 51 for each sheet as primary feature vector data, and the primary feature vector data is stored in the storage medium 55. Memorize.
- the initial registration unit 52 applies the primary feature vector data calculated for each sheet and the carbon sheets CA cut out in the anode side trim step S1 in a predetermined order to each sheet.
- the production number and unique information unique to the target carbon paper CA are stored in the storage medium 55 in association with each other.
- the unique information includes, for example, a manufacturing date of the carbon paper CA, a roll number given to each one of the base paper rolls RA1, a cutting position of the carbon paper CA, and the like.
- the management database as shown by the broken line 6a in FIG. 6 is constructed in the storage medium 55.
- the base paper roll RC1 is cut by a trim device (not shown) Cut out a rectangular carbon paper CC.
- FIG. 5 illustrates a case where four carbon papers CC are cut out in a row along the width direction of the base paper roll RC1 at one time, but the present invention is not limited to this.
- the plurality of carbon papers CC cut out from the base paper roll RC1 by the cathode side trim step S4 become the second gas diffusion layer 26b.
- the cathode side initial registration step S5 by using the management system S, information on each of the plurality of carbon papers CC cut out in the cathode side trim step S4 is registered, and a management database as illustrated in FIG. 6 is newly added.
- the camera C captures an image of a specific location PC predetermined for each carbon paper CC as shown in FIG.
- the specific portion PC is defined as a portion of the carbon paper CC where the cathode electrode material is not joined in the joining step S8 described later. More specifically, the specific portion PC is set to a part of the second surface CCb of the carbon paper CC opposite to the first surface CCa to which the cathode electrode material is bonded in the bonding step S8.
- the feature vector calculation unit 51 calculates the feature vector data based on the image data of the specific portion PC imaged by the camera C.
- the initial registration unit 52 sets the feature vector data calculated for each sheet by the feature vector calculation unit 51 as primary feature vector data, and the primary feature vector data is stored in the storage medium 55. Memorize.
- the initial registration unit 52 applies the primary feature vector data calculated for each sheet and the carbon sheets CC that are cut out in the cathode side trim step S4 to each other in a predetermined order.
- the manufacturing number and unique information unique to the target carbon paper CC are stored in the storage medium 55 in association with each other.
- the unique information includes, for example, a manufacturing date of the carbon paper CC, a roll number given to each one of the base paper rolls RC1, a cutting position of the carbon paper CC, and the like.
- the management database as shown by the broken line 6b in FIG. 6 is constructed in the storage medium 55.
- the anode electrodes cut out from the base paper roll RA2 of the anode electrode material prepared in advance for the first electrode catalyst layer 25a of the anode electrode body 25 are formed on both surfaces of the sheet-shaped electrolyte membrane 24 prepared in advance.
- the material and the cathode electrode material cut out from the base paper roll RC2 of the cathode electrode material prepared in advance for the second electrode catalyst layer 26a of the cathode electrode body 26 are joined by, for example, thermal transfer.
- the first electrode catalyst layer 25a and the second electrode catalyst layer 26a are bonded to both surfaces of the electrolyte membrane 24, respectively.
- the MEA 21 configured by being sandwiched between the body 25 and the cathode electrode body 26 is manufactured. As described above, the specific locations PA and PC are set on the second surfaces CAb and CCb of the carbon papers CA and CC. Therefore, these specific parts PA and PC can be visually recognized from the outside even in the MEA 21 manufactured through the joining step S8.
- the management system S is used to register the information of each of the MEAs 21 manufactured through the joining step S8, and update the management database as illustrated in FIG. More specifically, in the cell management step S9, the camera C is set to a portion other than the portion to which the first electrode catalyst layer 25a is joined, of the first gas diffusion layer 25b forming the MEA 21 as shown in FIG. The camera C captures an image of the specified specific area PA, and the camera C displays an image of the specific area PC set in a part of the second gas diffusion layer 26b forming the MEA 21 other than the part to which the second electrode catalyst layer 26a is joined. Image.
- the feature vector calculation unit 51 calculates the feature vector data based on the image data of the specific portions PA and PC captured by the camera C, respectively.
