CN113688224B - Self-adaptive processing method for complex equipment delivery problem based on grey correlation - Google Patents
Self-adaptive processing method for complex equipment delivery problem based on grey correlation Download PDFInfo
- Publication number
- CN113688224B CN113688224B CN202111243947.1A CN202111243947A CN113688224B CN 113688224 B CN113688224 B CN 113688224B CN 202111243947 A CN202111243947 A CN 202111243947A CN 113688224 B CN113688224 B CN 113688224B
- Authority
- CN
- China
- Prior art keywords
- delivery
- original
- degree
- dictionary
- question
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3322—Query formulation using system suggestions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a self-adaptive processing method of complex equipment delivery problems based on grey correlation, which specifically comprises the following operations: step 1, formatting the delivery problem based on a data dictionary; step 2, selecting an optimal disposal scheme of the delivery problem based on FMEA; and 3, realizing automatic matching and learning of the delivery problems based on grey correlation, extracting the original problems with the maximum correlation degree with the new problems, and taking the optimal treatment scheme as the treatment scheme of the new problems. The invention takes the data dictionary, FMEA and grey correlation as the theoretical basis, realizes the standardized processing of the delivery problem and the automatic matching with the original problem, improves the efficiency of solving the delivery problem and advances the delivery progress.
Description
Technical Field
The invention belongs to the technical field of intelligent manufacturing complex equipment delivery, and particularly relates to a self-adaptive processing method for complex equipment delivery problems based on grey correlation.
Background
As a high-precision and extremely complex product, the complex equipment often encounters various problems during the delivery process, and the reasons for these quality problems are mainly the quality problems of the complex equipment itself.
From a flow analysis of complex equipment inspection: due to the complexity of complex equipment, delivery problems have multi-source, multi-volume, multi-class, and multi-dimensional characteristics. In the actual delivery process, the cooperation connects the record personnel of connecing the dress inspection comparatively to be in short supply, mostly still adopts paper pen record pattern moreover, and the recording speed is slow, has the problem of unable accurate understanding and timely record, and this kind of problem record mode of irregularity has reduced the problem arrangement, analysis and investigation work efficiency of delivery acceptance department. Meanwhile, the positioning of the delivery and acceptance department in the whole group is a functional department, technical personnel are not provided, in the subsequent troubleshooting process, the technical personnel need to be extracted and dispatched from other departments in the group, the process is complex, the efficiency is low, and the problem is difficult to solve in real time. These factors combine to increase the uncertainty of the overall delivery process, making the overall delivery cycle redundant, extending the delivery cycle, and reducing delivery efficiency.
From institutional analysis: the delivery acceptance department does not form a unified problem recording standard and a system standard, the forms and the formats of the problems recorded by different personnel in the tipping process are inconsistent, the records of the same quality problem by different personnel are different, and the problem identification obstacle and the difficulty of subsequent troubleshooting are increased.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides an adaptive processing method of complex equipment delivery problems based on grey correlation, which is specifically operated as follows: step 1, formatting the delivery problem based on a data dictionary; step 2, selecting an optimal disposal scheme of the delivery problem based on FMEA; and 3, realizing automatic matching and learning of the delivery problems based on grey correlation, extracting the original problems with the maximum correlation degree with the new problems, and taking the optimal treatment scheme as the treatment scheme of the new problems. The invention takes the data dictionary, FMEA and grey correlation as the theoretical basis, realizes the standardized processing of the delivery problem and the automatic matching with the original problem, improves the efficiency of solving the delivery problem and advances the delivery progress.
The specific implementation content of the invention is as follows:
step 1: formatting delivery problems based on a data dictionary
The basic framework of the data dictionary comprises 5 levels: the first layer is a complex equipment cabin dictionary, the second layer is a part number dictionary (also called a figure number dictionary, and the following figure number dictionaries are used), the third layer is a primary classification dictionary for quality problem description, the fourth layer is a secondary classification dictionary for quality problem description, and the fifth layer is a tertiary classification dictionary for quality problem description:
1) complex equipment cabin dictionary: the complex equipment is divided according to the compartment space, one compartment space is a sub-dictionary, all the sub-dictionaries are combined together to form a compartment space dictionary, and the primary positioning of the delivery problem is realized according to the compartment space dictionary;
2) drawing number dictionary: in order to more accurately position the problem, the occurrence position of the delivery problem is directly positioned to each part on the complex equipment, each part in the cabin of the complex equipment needs to be numbered, the number of each part is defined as a figure number, and the figure numbers form a figure number dictionary;
3) a first-level classification dictionary: including design, manufacturing, and finishing issues;
4) a secondary classification dictionary: the second-level classification dictionary is a subdivision of problem descriptions in the first-level classification dictionary and comprises problems of cable manufacturing, cable installation, conduit manufacturing, conduit installation, surface manufacturing, surface assembly and the like;
5) a third-level classification dictionary: the third class dictionary is a subdivision of problem descriptions in the second class dictionary, for example, cable manufacturing in the second class includes cable protection, cable damage, label sleeve loosening and other third classes, cable installation in the second class includes cable trend, cable clearance, bending radius and other third classes, conduit manufacturing in the second class includes conduit protection, conduit damage, conduit bending radius, conduit connector damage and other third classes, conduit installation in the second class includes conduit clearance, conduit tightness, conduit loosening, conduit connector damage and other third classes, surface manufacturing in the second class includes damage, the surface assembly in the second-level classification comprises three-level classification such as step difference, gap, connecting piece quality, finished product installation, sealing performance and missing installation;
step 2: optimal handling scheme for delivery problems based on FMEA
Step 2.1: FMEA analysis team for establishing coping plans
Leaders and quality analyzers and related post technicians of all departments of the team members not only have the statistical analysis of the quality managers and the technical support of professional technicians, but also have the tripod force of the management team to assist;
step 2.2: delivery issue analysis
Analyzing the quality problems by team members by adopting analysis tools such as a brainstorming storm, a fishbone diagram, 5Why and the like aiming at the problems, identifying risk items and forming a process fault mode and influence analysis table;
