CN113592038B - Method and system for dividing solid wood customized cabinet door part family - Google Patents
Method and system for dividing solid wood customized cabinet door part family Download PDFInfo
- Publication number
- CN113592038B CN113592038B CN202111001660.8A CN202111001660A CN113592038B CN 113592038 B CN113592038 B CN 113592038B CN 202111001660 A CN202111001660 A CN 202111001660A CN 113592038 B CN113592038 B CN 113592038B
- Authority
- CN
- China
- Prior art keywords
- attribute
- parts
- classifying
- classified
- cabinet door
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
Abstract
The invention discloses a method and a system for dividing a solid wood customized cabinet door part family, and belongs to the technical field of furniture production. The method comprises the steps of obtaining attribute indexes of a solid wood customized cabinet door part family; sorting the attribute indexes according to the importance degree; classifying the attribute indexes, wherein the sizes are classified by adopting a hierarchical clustering method, the processes are classified by adopting a fuzzy clustering method, the mortises are classified according to the early-stage classification conditions of the mortises, and the bending degrees are classified according to the bending radii; and sorting the clustering results of the attribute indexes according to the classified character strings. The system comprises an attribute index acquisition unit, an attribute index sorting unit, an attribute index classifying unit and a clustering result sorting unit, wherein the attribute index classifying unit comprises a size classifying unit, a process classifying unit, a tongue-and-groove structure classifying unit and a bending structure classifying unit. The invention can divide the solid wood customized cabinet door part families, thereby overcoming the defect of single index classification.
Description
Technical Field
The invention belongs to the technical field of furniture production, and particularly relates to a method and a system for dividing a solid wood customized cabinet door part family.
Background
In the large-scale customization industry environment, the solid wood customization furniture industry needs to balance the relationship between multiple varieties and mass production. The grouping technology is to cluster the parts into groups according to a certain classification principle by utilizing the attribute similarity among the parts, and participate in design, production and manufacture in the form of a part family. The reasonable part grouping mode is the most critical ring in the grouping technology, part of the existing researches are developed on the combination of the attributes such as shapes and processes and the like with the methods such as genetic algorithm, fuzzy clustering algorithm, graph theory, neural network and the like, but most of the researches are carried out on single attributes, and the various attributes involved in grouping are not brought together for analysis.
The parts of the solid wood customized cabinet door furniture need to consider various factors such as size, modeling, process, tongue-and-groove structure and the like during production and manufacturing, and the single-index grouping mode has limitation in production.
Disclosure of Invention
The technical problem is as follows: the invention provides a method and a system capable of performing multi-attribute division on a part family of solid wood customized furniture, so that the one-sidedness of single-attribute classification in the prior art is overcome.
The technical scheme is as follows: on one hand, the invention provides a method for dividing a solid wood customized cabinet door part family, which comprises the following steps:
obtaining attribute indexes of a solid wood customized cabinet door part family, wherein the attribute indexes comprise size, process, tongue-and-groove structure and bending degree;
sorting the attribute indexes according to the importance degree;
classifying the attribute indexes, wherein the sizes are classified by adopting a hierarchical clustering method, the processes are classified by adopting a fuzzy clustering method, the mortises are classified according to the early-stage classification conditions of the mortises, and the bending degrees are classified according to the bending radii;
and sorting the clustering results of the attribute indexes according to the classified character strings.
Further, the method for classifying the size by adopting the hierarchical clustering method comprises the following steps:
(1) defining the length, width and thickness of the part after the disassembly of the customized solid wood cabinet door as x, y and z values respectively;
(2) normalizing the data by a maximum normalization method, wherein:
wherein x isSign board、ySign board、zSign boardRespectively representing the values of length, width and thickness of the standardized part, Xmax、Ymax、ZmaxRespectively representing the maximum values of the length, the width and the thickness of the part;
(3) measuring the similarity between any two parts after standardized treatment by adopting a squared Euclidean distance;
(4) data points or categories with similar distances are combined into a group, the distance between the groups is measured by using the average distance between the groups, the similar categories are combined, and a classification result is determined according to requirements.
