A method for selecting cutting tools based on fuzzy ontology analysis in machining

Abstract

This invention relates to the field of machining and manufacturing technology, and discloses a method for selecting cutting tools based on fuzzy ontology analysis in machining. The method includes receiving physical parameter requirements and extracting objective physical continuous quantities; using a mathematical trapezoidal mapping function and modifiers to transform and exponentiate the objective physical continuous quantities to obtain continuous index membership values; associating the continuous index membership values ​​with data attribute nodes according to a machining semantic hierarchy model to generate an ontology constraint assertion set; extracting feature variables based on a logical deduction instruction chain, performing multiplication calculations and threshold value comparisons, filtering and matching tool number records, and integrating them to generate a sequence of selected tool instances for output display. By transforming continuous physical parameter requirements into quantifiable values, encapsulating numerical variables as structured constraints, and combining Boolean deduction rules to perform collaborative reasoning, the method improves the efficiency of multiple process parameter matching calculations and the objectivity of generating the selected tool instance sequence.

Description

Technical Field

[0001] This invention relates to the field of machining and manufacturing technology, specifically to a method for selecting cutting tools using fuzzy ontology analysis in machining. Background Technology

[0002] The machining and manufacturing process encompasses numerous technological parameters and physical conditions, and the matching of cutting tools directly affects the final quality of the machined product. Existing cutting tool selection methods often employ fixed numerical divisions when dealing with objective, continuous physical quantities in the machining environment. Faced with complex physical parameter requirements, current methods lack a mechanism to convert objective, continuous physical quantities such as thermal hardness, wear resistance, machining accuracy, and toughness into machine-processable quantifiable values. This fixed division method cannot perform trapezoidal mapping transformations and exponential modification operations on continuous performance index values, thus limiting the computational efficiency of the underlying data extraction and transformation stages.

[0003] Tool selection requires a comprehensive evaluation of multiple process indicators, which are closely interdependent. When dealing with the matching of multi-dimensional process parameters, existing methods typically treat each numerical variable in isolation. Due to the lack of a unified semantic hierarchy model for machining, existing data processing workflows cannot encapsulate isolated numerical variables into structured constraints with topological hierarchical relationships. The lack of structured knowledge representation between multi-source performance parameter entities and tool instances increases the complexity of matching multi-dimensional process parameters, making it impossible to establish mapping relationships within the system and generate an effective set of ontology constraint assertions.

[0004] In the final tool matching decision stage, most existing data decision-making processes rely on single-dimensional condition filtering and lack a comprehensive collaborative reasoning mechanism for multiple cutting performance aspects. Due to the lack of verification through logical derivation instruction chains and Boolean operations, existing schemes struggle to conduct objective joint comparisons when faced with multiple independent membership variables. Furthermore, existing evaluation schemes lack quantitative support from a comprehensive evaluation score algebraic model in the output matching decision stage and lack objective screening criteria for descending ordering of candidate tool entities. This results in insufficient objectivity in the generation of tool selection instance sequences, often relying on human experience and judgment. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a fuzzy ontology analysis method for tool selection in machining. This method solves the problems of rigid boundaries in traditional database matching methods during tool selection, which cannot effectively handle the fuzzy semantic requirements of continuous physical performance indicators, and the lack of deep integration and automated reasoning decision-making with multi-dimensional complex working parameters and domain knowledge graphs.

[0006] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for selecting cutting tools based on fuzzy ontology analysis in machining, comprising: Receive physical parameter requirements and extract the basic material information of the workpiece and the objective physical continuity of the cutting process; Calculate the membership value, call the mathematical trapezoidal mapping function to transform the objective physical continuous quantity to obtain the initial membership value, obtain the preset modification operator, and use the modification operator as an exponent to perform a power operation on the initial membership value to obtain the continuous index membership value. Generate ontology assertions, associate continuous index values ​​with data attribute nodes according to the semantic hierarchy model of machining, establish a mapping relationship between performance parameter entities and tool instances, and generate a set of ontology constraint assertions with topological structure based on the mapping relationship. Execute Boolean logic deduction, load the ontology constraint assertion set, extract the feature variables in the ontology constraint assertion set according to the logic deduction instruction chain, perform mathematical multiplication calculation on the feature variables, compare the multiplication calculation result with the threshold value, and filter out the matching tool number records that are greater than the threshold value. Obtain the entity specification parameters corresponding to the matching tool number record, and integrate the entity specification parameters with the matching tool number record to generate a sequence of optional tool instances for output display.