- the update registration unit 53 sets the feature vector data calculated based on the image data of the specific portion PA as the secondary feature vector data, and the secondary feature vector data is stored in the storage medium 55.
- the carbon sheet manufacturing number of the first gas diffusion layer 25b that constitutes the target MEA 21 is specified by collating with the management database. More specifically, the update registration unit 53 reads out a plurality of primary feature vector data included in the management database, and calculates 2 based on the image data of the specific portion PA among the plurality of primary feature vector data.
- the one having the highest degree of coincidence with the next feature vector data is extracted, and the carbon sheet manufacturing number associated with the primary feature vector data having the highest degree of coincidence is specified.
- the degree of coincidence between the two feature vector data can be digitized by the norm value of the difference vector data whose component is the difference between the respective components of the two feature vector data as described above. That is, the closer the two feature vector data are, the smaller the norm value of the difference vector data is, and therefore the combination having the smallest norm value can be determined to have a high degree of coincidence.
- the update registration unit 53 stores the secondary feature vector data calculated based on the image data of the specific portion PA in the storage medium 55 in association with the carbon sheet manufacturing number and the primary feature vector data identified as described above.
- the update registration unit 53 sets the feature vector data calculated based on the image data of the specific location PC as the secondary feature vector data, and collates the secondary feature vector data with the management database stored in the storage medium 55.
- the carbon sheet manufacturing number of the second gas diffusion layer 26b forming the target MEA 21 is specified by the same procedure as described above.
- the update registration unit 53 stores the secondary feature vector data calculated based on the image data of the specific location PC in the storage medium 55 in association with the carbon sheet manufacturing number and the primary feature vector data identified as described above. Let In addition, when the update registration unit 53 stores the secondary feature vector data given to each MEA 21 in the above-described manner in the storage medium 55, two sets of the secondary feature vector data and the joining process are performed.
- the MEA 21 manufactured in S8 is stored in the storage medium 55 in association with the cell manufacturing number assigned in a predetermined order for each set.
- the management database as shown by the broken line 6c in FIG. 6 is constructed in the storage medium 55.
- the first gas diffusion layer having the carbon sheet manufacturing number “An*****1” is the carbon to which the same cell manufacturing number “***1” is given.
- the sheet manufacturing number is associated with the second gas diffusion layer having “Ca*****1”. Therefore, according to the example shown in FIG. 6, the MEA having the cell serial number “***1” has the carbon sheet serial number “An*****1” as the first gas diffusion layer. Then, it can be specified that the carbon sheet having the carbon sheet production number "Ca*****1” is assembled as the second gas diffusion layer. As shown in FIG. 6, there is a slight deviation between the primary feature vector data and the secondary feature vector data associated therewith.
- This deviation is caused by performing an anode-side cleaning process, a cathode-side cleaning process, a bonding process, etc. between the acquisition of the primary feature vector data and the acquisition of the secondary feature vector data. This is due to the change in the pattern at the specific places PA and PC. Therefore, when the feature vector data calculated based on the image captured by the camera C in the process after the cell management process is collated with the management database, the newly acquired feature vector data and the latest state of the carbon paper are obtained. It is preferable to calculate the degree of coincidence with the secondary feature vector data that is considered to reflect.
- the cell inspection step S11 various inspections are performed on the MEA 21 that has undergone the cell cleaning step S10 to determine whether or not sufficient performance is ensured.
- the management system S is used to register information regarding the result of this inspection and update the management database. More specifically, in the cell inspection step S11, the camera C captures an image of the specific portion PA of the MEA 21 that has undergone the cell cleaning step S10.
- the feature vector calculation unit 51 calculates feature vector data based on the image data of the specific portion PA. Further, the update registration unit 53 collates the feature vector data calculated based on the image data of the specific portion PA with the management database stored in the storage medium 55, and the secondary feature having the highest degree of coincidence with the feature vector data.
- the update registration unit 53 stores the information on the inspection result as described above in the storage medium 55 in association with the cell manufacturing number specified as described above.
