Product quality judgment rule management method and system based on digital twinning

Abstract

The invention provides a product quality judgment rule management method and system based on digital twinning, and relates to the technical field of artificial intelligence, and the method comprises the following steps: a rule configuration step: according to the category of a to-be-judged product, establishing a rule group and configuring a rule group access condition; a rule execution step: screening product parameters through a rule group admission condition, obtaining an execution process of the rule group set for judgment, and outputting a judgment result; and a visual platform construction step: constructing a product digital twin production line, and displaying the quality judgment result of the product in each process section. According to the invention, real-time response, intelligent management and high-stability operation can be realized, and precipitation is accelerated to form a business knowledge base.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a product quality determination rule management method and system based on digital twins. Background Art

[0002] At present, the application of product quality determination in process manufacturing industry generally has the following problems. Product quality determination relies on manual subjective judgment, lacks unified standards, and business knowledge is difficult to accumulate and pass on; product quality determination parameters are huge and frequently called, and real-time response is slow; given the variability and complexity of product quality determination requirements in process manufacturing industry, the use of traditional language hard coding is not conducive to expansion and has the risk of focusing on compilation and restart; there is a lack of a full-link visualization platform for the determination process, which cannot form an effective tracking and analysis of data and optimization of rules.

[0003] The invention patent with the publication number CN114862099B discloses an online system of continuous casting quality prediction model based on rule engine, including: rule engine module, used to refine process quality judgment rules, edit core strategies, and form continuous casting quality prediction model; based on the core strategy, make real-time decisions and output decision data for the input L1 and L2 system data; generate decision reports to display the decision data; online deployment module, used to deploy the continuous casting quality prediction model to the online system; adjust the continuous casting quality prediction model through manual inspection results of on-site slabs, and deploy it online again; decision data analysis module, used to judge the quality of slabs predicted by the continuous casting quality prediction model to be fault-free; for slabs predicted by the core strategy to have faults, guide the cutting optimization of slabs according to decision data. Although this patent can meet the continuous casting quality prediction in general scenarios with the help of rule engine, it still does not solve the problem of non-stop update, does not consider the performance problem of large data volume and high-frequency judgment scenarios, and does not establish a full-process visual digital twin platform. Summary of the invention

[0004] In view of the defects in the prior art, the present invention provides a product quality judgment rule management method and system based on digital twins.

[0005] According to a product quality determination rule management method and system based on digital twin provided by the present invention, the scheme is as follows: In a first aspect, a product quality determination rule management method based on digital twins is provided, the method comprising: Rule configuration steps: Create a rule group and configure the rule group access conditions according to the product category to be determined; Rule execution steps: filter product parameters through rule group access conditions, obtain the execution process of the rule group set for judgment, and output the judgment result; Steps for building a visualization platform: Build a digital twin production line for the product to display the quality judgment results of the product in each process stage.

[0006] Preferably, the rule configuration step includes: Step S1.1: Create rule groups for products to be judged by category, and configure the rule group access conditions, using hierarchical AND / OR logic, graphical configuration and natural language description; Step S1.2: Develop a groovy business rule script for tracking rule parameters according to different modes of product quality determination, i.e., determination items; Step S1.3: Construct a rule model for product quality judgment logic. The rule is composed of main and sub-rules. There are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under one sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by natural language description. Step S1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process.

[0007] Preferably, the step S1.3 comprises: Step S1.3.1: The main rule configuration manages the rule parameters such as directory, output code, disposal code, etc. that do not participate in the core calculation logic; Step S1.3.2: The sub-rule configuration includes the rule parameters involved in the core calculation logic, including the judgment item, actual value, comparison symbol, and standard value; among which the actual value and standard value support secondary processing of function expressions.

[0008] Preferably, the rule execution step includes: Step S2.1: filter the product parameters of the digital twin modeling product that are passed in through the rule group access conditions to obtain the rule group set that needs to be judged; Step S2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time; Step S2.3: sequentially obtain the execution process of the rule set, i.e., the decision tree, to make a product quality judgment; Step S2.4: Execute the cleanup Groovy script to output the determination result and determination history of the rule set.

