KR20220113009A - Failure anaysis system and method for lamination process - Google Patents

Abstract

A defect analysis system for a lamination process includes: a vision data pre-processing unit that pre-processes the vision data of a vision inspection file necessary for determining pass/fail to create a plurality of vision data groups and determines whether or not a product is defective according to product inspection conditions; a process data pre-processing unit that pre-processes process data including a process equipment event history, material replacement history, etc. according to vision data; a mapping unit for mapping corresponding information among process equipment events and material replacement records based on the number of units and manufacturing dates of the corresponding equipment for an n number of vision data groups among the plurality of vision data groups; and an analysis unit that visualizes the mapped data based on the mapping result, analyzes data singularities, and derives a defect rate and defect type based on the process equipment events and material replacement history. It is possible to automatically classify lamination defect types.

Description

FAILURE ANAYSIS SYSTEM AND METHOD FOR LAMINATION PROCESS

The present disclosure relates to a failure analysis system and method for a lamination process.

The lamination process is a process of attaching an anode electrode, a cathode electrode, and a separator and cutting according to specifications. The inspection target part of the electrode is measured by the inspection facility through the vision inspection at the end of the process to see if the electrode is positioned according to the standard, and a judgment of good or bad is performed according to the measurement result.

In the prior art, there is a method in which a defect type is automatically determined through Excel. However, using Excel for analyzing a large amount of data has limitations in the amount and speed of data processing. Practically, it is difficult to analyze even a day's worth of data generated in the process with the prior art.

An object of the present invention is to provide a defect analysis system and method for a lamination process that can automatically classify lamination defect types.

The failure analysis system for the lamination process according to one aspect of the invention pre-processes the vision data of the vision inspection file necessary for determining whether or not the product is defective, and creates a plurality of vision data groups, and determines whether the product is defective according to the product inspection conditions. a vision data preprocessing unit that preprocesses process data including process facility event history, material replacement history, etc. according to vision data, and a corresponding facility for n vision data groups among the plurality of vision data groups A mapping unit that maps corresponding information among process equipment events and material replacement histories based on the expiration date and manufacturing date of It may include an analysis unit for deriving the defective rate and defective type by .

The vision data preprocessor calculates a defect rate of each of the plurality of vision data groups according to the product inspection condition, and based on the calculated defect rate, a defect type for the n vision data groups among the plurality of vision data groups and corresponding The defect rate can be derived.

The vision data preprocessor may select n vision data groups having a relatively high defect rate among the plurality of vision data groups, or select n vision data groups having a higher defect rate than a predetermined reference value as the n vision data groups.

The defect types may include an X-axis defect, a Y-axis defect, a Tab 　 defect, a mismatch defect, a multiple defect mismeasurement defect, a misrecognition defect, multiple mismeasurement defect, multiple misrecognition defect, and a misclassification defect.

The process data preprocessor may classify the process facility event history and material replacement history based on the number of units generated and the manufacturing date of the vision data of the vision inspection file.

The failure analysis system may further include a process data server for storing data on production status for each process, process facility event history, material replacement history, and the like.

The failure analysis system may further include a vision data server that stores a vision inspection file including vision data obtained by vision measurement of an inspection cell that has completed the lamination process and data on product inspection conditions.

The defect analysis method for the lamination process according to another feature of the invention includes a step in which data about the production status by process, process facility event history, material replacement history, etc. are collected, vision measurement data of the inspection cell that has completed the lamination process A step in which data on the vision inspection file and product inspection conditions are collected, the vision data of the vision inspection file is generated based on the production unit and manufacturing date, and the inspection cell corresponding to the vision data is manufactured based on the manufacturing unit unit and manufacturing date. classifying equipment event history and material replacement history; grouping the vision data of the vision inspection file into a predetermined number unit to generate a plurality of vision data groups, and calculating a defect rate of each of the plurality of vision data groups; and and mapping corresponding information of process equipment events and material replacement histories to each of the n vision data groups among the plurality of vision data groups based on the unit number and manufacturing date of each equipment.

The failure analysis method may further include selecting the n vision data groups from among the plurality of vision data groups based on the calculated failure rate.

The defect analysis method may further include visualizing mapped data based on the mapping result, analyzing data outliers, and deriving a defect rate and defect type in a process facility event and analysis system.

A defect analysis system and method for a lamination process that can automatically classify lamination defect types are provided.

1 is a diagram illustrating a failure analysis system according to an exemplary embodiment.
2 is a view showing a vision inspection for the lower surface of the inspection cell.
3 is a flowchart illustrating a defect analysis method according to an exemplary embodiment.
4 and 5 are diagrams illustrating a trend analysis result by an analysis unit implemented through a SAS Viya program according to an exemplary embodiment.

