Industrial production diagnosis method of tree structure

Abstract

The invention relates to an industrial production diagnosis method of a tree structure. The method comprises the following steps: acquiring different types of production parameter values in a production process; analyzing the corresponding relationship between the production problem and the production parameter in the production record; a multi-level tree-shaped corresponding relation graph is established according to the corresponding relation between the production problems and the production parameters; and the production parameter subtypes corresponding to the faults are analyzed and identified from top to bottom, and the production parameter subtypes are checked in sequence to find fault causes. According to the method, the corresponding relations between the different sub-types of production parameters and the different types of production problems are quantitatively compared, so that the influence capability of the different sub-types of production parameters on the same production problem can be compared, and the visual level of the corresponding relations between the production parameters and the production problems is improved; the tree-shaped corresponding relation graph is established, the logic level of the troubleshooting process is enhanced, the content and the structure of the tree-shaped corresponding relation graph are continuously and dynamically optimized, the method adapts to the actual production situation, the effectiveness of the troubleshooting method is guaranteed, the troubleshooting accuracy is improved, and the troubleshooting efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of big data industry, and particularly relates to an industrial production diagnosis method with a tree structure. Background Art

[0002] CN201410373937.3 discloses a fault diagnosis method based on fault tree and database technology. Its technical solution is as follows: receive user input, where the user input includes a fault symptom keyword; the fault type is divided into: bottom - layer fault, intermediate fault, and top - layer fault according to the position of the current fault node in the fault tree. The top - layer fault corresponds to the root node, the bottom - layer fault corresponds to the leaf node, and the intermediate fault is the fault between the bottom - layer and top - layer faults; retrieve the fault diagnosis information table in the fault diagnosis database according to the fault symptom keyword, and determine whether the fault is a bottom - layer fault. If it is a bottom - layer fault, it means that the diagnosis result is already at the bottom layer of the fault tree, then display the content of the "repair suggestion" field corresponding to it, and save the diagnosis result, and this diagnosis process ends; obtain its fault type and fault symptom keyword according to the fault node; in each step of diagnosis, the "document number", "diagnosis serial number", "node code", "node name", "node position", "node type", "diagnosis content", "project code", "number of measuring points", "data length of single measuring point", "measuring point data", "number of indicators", "technical indicator data", "repair suggestion" information of this step of diagnosis will be stored in the fault diagnosis process information table. When the final diagnosis result is output, the information of each single - step diagnosis in the fault diagnosis process information table in the fault diagnosis database will be retrieved and output in the form of a report.

[0003] The fault diagnosis method based on fault tree and database technology has the following advantages: reasoning is carried out layer by layer according to the response relationship between the fault event and the fault cause; define the knowledge information and detection process information of each fault node and store them in the fault diagnosis database, and perform manual detection or automatic detection on each fault node in the fault tree, effectively allocating detection resources. When a complex system fails, it can quickly perform fault diagnosis and improve the fault diagnosis efficiency.

[0004] However, the fault diagnosis method based on fault tree and database technology also has the following disadvantages: the fault tree structure is relatively fixed and cannot be adjusted according to the actual production situation, which may cause waste of more fault troubleshooting work and result in lower efficiency in the actual fault troubleshooting process; the content of the fault tree cannot be updated in real time, which may cause information lag and make a large difference between the fault troubleshooting situation and the actual situation; the response relationship between the fault event and the fault cause is relatively fuzzy and cannot be intuitively compared.

[0005] Therefore, there is a need for a method or system that can flexibly adjust the fault diagnosis structure and content according to the actual production situation and visually compare the relationship between fault events and fault causes. Summary of the Invention

[0006] To solve the problems of inability to adjust the structure and content according to the actual production situation and inability to visually compare the relationship between fault events and fault causes, the present application provides an industrial production diagnosis method and system with a tree structure.

[0007] The present application provides an industrial production diagnosis method with a tree structure, including the following steps:

[0008] Step S1, collect parameter data. The production monitoring device collects different types of production parameter values during the production process, marks the time, and uploads them to the big data platform.

[0009] Step S2, confirm the corresponding relationship. Respectively retrieve the production record database and the production parameter database, and analyze the corresponding relationship between the production problems in the production records and the production parameters.

[0010] Step S3, establish a tree-shaped corresponding relationship diagram. Establish a multi-level tree-shaped corresponding relationship diagram according to the corresponding relationship between the production problems and the production parameters.