- the management database as shown by the broken line 6d in FIG. 6 is constructed in the storage medium 55.
- the MEA having the cell serial number “***1” has the carbon sheet having the carbon sheet serial number “An*****1” as the first gas diffusion layer.
- a carbon sheet whose serial number is "Ca*****1” was assembled as a second gas diffusion layer, and the inspection result is that the result of inspection A is "aa point”. It is possible to specify that the result of the inspection B is “ba point” and the result of the inspection C is “ca point”.
- the management system S is used to register information about the fuel cell stack 1 and update the management database as illustrated in FIG.
- the camera C captures an image of each specific portion PA of the m MEAs 21 selected for assembling one set of fuel cell stacks 1.
- the feature vector calculation unit 51 calculates the feature vector data based on the image data of the specific locations PA of these m MEAs 21.
- the update registration unit 53 collates the feature vector data calculated based on the image data of the specific locations PA of the plurality of MEAs 21 with the management database stored in the storage medium 55, and finds that the degree of coincidence with each feature vector data is the highest. Identify the cell serial number associated with the high secondary feature vector data.
- the update registration unit 53 targets the stack manufacturing number assigned to each set of the fuel cell stacks 1 in the stack assembly/stack management step S12 in a predetermined order and the set of the fuel cell stacks 1.
- the layer number corresponding to the position where the MEA 21 is inserted is stored in the storage medium 55 in association with the cell manufacturing number specified as described above.
- the management database as shown by the broken line 6e in FIG. 6 is constructed in the storage medium 55.
- the fuel cell stack having the stack production number “***1” has the MEA having the cell production number “***1” and the cell production number “***1”. It can be specified that the MEA that is "*2" is used. Further, according to the example of FIG. 6, in the fuel cell stack having the stack production number “***1”, the MEA having the cell production number “***1” is used as the “a1st layer”. Therefore, it can be specified that the MEA having the cell manufacturing number “***2” is used as the “a2 layer”.
- the present invention is not limited to this.
- the detailed configuration may be appropriately changed within the scope of the gist of the present invention.
- the anode electrode body 25 and the cathode electrode body 26 are formed by joining the first gas diffusion layer 25b and the second gas diffusion layer 26b to the first electrode catalyst layer 25a and the second electrode catalyst layer 26a, respectively.