step 2.3: calculation of delivery issue handling plan RPN
The FEMA analysis is to calculate a risk priority index (RPN), wherein the larger the risk priority index (RPN) is, the more serious the consequences caused by quality problems are, and the optimal treatment scheme is determined by comparing the sizes of the risk priority index (RPN);
the risk priority index (RPN) is calculated as:
in the formula: s represents the influence of the delivery problem on the safety of the complex equipment, O represents the repeatability of the delivery problem, and D represents the processing period of the delivery problem;
the grade classification standard of the safety, the repeatability and the processing period of the complex equipment is as follows:
1) safety (S): if the influence of the delivery problem on the complex equipment is extremely small, the safety of the problem is extremely small, and the safety level of the problem is 1; if the delivery problem has a slight effect on the complex equipment, indicating that the safety of the problem is small, the safety level is 2; if the delivery problem has a moderate impact on the complex equipment, the safety of the problem is medium, the safety level is 3; if the delivery problem has a severe impact on the complex equipment, indicating that the security of the problem is greater, the security level for this problem is 4; if the delivery problem has a very severe impact on the complex equipment, indicating that the safety of the problem is very great, the safety rating of this problem is 5;
2) degree of repetition (O): if the delivery problem happens rarely, and the repeatability of the problem is extremely low, the repeatability grade of the problem is 1; if the frequency of the delivery problem is low, which indicates that the problem repeatability is low, the repeatability grade of the problem is 2; if the frequency of the delivery problems is medium, which indicates that the problem repeatability is medium, the repeatability grade of the problem is 3; if the delivery problem occurs frequently, the repeatability grade of the problem is 4, which indicates that the repeatability of the problem is higher; if the delivery problem occurs frequently, the repeatability grade of the problem is 5 when the repeatability of the problem is extremely high;
3) treatment cycle (D): if the delivery problem is very easy to handle and the processing cycle to explain the problem is very short, the processing cycle rating of the problem is 1; if the delivery problem is easier to handle and the processing cycle for describing the problem is shorter, the processing cycle grade of the problem is 2; if the processing difficulty of the delivery problem is general and the processing period of the problem is in a medium level, the processing period grade of the problem is 3; if the delivery problem is difficult to process, and the processing period for explaining the problem is long, the processing period grade of the problem is 4; if the delivery problem is very difficult to process, and the processing period for explaining the problem is extremely long, the processing period grade of the problem is 5;
step 2.4: determining an optimal treatment plan
After the treatment is carried out for a certain time, collecting and sorting the basic data, and comparing risk indexes (RPN) of different treatment schemes, wherein the treatment scheme corresponding to the minimum risk index (RPN) is the optimal treatment scheme;
and step 3: automatic matching and learning for delivery problems based on grey correlation
Step 3.1: automatic recommendation of delivery issue handling schemes based on grey correlation
Step 3.1.1: converting data dictionary-based delivery issue formatted text to numbers
1) And (3) language variable conversion of the cabin dictionary: assuming a complex configuration with 20 bays, each bay corresponds to a numerical value, specifically: the number 1 corresponding to the cabin space 1, the number 2 corresponding to the cabin space 2, the numbers 3 and … corresponding to the cabin space 3, and the number 20 corresponding to the cabin space 20;
2) language variable conversion of a figure number dictionary: assuming that the parts on the complex equipment only have 20 figure numbers, each figure number corresponds to a numerical value, and the specific steps are as follows: figure 1 corresponds to number 1, figure 2 corresponds to number 2, figure 3 corresponds to numbers 3 and …, and figure 20 corresponds to number 20;
3) language variable conversion of the primary classification dictionary: defining the number of the first-level classification dictionaries as 3, wherein the specific corresponding relation is as follows: the first-level classification dictionary 1 corresponds to a number 1, the first-level classification dictionary 2 corresponds to a number 2, and the first-level classification dictionary 3 corresponds to a number 3;
4) and (3) language variable conversion of the secondary classification dictionary: defining the number of the secondary classification dictionaries as 10, wherein the specific corresponding relation is as follows: the secondary classification dictionary 1 corresponds to a number 1, the secondary classification dictionary 2 corresponds to numbers 2 and …, and the secondary classification dictionary 10 corresponds to a number 10;
5) and (3) language variable conversion of the three-level classification dictionary: defining the number of the three-level classification dictionaries as 20, wherein the specific corresponding relation is as follows: the three-level classification dictionary 1 corresponds to a number 1, the three-level classification dictionary 2 corresponds to numbers 2 and …, and the three-level classification dictionary 20 corresponds to a number 20;
based on this, each delivery issue can be located with 5 numbers;
step 3.1.2: the initial value image (mean value image) of each sequence is obtained by the following formula:
in the formula:
is the ith orderAn initial image of the column;
is the original value of the ith sequence;
representing the component values of the ith sequence initial value image; m is the total number of sequences or the total number of original problems;
a first component value being the original value of the ith sequence; n is the total number of components contained in each sequence; i is a sequence corresponding to the ith original problem;
step 3.1.3: calculating the original value of the new problem sequence
With the original values of the original problem sequence
The specific calculation formula of the absolute value sequence of the difference between the corresponding components of the initial value image is as follows:
in the formula:
is composed of
And
the difference between the corresponding components of the initial value image;
is composed of
And
the initial value of (a) is a sequence composed of the differences between the corresponding components,
is the specific individual component difference in the sequence of component differences;
is composed of
The kth component value of the initial image;
is composed of
The kth component value of the initial image;
；
step 3.1.4: calculating the component difference
The specific calculation formula is as follows:
in the formula:
is composed of
Maximum value of (1);
is composed of
Minimum value of (1);
step 3.1.5: calculating the correlation coefficient, wherein the specific calculation formula is as follows:
in the formula:
the correlation coefficient between the new problem sequence o and the original problem sequence i is obtained;
step 3.1.6: the average value of the correlation coefficients, i.e., the desired degree of correlation, is obtained by the following formula:
in the formula:
the correlation degree between the new problem o and the original problem i is obtained;
step 3.1.7: and extracting the original problem with the maximum relevance due to the new problem, and taking the optimal treatment scheme of the extracted original problem as the treatment scheme of the new problem.