[0009] Preferably, the step S2.3 comprises: Step S2.3.1: Starting from the main rule of the root node of the decision tree, determine the main rules in sequence; Step S2.3.2: According to the judgment result of the main rule, the next node in the decision tree is selected to continue the judgment, that is, enter step S2.3.1; Step S2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output.

[0010] Preferably, the step S2.3.1 comprises: Step S2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rule under the main rule in sequence, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations; Step S2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment and enter step S2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, enter step S2.3.1.1.

[0011] Preferably, the step S2.3.1.1 comprises: Try to load the Groovy script corresponding to the determined project. If the script has been loaded and is the latest version, directly return the Class information and execute the script. Try to load the Groovy script corresponding to the judgment project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script, return the Class information and execute the script.

[0012] In a second aspect, a product quality determination rule management system based on digital twins is provided, the system comprising: Rule configuration module: establish rule groups and configure rule group access conditions according to the product category to be determined; Rule execution module: filters product parameters through rule group access conditions, obtains the execution process of the rule group set for judgment, and outputs the judgment results; Visualization platform construction module: Build a product digital twin production line to display the quality judgment results of the product in each process stage.

[0013] Preferably, the rule configuration module includes: Module M1.1: Establish rule groups for products to be judged by category, and configure the entry conditions of the rule groups, using hierarchical AND / OR logic, graphical configuration and natural language description; Module M1.2: Develop groovy business rule scripts for tracking rule parameters based on different modes of product quality determination, i.e., determination items; Module M1.3: Construct a rule model for product quality judgment logic. The rule is composed of main and sub-rules. There are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under one sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by natural language description. Module M1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process; The module M1.3 includes: Module M1.3.1: Main rule configuration management directory, output code, disposal code and other rule parameters that do not participate in the core calculation logic; Module M1.3.2: Sub-rule configuration includes the rule parameters involved in the core calculation logic, including judgment items, actual values, comparison symbols, and standard values; among which the actual values ​​and standard values ​​support secondary processing of function expressions.

[0014] Preferably, the rule execution module includes: Module M2.1: The product parameters of the digital twin modeling product are filtered through the rule group access conditions to obtain the rule group set that needs to be judged; Module M2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time; Module M2.3: Obtain the execution process of the rule set, i.e., the decision tree, in sequence to make a product quality judgment; The module M2.3 includes: Module M2.3.1: Starting from the main rule of the root node of the decision tree, determine the main rules in sequence; Module M2.3.1 includes: Module M2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rules under the main rule in sequence, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations; Module M2.3.1.1 includes: Try to load the Groovy script corresponding to the determined project. If the script has been loaded and is the latest version, directly return the Class information and execute the script. Try to load the Groovy script corresponding to the determined project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script, return the Class information and execute the script; Module M2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment and enter module M2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, entering module M2.3.1.1; Module M2.3.2: According to the judgment result of the main rule, select the next node in the decision tree to continue the judgment, that is, enter module M2.3.1; Module M2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output; Module M2.4: Execute the cleanup Groovy script to output the determination results and determination history of the rule set.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention adopts the rule group-rule-process model to construct product quality determination rule groups, rules, and process logic in different categories, thereby solving the problem that business knowledge cannot be accumulated and passed on; 2. The present invention adopts decision tree classification to construct the judgment process logic, overcomes the unnecessary performance consumption caused by judgment data redundancy, greatly reduces the time required for complex judgment, and solves the problem of real-time response in the process; 3. The present invention adopts Groovy as the rule scripting language to separately encapsulate the rule calculation logic, thereby solving the problem of type expansion of product quality judgment; dynamic script loading and hot update solve the problem of restart risk; the agile development language based on JVM integrates the powerful features of many high-level languages ​​and solves the problem of real-time response in terms of execution efficiency; 4. The present invention adopts visual monitoring technology to build a full-link visualization platform for the determination process, realizes integrated intelligent management of product quality determination, and solves the problem of failing to form a feedback closed loop and continuous optimization.