Further, the method for classifying the processes by adopting the fuzzy clustering method comprises the following steps:
(1) constructing a part-process matrix; defining a set A of furniture parts to be classifiedn＝﹛a1，a2，……，anCompare, enterprise's cabinet door production process set Bm＝﹛b1，b2，……，bmPartition, build part-process matrix as:
(2) constructing a part similarity coefficient matrix; determining the similarity coefficient between the two parts by adopting a maximum and minimum method, wherein the formula is as follows:
wherein x isijRepresenting the similarity factor of part i and part j belonging to set AnTwo parts with different inner parts, process k belongs to the production process set Bm，eikValue of part i on Process k, ejkIs the value of part j on process k; the similarity matrix between parts is:
(3) calculating a transfer closure, and solving a component similarity coefficient matrix X by using a transfer closure methodijThe fuzzy equivalence matrix T;
(4) intercepting the fuzzy equivalent matrix T by using a threshold lambda to obtain grouping conditions of different thresholds, determining the grouping number, and counting different clustering results to obtain a dynamic clustering chart of the fuzzy equivalent matrix T;
(5) calculating the similarity coefficients within the groups by constructing the average similarity coefficient C within each groupzDetermining the mean intra-group similarity coefficient CpBy using CpAnd judging the process similarity of the parts in the group, wherein the formula is as follows:
Cz＝(∑q∈H∑s∈HXqs)/h2，
wherein g represents the number of groups intercepted under a certain threshold lambda, H represents the number of parts in each group after grouping, H is a part set in the group after grouping, q and s represent two parts of the H set, q, s belongs to H, XqsA similarity matrix representing part q and part s.
Further, the method for classifying the mortises according to the early classification condition of the mortises comprises the following steps:
according to the connecting mode of the parts, the connecting structure is divided and comprises four edges and a core plate, four edges and lines, the mortise structure of the obtained part comprises mortises of the four edges, core plate mortises and line mortises, and the mortises of the same type are classified into one type.
Further, the method for classifying the bending degree according to the bending radius and the corner radian radius comprises the following steps: and analyzing the parts with the bent shapes in the parts, and dividing the parts into classes according to the bending radius and the corner radian radius.
Further, the method for sorting the clustering results of the attribute indexes according to the divided category character strings comprises the following steps: all attribute indexes are brought into multi-attribute cross clustering, parts corresponding to character strings with the same sequence are classified into a group, and the group number formed after removing the empty set is the total group number of the part group division.
Further, the attribute indexes are sorted by adopting an AHP analytic hierarchy process.
On the other hand, the invention provides a system for dividing a part family of a solid wood customized cabinet door, which is characterized by comprising the following steps of:
the attribute index acquisition unit is used for acquiring attribute indexes of a solid wood customized cabinet door part family, wherein the attribute indexes comprise size, process, tongue-and-groove structure and bending degree;
the attribute index sorting unit is used for sorting the attribute indexes according to the importance degree;
the attribute index classification unit is used for classifying each attribute index;
and the clustering result sorting unit is used for sorting the clustering results of the attribute indexes according to the divided category character strings.
Further, the attribute index classification unit includes:
the size classification unit is used for classifying the sizes by adopting a hierarchical clustering method;
the process classification unit is used for classifying the processes by adopting a fuzzy clustering method;
the mortise structure classification unit is used for classifying the mortises according to the early classification conditions of the mortises;
and the bent structure classifying unit is used for classifying the bending degree according to the bending radius.
Compared with the prior art, the invention has the following advantages: selecting four main classification attributes (size, process, tongue-and-groove structure and bending degree), sorting the importance degrees of the four main classification attributes by adopting an AHP method, classifying the categories of single indexes by adopting different methods (size-hierarchical clustering, process-fuzzy clustering, tongue-and-groove structure-connection mode structure, bending degree-bending radius and corner radian radius), formulating category codes, and counting classification results by adopting a mode of cross mixing the category codes. And the results of the specific examples show that: the method has clear classification result, overcomes the one-sidedness of single-index classification, can explain the attribute characteristics and significance of a part family in multiple aspects, and can select attribute classification on individual working sections according to actual production requirements to meet special requirements.