Step 3.2: adaptive learning of delivery problem template indices based on grey correlation; establishing delivery indexes for all delivery problems, evaluating the influence degree of the delivery problems under each index based on grey target decision to obtain an influence degree interval under each delivery index, taking the maximum value of the influence degree interval as a positive target center of the corresponding delivery index, and taking the minimum value of the influence degree interval as a negative target center of the corresponding delivery index; then, respectively calculating to obtain the target center distance from the delivery problem to the positive target center and the target center distance from the negative target center under the corresponding delivery indexes, and then taking the value of the target center distance from the positive target center/(the target center distance from the positive target center + the target center distance from the negative target center) as a comprehensive target center distance, wherein the comprehensive target center distance is a template index;
step 3.2.1: determining the weight of each index of each original problem based on grey correlation
Determining the contribution degree of each original problem to the new problem according to the relative magnitude of the relevance degree, so that the weights corresponding to each index of each original problem are respectively as follows:
in the formula:
ranking the relevance between the new question and the grey relevance between the first original question;
ranking the degree of association between the new question and the second original question for the degree of association of the gray;
ranking the relevance between the new question and the grey relevance between the third original question;
representing the grey relevance degree to rank the contribution degree of the first original problem to the new problem;
representing the grey relevance degree to rank the contribution degree of the second original problem to the new problem;
representing the grey relevance rank ordering the contribution degree of the third original question to the new question;
step 3.2.2: calculating each parameter of new problem based on weight of each index of original problem
According to the grey correlation model and the weight corresponding to each index of each original problem, each parameter of the new problem can be expressed as:
in the formula:
degree of safety impact for new issues;
ranking the security of the first original question for gray relevance with the new question;
ranking the security of the second original question for gray relevance with the new question;
is and newGrey relevance between questions ranks the security of the third original question;
in the formula:
the degree of influence for the repeatability of the new problem;
ordering the degree of repetition of the first original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the second original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the third original question for the degree of grey correlation with the new question;
in the formula:
degree of impact for the processing cycle of the new problem;
ordering the processing period of the first original question for the grey correlation degree with the new question;
ordering the processing period of the second original question for the grey correlation degree with the new question;
sequencing the processing period of the third original problem for the grey correlation degree with the new problem;
in the formula:
a template index for a new question;
ranking the template indices of the first original question for gray relevance to the new question;
ranking the template indices of the second original question for gray relevance to the new question;
the template indices of the third original question are ranked for gray relevance to the new question.
Compared with the prior art, the invention has the beneficial effects that:
1. the delivery problem is formatted based on the data dictionary, so that the accuracy and the integrity of the data are guaranteed, the normalization and the standardization of the data are guaranteed, and the data information has shareability;
2. automatic matching of delivery problems and template index adaptive learning are achieved based on grey correlation. According to the grey correlation, the disposal scheme of the original problem with the highest grey correlation degree with the new problem is selected as the optimal disposal scheme of the new problem, and the efficiency of solving the delivery problem is greatly improved. And selecting the original problems with the first three in the relevance ranking, determining the weight of each original problem according to the relative size of the relevance, and then obtaining the safety, the repeatability, the processing period and the template index of the new problem according to the weight and storing the safety, the repeatability, the processing period and the template index, so that the link of expert evaluation is avoided, and the problem storage efficiency is improved.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a basic frame diagram of a data dictionary.