- the present invention is not limited to this. Even if the anode electrode body 25 and the cathode electrode body 26 are manufactured by applying the first electrode catalyst layer 25a and the second electrode catalyst layer 26a to the first gas diffusion layer 25b and the second gas diffusion layer 26b, respectively. Good.
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Manufacturing & Machinery (AREA)
- General Chemical & Material Sciences (AREA)
- Electrochemistry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Sustainable Energy (AREA)
- Sustainable Development (AREA)
- Quality & Reliability (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Fuel Cell (AREA)
- Inert Electrodes (AREA)
Abstract
Description
例えば上記実施形態では、第1ガス拡散層25b及び第2ガス拡散層26bと第1電極触媒層25a及び第2電極触媒層26aとをそれぞれ接合することによってアノード電極体25及びカソード電極体26を製造する場合について説明したが、本発明はこれに限らない。これら第1ガス拡散層25b及び第2ガス拡散層26bには、第1電極触媒層25a及び第2電極触媒層26aを塗工することによってアノード電極体25及びカソード電極体26を製造してもよい。
C…カメラ
1…燃料電池スタック(製品)
2…燃料電池セル(製品)
21…MEA(構造体)
24…電解質膜(イオン交換層)
25…アノード電極体(構造体、アノード電極体)
25a…第1電極触媒層(材料、アノード電極材)
25b…第1ガス拡散層(ワーク、第1ガス拡散層)
26…カソード電極体(構造体、カソード電極体)
26a…第2電極触媒層(材料、カソード電極材)
26b…第2ガス拡散層(ワーク、第2ガス拡散層)
5…コンピュータ
51…特徴ベクトル演算部
52…初期登録部
53…更新登録部
55…記憶媒体
CA,CC…カーボンペーパ(ワーク)
Claims (6)
- カーボンペーパを基材とするワークに材料を接合又は塗工する材料形成工程を経て製造される構造体又は当該構造体を部品とする製品をコンピュータによって管理する製品管理方法であって、
複数枚のワークの各々に対し前記材料が接合又は塗工されない部分における特定箇所の画像を撮像することによって1次ワークデータを取得し、取得した各ワークの1次ワークデータを記憶媒体に記憶させる登録工程と、
前記登録工程を経た後のワークの前記特定箇所の画像を撮像することによって参照ワークデータを取得し、当該参照ワークデータを前記記憶媒体に記憶されたデータと照合する照合工程と、を備えることを特徴とする製品管理方法。 - 前記特定箇所は、前記ワークのうち前記材料が接合又は塗工されない面の一部であることを特徴とする請求項1に記載の製品管理方法。
- 前記登録工程及び前記材料形成工程を経た後のワークの前記特定箇所の画像を撮像することによって2次ワークデータを取得し、当該2次ワークデータと前記記憶媒体に記憶された複数の1次ワークデータとを照合し、最も一致度合いが高い1次ワークデータと前記2次ワークデータとを関連付けて前記記憶媒体に記憶させる更新工程をさらに備えることを特徴とする請求項1又は2に記載の製品管理方法。
- 第1ガス拡散層にアノード電極材が接合又は塗工されたアノード電極体と、第2ガス拡散層にカソード電極材が接合又は塗工されたカソード電極体と、前記アノード電極体と前記カソード電極体との間に挟まれたイオン交換層と、を有する一体電極構造体又は当該一体電極構造体を部品とする製品を管理する製品管理方法であって、
複数枚の第1ガス拡散層の各々に対し前記アノード電極材が接合又は塗工されない部分における第1特定箇所の画像を撮像することによって1次アノード電極データを取得し、取得した各第1ガス拡散層の1次アノード電極データを記憶媒体に記憶させるアノード側登録工程と、
複数枚の第2ガス拡散層の各々に対し前記カソード電極材が接合又は塗工されない部分における第2特定箇所の画像を撮像することによって1次カソード電極データを取得し、取得した各第2ガス拡散層の1次カソード電極データを前記記憶媒体に記憶させるカソード側登録工程と、
前記アノード側登録工程を経た第1ガス拡散層の前記第1特定箇所又は前記カソード側登録工程を経た第2ガス拡散層の前記第2特定箇所の画像を撮像することによって参照電極データを取得し、当該参照電極データを前記記憶媒体に記憶されたデータと照合する照合工程と、を備えることを特徴とする製品管理方法。 - 前記一体電極構造体に含まれる第1ガス拡散層の1次アノード電極データと当該一体電極構造体に含まれる第2ガス拡散層の1次カソード電極データとを関連付けて前記記憶媒体に記憶させるセル管理工程をさらに備えることを特徴とする請求項4に記載の製品管理方法。
- 複数の前記一体電極構造体を積層して構成される積層体に含まれる複数枚の第1ガス拡散層の各々の1次アノード電極データと当該積層体に含まれる複数枚の第2ガス拡散層の各々の1次カソード電極データとを関連付けて前記記憶媒体に記憶させるスタック管理工程をさらに備えることを特徴とする請求項4又は5に記載の製品管理方法。
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201980086314.9A CN113228357B (zh) | 2018-12-28 | 2019-12-26 | 产品管理方法 |
| US17/418,228 US12045976B2 (en) | 2018-12-28 | 2019-12-26 | Product management method |
| JP2020562437A JP7128295B2 (ja) | 2018-12-28 | 2019-12-26 | 製品管理方法 |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2018247340 | 2018-12-28 | ||
| JP2018-247340 | 2018-12-28 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020138339A1 true WO2020138339A1 (ja) | 2020-07-02 |
Family