Drawings
Fig. 1 is a diagram illustrating a door panel of a cabinet according to an embodiment of the present invention;
FIG. 2 is a flow chart of a method for dividing a family of parts of a customized solid wood cabinet door according to an embodiment of the present invention;
FIG. 3 is a tree spectrum of sub-threshold grouping cases in an embodiment of the present invention;
FIG. 4 is a graph of the number of packets and the average similarity coefficient within a group at different thresholds according to an embodiment of the present invention;
FIG. 5 is a schematic view of a tongue and groove configuration according to an embodiment of the present invention;
fig. 6 is a block diagram of a system for dividing a family of parts of a customized solid wood cabinet door according to an embodiment of the present invention.
Detailed Description
For better explanation of the present invention, the embodiments of the present invention are explained in detail based on a specific example. In this specific example, a test sample is provided, specifically: 40 model customization solid wood cabinet door (26 types of log cabinet door, 14 types of compound solid wood door), flat core and compound core board, compound flat board and log flat board cabinet door are as the sample, and the product model is as shown in figure 1. After the customized solid wood cabinet door is disassembled, the parts are as follows: the method comprises the following steps of feeding, blanking, left feeding, right feeding, a core plate, lines, louver strips, louver frame transverse and vertical strips and a whole plate, 458 parts are obtained after disassembly, 314 solid wood parts are obtained after plastic clamping strips and hardware decoration strips are removed, and 205 solid wood parts with different modeling structures are obtained, and the three parts are used as test samples.
Fig. 2 shows a flow chart of a method for dividing a family of parts of a customized solid wood cabinet door according to an embodiment of the present invention. As shown in fig. 2, the method includes:
s100: and obtaining the attribute indexes of the solid wood customized cabinet door part family. In the embodiment of the invention, the obtained attribute indexes comprise dimensions, processes, tongue-and-groove structures and bending degrees, and the attribute indexes can be selected on individual sections according to actual production requirements.
S200: and sequencing the attribute indexes according to the importance degree. The degree of attention that may be paid to by designers may vary during the production of a solid wood custom door panel for the aforementioned attribute goals. Therefore, in the embodiment of the present invention, it is necessary to sort the attribute indexes according to the degree of importance. Specifically, in a specific example of the present invention, the attribute indexes are sorted by using an AHP analytic hierarchy process, and a manner of expert scoring, for example, a scoring system of 1 to 5 may be adopted, where 1 represents the least important and 5 represents the most important, and scoring evaluation is performed by a plurality of experts in the field, so as to count the importance degree of each attribute index and rank the attribute indexes. In the present example, the results are shown in Table 1, with a score of 10 experts.
Table 110 expert scoring
Then, AHP analytic hierarchy process is used for analysis, the result of AHP analysis is shown in table 2, and the result is subjected to consistency test and passes. The sizes, the processes, the tongue-and-groove structures and the bending degrees are determined according to the weight values (the size: 31.25%, the tongue-and-groove structure: 26.25%, the process: 27.50% and the bending degree: 15.00%).
TABLE 2 AHP level analysis results
S300: and classifying the attribute indexes. In the embodiment of the invention, the size is classified by adopting a hierarchical clustering method, and the classification method comprises the following steps:
(1) defining three dimensions of length, width and thickness of parts of the disassembled solid wood custom cabinet door as x, y and z values respectively;
(2) the data is normalized using a maximum normalization method, wherein,
(3) measuring the similarity between any two parts after standardized treatment by adopting a squared Euclidean distance; for example, part A (x) after standardized processingA,yA,zA) And part B (x)B,yB,zB) The similarity between A and B is defined as the distance d between A and BABAnd then:
dAB＝(xA-xB)2+(yA-yB)2+(zA-zB)2，dAB∈(0,3)
(4) data points or categories with similar distances are combined into a group, the distance between the groups is measured by using the average distance between the groups, the similar categories are combined, and a classification result is determined according to requirements.
In a specific example, the size modes of various models are taken as test data, the test data are taken into the sizes of parts (feeding, blanking, left feeding, right feeding, core plates, lines, louver strips, louver frame horizontal and vertical strips and whole plates) and then are disassembled, the size data of the parts are defined as a point set of a three-dimensional space, the similarity between the parts is determined by calculating the distance between the points, and the smaller the distance is, the higher the similarity is. Data points or categories that are close in distance are combined into groups.