[0016] Other beneficial effects of the present invention will be explained in the specific implementation manner through the introduction of specific technical features and technical solutions. Through the introduction of these technical features and technical solutions, those skilled in the art should be able to understand the beneficial technical effects brought about by the technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a system architecture diagram of the present invention; Figure 2 This is a schematic diagram of the rule decision tree interface of the present invention; Figure 3This is a schematic diagram of the rule determination process of the present invention; Figure 4 This is a schematic diagram of the rule script execution of the present invention; Figure 5 This is a schematic diagram of the script loading principle of the present invention. DETAILED DESCRIPTION

[0018] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0019] The embodiment of the present invention provides a product quality judgment rule management method based on digital twins, referring to Figure 1 As shown, the method specifically includes: Rule configuration steps: Create a rule group and configure the rule group access conditions based on the product category to be determined.

[0020] Rule execution steps: filter product parameters through rule group access conditions, obtain the execution process of the rule group set for judgment, and output the judgment result.

[0021] Steps for building a visualization platform: Build a digital twin production line for the product to display the quality judgment results of the product in each process stage.

[0022] Specifically, the rule configuration steps include: Step S1.1: Create rule groups for products to be judged by category and configure access conditions for the rule groups. The basic unit of the access conditions is the "product parameter-target value" correspondence, using hierarchical AND or logic. Graphical configuration is accompanied by natural language description.

[0023] For example: (APN [in] [“X”, “Y”] AND UNIT_CODE [=] “Z001”) OR (LEVEL [=] “1”).

[0024] Step S1.2: According to different modes of product quality judgment, such as all points of the curve, single point, mean, etc., develop a groovy business rule script with rule parameters embedded, hereinafter referred to as the judgment project (preset scripts are included, and the business can be customized).

[0025] Step S1.3: Construct a rule model for product quality judgment logic. The rules are composed of main sub-rules, that is, there are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under 1 sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by a natural language description.

[0026] Specifically, the calculation expressions, inputs and outputs in the product quality judgment logic are mapped into rule parameters, thereby extracting specific master and sub-rules. For example: The rule judgment logic for judging the level of liquid deviation as level 1 is as follows: 1) The maximum actual value of the crystallizer liquid level - the average value of the crystallizer liquid level setting < the upper limit of the liquid level deviation level 1 standard; 2) The average value of the crystallizer liquid level setting - the minimum actual value of the crystallizer liquid level < the upper limit of the liquid level deviation level 1 standard; 3) The minimum actual value of the crystallizer liquid level is greater than the lower limit of the liquid level deviation level 1 standard; 1) 2) 3) are all sub-rules. 1) 2) together form a sub-rule group, and 3) alone forms a sub-rule group. The main rule consists of these two sub-rule groups.

[0027] If condition 3) is met or conditions 1) and 2) are met at the same time, the page deviation level is determined to be level 1 and the corresponding code is output.

[0028] Step S1.3.1: The main rule configuration manages the rule parameters such as directory, output code, disposal code, etc. that do not participate in the core calculation logic.

[0029] Step S1.3.2: Sub-rule configuration includes the rule parameters such as judgment items, actual values ​​(product quality judgment parameter data for digital twin modeling), comparison symbols, standard values ​​(set target values) and so on, which participate in the core calculation logic. The actual values ​​and standard values ​​support secondary processing of function expressions to achieve more complex judgment logic.

[0030] Step S1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process. Figure 2 As shown, it supports convenient operations such as subtree copying, node replacement, and node migration, helping users to quickly build decision trees in the early stages; the interface is simple, clear, and intuitive, which is conducive to user learning, adjustment, optimization, and subsequent operation and maintenance.