[0011] Step S4, fault analysis. According to the generated tree-shaped corresponding relationship diagram, analyze and identify the sub-types of production parameters corresponding to the fault from top to bottom, and sequentially check the conditions corresponding to the sub-types of production parameters until the fault cause is found.

[0012] Further, the step S1 includes: step S11, collect real-time environmental parameters. The production monitoring device produces sub-type parameters of the production environment, then marks the production time, and uploads the sub-type parameters of the production environment to the big data platform; step S12, collect real-time equipment parameters, then mark the production time, the production monitoring device produces sub-type parameters of the production equipment, and uploads the sub-type parameters of the production equipment to the big data platform; step S13, collect real-time raw material parameters. The production monitoring device produces sub-type parameters of the production raw materials, then marks the production time, and uploads the sub-type parameters of the production raw materials to the big data platform.

[0013] The sub-type parameters of the production environment include but are not limited to: production environment temperature, production environment humidity, production environment noise, production environment atmosphere concentration.

[0014] The sub-type parameters of the production equipment include but are not limited to: production equipment temperature, production equipment noise intensity, production equipment vibration frequency.

[0015] The sub-type parameters of the production raw materials include but are not limited to: production raw material performance, production raw material feeding speed.

[0016] Further, the step S2 includes: step S21, summarizing production problems. The production problem recognition module retrieves the production record database, identifies the types of production problems in the production records, and marks the production time; step S22, analyzing the single correspondence relationship. Extract the production parameters in the production parameter database and the production problems in the production record database at the same production time, and analyze the correspondence relationship between the production parameter subtypes and the occurrence of production problems; step S23, summarizing the correspondence relationship. Analyze one by one the correspondence relationships between the production parameter subtypes and the occurrence of production problems at all times according to the production time. According to the analysis results, summarize the correspondence relationship between the production problem types and the production parameter subtypes.

[0017] Further, in the step S23, the correspondence analysis method is as follows:

[0018]

[0019] Among them, ω represents the influence ability index of a certain subtype of production parameter on a certain type of production problem;

[0020] X represents the value of this subtype of production parameter at a certain production time; X′ represents the ideal value of this subtype of production parameter, which is analyzed and set by production management personnel according to the specific production parameters; Y represents the occurrence parameter of this type of production problem at the same production time. If this type of production problem occurs at the same production time, then Y is 1. If this type of production problem does not occur at the same production time, then Y is 0; Y′ represents the ideal value of the occurrence index of this type of production problem, which is 0;

[0021] When ω > 0, the larger ω is, the greater the influence ability of this subtype of production parameter on this type of production problem, that is, the stronger the correspondence relationship; the smaller ω is, the smaller the influence ability of this subtype of production parameter on this type of production problem, that is, the weaker the correspondence relationship;

[0022] When ω ≤ 0, this subtype of production parameter has no positive influence ability on this type of production problem, that is, the correspondence relationship is weak.

[0023] By adopting the above technical solution, the correspondence relationships between different subtypes of production parameters and different types of production problems are quantitatively compared, the influence abilities of different subtypes of production parameters on the same production problem can be compared, and data reference is provided during the troubleshooting process, improving the intuitive level of the correspondence relationship between production parameters and production problems and improving the troubleshooting efficiency.

[0024] Further, the step S3 includes: step S31, correspondence classification. According to the influencing capabilities of different sub-type production parameters on different types of production problems in step S23, extract the sub-types of production parameters with positive influencing capability indices corresponding to each type of production problem one by one; S32, generate a tree-shaped correspondence graph. For the sub-types of production parameters with positive influencing capability indices extracted for each production problem type, generate a tree-shaped correspondence graph corresponding to the production problem type. The production problem type is the first level, the production parameter type is the second level, and the sub-types of production parameters included in the production parameter type are the third level; S33, adjust the tree-shaped correspondence graph. According to the influencing capabilities of different sub-type production parameters on different types of production problems, adjust the third-level sequence and further adjust the second-level sequence; step S34, dynamic optimization. The big data platform 1 continuously collects and updates the production record database and the production parameter database, updates the production problem type, calculates the influencing capability index of the sub-type of production parameter on the newly generated production problem type, and dynamically updates the influencing capability index of the existing production problem type at the same time.

[0025] By adopting the above technical solution, a tree-shaped correspondence graph is established, enhancing the logical level of the fault troubleshooting process and improving the fault troubleshooting efficiency; the content of the tree-shaped correspondence graph is continuously and dynamically optimized with data updates, making the content of the tree-shaped correspondence graph always adapt to the actual production situation, ensuring the effectiveness of the fault troubleshooting method and improving the accuracy of the fault troubleshooting.