ID=71127789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2019/051245 Ceased WO2020138339A1 (ja) | 2018-12-28 | 2019-12-26 | 製品管理方法 |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US12045976B2 (ja) |
| JP (1) | JP7128295B2 (ja) |
| CN (1) | CN113228357B (ja) |
| WO (1) | WO2020138339A1 (ja) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210264153A1 (en) * | 2020-02-21 | 2021-08-26 | CACI, Inc.- Federal | Machine learning method and apparatus for detection and continuous feature comparison |
| CN118023057B (zh) * | 2024-04-15 | 2024-06-21 | 湖南隆深氢能科技有限公司 | 应用于连续化卷对卷mpl涂布生产线的自寻优方法及其系统 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09173935A (ja) * | 1995-12-25 | 1997-07-08 | Matsushita Electric Works Ltd | フローコート塗装状態の検出方法およびその装置 |
| JP2007291535A (ja) * | 2006-04-21 | 2007-11-08 | Toray Ind Inc | 炭素繊維布帛の検査装置および検査方法 |
| JP2008089534A (ja) * | 2006-10-05 | 2008-04-17 | Toray Ind Inc | 炭素繊維布帛の検査方法および検査装置 |
| JP2010169528A (ja) * | 2009-01-22 | 2010-08-05 | Japan Gore Tex Inc | 蛍光x線分析(xrf)を用いた積層体の表裏を判別する方法 |
| JP2019003824A (ja) * | 2017-06-15 | 2019-01-10 | 東レ株式会社 | 多孔質炭素シート、それを有するガス拡散電極基材、及び多孔質炭素シートの製造方法 |
| JP2019155238A (ja) * | 2018-03-09 | 2019-09-19 | 株式会社Screenホールディングス | 塗工装置および塗工方法 |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000012758A1 (en) * | 1998-09-02 | 2000-03-09 | Diadexus Llc | A novel method of diagnosing, monitoring, staging, imaging and treating various cancers |
| JP2010016952A (ja) | 2008-07-02 | 2010-01-21 | Daikin Ind Ltd | 界磁子、圧縮機、送風機及び空気調和機 |
| KR101230993B1 (ko) * | 2010-12-02 | 2013-02-07 | 기아자동차주식회사 | 연료전지 스택 부품 핀홀 검출 시스템 |
| KR20150045071A (ko) * | 2013-10-18 | 2015-04-28 | 현대자동차주식회사 | 막전극접합체 제조 장치 |
| EP3076366A4 (en) * | 2014-02-03 | 2017-05-24 | Prosper Creative Co., Ltd. | Image inspecting device and image inspecting program |
| JP6217614B2 (ja) | 2014-12-11 | 2017-10-25 | トヨタ自動車株式会社 | 膜電極接合体の検査方法 |
| JP6729111B2 (ja) | 2015-10-21 | 2020-07-22 | 日本電気株式会社 | 識別装置、トレーサビリティシステムおよび識別方法 |
| JP6852488B2 (ja) | 2017-03-21 | 2021-03-31 | 日本電気株式会社 | 識別装置、識別方法およびプログラム |
| US10789705B2 (en) * | 2018-03-26 | 2020-09-29 | Kabushiki Kaisha Toshiba | Quality monitoring system |
| US10769770B2 (en) * | 2018-05-07 | 2020-09-08 | Cummins Enterprise Llc | Quality monitoring system and quality monitoring method for fuel cell manufacturing line and quality monitoring system for manufacturing line |
| US11088382B2 (en) * | 2018-11-30 | 2021-08-10 | Bloom Energy Corporation | Fuel cell stack including witness marks and inspection method thereof |
-
2019
- 2019-12-26 WO PCT/JP2019/051245 patent/WO2020138339A1/ja not_active Ceased
- 2019-12-26 CN CN201980086314.9A patent/CN113228357B/zh active Active
- 2019-12-26 JP JP2020562437A patent/JP7128295B2/ja active Active
- 2019-12-26 US US17/418,228 patent/US12045976B2/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09173935A (ja) * | 1995-12-25 | 1997-07-08 | Matsushita Electric Works Ltd | フローコート塗装状態の検出方法およびその装置 |
| JP2007291535A (ja) * | 2006-04-21 | 2007-11-08 | Toray Ind Inc | 炭素繊維布帛の検査装置および検査方法 |