The data in the specific example of the present invention were sorted using the method described above, and the length, width, and thickness dimensions of 205 parts are shown in table 3. And (3) carrying out standardization processing by using SPSS software, measuring the distance between each part, determining the hierarchical relationship according to the similarity, then, initially setting the clustering number to be 8, selecting the clustering number to be 4 by combining with the actual functional types, and carrying out classification shape description on each part.
TABLE 3 parts dimensional data (units: mm)
In this example, the dimensions are divided into 4 categories and the parts are distributed in each category as shown in Table 4. The A1 category comprises all four-edge materials, partial lines and whole plates, and the parts have larger difference in length, width and thickness and are wide and long materials; the B1 category comprises partial lines, louver strips and louver frame transverse and vertical materials, the parts are similar in width and thickness, have larger difference between length and width, and are slender materials; the C1 category comprises all core plates and partial whole plates, the width and length of the parts are similar, and the difference between the width and the length of the parts is larger, so that the parts are plate materials; the D1 category is a square bar.
TABLE 4 results of size classification
Further, in the embodiment of the present invention, the process is classified by fuzzy clustering, which includes:
(1) constructing a part-process matrix; defining a set A of furniture parts to be classifiedn＝﹛a1，a2，……，anCompare, enterprise's cabinet door production process set Bm＝﹛b1，b2，……，bmPartition, build part-process matrix as:
(2) a component part similarity coefficient matrix;
and determining the similarity coefficient between the two parts by using the collected part-process matrix, and determining the similarity coefficient by using a maximum and minimum method. x is the number ofijFor similarity coefficients of part i and part j, part i and part j belong to set AnInternal differenceTwo parts, Process k belonging to the production Process set Bm，eikValue of part i on Process k, ejkIs the value of part j on process k. The similarity factor for the two parts is therefore:
a similar coefficient may be understood as xijThe number of processes shared by the component i and the component j/(the number of processes shared by the component i and the component j + the process owned by the component i but not owned by the component j + the process owned by the component j but not owned by the component i). Therefore, the more process ratios both parts have, the larger the similarity coefficient. The similarity matrix between the parts is constructed by the following steps:
(3) computation transitive closure
Constructed part similarity coefficient matrix XijSatisfies symmetry and reflexibility, and XijThe fuzzy equivalent matrix T is obtained by using a transfer closed-packet method. Specifically, starting from the fuzzy similarity matrix X, the square X is solved in sequence2，X4,X6,X8,……，XRWhen X is presentR＝X2RWhen, XRFor the fuzzy equivalence matrix T, T ═ Tij)n×n。
(4) Intercepting the fuzzy equivalent matrix T by using a threshold lambda to obtain grouping conditions of different thresholds, determining the grouping number, and counting different clustering results to obtain a dynamic clustering chart of the fuzzy equivalent matrix T; in particular, the method comprises the following steps of,
λ truncated matrix for T, i.e.:
(5) calculating intra-group similarity coefficients
By intercepting the threshold lambda, a plurality of grouping modes can be obtained, and an average similarity coefficient C in each group is constructedzAnd use of CzConstruction of the mean intra-group similarity coefficient CpBy using CpThe process similarity of the parts in the group can be visually judged. CpThe larger the value, the higher the similarity coefficient in the average group, i.e. the better the grouping effect, wherein:
Cz＝(∑q∈H∑s∈HXqs)/h2，
wherein g represents the number of groups obtained by truncation under a certain threshold lambda, and g average similarity coefficients are obtained in total, that is to say
H is the number of parts in each group after grouping, H is a part set in each group after grouping, q and s represent two parts of the H set, q and s belong to H, XqsA similarity matrix representing part q and part s.
The processes in the specific example were classified using the method described above, and in this specific example, 33 statistical process routes were calculated, see table 5, using C language programming to assist in the calculation.