[0026] Further, in the step S33, first mark the influencing capability index in the tree-shaped correspondence graph, and sort the third level in descending order of value according to the production parameter type to which it belongs; then extract the sub-types of production parameters in the tree-shaped correspondence graph whose influencing capability index exceeds the set threshold, count the number of sub-types of production parameters whose influencing capability index exceeds the set threshold included in each production parameter type at the second level, and sort the second level in descending order of the included number.

[0027] By adopting the above technical solution, according to the influencing capabilities of different sub-type production parameters on different types of production problems, the structure of the tree-shaped correspondence graph is flexibly adjusted, making the structure of the tree-shaped correspondence graph always adapt to the actual production situation, ensuring the effectiveness of the fault troubleshooting method and improving the accuracy of the fault troubleshooting.

[0028] Further, step S4 includes: step S41, production problem identification, where a production problem identification module identifies production problems occurring during the production process and extracts the corresponding tree-shaped correspondence diagram for the production problem; step S42, fault troubleshooting, which corresponds to the first level according to the identified production problem type, and corresponds to the production parameter type according to the sequence of the second level. The production parameter subtypes in the third level included in the production parameter type are checked in descending order of the influence ability index. If the cause of the fault is not found, the second level of the next sequence is checked until the cause of the fault is found.

[0029] By adopting the above technical solution, fault troubleshooting work is carried out according to the tree-shaped correspondence diagram, and according to the influence ability of different subtypes of production parameters on different types of production problems, the troubleshooting is carried out level by level to prevent omission of the cause of the fault, enhance the logical level of the fault troubleshooting process, and improve the fault troubleshooting efficiency.

[0030] An industrial production diagnosis system with a tree structure includes: a big data platform and a production monitoring device connected to the big data platform;

[0031] The big data platform includes: a memory; a processor connected to the memory; a production record database provided in the memory for storing production record data; a production parameter database provided in the memory for receiving and storing different types of production parameters collected by the production monitoring device;

[0032] The production monitoring device includes: a production environment monitoring device for collecting production environment parameters and uploading them to the production parameter database of the big data platform; a production equipment monitoring device for collecting production equipment parameters and uploading them to the production parameter database of the big data platform; a production raw material monitoring device for collecting production raw material parameters and uploading them to the production parameter database of the big data platform.

[0033] Further, it further includes: a production problem identification module running on the processor for retrieving production records in the production record database and identifying production problems and identifying production problem types; a production correspondence analysis module running on the processor for analyzing and calculating the correspondence between production parameters and production problems.

[0034] In summary, the present application includes the following beneficial technical effects:

[0035] 1. Quantitatively comparing the correspondence between different subtypes of production parameters and different types of production problems can compare the influence ability of different subtypes of production parameters on the same production problem, and provide data reference during the fault troubleshooting process, improving the intuitive level of the correspondence between production parameters and production problems;

[0036] 2. Establish a tree - shaped correspondence diagram, conduct fault troubleshooting based on the tree - shaped correspondence diagram, and sequentially troubleshoot by level according to the influencing ability of different sub - type production parameters on different types of production problems, preventing omission of fault causes, enhancing the logical level of the fault - troubleshooting process, and improving the fault - troubleshooting efficiency;

[0037] 3. According to the update of the collected parameters, the content of the tree - shaped correspondence diagram is continuously and dynamically optimized, enabling the content of the tree - shaped correspondence diagram to always adapt to the actual production situation, ensuring the effectiveness of the fault - troubleshooting method, and improving the accuracy of fault troubleshooting;

[0038] 4. According to the influencing ability of different sub - type production parameters on different types of production problems, flexibly adjust the structure of the tree - shaped correspondence diagram, making the structure of the tree - shaped correspondence diagram always adapt to the actual production situation and making the fault - troubleshooting sequence adapt to the influencing ability of production parameters on production problems, thereby improving the fault - troubleshooting efficiency. Description of the Drawings

[0039] Figure 1 is a step diagram of an industrial production diagnosis method with a tree - shaped structure according to an embodiment of the present application.

[0040] Figure 2 is a structure diagram of an industrial production diagnosis system with a tree - shaped structure according to an embodiment of the present application.