[0031] Specifically, refer to Figure 3 The rule execution steps include: Step S2.1: The product parameters of the digital twin modeling product are filtered through the rule group access conditions to obtain the rule group set that needs to be judged.

[0032] Step S2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time to improve the judgment efficiency.

[0033] Step S2.3: Obtain the execution process of the rule group set, i.e., the decision tree, in sequence to make a product quality judgment.

[0034] Step S2.3.1: Starting from the main rule of the decision tree root node, determine the main rules in sequence; Step S2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rules under the main rule in turn, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations. Figure 4 As shown. The comparison operator is used in the execution of the Groovy script to compare the actual value (the left value of the expression) with the standard value (the right value of the expression) to obtain the rule calculation result.

[0035] The following are examples of core calculation descriptions for three basic judgment items: 1. All points: The actual value of each point on the product production process curve is compared with the standard value through the comparison symbol. If and only if all expressions are established, the judgment is passed, otherwise, it is judged to fail.

[0036] 2. Single point: As long as there is an actual value of a point on the product production process curve, the expression can be established by comparing it with the standard value through the comparison symbol, and the judgment is passed. Otherwise, the judgment is failed.

[0037] 3. Mean: First calculate the average value of the actual values ​​of all points on the product production process curve, and then compare it with the standard value through the comparison symbol. If the expression is established, the judgment is passed, otherwise, the judgment is failed.

[0038] Through the above calculations and calculations of other determination items, it is possible to determine high-frequency process data such as temperature, width, thickness, etc. of the product production process.

[0039] Try to load the Groovy script corresponding to the judgment project. If the script has been loaded and the script status is the latest version, directly return the Class information and execute the script to improve the judgment efficiency.

[0040] Try to load the Groovy script corresponding to the judgment project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script. Figure 5 As shown, Class information is returned and the script is executed.

[0041] Step S2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment to improve the judgment efficiency and enter step S2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, enter step S2.3.1.1.

[0042] Step S2.3.2: According to the judgment result of the main rule, the next node in the decision tree is selected to continue the judgment, that is, enter step S2.3.1; Step S2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output.

[0043] Step S2.4: Execute the cleanup Groovy script to output the judgment results and judgment history of the rule group set.

[0044] Specifically, the steps for building a visualization platform include: Step S3.1: Build a digital twin production line for process manufacturing products; Step S3.2: Display the quality determination results of the product in each process stage on the product digital twin production line; Step S3.3: Output the quality judgment results in the form of a legend. Click the result legend to view the quality judgment parameter data in the form of a histogram, line chart, pie chart, etc., and display the judgment process rules and judgment log in a dynamic diagram + decision tree.

[0045] The present invention also provides a product quality determination rule management system based on digital twins, which can be implemented by executing the process steps of the product quality determination rule management method based on digital twins, that is, those skilled in the art can understand the product quality determination rule management method based on digital twins as a preferred implementation of the product quality determination rule management system based on digital twins. The system specifically includes: Rule configuration module: Create rule groups and configure rule group access conditions based on the product category to be determined.

[0046] Rule execution module: filter product parameters through rule group access conditions, obtain the execution process of the rule group set for judgment, and output the judgment result.

[0047] Visualization platform construction module: Build a product digital twin production line to display the quality judgment results of the product in each process stage.

[0048] Specifically, the rule configuration module includes: Module M1.1: Create rule groups for products to be judged by category and configure access conditions for the rule groups. The basic unit of the access conditions is the "product parameter-target value" correspondence, using hierarchical AND or logic. Graphical configuration is accompanied by natural language description.

[0049] For example: (APN [in] [“X”, “Y”] AND UNIT_CODE [=] “Z001”) OR (LEVEL [=] “1”).

[0050] Module M1.2: According to different modes of product quality judgment, such as all points of the curve, single point, mean, etc., develop a groovy business rule script with rule parameters embedded, hereinafter referred to as the judgment project (preset scripts are included, and the business can be customized).