Table 533 representation part-Process matrix
Constructing a 33 x 33 similarity coefficient matrix and computing a transitive closure matrix, R10＝R12Therefore, 5 times of calculation are carried out to obtain a fuzzy equivalent matrix, the matrix is intercepted by using different threshold values lambda to obtain 15 classification results in total, and different threshold values are obtainedThe number of subgroups under value and the average similarity coefficient within a group are shown in fig. 3 and fig. 4. It can be seen from fig. 4 that as the threshold value increases, the number of groups and the average similarity coefficient within the groups gradually increase. In combination with actual production of a data source enterprise, only one cabinet door production line is provided, wherein the number of devices with the same function on the production line cannot exceed two, and therefore a scheme with small group number is selected from nodes with the highest average similarity coefficient acceleration rate in the group. Therefore, on the process group, the data node when λ is 0.71 is selected, i.e. the group number is 5, and the grouping result is shown in table 6.
TABLE 6 Process Classification results
The process is divided into 5 types, the A2 type comprises all four side materials, and the conventional side material process route is followed; b2 type is partial line, and the line nailing process route is directly followed; c2 type is partial lines and square lattice bars, and the line nailing route is performed after the county wonderful color processing; d2 type is a whole plate CYS051, and a metal groove route is opened; the E2 category comprises all core plates, louver strips, louver frame horizontal and vertical materials and flat plate whole plates, and the shaping route is milled.
Further, in the embodiment of the present invention, the method for classifying the mortise according to the previous classification condition of the mortise comprises: according to the connecting mode of the parts, the connecting structures are divided and comprise four-edge materials and a core board, four-edge materials and lines, the mortise structure of the obtained part comprises mortises of the four-edge materials, core board mortises and line mortises, and the mortises of the same type are classified into one type.
After integrating and summarizing the structures of 40 types of door-shaped parts with the same function, the tongue-and-groove structure is found to mainly refer to the connection mode of four-edge materials, a core plate, clamping strips and lines. The four-edge material is used as the most key main part in the cabinet door part, and the mortise structure of the cabinet door part is determined by whether lines exist or not and the line style. As shown in fig. 5, the lines are divided into two types, i.e., F3 buckle lines and G3 press lines; therefore, the four-side material tongue-and-groove structure connected with the material can be divided into three types, namely, A3 no-line, B3 line-buckling line and C3 line-pressing line; d3 is a whole plate structure; e3 is a core plate structure, although the core plate has two thicknesses of a flat core plate 5mm and a modeling core plate 15mm, the thickness of the joint connected with the rim charge is 5mm, so the tongue-and-groove structures can be unified into one; the special parts of the grille door type and the shutter door type are different in structure and are classified as H3. The classification results are shown in Table 7.
TABLE 7 tongue and groove Structure Classification results
Further, the method for classifying the bending degree according to the bending radius and the corner radian radius comprises the following steps: and analyzing the parts with the bent shapes in the parts, and dividing the parts into classes according to the bending radius and the corner radian radius. The feeding of a part of door shapes in the part has a bending shape, the bending shape only has one side of the feeding and needs to be specially processed by a tenoning machine, and the four side materials of the part of door shapes are assembled and then have corner radian shapes, so that the bending radius and the corner radian radius of the feeding of the bending shape are counted, and under the condition that the shapes are approximately unchanged, the sizes are integrated in a similar way, wherein the corner radian radii are just one, so that the bending shape is reserved. See table 8 for details.
TABLE 8 summary of bend radii for the bends
The parts with bending shapes are classified into three types of A4, B4 and C4 according to the bending radius, and the straight materials are uniformly summarized into D4, see Table 9.
TABLE 9 classification of degree of bending
S400: and sorting the clustering results of the attribute indexes according to the classified character strings. Sorting the clustering results of all attribute indexes according to the classified character strings, bringing all indexes into multi-attribute cross clustering, classifying the parts with the same sorting character strings into a family, removing empty sets, and obtaining the number of the families sorted by different character strings, which is the total number of the part family division. Specifically, in the specific example, the category codes of the attribute indexes are subjected to cross comprehensive sequencing, empty sets are removed, and a total of 17 part families are obtained in the test, which is shown in table 10.
TABLE 10 results of classification of family of parts
By the method, multi-attribute division of the solid wood customized cabinet door part family is realized.