[0007] The objective physical continuous quantities of this invention include thermal hardness, wear resistance, machining accuracy, and toughness values. During the conversion to obtain the initial membership degree value, a trapezoidal membership function model with four node constant parameters is read. These four node constant parameters divide the performance domain into five computational sub-intervals. The objective physical continuous quantity is compared with the four node constant parameters to determine the specific computational sub-interval the objective physical continuous quantity currently falls into. Algebraic evaluation of the corresponding branch is initiated according to the matched specific computational sub-interval, thereby outputting the initial membership degree value. The specific operation logic is as follows: when the objective physical continuous quantity falls into the first or fifth computational sub-interval, a zero constant is directly output; when it falls into the third computational sub-interval, a one constant is directly output; when it falls into the second computational sub-interval, a first-order linear increasing division operation is performed; and when it falls into the fourth computational sub-interval, a first-order linear decreasing division operation is performed.

[0008] During the calculation of continuous index membership values, the modifier operator is extracted and compared logically with the standard value. Based on the comparison result, the modifier logic type is set. If the modifier operator is determined to be greater than the standard value, it is identified as an emphasized operator, and the emphasized operator is invoked to perform a power algebra operation greater than one on the initial membership value. If the modifier operator is determined to be less than the standard value, it is identified as a weakened operator, and the weakened operator is invoked to perform a power algebra operation less than one on the initial membership value.

[0009] In the process of generating ontology assertions, a semantic hierarchy model for machining is established. This model includes part entity classes representing the physical workpiece to be machined, performance parameter entity classes covering the boundary conditions of physical parameters required for the cutting process, material entity classes representing the properties of metal raw materials, tool instance classes storing physical tool information, and fuzzy function classes and fuzzy constraint classes that handle mathematically quantified variables. Object attribute connections and data attribute connections are established between different entity class nodes, and boundary constraints including domain and range restrictions are applied to all attribute connections. Continuous index membership values ​​are bound to data attribute connections, and a machine-readable file containing nodes and connections is generated based on these bindings. A set of fact assertions is generated within this machine-readable file. Further, a set of requirement assertions between part entities and performance factors is established and populated with fact triples; a set of modal assertions is generated and populated with modal fact triples; a set of constraint relationship assertions is generated based on the membership values ​​of continuous indices and populated with tool performance value triples. The requirement assertion set, modal assertion set, and constraint relationship assertion set are integrated to form a fact derivation environment.

[0010] During the Boolean logic derivation process, individual variables of parts and tool instances are extracted from the factual derivation environment. A single-dimensional basic derivation rule containing derivation conditions (antecedents) and derivation conclusions (consequences) is loaded. A logic AND operation operator is invoked to perform a chain judgment on the pre-assertional units containing object attribute association units and data attribute association units. The data attribute association units are parsed to obtain single performance dependency numerical variables. A Boolean greater than judgment operator is invoked to compare the single performance dependency numerical variable with a preset benchmark threshold. When the single performance dependency numerical variable is determined to be greater than the preset benchmark threshold, the derivation condition antecedent is fully satisfied, the derivation trigger operation operator is activated, and a derivation conclusion consequent establishing a one-to-one matching relationship between individual variables of parts and tool instances is generated. Simultaneously, multiple independent dependency numerical variables of various cutting performance characteristics are extracted in parallel from the factual derivation environment. A multi-dimensional collaborative derivation rule containing multiple derivation condition antecedents and a single derivation conclusion consequent is loaded. A logic AND operation operator is invoked to chain multiple independent data attribute association instructions. The Boolean multiplication operator is activated to perform a series algebraic calculations on the extracted independent membership degree numerical variables, and the result of the series algebraic calculation is assigned to the comprehensive performance membership degree numerical variable. The system's built-in comprehensive comparison threshold is extracted, and the Boolean greater than decision operator is called. When the comprehensive performance membership degree numerical variable exceeds the comprehensive comparison threshold, the derivation trigger operation operator is triggered, generating a single derivation conclusion consequent that establishes a global mapping relationship between individual part variables and individual variables of the optimal tool instance.

[0011] In the process of extracting physical parameter requirements, the scene's physical parameters are received and decomposed into independent data dimension variables. Basic material information containing material attribute identifiers is parsed, and the hardness index constants contained within the specific metal grade data are extracted. Machining performance constraints, including target surface roughness and dimensional tolerance values, are extracted and transformed into comprehensive accuracy requirement values. Cutting environment constraint variables, encompassing spindle speed and feed rate limit constants, are established. Based on these constraints, continuous quantities of hot hardness and wear resistance requirements are extracted as objective physical continuous quantities.

[0012] During the generation of the selection sequence, candidate tool entities are extracted from the initial matching pool formed by the matching conditions. A comprehensive evaluation score is calculated for each candidate tool entity. An algebraic model of the comprehensive evaluation score is established, and specific numerical variables and weight constants are input into the model to obtain the comprehensive evaluation score variable value for each candidate tool entity. The candidate tool entities are then sorted in descending order according to the value of their comprehensive evaluation score variable, and the candidate tool entity with the highest ranking is selected as the recommended selection scheme. A final decision list containing tool number characters and comprehensive evaluation score variable values ​​is generated and sent to the human-computer interaction interface for display.