FIG. 3 is a flow diagram of an automatic handling of delivery problems and an adaptive learning mechanism.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and therefore should not be considered as a limitation to the scope of protection. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1:
the embodiment provides an adaptive processing method for complex equipment delivery problems based on gray correlation, and as shown in fig. 1, fig. 2, and fig. 3, the method specifically includes the following steps:
step 1: formatting delivery problems based on a data dictionary
The basic framework of the data dictionary comprises 5 levels: the first layer is a complex equipment cabin dictionary, the second layer is a part number dictionary (also called a figure number dictionary, and the following figure number dictionaries are used), the third layer is a primary classification dictionary for quality problem description, the fourth layer is a secondary classification dictionary for quality problem description, and the fifth layer is a tertiary classification dictionary for quality problem description:
1) complex equipment cabin dictionary: the complex equipment is divided according to the compartment space, one compartment space is a sub-dictionary, all the sub-dictionaries are combined together to form a compartment space dictionary, and the primary positioning of the delivery problem is realized according to the compartment space dictionary;
2) drawing number dictionary: in order to more accurately position the problem, the occurrence position of the delivery problem is directly positioned to each part on the complex equipment, each part in the cabin of the complex equipment needs to be numbered, the number of each part is defined as a figure number, and the figure numbers form a figure number dictionary;
3) a first-level classification dictionary: including design, manufacturing, and finishing issues;
4) a secondary classification dictionary: the second-level classification dictionary is a subdivision of problem descriptions in the first-level classification dictionary and comprises problems of cable manufacturing, cable installation, conduit manufacturing, conduit installation, surface manufacturing, surface assembly and the like;
5) a third-level classification dictionary: the third class dictionary is a subdivision of problem descriptions in the second class dictionary, for example, cable manufacturing in the second class includes cable protection, cable damage, label sleeve loosening and other third classes, cable installation in the second class includes cable trend, cable clearance, bending radius and other third classes, conduit manufacturing in the second class includes conduit protection, conduit damage, conduit bending radius, conduit connector damage and other third classes, conduit installation in the second class includes conduit clearance, conduit tightness, conduit loosening, conduit connector damage and other third classes, surface manufacturing in the second class includes damage, the surface assembly in the second-level classification comprises three-level classification such as step difference, gap, connecting piece quality, finished product installation, sealing performance and missing installation;
based on the above theory, we can classify the following examples according to the data dictionary: the delivery problem is 'paint dropping of liquid cooling pipes XXXXXXX-XXX-XXX in the left equipment cabin of the No. 1 oil tank', the cabin dictionary of the delivery problem is 'the left equipment cabin of the No. 1 oil tank', the drawing number dictionary is 'XXXXXXXXX-XXX-XXX', the first-level classification dictionary is 'manufacturing', the second-level classification dictionary is 'surface manufacturing', and the third-level classification dictionary is 'paint quality';
step 2: optimal handling scheme for delivery problems based on FMEA
Step 2.1: FMEA analysis team for establishing coping plans
Leaders and quality analyzers and related post technicians of all departments of the team members not only have the statistical analysis of the quality managers and the technical support of professional technicians, but also have the tripod force of the management team to assist;
step 2.2: delivery issue analysis
Analyzing the quality problems by team members by adopting analysis tools such as a brainstorming storm, a fishbone diagram, 5Why and the like aiming at the problems, identifying risk items and forming a process fault mode and influence analysis table; (ii) a For example, the risk item "paint dropping of liquid cooling pipes XXXXXXX-XXX-XXX of left equipment cabin of No. 1 oil tank" can be identified by the method;
step 2.3: calculation of delivery issue handling plan RPN
The FEMA analysis is to calculate a risk priority index (RPN), wherein the larger the risk priority index (RPN) is, the more serious the consequences caused by quality problems are, and the optimal treatment scheme is determined by comparing the sizes of the risk priority index (RPN);
the risk priority index (RPN) is calculated as:
in the formula: s represents the influence of the delivery problem on the safety of the complex equipment, O represents the repeatability of the delivery problem, and D represents the processing period of the delivery problem;
the grade classification standard of the safety, the repeatability and the processing period of the complex equipment is as follows:
1) safety (S): if the influence of the delivery problem on the complex equipment is extremely small, the safety of the problem is extremely small, and the safety level of the problem is 1; if the delivery problem has a slight effect on the complex equipment, indicating that the safety of the problem is small, the safety level is 2; if the delivery problem has a moderate impact on the complex equipment, the safety of the problem is medium, the safety level is 3; if the delivery problem has a severe impact on the complex equipment, indicating that the security of the problem is greater, the security level for this problem is 4; if the delivery problem has a very severe impact on the complex equipment, indicating that the safety of the problem is very great, the safety rating of this problem is 5;
2) degree of repetition (O): if the delivery problem happens rarely, and the repeatability of the problem is extremely low, the repeatability grade of the problem is 1; if the frequency of the delivery problem is low, which indicates that the problem repeatability is low, the repeatability grade of the problem is 2; if the frequency of the delivery problems is medium, which indicates that the problem repeatability is medium, the repeatability grade of the problem is 3; if the delivery problem occurs frequently, the repeatability grade of the problem is 4, which indicates that the repeatability of the problem is higher; if the delivery problem occurs frequently, the repeatability grade of the problem is 5 when the repeatability of the problem is extremely high;
3) treatment cycle (D): if the delivery problem is very easy to handle and the processing cycle to explain the problem is very short, the processing cycle rating of the problem is 1; if the delivery problem is easier to handle and the processing cycle for describing the problem is shorter, the processing cycle grade of the problem is 2; if the processing difficulty of the delivery problem is general and the processing period of the problem is in a medium level, the processing period grade of the problem is 3; if the delivery problem is difficult to process, and the processing period for explaining the problem is long, the processing period grade of the problem is 4; if the delivery problem is very difficult to process, and the processing period for explaining the problem is extremely long, the processing period grade of the problem is 5;
based on the grade classification standard and professional evaluation of the three indexes, the delivery problem has two disposal schemes by taking the delivery problem of 'No. 1 oil tank left equipment cabin liquid cooling pipe XXXXXXX-XXX-XXX paint removal' as an example: the first is repainting, which results in a safety (S) of 2, a degree of repetition (O) of 3 and a treatment period (D) of 3, so that the risk index (RPN) of this solution is 18; the second is a renewal, which results in a safety (S) of 2, a repetition (O) of 2 and a processing period (D) of 3, so that the risk index (RPN) of the solution is 12;