The present invention can reduce the processing time of a large amount of data by using the analysis platform. By reducing the processing time, it is possible to analyze daily, weekly, monthly and unit-by-unit. As the analysis platform, a SAS program, which is one of the analysis programs, may be used.

The present invention can automatically classify defective types into a predetermined number based on measurement data. According to the present invention, data in which classification is completed and data in a specific server (eg, an in-house server) can be linked. In the present invention, information such as production performance and facility alarm history can be added in matching server data and classified data. These information may be information to assist the operator in determining whether or not the lining process is successful. Data that has been classified and matched can be visualized and analyzed through a GUI platform. A SAS Viya program can be used as a GUI platform.

Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but identical or similar components are given the same and similar reference numerals, and overlapping descriptions thereof will be omitted. The suffixes “module” and/or “part” for the components used in the following description are given or mixed in consideration of only the ease of writing the specification, and do not have distinct meanings or roles by themselves. In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in the present specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical spirit disclosed herein is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present invention , should be understood to include equivalents or substitutes.

Terms including ordinal numbers such as first, second, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.

When a component is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but it is understood that other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.

In the present application, terms such as “comprises” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It is to be understood that this does not preclude the possibility of the presence or addition of numbers, steps, operations, components, parts, or combinations thereof.

1 is a diagram illustrating a failure analysis system according to an exemplary embodiment.

As shown in FIG. 1 , the failure analysis system 1 includes a process data server 2 , a vision data server 3 , a vision data preprocessor 10 , a process data preprocessor 20 , and a mapping unit 30 . , and an analysis unit 40 .

The process data server 2 stores data on production status by process, process facility event history, material replacement history, and the like. The production status for each process is collected in the process data server 2, and whenever an event history such as a facility alarm occurs, a process facility event history including detailed information about the process facility event is collected in the process data server 2 And, whenever an event in which a material is replaced occurs, a material replacement event history including information about the material after replacement may be collected in the process data server 2 . A facility alarm may occur when an abnormality occurs in the facility.

The vision data server 3 includes a vision inspection file including vision measurement data (hereinafter, vision data) of a bi-cell or mono-cell that has completed the lamination process and data on product inspection conditions save the In one embodiment, it will be described that a vision inspection is performed on a bi-cell or a mono-cell that has completed the lamination process. However, the present invention is not limited thereto, and the present invention may be applied to other types of cells other than bi-cell or mono-cell. Hereinafter, the bi-cell or mono-cell is substituted for the bi-cell or mono-cell as a test cell collectively referred to as a bi-cell or mono-cell.

2 is a view showing a vision inspection for the upper / lower surface of the inspection cell.

As shown in (a) and (b) of FIG. 2 , a plurality of inspection regions may be specified on a photographing screen for an upper surface and a lower surface, and a length of an electrode may be measured in each inspection region. The vision inspection file may include vision data including electrode lengths measured in a plurality of inspection areas for the upper and lower surfaces of the inspection cell.

The vision data pre-processing unit 10 pre-processes vision data of a vision file necessary for determining whether or not it is acceptable. Determination of failure or failure is performed by acquiring vision data for an inspection cell that has completed a lamination process during a battery production process, and comparing values corresponding to a plurality of measurement items in the vision data with a predetermined reference value. The battery cell is configured by packing a plurality of test cells. The test cell includes at least one unit structure in which an anode, a separator, and a cathode are stacked. The test cell may be implemented by stacking a plurality of unit structures, and a separator may be positioned between the unit structures.

The vision data pre-processing unit 10 derives the vision inspection file and product inspection conditions for judging whether or not it is acceptable from the vision data server 3, and groups the vision data of the vision inspection file in a predetermined number unit to generate a plurality of vision data groups. and calculates the defective rate of each of the plurality of vision data groups, and derives a defect type and a corresponding defective rate for n (a natural number greater than or equal to 1) vision data groups among the plurality of vision data groups based on the calculated defective rate. The vision data preprocessor 10 may set n pieces of the plurality of vision data groups having a relatively high defect rate or n pieces having a defect rate greater than or equal to a predetermined reference value as the n vision data groups.

The defect types are shown in Table 1 below.