On the other hand, the embodiment of the invention also provides a system for dividing the family of the solid wood customized cabinet door parts, and the system divides the family of the solid wood customized cabinet door parts by using the proposed method. As shown in connection with fig. 6, the system includes: the device comprises an attribute index acquisition unit, an attribute index sorting unit, an attribute index classification unit and a clustering result sorting unit; the attribute index acquisition unit is used for acquiring attribute indexes of a solid wood customized cabinet door part family, wherein the attribute indexes comprise size, process, tongue-and-groove structure and bending degree; the attribute index sorting unit is used for sorting the attribute indexes according to the importance degree; the attribute index classification unit is used for classifying each attribute index; and the clustering result sorting unit is used for sorting the clustering results of the attribute indexes according to the divided category character strings.
The attribute index classification unit comprises a size classification unit, a process classification unit, a tongue-and-groove structure classification unit and a bending structure classification unit, wherein the size classification unit is used for classifying sizes by adopting a hierarchical clustering method; the process classification unit is used for classifying the processes by adopting a fuzzy clustering method; the mortise structure classification unit is used for classifying the mortises according to the early classification conditions of the mortises; the bent structure classifying unit is used for classifying the bending degree according to the bending radius.
How to implement the corresponding functions in each unit is provided in the proposed method, and the details are not described herein.
The method provided by the invention is used for dividing the solid wood customized cabinet door into the part families, aiming at the influence of various factors on the classification of the solid wood customized cabinet door parts. Selecting four main classification attributes (size, process, tongue-and-groove structure and bending degree), sorting the importance degrees of the four main classification attributes by adopting an AHP method, classifying the categories of single indexes by adopting different methods (size-hierarchical clustering, process-fuzzy clustering, tongue-and-groove structure-connection mode structure, bending degree-bending radius and corner radian radius), formulating category codes, and counting classification results by adopting a mode of cross mixing the category codes. And the results of the specific examples show that: the method has clear classification result, overcomes the one-sidedness of single-index classification, can explain the attribute characteristics and significance of a part family in multiple aspects, and can select attribute classification on individual working sections according to actual production requirements to meet special requirements. The above examples are only preferred embodiments of the present invention, it should be noted that: it will be apparent to those skilled in the art that various modifications and equivalents can be made without departing from the spirit of the invention, and it is intended that all such modifications and equivalents fall within the scope of the invention as defined in the claims.
Claims (6)
1. A solid wood customized cabinet door part family dividing method is characterized by comprising the following steps:
obtaining attribute indexes of a solid wood customized cabinet door part family, wherein the attribute indexes comprise size, process, tongue-and-groove structure and bending degree;
sorting the attribute indexes according to the importance degree;
classifying the attribute indexes, wherein the sizes are classified by adopting a hierarchical clustering method, the processes are classified by adopting a fuzzy clustering method, the mortises are classified according to the early-stage classification conditions of the mortises, and the bending degrees are classified according to the bending radii; the method for classifying the sizes by adopting the hierarchical clustering method comprises the following steps:
(1) defining the length, width and thickness of the part after the disassembly of the customized solid wood cabinet door as x, y and z values respectively;
(2) normalizing the data by a maximum normalization method, wherein:
wherein x isSign board、ySign board、zSign boardRespectively representing the values of length, width and thickness of the standardized part, Xmax、Ymax、ZmaxRespectively representing the maximum values of the length, the width and the thickness of the part;
(3) measuring the similarity between any two parts after standardized treatment by adopting a squared Euclidean distance;
(4) combining the data points or categories with similar distances into a group, measuring the distance between the groups by using the average distance between the groups, combining the similar categories, and determining a classification result according to the requirement;