step 2.4: determining an optimal treatment plan
After the treatment is carried out for a certain time, collecting and sorting the basic data, and comparing risk indexes (RPN) of different treatment schemes, wherein the treatment scheme corresponding to the minimum risk index (RPN) is the optimal treatment scheme; therefore, according to the theory that the RPN is the minimum optimal disposal scheme, the optimal disposal scheme of the liquid cooling pipe XXXXXXX-XXX-XXX of the left equipment compartment of the No. 1 oil tank in the last step is changed;
and step 3: automatic matching and learning for delivery problems based on grey correlation
Step 3.1: automatic recommendation of delivery issue handling schemes based on grey correlation
Step 3.1.1: converting data dictionary-based delivery issue formatted text to numbers
1) And (3) language variable conversion of the cabin dictionary: assuming a complex configuration with 20 bays, each bay corresponds to a numerical value, specifically: the number 1 corresponding to the cabin space 1, the number 2 corresponding to the cabin space 2, the numbers 3 and … corresponding to the cabin space 3, and the number 20 corresponding to the cabin space 20;
2) language variable conversion of a figure number dictionary: assuming that the parts on the complex equipment only have 20 figure numbers, each figure number corresponds to a numerical value, and the specific steps are as follows: figure 1 corresponds to number 1, figure 2 corresponds to number 2, figure 3 corresponds to numbers 3 and …, and figure 20 corresponds to number 20;
3) language variable conversion of the primary classification dictionary: defining the number of the first-level classification dictionaries as 3, wherein the specific corresponding relation is as follows: the first-level classification dictionary 1 corresponds to a number 1, the first-level classification dictionary 2 corresponds to a number 2, and the first-level classification dictionary 3 corresponds to a number 3;
4) and (3) language variable conversion of the secondary classification dictionary: defining the number of the secondary classification dictionaries as 10, wherein the specific corresponding relation is as follows: the secondary classification dictionary 1 corresponds to a number 1, the secondary classification dictionary 2 corresponds to numbers 2 and …, and the secondary classification dictionary 10 corresponds to a number 10;
5) and (3) language variable conversion of the three-level classification dictionary: defining the number of the three-level classification dictionaries as 20, wherein the specific corresponding relation is as follows: the three-level classification dictionary 1 corresponds to a number 1, the three-level classification dictionary 2 corresponds to numbers 2 and …, and the three-level classification dictionary 20 corresponds to a number 20;
based on this, each delivery question can be located by 5 numbers, we take four original questions and one new question as an example, and from this step onwards, we take these five delivery questions as an example to calculate, and their numbers under the corresponding classification dictionary are: problem 1: the cabin is 10, the figure number is 11, the first class is 2, the second class is 7 and the third class is 9; problem 2: the cabin space is 4, the figure number is 15, the first-level classification is 3, the second-level classification is 2 and the third-level classification is 10; problem 3: the cabin space is 6, the figure number is 13, the first-level classification is 2, the second-level classification is 7 and the third-level classification is 7; problem 4: the cabin space is 6, the figure number is 14, the first-level classification is 1, the second-level classification is 4 and the third-level classification is 14; the new problem is as follows: the cabin space is 8, the figure number is 15, the first-level classification is 2, the second-level classification is 5 and the third-level classification is 10;
to simplify the presentation, we write the numbers of each delivery question above under the corresponding classification dictionary as a sequence:
the new problem is as follows:
problem 1:
problem 2:
problem 3:
problem 4:
step 3.1.2: the initial value image (or the mean value image) of each sequence is obtained, and the formula is as follows:
in the formula:
is the initial value image of the ith sequence;
is the original value of the ith sequence;
representing the component values of the ith sequence initial value image; m is the total number of sequences or the total number of original problems;
a first component value being the original value of the ith sequence; n is the total number of components contained in each sequence; i is a sequence corresponding to the ith original problem;
the initial values of the sequence of the new problem and the four original problems in the previous step are like:
step 3.1.3: calculating the original value of the new problem sequence
With the original values of the original problem sequence
The specific calculation formula of the absolute value sequence of the difference between the corresponding components of the initial value image is as follows:
in the formula:
is composed of
And
the difference between the corresponding components of the initial value image;
is composed of
And
the initial value of (a) is a sequence composed of the differences between the corresponding components,
is the specific individual component difference in the sequence of component differences;
is composed of
The kth component value of the initial image;
is composed of
The kth component value of the initial image;
；
then, the sequence formed by the difference between the components of the initial image of the sequence of the four original problems in the previous step and the components corresponding to the initial image of the sequence of the new problem is respectively:
step 3.1.4: to find
Maximum and minimum values of (a):
in the formula:
is composed of
Maximum value of (1);
is composed of
Minimum value of (1);
calculating the maximum value and the minimum value in the sequence formed by the difference between each component of the initial value image of the sequence of the four original problems and each component corresponding to the initial value image of the sequence of the new problem in the previous step to obtain 1.88 and 0 respectively;
step 3.1.5: calculating a correlation coefficient, wherein the formula is as follows:
in the formula:
the correlation coefficient between the new problem sequence o and the original problem sequence i is obtained;
here, take
Therefore, the correlation coefficients between the four original problems and the new problem in the previous step can be obtained:
step 3.1.6: the average value of the correlation coefficients, i.e., the desired degree of correlation, is obtained by the following formula:
in the formula:
the correlation degree between the new problem o and the original problem i is obtained;
therefore, the correlation degrees between the four original problems and the new problem in the previous step can be obtained as follows:
step 3.1.7: the relevance of the new question and the question 3 is the maximum, so that the optimal treatment scheme corresponding to the question 3 can be recommended as the treatment scheme of the new question;
step 3.2: adaptive learning of delivery problem template indices based on grey correlation; establishing delivery indexes for all delivery problems, evaluating the influence degree of the delivery problems under each index based on grey target decision to obtain an influence degree interval under each delivery index, taking the maximum value of the influence degree interval as a positive target center of the corresponding delivery index, and taking the minimum value of the influence degree interval as a negative target center of the corresponding delivery index; and then, respectively calculating to obtain the target center distance from the delivery problem to the positive target center and the target center distance from the negative target center under the corresponding delivery indexes, and then taking the value of the target center distance from the positive target center/(the target center distance from the positive target center + the target center distance from the negative target center) as a comprehensive target center distance, wherein the comprehensive target center distance is the template index.