| JP2008089534A (ja) * | 2006-10-05 | 2008-04-17 | Toray Ind Inc | 炭素繊維布帛の検査方法および検査装置 |
| JP2010169528A (ja) * | 2009-01-22 | 2010-08-05 | Japan Gore Tex Inc | 蛍光x線分析(xrf)を用いた積層体の表裏を判別する方法 |
| JP2019003824A (ja) * | 2017-06-15 | 2019-01-10 | 東レ株式会社 | 多孔質炭素シート、それを有するガス拡散電極基材、及び多孔質炭素シートの製造方法 |
| JP2019155238A (ja) * | 2018-03-09 | 2019-09-19 | 株式会社Screenホールディングス | 塗工装置および塗工方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| JPWO2020138339A1 (ja) | 2021-11-04 |
| US20220085385A1 (en) | 2022-03-17 |
| JP7128295B2 (ja) | 2022-08-30 |
| CN113228357B (zh) | 2025-01-03 |
| US12045976B2 (en) | 2024-07-23 |
| CN113228357A (zh) | 2021-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5415017B2 (ja) | 二次電池、二次電池の製造方法、及び製造システム | |
| US8288059B2 (en) | Processing methods and systems for assembling fuel cell perimeter gaskets | |
| US8012284B2 (en) | Method and apparatus for fabricating roll good fuel cell subassemblies | |
| CN112771691B (zh) | 电极组件及其检查方法 | |
| JP7128295B2 (ja) | 製品管理方法 | |
| US9425476B2 (en) | Manufacture method of membrane electrode assembly for fuel cell | |
| US20160329591A1 (en) | Membrane-electrode assembly manufacturing apparatus of fuel cell | |
| JP5348388B2 (ja) | 燃料電池の製造方法 | |
| US11670793B2 (en) | Fuel cell stack including witness marks and inspection method thereof | |
| KR20210134166A (ko) | 이차전지의 전극셀 제조방법 | |
| US7569082B2 (en) | Manufacture of unitized electrode assembly for PEM fuel cells | |
| JP2014041806A (ja) | 燃料電池スタックの自動積層整列装置及び方法 | |
| EP0996186A2 (en) | Electrolyte for solid oxide fuel cells | |
| DE102015220769B4 (de) | System zum Prüfen der Qualität einer Membranelektrodenanordnung einer Brennstoffzelle und Qualitätsprüfungsverfahren desselben | |
| JP7315789B2 (ja) | 膜-電極アセンブリの製造のための方法及び装置 | |
| Schäfer et al. | Genetic algorithm for the optimization of vision acquisition for on-the-fly position measurement of individual layers in fuel cell stack assembly | |
| KR20210078380A (ko) | 전극의 교대 공급에 의한 단위셀의 제조방법 | |
| DE102012215657A1 (de) | Verfahren zur Herstellung einer Wasserdampftransfermembran | |
| DE112007000899T5 (de) | Verfahren zum Entfernen einer Dichtplatte für eine Brennstoffzelle, eine Dichtplatte und Vorrichtung zum Entfernen derselben | |
| JP6834856B2 (ja) | 膜電極接合体の製造方法 | |
| JP4960903B2 (ja) | 積層体の検査方法、検査装置及び検査プログラム | |
| CN111707674B (zh) | 一种电堆元件识别方法、电堆元件识别装置和系统 | |
| JP2009277501A (ja) | 燃料電池の検査装置 | |
| CN118522931A (zh) | 用于自动堆叠电化学系统的不同部件的生产系统 | |
| KR20250066939A (ko) | 전극 검사장치, 전극 코팅시스템, 및 그것을 사용한 전극 검사방법 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19905005 Country of ref document: EP Kind code of ref document: A1 |
|
| ENP | Entry into the national phase |
Ref document number: 2020562437 Country of ref document: JP Kind code of ref document: A |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19905005 Country of ref document: EP Kind code of ref document: A1 |
|
| WWG | Wipo information: grant in national office |
Ref document number: 201980086314.9 Country of ref document: CN |