[0051] Module M1.3: Construct a rule model for product quality judgment logic. The rules are composed of main and sub-rules, that is, there are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under 1 sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by natural language description.

[0052] Specifically, the calculation expressions, inputs and outputs in the product quality judgment logic are mapped into rule parameters, thereby extracting specific master and sub-rules. For example: The rule judgment logic for judging the level of liquid deviation as level 1 is as follows: 1) The maximum actual value of the crystallizer liquid level - the average value of the crystallizer liquid level setting < the upper limit of the liquid level deviation level 1 standard; 2) The average value of the crystallizer liquid level setting - the minimum actual value of the crystallizer liquid level < the upper limit of the liquid level deviation level 1 standard; 3) The minimum actual value of the crystallizer liquid level is greater than the lower limit of the liquid level deviation level 1 standard; 1) 2) 3) are all sub-rules. 1) 2) together form a sub-rule group, and 3) alone forms a sub-rule group. The main rule consists of these two sub-rule groups.

[0053] If condition 3) is met or conditions 1) and 2) are met at the same time, the page deviation level is determined to be level 1 and the corresponding code is output.

[0054] Module M1.3.1: Main rule configuration management directory, output code, disposal code and other rule parameters that do not participate in the core calculation logic.

[0055] Module M1.3.2: Sub-rule configuration judgment items, actual values ​​(product quality judgment parameter data for digital twin modeling), comparison symbols, standard values ​​(set target values) and other rule parameters involved in the core calculation logic. The actual values ​​and standard values ​​support secondary processing of function expressions to achieve more complex judgment logic.

[0056] Module M1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process, refer to Figure 2 As shown, it supports convenient operations such as subtree copying, node replacement, and node migration, helping users to quickly build decision trees in the early stages; the interface is simple, clear, and intuitive, which is conducive to user learning, adjustment, optimization, and subsequent operation and maintenance.

[0057] Specifically, refer to Figure 3The rule execution module includes: Module M2.1: Filter the product parameters of the incoming digital twin modeling product through the rule group access conditions to obtain the rule group set that needs to be judged.

[0058] Module M2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time to improve the determination efficiency.

[0059] Module M2.3: Obtain the execution process of the rule set, i.e., the decision tree, in sequence to make a product quality judgment.

[0060] Module M2.3.1: Starting from the main rule of the decision tree root node, determine the main rules in turn; Module M2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rules under the main rule in turn, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations. Figure 4 As shown. The comparison operator is used in the execution of the Groovy script to compare the actual value (the left value of the expression) with the standard value (the right value of the expression) to obtain the rule calculation result.

[0061] The following are examples of core calculation descriptions for three basic judgment items: 1. All points: The actual value of each point on the product production process curve is compared with the standard value through the comparison symbol. If and only if all expressions are established, the judgment is passed, otherwise, it is judged to fail.

[0062] 2. Single point: As long as there is an actual value of a point on the product production process curve, the expression can be established by comparing it with the standard value through the comparison symbol, and the judgment is passed. Otherwise, the judgment is failed.

[0063] 3. Mean: First calculate the average value of the actual values ​​of all points on the product production process curve, and then compare it with the standard value through the comparison symbol. If the expression is established, the judgment is passed, otherwise, the judgment is failed.

[0064] Through the above calculations and calculations of other determination items, it is possible to determine high-frequency process data such as temperature, width, thickness, etc. of the product production process.

[0065] Try to load the Groovy script corresponding to the judgment project. If the script has been loaded and the script status is the latest version, directly return the Class information and execute the script to improve the judgment efficiency.

[0066] Try to load the Groovy script corresponding to the judgment project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script. Figure 5 As shown, Class information is returned and the script is executed.

[0067] Module M2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment to improve the judgment efficiency and enter module M2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, enter module M2.3.1.1.

[0068] Module M2.3.2: According to the judgment result of the main rule, select the next node in the decision tree to continue the judgment, that is, enter module M2.3.1; Module M2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output.