Step 3.2.1: determining the weight of each index of each original problem based on grey correlation
Determining the contribution degree of each original problem to the new problem according to the relative magnitude of the relevance degree, so that the weights corresponding to each index of each original problem are respectively as follows:
in the formula:
ranking the relevance between the new question and the grey relevance between the first original question;
ranking the degree of association between the new question and the second original question for the degree of association of the gray;
ranking the relevance between the new question and the grey relevance between the third original question;
representing the grey relevance degree to rank the contribution degree of the first original problem to the new problem;
representing the grey relevance degree to rank the contribution degree of the second original problem to the new problem;
representing the grey relevance rank ordering the contribution degree of the third original question to the new question;
in this example, the three original questions that are most highly related to the new question from the previous step are: therefore, according to the above theory of relative magnitudes of the degrees of relevance, the degrees of contribution of the problem 1, the problem 3, and the problem 4 to the new problem are:
in the formula:
representing the grey relevance degree to rank the contribution degree of the first original problem to the new problem;
representing grey relevance rankingThe degree of contribution of the second original question to the new question;
representing the grey relevance rank ordering the contribution degree of the third original question to the new question;
step 3.2.2: calculating each parameter of new problem based on weight of each index of original problem
According to the grey correlation model and the weight corresponding to each index of each original problem, each parameter of the new problem can be expressed as: :
in the formula:
degree of safety impact for new issues;
ranking the security of the first original question for gray relevance with the new question;
ranking the security of the second original question for gray relevance with the new question;
ranking the security of the third original question for gray relevance with the new question;
in the formula:
the degree of influence for the repeatability of the new problem;
ordering the degree of repetition of the first original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the second original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the third original question for the degree of grey correlation with the new question;
in the formula:
degree of impact for the processing cycle of the new problem;
ordering the processing period of the first original question for the grey correlation degree with the new question;
ordering the processing period of the second original question for the grey correlation degree with the new question;
sequencing the processing period of the third original problem for the grey correlation degree with the new problem;
in the formula:
a template index for a new question;
ranking the template indices of the first original question for gray relevance to the new question;
ranking the template indices of the second original question for gray relevance to the new question;
the template indices of the third original question are ranked for gray relevance to the new question.
In this example, the three original questions that are most highly related to the new question from the previous step are: problem 1, problem 3 and problem 4, wherein the contribution degree of problem 1 to the new problem is 0.33, the contribution degree of problem 3 to the new problem is 0.34, and the contribution degree of problem 4 to the new problem is 0.33;
the former steps obtain expert scores of the three original problems under three evaluation indexes of 'safety', 'repeatability' and 'processing period', and the expert scores are as follows: the security of question 1 is 2, the degree of repetition is 3, the processing cycle is 2, the template index is 0.56, and the degree of grey correlation with the new question is 0.83; the security of question 3 is 3, the degree of repetition is 2, the processing cycle is 3, the template index is 0.54, and the degree of grey correlation with the new question is 0.85; the security of question 4 is 3, the degree of repetition is 2, the processing cycle is 2, the template index is 0.65, and the degree of grey correlation with the new question is 0.80;
therefore, according to the idea related to the adaptive learning model in this step, the values of the new problem under the three evaluation indexes of "safety", "repeatability", and "processing period" and the size of the template index can be obtained in this example, as follows:
safety:
the repetition degree:
and (3) treatment period:
template index:
therefore, based on the above model, the values of the new problem under "safety", "repeatability", "processing period" and "template index" are: 2.67, 2.33, 2.34, 0.55, realizing the adaptive learning of new problems.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications and equivalent variations of the above embodiments according to the technical spirit of the present invention are included in the scope of the present invention.
Claims (3)
1. An adaptive processing method for complex equipment delivery problems based on gray correlation is characterized by comprising the following steps:
step 1: constructing a data dictionary, and formatting the delivery problem based on the data dictionary;
step 2: selecting an optimal disposal scheme for the delivery problem based on FMEA analysis;
and step 3: automatically matching and learning the selected delivery problem based on the grey correlation;
the specific operation of the step 3 is as follows:
step 3.1: automatically recommending a disposal scheme for the delivery problem based on the grey correlation;
step 3.2: performing adaptive learning of template indexes of delivery problems based on grey correlation; the concrete calculation method of the template index comprises the following steps:
establishing delivery indexes for all delivery problems, evaluating the influence degree of the delivery problems under each index based on grey target decision to obtain an influence degree interval under each delivery index, taking the maximum value of the influence degree interval as a positive target center of the corresponding delivery index, and taking the minimum value of the influence degree interval as a negative target center of the corresponding delivery index; then, respectively calculating to obtain the target center distance from the delivery problem to the positive target center and the target center distance from the negative target center under the corresponding delivery indexes, and then taking the value of the target center distance from the positive target center/(the target center distance from the positive target center + the target center distance from the negative target center) as a comprehensive target center distance, wherein the comprehensive target center distance is a template index;
the specific operation of the step 3.1 is as follows:
step 3.1.1: converting delivery problem formatted text based on a data dictionary into numbers; writing the number of each delivery question under the corresponding classification dictionary into a sequence;
step 3.1.2: calculating the initial value image of each sequence, wherein the specific calculation formula is as follows:
in the formula:
is the initial value image of the ith sequence;
is the original value of the ith sequence;