the method for classifying the process by adopting the fuzzy clustering method comprises the following steps:
(1) constructing a part-process matrix; defining a set A of furniture parts to be classifiedn＝﹛a1，a2，……，anCompare, enterprise's cabinet door production process set Bm＝﹛b1，b2，……，bmPartition, build part-process matrix as:
(2) constructing a part similarity coefficient matrix; determining the similarity coefficient between the two parts by adopting a maximum and minimum method, wherein the formula is as follows:
wherein x isijRepresenting the similarity factor of part i and part j belonging to set AnTwo parts with different inner parts, process k belongs to the production process set Bm，eikValue of part i on Process k, ejkIs the value of part j on process k; the similarity matrix between parts is:
(3) calculating a transfer closure, and solving a component similarity coefficient matrix X by using a transfer closure methodijThe fuzzy equivalence matrix T;
(4) intercepting the fuzzy equivalent matrix T by using a threshold lambda to obtain grouping conditions of different thresholds, determining the grouping number, and counting different clustering results to obtain a dynamic clustering chart of the fuzzy equivalent matrix T;
(5) calculating the similarity coefficients within the groups by constructing the average similarity coefficient C within each groupzDetermining the mean intra-group similarity coefficient CpBy using CpAnd judging the process similarity of the parts in the group, wherein the formula is as follows:
Cz＝(∑q∈H∑s∈HXqs)/h2，
wherein g represents the number of groups intercepted under a certain threshold lambda, H represents the number of parts in each group after grouping, H is a part set in the group after grouping, q and s represent two parts of the H set, q, s belongs to H, XqsA similarity matrix representing part q and part s;
sorting the clustering results of the attribute indexes according to the classified character strings, wherein the method comprises the following steps: all attribute indexes are brought into multi-attribute cross clustering, parts corresponding to character strings with the same sequence are classified into a group, and the group number formed after removing the empty set is the total group number of the part group division.
2. The method of claim 1 wherein the method of classifying the mortise according to the preliminary mortise classification comprises:
according to the connecting mode of the parts, the connecting structure is divided and comprises four edges and a core plate, four edges and lines, the mortise structure of the obtained part comprises mortises of the four edges, core plate mortises and line mortises, and the mortises of the same type are classified into one type.
3. The method of claim 1, wherein the classifying the degree of curvature according to the radius of curvature and the radius of corner radians is by: and analyzing the parts with the bent shapes in the parts, and dividing the parts into classes according to the bending radius and the corner radian radius.
4. The method of any one of claims 1-3, wherein the attribute indices are ranked using AHP analytic hierarchy process.
5. A system for dividing a family of parts of a solid wood custom cabinet door by using the method of any one of claims 1 to 4, comprising:
the attribute index acquisition unit is used for acquiring attribute indexes of a solid wood customized cabinet door part family, wherein the attribute indexes comprise size, process, tongue-and-groove structure and bending degree;
the attribute index sorting unit is used for sorting the attribute indexes according to the importance degree;
the attribute index classification unit is used for classifying each attribute index;
and the clustering result sorting unit is used for sorting the clustering results of the attribute indexes according to the divided category character strings.
6. The system according to claim 5, wherein the attribute index classification unit comprises:
the size classification unit is used for classifying sizes by adopting a hierarchical clustering method;
the process classification unit is used for classifying the processes by adopting a fuzzy clustering method;
the mortise structure classification unit is used for classifying the mortises according to the early classification conditions of the mortises;
and the bent structure classifying unit is used for classifying the bending degree according to the bending radius.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111001660.8A CN113592038B (en) | 2021-08-30 | 2021-08-30 | Method and system for dividing solid wood customized cabinet door part family |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111001660.8A CN113592038B (en) | 2021-08-30 | 2021-08-30 | Method and system for dividing solid wood customized cabinet door part family |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113592038A CN113592038A (en) | 2021-11-02 |
| CN113592038B true CN113592038B (en) | 2022-04-05 |
Family