X-axis defective Defective X-axis item (within the product inspection condition range 　2 times) Y-axis defective Y-axis item defective (within product inspection condition range 　2 times) Tab　Bad Tab 　 Item defective (within product inspection condition range 　 2 times) bad mismatch Bad mismatch multiple bad X or Y or Tab or 　 mismatch is bad at the same time wrong measurement Defects outside the product inspection condition range 　2 times misrecognition Defects that are out of ROI (Region of Interest) 　 ‘0’ 　 Multiple 　 Mismeasurement 　 Defective Multiple items 　 Measured 　 Defective Multiple 　 Misrecognition 　 Bad Multiple items 　 Misrecognized 　 Defective unclassified poor Defects that are not included in the above defect types

In Table 1 above, “within twice the range of product inspection conditions” is classified as an erroneous measurement when it exceeds the product inspection conditions twice. it means to be For example, when the difference between the vision data of a specific part of the inspection cell in the X-axis direction and the reference value is A, if A is more than twice as high as the product inspection condition B, the vision data obtained by erroneous measurement Since it is judged by vision data, the type of defect is classified as a mis-measurement defect, not an X-axis defect. Referring to (a) and (b) of Figure 2, the X-axis defect is a type in which the vision measurement values of A1U-A4U and A1L-A4L are out of the normal range, and the Y-axis defect is A5U, A6U, A5L, and A6L. The vision measurement value of the part is out of the normal range. Tab failure is the type where the vision measurement value of the TW, TW-I, TS, TSU, TH_A part is out of the normal range. L, and the vision measurement value of the SF2U/L part may be of a type outside the normal range.

The process data pre-processing unit 20 pre-processes process data including a production status for each battery production process, a process facility event history, a material replacement history, and the like according to the vision data. The process facility event history may include various information regarding the operation of the facility, such as the operation history and failure of each facility. For example, when an alarm is generated by detecting an abnormality in the equipment, when the equipment is restarted after being stopped, etc. may be included in the process equipment event history.

The process data preprocessing unit 20 derives the vision inspection file from the vision data server 3 and the process data from the process data server 2, and the process data based on the number of units and the manufacturing date in which the vision data of the vision inspection file was generated. classify The process data preprocessor 20 may classify the process equipment event history and material replacement history based on the equipment number and manufacturing date in which the inspection cell corresponding to the vision data is manufactured.

The mapping unit 30 maps corresponding information among process facility events and material replacement histories based on the expiration date and manufacturing date of each facility for each of the n vision data groups. The mapping unit 30 maps process facility event history and material history to all vision data constituting each of the n vision data groups, based on the expiration date and manufacturing date of the facility in which the corresponding inspection cell is manufactured. For example, the mapping unit 30 maps the process facility event history in the facility expiration date and manufacturing date in which the inspection cell in which the X-axis defect occurred, to the corresponding vision data. Alternatively, the mapping unit 30 maps the material replacement history in the equipment unit number and manufacturing date in which the inspection cell in which the X-axis defect occurred, to the corresponding vision data.

The analysis unit 40 may visualize the mapped data based on the mapping result and analyze the data singularity to derive the defect rate and defect type due to the process facility event and material replacement history.

3 is a flowchart illustrating a defect analysis method according to an exemplary embodiment.

By the process data server 2, data on the production status by process, process facility event history, material replacement history, etc. are collected (S1).

By the vision data server 3, a vision inspection file including data obtained by vision measurement of an inspection cell that has completed the lamination process and data on product inspection conditions are collected (S2).

By the process data pre-processing unit 20, the process equipment event history and material replacement are based on the number of units and the manufacturing date in which the vision data of the vision inspection file is generated, and the unit number and manufacturing date of the inspection cell corresponding to the vision data The history is classified (S3).

A plurality of vision data groups are generated by grouping the vision data of the vision inspection file by a predetermined number by the vision data pre-processing unit 10 , and a defect rate of each of the plurality of vision data groups is calculated, based on the calculated defect rate Defect types and corresponding defect rates for at least n (a natural number equal to or greater than 1) vision data groups among the plurality of vision data groups are derived (S4). Defect types may include X-axis defective, Y-axis defective, Tab 　 defective, mismatch defect, multiple defective mismeasurement defect, misrecognition defect, multiple misrecognition defect, multiple misrecognition defect, and unclassified defect.

By the mapping unit 30 , corresponding information among process equipment events and material replacement histories is mapped to each of the n vision data groups based on the expiration date and manufacturing date of each equipment ( S5 ).

By the analysis unit 40, the mapped data is visualized based on the mapping result, and the data singularity is analyzed to derive the defect rate and defect type by the process equipment event history and material replacement history (S6). As a result of the analysis, a defect type trend can be derived.