[0069] Module M2.4: Execute the cleanup Groovy script and output the judgment results and judgment history of the rule group set.

[0070] Specifically, the visualization platform building blocks include: Module M3.1: Build a digital twin production line for process manufacturing products; Module M3.2: Display the quality judgment results of the product in each process stage on the product digital twin production line; Module M3.3: Output the quality judgment results in the form of a legend. Click the result legend to view the quality judgment parameter data in the form of a histogram, line chart, pie chart, etc., and display the judgment process rules and judgment log in a dynamic diagram + decision tree.

[0071] The embodiment of the present invention provides a product quality judgment rule management method and system based on digital twins. After the product quality data is modeled by digital twin technology, the rule group-rule-decision tree model is used, combined with a rule scripting language with hot update characteristics to perform quality judgment and display the judgment results on the digital twin visualization platform of the process manufacturing production line. Product quality judgment realizes real-time response, intelligent management, and highly stable operation, accelerating the precipitation to form a business knowledge base.

[0072] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0073] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A product quality judgment rule management method based on digital twins, characterized in that: include: Rule configuration steps: Create a rule group and configure the rule group access conditions according to the product category to be determined; Rule execution steps: filter product parameters through rule group access conditions, obtain the execution process of the rule group set for judgment, and output the judgment result; Steps for building a visualization platform: Build a digital twin production line for the product to display the quality judgment results of the product in each process stage.

2. The product quality determination rule management method based on digital twin according to claim 1 is characterized in that: The rule configuration step includes: Step S1.1: Create rule groups for products to be judged by category, and configure the rule group access conditions, using hierarchical AND / OR logic, graphical configuration and natural language description; Step S1.2: Develop a groovy business rule script for tracking rule parameters according to different modes of product quality determination, i.e., determination items; Step S1.3: Construct a rule model for product quality judgment logic. The rule is composed of main and sub-rules. There are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under one sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by natural language description. Step S1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process.

3. The product quality determination rule management method based on digital twin according to claim 2 is characterized in that: The step S1.3 comprises: Step S1.3.1: The main rule configuration manages the rule parameters that do not participate in the core calculation logic, including the directory, output code, and disposal code; Step S1.3.2: The sub-rule configuration includes the rule parameters involved in the core calculation logic, including the judgment item, actual value, comparison symbol, and standard value; among which the actual value and standard value support secondary processing of function expressions.

4. The product quality determination rule management method based on digital twin according to claim 3 is characterized in that: The rule execution steps include: Step S2.1: filter the product parameters of the digital twin modeling product that are passed in through the rule group access conditions to obtain the rule group set that needs to be judged; Step S2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time; Step S2.3: sequentially obtain the execution process of the rule set, i.e., the decision tree, to make a product quality judgment; Step S2.4: Execute the cleanup Groovy script to output the determination result and determination history of the rule set.

5. The product quality determination rule management method based on digital twin according to claim 4 is characterized in that: The step S2.3 comprises: Step S2.3.1: Starting from the main rule of the root node of the decision tree, determine the main rules in sequence; Step S2.3.2: According to the judgment result of the main rule, the next node in the decision tree is selected to continue the judgment, that is, enter step S2.3.1; Step S2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output.

6. The product quality determination rule management method based on digital twin according to claim 5 is characterized in that: The step S2.3.1 comprises: Step S2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rule under the main rule in sequence, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations; Step S2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment and enter step S2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, enter step S2.3.1.

1.

7. The product quality determination rule management method based on digital twin according to claim 6 is characterized in that: The step S2.3.1.1 comprises: Load the Groovy script corresponding to the judgment project. If the script has been loaded and the script status is the latest version, the Class information is directly returned and the script is executed; Load the Groovy script corresponding to the judgment project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script, return the Class information and execute the script.