representing the component values of the ith sequence initial value image; m is the total number of sequences or the total number of original problems;
a first component value being the original value of the ith sequence; n is the total number of components contained in each sequence; i is a sequence corresponding to the ith original problem;
step 3.1.3: calculating the original value of the new problem sequence
Original to original problem sequenceValue of
The specific calculation formula of the absolute value sequence of the difference between the corresponding components of the initial value image is as follows:
in the formula:
is composed of
And
the difference between the corresponding components of the initial value image;
is composed of
And
the initial value of (a) is a sequence composed of the differences between the corresponding components,
is the specific individual component difference in the sequence of component differences;
is composed of
The kth component value of the initial image;
is composed of
The kth component value of the initial image;
；
step 3.1.4: calculating the component difference
The specific calculation formula is as follows:
in the formula:
is composed of
Maximum value of (1);
is composed of
Minimum value of (1);
step 3.1.5: calculating the correlation coefficient, wherein the specific calculation formula is as follows:
in the formula:
the correlation coefficient between the new problem sequence o and the original problem sequence i is obtained;
step 3.1.6: the average value of the correlation coefficients, i.e., the desired degree of correlation, is obtained by the following formula:
in the formula:
the correlation degree between the new problem o and the original problem i is obtained;
step 3.1.7: extracting the original problem with the maximum relevance degree from the new problem, and taking the optimal treatment scheme of the extracted original problem as the treatment scheme of the new problem;
the specific steps of the step 3.2 are as follows:
step 3.2.1: determining the weight of each index of each original problem based on the grey correlation degree;
in the formula:
ranking the relevance between the new question and the grey relevance between the first original question;
ranking the degree of association between the new question and the second original question for the degree of association of the gray;
ranking the relevance between the new question and the grey relevance between the third original question;
representing the grey relevance degree to rank the contribution degree of the first original problem to the new problem;
representing the grey relevance degree to rank the contribution degree of the second original problem to the new problem;
representing the grey relevance rank ordering the contribution degree of the third original question to the new question;
step 3.2.2: calculating each parameter of the new problem based on the weight of each index of each original problem, wherein the specific calculation formula is as follows:
in the formula:
degree of safety impact for new issues;
ordering first primitive for grey relevance to New problemThe safety of the problem;
ranking the security of the second original question for gray relevance with the new question;
ranking the security of the third original question for gray relevance with the new question;
in the formula:
the degree of influence for the repeatability of the new problem;
ordering the degree of repetition of the first original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the second original question for the degree of grey correlation with the new question;
ranking the degree of repetition of the third original question for the degree of grey correlation with the new question;
in the formula:
degree of impact for the processing cycle of the new problem;
ordering the processing period of the first original question for the grey correlation degree with the new question;
ordering the processing period of the second original question for the grey correlation degree with the new question;
sequencing the processing period of the third original problem for the grey correlation degree with the new problem;
in the formula:
a template index for a new question;
ranking the template indices of the first original question for gray relevance to the new question;
ranking the template indices of the second original question for gray relevance to the new question;
the template indices of the third original question are ranked for gray relevance to the new question.
2. The adaptive processing method for the complex equipment delivery problem based on gray correlation as claimed in claim 1, wherein in the step 1, the specific operation of constructing the data dictionary is as follows:
dividing the data dictionary into 5 layers, wherein the first layer is a complex equipment cabin dictionary; the second layer is a part number dictionary with quality problems, also called a figure number dictionary; the third layer is a first-level classification dictionary for quality problem description; the fourth layer is a secondary classification dictionary for quality problem description; the fifth layer is a three-level classification dictionary for quality problem description;
for the complex equipment bay dictionary: dividing the complex equipment according to the compartment space, wherein one compartment space is a sub-dictionary, all the sub-dictionaries are combined together to form the complex equipment compartment space dictionary, and the delivery problem is preliminarily positioned according to the complex equipment compartment space dictionary;
for the graph number dictionary: numbering each part in the complex equipment compartment, defining the number of each part as a figure number, and forming a figure number dictionary by the figure number; the method comprises the steps that a map number dictionary is used for more accurately positioning delivery problems relative to a complex equipment cabin dictionary, and the occurrence positions of the delivery problems are directly positioned on each part on the complex equipment;
for the primary classification dictionary: the first-level classification dictionary comprises design problems, manufacturing problems and finished product problems;
for the secondary classification dictionary: the second class dictionary is a subdivision of problem descriptions in the first class dictionary, and comprises cable manufacturing problems, cable installation problems, conduit manufacturing problems, conduit installation problems, surface manufacturing problems and surface assembly problems;
for the three-level classification dictionary: the third-level classification dictionary is the subdivision of the problem description in the second-level classification dictionary; the second-level classification cable manufacturing problems comprise cable protection problems, cable damage problems and mark sleeve loosening problems; the cable installation problem comprises a cable trend problem, a cable clearance problem and a bending radius problem; the conduit manufacturing problems include conduit protection problems, conduit damage problems, conduit bend radius problems, conduit connector damage problems; the problems of conduit installation include conduit clearance, conduit sealability, conduit looseness and conduit connection damage; surface manufacturing problems include damage problems, corrosion/erosion problems, paint layer quality problems, adhesive quality problems, film layer quality problems, smudging problems; surface mounting problems include step problems, clearance problems, connector quality problems, finished product installation problems, sealability problems, and lack of packaging problems.