The vision data preprocessor 10 , the process data preprocessor 20 , and the mapping unit 30 may be implemented through a SAS EG program. The analysis unit 40 may be implemented through a SAS Viya program.

4 and 5 are diagrams illustrating a trend analysis result by an analysis unit implemented through a SAS Viya program according to an exemplary embodiment.

As shown in FIG. 4 , as a result of performing a vision inspection on 550,000 via cells, 9,700 defective via cells were generated. The defect rate is 1.76%.

At this time, the result was analyzed that the defect rate caused by the process facility event history “start after stop” was 62%, and the defect rate caused by the process facility event history “normal operation” was 38%.

In addition, it can be seen that “continuous defect” is 68.8% and “single defect” is 31.2%, and the defect rate for each defect type can be easily known.

In addition, defect types and defect rates are also visualized according to the manufacturing date, so that the defect types and trends can be easily identified.

As shown in Figure 5, the defect rate at each facility, the defect rate according to the process facility event history (startup after stop, normal operation) at each facility, and the continuous defect rate and single defect rate at each facility are visualized as graphs. have. In addition, the types of defects by manufacturing date in a specific facility are shown in a graph.

As such, the defect analysis system according to an embodiment derives the defect type and defect rate for a large amount of via cell by automatically analyzing the vision data, and uses the process facility event and material history and vision data based on the equipment number and manufacturing date. By mapping, it is possible to analyze the defect trend.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those of ordinary skill in the field to which the present invention belongs are also entitled to the rights of the present invention. belong to the scope

1: Defect analysis system
2: Process data server
3: Vision Data Server
10: Vision data preprocessor
20: process data preprocessor
30: mapping unit
40: analysis unit

Claims (10)

In the failure analysis system for the lamination process,
a vision data pre-processing unit that pre-processes vision data of a vision inspection file necessary for determining whether a product is defective, generates a plurality of vision data groups, and determines whether a product is defective according to product inspection conditions;
a process data preprocessor for preprocessing process data including process facility event history and material replacement history according to vision data;
a mapping unit for mapping n vision data groups from among the plurality of vision data groups to corresponding information among process equipment events and material replacement histories based on the number of units and manufacturing dates of the corresponding equipment; and
and an analysis unit that visualizes mapped data based on the mapping result and analyzes data singularities to derive a defect rate and defect type based on a process facility event and material replacement history. According to claim 1,
The vision data preprocessor,
Calculating the defect rate of each of the plurality of vision data groups according to the product inspection condition, and deriving a defect type and a corresponding defect rate for the n vision data groups among the plurality of vision data groups based on the calculated defect rate analysis system. 3. The method of claim 2,
The vision data preprocessor,
and selecting n vision data groups having a relatively high defect rate among the plurality of vision data groups or selecting n units having a higher defect rate than a predetermined reference value as the n vision data groups. According to claim 1,
The defect type includes an X-axis defect, a Y-axis defect, a tab defect, a mismatch defect, a multiple defect mismeasurement defect, a misrecognition defect, multiple mismeasurement defect, multiple misrecognition defect, and an unclassified defect. According to claim 1,
The process data preprocessor,
A failure analysis system for classifying a process facility event history and material replacement history based on the number of units and the manufacturing date in which the vision data of the vision inspection file is generated. According to claim 1,
Further comprising a process data server for storing data on the production status for each process, process equipment event history, material replacement history, etc., failure analysis system. According to claim 1,
The failure analysis system, further comprising a vision inspection file containing vision data obtained by vision measurement of the inspection cell that has completed the lamination process, and a vision data server for storing data on product inspection conditions. In the failure analysis method for the lamination process,
collecting data on production status by process, process facility event history, material replacement history, etc.;
collecting a vision inspection file including vision measurement data of an inspection cell that has completed the lamination process and data on product inspection conditions;
classifying the process facility event history and material replacement history based on the number of units and the manufacturing date in which the inspection cell corresponding to the vision data was manufactured and the manufacturing date based on the unit number and manufacturing date in which the vision data of the vision inspection file was generated;
generating a plurality of vision data groups by grouping the vision data of the vision inspection file by a predetermined number, and calculating a defect rate of each of the plurality of vision data groups; and
and mapping corresponding information of process equipment events and material replacement histories to each of the n vision data groups among the plurality of vision data groups based on the unit number and manufacturing date of each equipment. 9. The method of claim 8,
and selecting the n vision data groups from among the plurality of vision data groups based on the calculated defect rate. 9. The method of claim 8,
The method of claim 1, further comprising: visualizing mapped data based on the mapping result and analyzing data outliers to derive a defective rate and a defective type in a process facility event and analysis system.

Images