ID=78240278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111001660.8A Active CN113592038B (en) | 2021-08-30 | 2021-08-30 | Method and system for dividing solid wood customized cabinet door part family |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113592038B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1355240A1 (en) * | 2002-04-18 | 2003-10-22 | BRITISH TELECOMMUNICATIONS public limited company | Data analysis method |
| CN102081706A (en) * | 2011-02-28 | 2011-06-01 | 同济大学 | Process planning method based on similarity theory |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10956779B2 (en) * | 2015-03-26 | 2021-03-23 | Oracle International Corporation | Multi-distance clustering |
| CN107545133A (en) * | 2017-07-20 | 2018-01-05 | 陆维嘉 | A kind of Gaussian Blur cluster calculation method for antidiastole chronic bronchitis |
| CN108717551A (en) * | 2018-05-08 | 2018-10-30 | 北京理工大学 | A kind of fuzzy hierarchy clustering method based on maximum membership degree |
| CN109117560B (en) * | 2018-08-17 | 2020-03-31 | 武汉理工大学 | Three-dimensional process design method and platform for machining parts of automotive typical machine based on MBD |
-
2021
- 2021-08-30 CN CN202111001660.8A patent/CN113592038B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1355240A1 (en) * | 2002-04-18 | 2003-10-22 | BRITISH TELECOMMUNICATIONS public limited company | Data analysis method |
| CN102081706A (en) * | 2011-02-28 | 2011-06-01 | 同济大学 | Process planning method based on similarity theory |
Non-Patent Citations (3)
| Title |
|---|
| "A diversity-driven structure learning algorithm for building hierarchical neuro-fuzzy classifiers";N.E. Mitrakis等;《Information Sciences》;20120331;第186卷(第1期);第40-58页 * |
| "儿童家具实木零部件成组技术的标准化设计研究";熊先青等;《林业工程学报》;20180131;第3卷(第1期);第135-140页 * |
| "木质柜类定制家具零件编码的研究与应用";龚瑶等;《林产工业》;20190531;第46卷(第5期);第59-61、64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113592038A (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Guha et al. | ROCK: A robust clustering algorithm for categorical attributes | |
| US8515958B2 (en) | System and method for providing a classification suggestion for concepts | |
| CN108564286A (en) | A kind of artificial intelligence finance air control credit assessment method and system based on big data reference | |
| US9514213B2 (en) | Per-attribute data clustering using tri-point data arbitration | |
| Hong et al. | Rapid identification of the optimal product configuration and its parameters based on customer-centric product modeling for one-of-a-kind production | |
| CN102081706B (en) | Process planning method based on similarity theory | |
| CN110751355A (en) | Scientific and technological achievement assessment method and device | |
| US6563952B1 (en) | Method and apparatus for classification of high dimensional data | |
| CN108470243A (en) | A kind of method and system of performance appraisal | |
| CN107944487B (en) | Crop breeding variety recommendation method based on mixed collaborative filtering algorithm | |
| CN109241446A (en) | A kind of position recommended method and system | |
| CN110737805A (en) | Method and device for processing graph model data and terminal equipment | |
| CN113592038B (en) | Method and system for dividing solid wood customized cabinet door part family | |
| CN111126865B (en) | Technology maturity judging method and system based on technology big data | |
| CN108664653A (en) | A kind of Medical Consumption client's automatic classification method based on K-means | |
| CN111353529A (en) | Mixed attribute data set clustering method for automatically determining clustering center | |
| CN107480426A (en) | From iteration case history archive cluster analysis system | |
| CN112270374B (en) | Clustering method of mathematical expression based on SOM (sequence of events) clustering model | |
| CN113312494A (en) | Vertical domain knowledge graph construction method, system, equipment and storage medium | |
| CN110134710B (en) | Cigarette formula rule extraction and formula maintenance method based on Bayesian network | |
| CN110097126B (en) | Method for checking important personnel and house missing registration based on DBSCAN clustering algorithm | |
| CN113205274A (en) | Quantitative ranking method for construction quality | |
| CN111311390A (en) | Method and system for evaluating financial risk of biological medicine enterprise | |
| Ozdemir et al. | Performance Comparison with Hierarchical and Partitional Clustering Methods | |
| CN115438101B (en) | Data feature construction system and method based on feature morphology and data relationship |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20211102 Assignee: DORAND FURNITURE Co.,Ltd. Assignor: NANJING FORESTRY University Contract record no.: X2022980017094 Denomination of invention: A Method and System for Partitioning Parts Family of Solid Wood Custom Cabinet Door Granted publication date: 20220405 License type: Common License Record date: 20220930 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20211102 Assignee: Anhui Yucai automation equipment Co.,Ltd. Assignor: NANJING FORESTRY University Contract record no.: X2022980017569 Denomination of invention: A Method and System for Partitioning Parts Family of Solid Wood Custom Cabinet Door Granted publication date: 20220405 License type: Common License Record date: 20221011 |
|
| EE01 | Entry into force of recordation of patent licensing contract |