[0041] Description of the Reference Numerals:

[0042] 1. Big data platform; 2. Memory; 3. Processor; 4. Production record database; 5. Production parameter database; 6. Production problem identification module; 7. Production correspondence analysis module;

[0043] 8. Production monitoring device; 81. Production environment monitoring device; 82. Production equipment monitoring device; 83. Production raw material monitoring device. Detailed Embodiment

[0044] The following further elaborates in detail on the specific implementation manners of the present application, such as the shapes, structures, mutual positions and connection relationships of the components involved, the functions and working principles of each part, the manufacturing process, and the operation and usage methods, etc., with reference to the drawings, so as to help those skilled in the art have a more complete, accurate, and in - depth understanding of the inventive concept and technical solution of the present invention. For the convenience of description, the directions mentioned in the present application are subject to the directions shown in the drawings.

[0045] Refer to Figure 1 - Figure 2 As shown, an industrial production diagnosis method with a tree - shaped structure includes the following steps:

[0046] Step S1: Collect parameter data. The production monitoring device 8 collects the values of different types of production parameters during the production process, marks the time, and uploads them to the big data platform 1.

[0047] Step S2: Confirm the correspondence. Retrieve the production record database 4 and the production parameter database 5 respectively, and analyze the correspondence between the production problems in the production records and the production parameters.

[0048] Step S3: Establish a tree-like correspondence diagram. Establish a multi-level tree-like correspondence diagram according to the correspondence between the production problems and the production parameters.

[0049] Step S4: Fault analysis. According to the generated tree-like correspondence diagram, analyze and identify the sub-types of production parameters corresponding to the faults from top to bottom, and sequentially check the conditions corresponding to the sub-types of production parameters until the cause of the fault is found.

[0050] The said Step S1 includes: Step S11: Collect real-time environmental parameters. The production monitoring device 8 produces sub-type parameters of the production environment, then marks the production time, and uploads the sub-type parameters of the production environment to the big data platform 1; Step S12: Collect real-time equipment parameters, then mark the production time, the production monitoring device 8 produces sub-type parameters of the production equipment, and uploads the sub-type parameters of the production equipment to the big data platform 1; Step S13: Collect real-time raw material parameters. The production monitoring device 8 produces sub-type parameters of the production raw materials, then marks the production time, and uploads the sub-type parameters of the production raw materials to the big data platform 1.

[0051] The sub-type parameters of the production environment include but are not limited to: production environment temperature, production environment humidity, production environment noise, production environment atmosphere concentration.

[0052] The sub-type parameters of the production equipment include but are not limited to: production equipment temperature, production equipment noise intensity, production equipment vibration frequency.

[0053] The sub-type parameters of the production raw materials include but are not limited to: production raw material performance, production raw material feeding speed.

[0054] The said Step S2 includes: Step S21: Summarize production problems. The production problem identification module 6 retrieves the production record database 4, identifies the types of production problems in the production records, and marks the production time; Step S22: Analyze the single correspondence. Extract the production parameters in the production parameter database 5 and the production problems in the production record database 4 at the same production time, and conduct a correspondence analysis on the sub-types of production parameters and the occurrence of production problems; Step S23: Summarize the correspondence. Analyze one by one the correspondence between the sub-types of production parameters and the occurrence of production problems at all times according to the production time, and summarize the correspondence between the types of production problems and the sub-types of production parameters according to the analysis results.

[0055] In step S23, the corresponding relationship analysis method is as follows:

[0056]

[0057] Among them, ω represents the influence ability index of a certain subtype production parameter on a certain type of production problem;

[0058] X represents the value of the subtype production parameter at a certain production time; X′ represents the ideal value of the subtype production parameter, which is analyzed and set by production management personnel according to the specific production parameters; Y represents the occurrence parameter of this type of production problem at the same production time. If this type of production problem occurs at the same production time, then Y is 1. If this type of production problem does not occur at the same production time, then Y is 0; Y′ represents the ideal value of the occurrence index of this type of production problem, that is, 0;

[0059] When ω > 0, the larger ω is, the greater the influence ability of the subtype production parameter on this type of production problem, that is, the stronger the corresponding relationship; the smaller ω is, the smaller the influence ability of the subtype production parameter on this type of production problem, that is, the weaker the corresponding relationship;

[0060] When ω ≤ 0, the subtype production parameter has no positive influence ability on this type of production problem, that is, the corresponding relationship is weak.