8. The product quality determination rule management method based on digital twin according to claim 6 is characterized in that: The visualization platform construction steps include: Step S3.1: Build a product digital twin production line; Step S3.2: Display the quality judgment results of the product in each process stage on the product digital twin production line; Step S3.3: Output the quality judgment results in the form of legends, check the quality judgment parameter data, and display the judgment process rules and judgment logs with dynamic diagrams + decision trees.

9. A product quality judgment rule management system based on digital twins, characterized in that: include: Rule configuration module: establish rule groups and configure rule group access conditions according to the product category to be determined; Rule execution module: filters product parameters through rule group access conditions, obtains the execution process of the rule group set for judgment, and outputs the judgment results; Visualization platform construction module: Build a product digital twin production line to display the quality judgment results of the product in each process stage.

10. The product quality determination rule management system based on digital twin according to claim 9 is characterized in that: The rule configuration module includes: Module M1.1: Establish rule groups for products to be judged by category, and configure the entry conditions of the rule groups, using hierarchical AND / OR logic, graphical configuration and natural language description; Module M1.2: Develop groovy business rule scripts for tracking rule parameters based on different modes of product quality determination, i.e., determination items; Module M1.3: Construct a rule model for product quality judgment logic. The rule is composed of main and sub-rules. There are 1 to n sub-rule groups under the main rule, and 1 to n sub-rules under one sub-rule group. The conditions within the group are AND, and the conditions between groups are OR. At least one group condition satisfies the main rule. The graphical configuration is accompanied by natural language description. Module M1.4: In the graphical interface, drag the main rule to the rule group to form a decision tree representing the rule execution process; The module M1.3 includes: Module M1.3.1: Main rule configuration management directory, output code, disposal code, and other rule parameters that do not participate in the core calculation logic; Module M1.3.2: Sub-rule configuration includes the rule parameters involved in the core calculation logic, including judgment items, actual value, comparison symbol, and standard value; the actual value and standard value support secondary processing of function expressions; The rule execution module includes: Module M2.1: The product parameters of the digital twin modeling product are filtered through the rule group access conditions to obtain the rule group set that needs to be judged; Module M2.2: Execute the initialization configuration Groovy script to obtain the actual value and standard value of the sub-rule configuration of all rules under the rule group set at one time; Module M2.3: Obtain the execution process of the rule set, i.e., the decision tree, in sequence to make a product quality judgment; The module M2.3 includes: Module M2.3.1: Starting from the main rule of the root node of the decision tree, determine the main rules in sequence; Module M2.3.1 includes: Module M2.3.1.1: Obtain the rule parameters including the performance value, comparison symbol, standard value, and judgment item of the sub-rules under the main rule in sequence, load the Groovy rule script corresponding to the judgment item, replace the parameter embedding points in the script with the obtained rule parameters, and perform core calculations; Module M2.3.1.1 includes: Load the Groovy script corresponding to the judgment project. If the script has been loaded and the script status is the latest version, the Class information is directly returned and the script is executed; Load the Groovy script corresponding to the judgment project. If the script has not been loaded or the script status is not the latest version, you need to load and parse the latest version of the script, return the Class information and execute the script; Module M2.3.1.2: According to the script execution result, if the main rule judgment result has been determined, the main rule will immediately end the judgment and enter module M2.3.2; if it is not determined, the remaining sub-rules under the main rule will continue to be executed, that is, entering module M2.3.1.1; Module M2.3.2: According to the judgment result of the main rule, select the next node in the decision tree to continue the judgment, that is, enter module M2.3.1; Module M2.3.3: When the judgment of the first terminal main rule node is completed, the judgment result and judgment history of the rule group are output; Module M2.4: execute the cleanup Groovy script to output the determination result and determination history of the rule group set; The visualization platform building blocks include: Module M3.1: Build a product digital twin production line; Module M3.2: Display the quality judgment results of the product in each process stage on the product digital twin production line; Module M3.3: Output the quality judgment results in the form of legends, check the quality judgment parameter data, and display the judgment process rules and judgment logs with dynamic diagrams + decision trees.

Images