3. The adaptive processing method for the complex equipment delivery problem based on the gray correlation as claimed in claim 1 or 2, wherein the step 2 specifically comprises the following steps:
step 2.1: establishing a team for analyzing the predetermined FMEA;
step 2.2: analyzing the delivery problem;
step 2.3: calculating the risk priority index RPN of the delivery problem handling scheme, wherein the specific calculation formula is as follows:
in the formula: s represents the influence of the delivery problem on the safety of the complex equipment, O represents the repeatability of the delivery problem, and D represents the processing period of the delivery problem;
step 2.4: implementing the processing scheme for delivering the problem for a period of time, then collecting and sorting basic data, then comparing risk priority indexes RPN corresponding to different processing schemes, and selecting the processing scheme with the minimum risk priority index as the optimal disposal scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243947.1A CN113688224B (en) | 2021-10-26 | 2021-10-26 | Self-adaptive processing method for complex equipment delivery problem based on grey correlation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243947.1A CN113688224B (en) | 2021-10-26 | 2021-10-26 | Self-adaptive processing method for complex equipment delivery problem based on grey correlation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113688224A CN113688224A (en) | 2021-11-23 |
| CN113688224B true CN113688224B (en) | 2022-01-25 |
Family
ID=78588045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111243947.1A Active CN113688224B (en) | 2021-10-26 | 2021-10-26 | Self-adaptive processing method for complex equipment delivery problem based on grey correlation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113688224B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105678438A (en) * | 2015-06-15 | 2016-06-15 | 温州大学 | Improved FMEA method based on linguistic weighted geometric operator and fuzzy priority sequence |
| CN110211675A (en) * | 2019-06-03 | 2019-09-06 | 上海大学 | One kind is based on uncertain language and the improved FMEA method of GRA-TOPSIS |
| CN112269124A (en) * | 2020-10-16 | 2021-01-26 | 厦门理工学院 | High-voltage circuit breaker fault diagnosis method based on self-adaptive grey correlation analysis |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101950382B (en) * | 2010-09-01 | 2013-03-06 | 燕山大学 | Method for optimal maintenance decision-making of hydraulic equipment with risk control |
| TW201413605A (en) * | 2012-09-18 | 2014-04-01 | Askey Computer Corp | Product quality improvement feedback method |
| CN108170566A (en) * | 2017-12-18 | 2018-06-15 | 新疆金风科技股份有限公司 | Product failure information processing method, system, equipment and collaboration platform |
| CN109345083A (en) * | 2018-09-07 | 2019-02-15 | 上海大学 | A method of the improvement FMEA based on section Two-tuple Linguistic Information Processing and grey correlation analysis |
| CN110795696A (en) * | 2019-10-29 | 2020-02-14 | 北京航空航天大学 | Risk priority number calculation method based on grey correlation |
| CN112288218B (en) * | 2020-09-22 | 2022-08-12 | 成都飞机工业（集团）有限责任公司 | Delivery method based on airplane acceptance process |
| CN112463201A (en) * | 2020-12-14 | 2021-03-09 | 四川长虹电器股份有限公司 | Software maintenance cost optimization method based on system clustering and grey correlation analysis |
| CN112884379A (en) * | 2021-04-02 | 2021-06-01 | 辽宁工程技术大学 | Method for predicting coal mine underground human factor safety risk |
-
2021
- 2021-10-26 CN CN202111243947.1A patent/CN113688224B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105678438A (en) * | 2015-06-15 | 2016-06-15 | 温州大学 | Improved FMEA method based on linguistic weighted geometric operator and fuzzy priority sequence |
| CN110211675A (en) * | 2019-06-03 | 2019-09-06 | 上海大学 | One kind is based on uncertain language and the improved FMEA method of GRA-TOPSIS |
| CN112269124A (en) * | 2020-10-16 | 2021-01-26 | 厦门理工学院 | High-voltage circuit breaker fault diagnosis method based on self-adaptive grey correlation analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113688224A (en) | 2021-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Barr et al. | Forecasting bank failure: A non-parametric frontier estimation approach | |
| CN111984788B (en) | Electric power system violation management method and device and electric power equipment | |
| CN102498445A (en) | Supervised fault learning using rule-generated samples for machine condition monitoring | |
| CN110555024B (en) | Accurate automobile model matching system based on artificial intelligence algorithm | |
| CN111798297A (en) | Financial risk early warning analysis method and device | |
| CN116342073B (en) | Book printing digital information management system and method thereof | |
| CN112785377B (en) | Data distribution-based order completion period prediction model construction method and prediction method | |
| Aslam et al. | Introduction to statistical process control | |
| CN111899090A (en) | Enterprise associated risk early warning method and system | |
| CN113688224B (en) | Self-adaptive processing method for complex equipment delivery problem based on grey correlation | |
| Chan | Quality management | |
| CN116862477A (en) | Intelligent maintenance auxiliary method, system, equipment and storage medium | |
| CN112258095B (en) | Standard normal distribution based scoring method, device, equipment and storage medium | |
| CN114611515A (en) | Method and system for identifying actual control person of enterprise based on enterprise public opinion information | |
| Wang et al. | Prediction of accrual expenses in balance sheet using decision trees and linear regression | |
| CN113222255A (en) | Method and device for contract performance quantification and short-term default prediction | |
| Whalen | Are Early Warning Models Still Useful Tools for Bank Supervisors ‘ | |
| CN110163566A (en) | It is a kind of can motivate student learn on-line operation read and make comments system and method | |
| CN117102738A (en) | Automatic recognition method for groove form and joint type of welding groove schematic diagram | |
| Stead | Toward a synthesis of accounting doctrine | |
| US20210374092A1 (en) | Computer-implemented method of providing troubleshooting support and maintenance instructions for servicing assets | |
| Roeder | Decision Support for Credit Risk Management Using Alternative Data | |
| MATHAD | Overview of Strategy and Productivity | |
| Dang | Factors affecting the selection of inventory cost flow assumptions: An evidence from listed companies on NASDAQ Helsinki Stock Exchange during 2015-2019 | |
| Liu et al. | Ergonomics reliability evaluation of HMI design considering special circumstances based on improved SPAR-H and CRITIC |
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 |