[0061] Step S3 includes: Step S31, corresponding relationship classification. According to the influence ability of different subtype production parameters on different types of production problems in step S23, extract the production parameter subtypes with positive influence ability indexes corresponding to each type of production problem one by one; S32, generate a tree - shaped corresponding relationship diagram. For the production parameter subtypes with positive influence ability indexes extracted for each production problem type, generate a tree - shaped corresponding relationship diagram corresponding to the production problem type. The production problem type is the first level, the production parameter type is the second level, and the production parameter subtypes included in the production parameter type are the third level; S33, adjust the tree - shaped corresponding relationship diagram. According to the influence ability of different subtype production parameters on different types of production problems, adjust the third - level sequence and further adjust the second - level sequence; Step S34, dynamic optimization. The big data platform 1 continuously collects and updates the production record database 4 and the production parameter database 5, updates the production problem type and calculates the influence ability index of the production parameter subtype on the newly generated production problem type, and at the same time dynamically updates the influence ability index of the existing production problem types.

[0062] In the step S33, first, mark the influencing ability indexes in the tree-shaped correspondence graph, and sort them in descending order of values in the third level according to their respective production parameter types; then, extract the production parameter subtypes in the tree-shaped correspondence graph whose influencing ability indexes exceed the set threshold, count the number of production parameter subtypes whose influencing ability indexes exceed the set threshold included in each production parameter type in the second level, and sort the second level in descending order of the included quantity.

[0063] The step S4 includes: step S41, production problem identification, where the production problem identification module 6 identifies the production problems occurring in the production process and extracts the corresponding tree-shaped correspondence graph of the production problems; step S42, fault troubleshooting, corresponding to the first level according to the identified production problem type, and corresponding to the production parameter type according to the sequence in the second level, check the production parameter subtypes in the third level included in the production parameter type in descending order of the influencing ability index. If the fault cause is not found, check the next sequence of the second level until the fault cause is found.

[0064] An industrial production diagnosis system with a tree structure includes: a big data platform 1 and a production monitoring device 8 connected to the big data platform 1;

[0065] The big data platform 1 includes: a memory 2; a processor 3 connected to the memory 2; a production record database 4 set in the memory 2 for storing production record data; a production parameter database 5 set in the memory 2 for receiving and storing different types of production parameters collected by the production monitoring device 8.

[0066] The production monitoring device 8 includes: a production environment monitoring device 81 for collecting production environment parameters and uploading them to the production parameter database 5 of the big data platform 1; a production equipment monitoring device 82 for collecting production equipment parameters and uploading them to the production parameter database 5 of the big data platform 1; a production raw material monitoring device 83 for collecting production raw material parameters and uploading them to the production parameter database 5 of the big data platform 1.

[0067] It further includes: a production problem identification module 6 running on the processor 3 for retrieving the production records in the production record database 4 to identify production problems and identify the production problem types; a production correspondence analysis module 7 running on the processor 3 for analyzing and calculating the correspondence between production parameters and production problems.

[0068] In the embodiment of the present application, the working principle of an industrial production diagnosis method and system with a tree structure is as follows: Quantitatively comparing the correspondence between different sub-types of production parameters and different types of production problems can compare the influence capabilities of different sub-types of production parameters on the same production problem, provide data reference during the troubleshooting process, improve the intuitive level of the correspondence between production parameters and production problems, and improve the troubleshooting efficiency; Establishing a tree-shaped correspondence diagram enhances the logical level of the troubleshooting process and improves the troubleshooting efficiency; Moreover, the content of the tree-shaped correspondence diagram is continuously and dynamically optimized as the data is updated, enabling the content of the tree-shaped correspondence diagram to always adapt to the actual production situation, ensuring the effectiveness of the troubleshooting method, and improving the accuracy of troubleshooting.

[0069] In the embodiment of the present application, according to the influence capabilities of different sub-types of production parameters on different types of production problems, the structure of the tree-shaped correspondence diagram is flexibly adjusted to make the structure of the tree-shaped correspondence diagram always adapt to the actual production situation, ensuring the effectiveness of the troubleshooting method and improving the accuracy of troubleshooting.

[0070] Carry out troubleshooting work according to the tree-shaped correspondence diagram, and sequentially troubleshoot according to the influence capabilities of different sub-types of production parameters on different types of production problems at different levels to prevent omission of the cause of the fault, enhance the logical level of the troubleshooting process, and improve the troubleshooting efficiency.

[0071] The above schematically describes the present invention and its implementation manners. This description is not restrictive, and what is shown in the drawings is only one of the implementation manners of the present invention. The actual structure is not limited to this. Therefore, if those of ordinary skill in the art are inspired by it and design, without creative efforts, a structural manner and an embodiment similar to this technical solution without departing from the purpose of the present invention, they shall fall within the protection scope of the present invention.

Claims

1. An industrial production diagnosis method with a tree structure, characterized in that, It includes the following steps: Step S1, collect parameter data. The production monitoring device (8) collects the values of different types of production parameters during the production process, marks the time, and uploads them to the big data platform (1); Step S2, confirm the correspondence. The production record database (4) and the production parameter database (5) are retrieved respectively, and the correspondence between the production problems in the production records and the production parameters is analyzed; Step S3, establish a tree-shaped correspondence diagram. A multi-level tree-shaped correspondence diagram is established according to the correspondence between the production problems and the production parameters; Step S4, fault analysis. According to the generated tree-shaped correspondence diagram, analyze and identify the sub-types of production parameters corresponding to the faults from top to bottom, and sequentially check the conditions corresponding to the sub-types of production parameters until the cause of the fault is found; The said Step S1 includes: Step S11, collect real-time environmental parameters. The production monitoring device (8) produces sub-type parameters of the production environment, then marks the production time, and uploads the sub-type parameters of the production environment to the big data platform (1); Step S12, collect real-time equipment parameters. After marking the production time, the production monitoring device (8) produces sub-type parameters of the production equipment, and uploads the sub-type parameters of the production equipment to the big data platform (1); Step S13, collect real-time raw material parameters. The production monitoring device (8) produces sub-type parameters of the production raw materials, then marks the production time, and uploads the sub-type parameters of the production raw materials to the big data platform (1); The sub-type parameters of the production environment include but are not limited to: production environment temperature, production environment humidity, production environment noise, production environment atmosphere concentration; The sub-type parameters of the production equipment include but are not limited to: production equipment temperature, production equipment noise intensity, production equipment vibration frequency; The sub-type parameters of the production raw materials include but are not limited to: production raw material performance, production raw material feeding speed; The said Step S2 includes: Step S21, summarize production problems. The production problem identification module (6) retrieves the production record database (4), and identifies the types of production problems in the production records and marks the production time; Step S22, analyze the single correspondence. Extract the production parameters in the production parameter database (5) and the production problems in the production record database (4) at the same production time, and conduct a correspondence analysis on the sub-types of production parameters and the occurrence of production problems; Step S23, summarize the correspondence. Analyze one by one the correspondence between the sub-types of production parameters and the occurrence of production problems at all times according to the production time. According to the analysis results, summarize the correspondence between the types of production problems and the sub-types of production parameters; In the said Step S23, the correspondence analysis method is: Among them, ω represents the influence ability index of a certain sub-type of production parameter on a certain type of production problem; X represents the value of the production parameters of this subtype at a certain production time; X' represents the ideal value of the production parameters of this subtype, which is analyzed and set by production management personnel according to the specific production parameters; Y represents the occurrence parameter of the production problems of this type at the same production time. If the production problems of this type occur at the same production time, then Y is 1. If the production problems of this type do not occur at the same production time, then Y is 0; Y' represents the ideal value of the occurrence index of the production problems of this type, that is, 0. When ω > 0, the larger ω is, the greater the influencing ability of the production parameters of this subtype on the production problems of this type, that is, the stronger the corresponding relationship; the smaller ω is, the smaller the influencing ability of the production parameters of this subtype on the production problems of this type, that is, the weaker the corresponding relationship. When ω ≤ 0, the production parameters of this subtype have no positive influencing ability on the production problems of this type, that is, the corresponding relationship is weak. The step S3 includes: Step S31, corresponding relationship classification. According to the influencing ability of different production parameter subtypes of different types of production problems in step S23, extract the production parameter subtypes with positive influencing ability indices corresponding to each type of production problem one by one. Step S32, generate a tree-shaped corresponding relationship diagram. For the production parameter subtypes with positive influencing ability indices extracted for each production problem type, correspond to the production problem type to generate a tree-shaped corresponding relationship diagram. The production problem type is the first level, the production parameter type is the second level, and the production parameter subtypes included in the production parameter type are the third level. Step S33, adjust the tree-shaped corresponding relationship diagram. According to the influencing ability of different production parameter subtypes on different types of production problems, adjust the third-level sequence and further adjust the second-level sequence. Step S34, dynamic optimization. The big data platform (1) continuously collects and updates the production record database (4) and the production parameter database (5), updates the production problem types and calculates the influencing ability indices of the production parameter subtypes on the updated production problem types, and at the same time dynamically updates the influencing ability indices of the existing production problem types.

Images