[0013] The innovative principle of this invention lies in combining fuzzy mathematical quantification with ontology semantic web technology to construct a hybrid numerical and logical analysis mechanism for tool selection. By introducing trapezoidal membership mapping and exponential modification operators, continuous physical variables in the cutting environment are analyzed and transformed into discrete membership values ​​that can be processed by machines. Based on the established semantic hierarchy model of machining, independent quantified values ​​are converted into constraint assertions with topological hierarchical relationships, and then Boolean derivation rules are used to perform cross-dimensional dataset multiplication judgments. The above process establishes data associations between process parameter acquisition, semantic assertion encapsulation, and Boolean logic derivation, and uses continuous algebraic operations and threshold filtering to perform matching judgments.

[0014] This invention provides a method for selecting cutting tools based on fuzzy ontology analysis in machining. It offers the following advantages: 1. This invention performs transformation and exponentiation operations on the objective physical continuous quantities of cutting processing conditions by calling the mathematical trapezoidal mapping function and preset modification operators to obtain continuous index membership values. This transforms the continuous mechanical processing physical parameter requirements into machine-processable quantitative values, thereby improving the computational efficiency of the underlying data extraction and transformation process.

[0015] 2. This invention establishes a mapping relationship between performance parameter entities and tool instances by associating continuous index values ​​with data attribute nodes according to the semantic hierarchy model of machining, generating an ontology constraint assertion set, and encapsulating isolated numerical variables into structured constraints with topological hierarchical relationships, thereby reducing the complexity of matching relationships between multi-dimensional process parameters.

[0016] 3. This invention extracts feature variables within the ontology constraint assertion set based on logical deduction instruction chain, calls Boolean operation operators to perform multiplication calculations and threshold value comparison operations, and combines a comprehensive evaluation score algebra model to perform descending order sorting of candidate tool entities, thereby realizing comprehensive collaborative reasoning for multiple cutting performance aspects and improving the objectivity of generating tool selection instance sequences. Attached Figure Description

[0017] Figure 1 This is a system architecture diagram of a tool fuzzy ontology analysis selection method for machining according to the present invention; Figure 2 This is a fuzzy set representation diagram of the continuous performance attributes of the present invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See attached document Figure 1 This invention provides a method for selecting cutting tools based on fuzzy ontology analysis in machining, comprising constructing a hardware and software platform to support system operation. The hardware platform is configured with a computer device having a processor, memory, and hard disk. The software platform deploys a Windows operating system and installs a Java runtime environment. The ontology construction and compilation programs are deployed in the computer device's memory and executed by the processor.

[0020] See attached document Figure 1 The system architecture of this invention is divided into multiple logical execution units. The system architecture includes a data acquisition and input module. This module receives feature data, material parameters, and various performance level requirements for shaft-type parts from the user interface.

[0021] The system architecture also includes a fuzzy processing engine. The fuzzy processing engine connects to the data acquisition input module. It performs mathematical quantization transformation on the received performance requirement information. The engine has a pre-defined fuzzy set representation mechanism and trapezoidal membership calculation formula. Finally, it outputs the calculated membership variables in floating-point numerical format.

[0022] The system architecture includes an ontology knowledge base construction module. This module is connected to the fuzzy processing engine. The ontology knowledge base construction module loads entity classes, subclasses, and object attributes from the machining domain. It maps the membership variables output by the fuzzy processing engine to the ontology model, generating a set of fact assertions that includes performance numerical constraints.

[0023] The system architecture integrates a Semantic Web rule-based language inference engine. This engine calls upon assertion sets generated by the ontology knowledge base construction module. Internally, the engine defines logical matching conditions. It then performs comparisons and determinations of the antecedents of these conditions.

[0024] The system architecture includes a decision output module. This module connects to a semantic web rule-based language inference engine. It extracts the result items from logical decisions and then transforms and displays these result items as a tool list adapted to the machining requirements.

[0025] Based on the above system architecture, the macroscopic workflow of this invention consists of a series of execution steps.

[0026] The data acquisition and input module performs structured extraction of basic part data and operating condition requirements. The acquired data circulates within the system bus and is transmitted to the fuzzy processing engine.

[0027] The fuzzy processing engine initiates a mathematical transformation process. It converts continuous performance requirements into numerical values ​​for variables within a fuzzy domain. After processing using mathematical formulas, the fuzzy processing engine then passes precise numerical membership features to the next level.

[0028] The ontology knowledge base construction module receives numerical membership features. Based on a pre-defined hierarchical framework, it loads instances of the knowledge graph. In this step, the ontology knowledge base construction module fully solidifies the relationships between entity objects and generates a standard ontology file format.

[0029] The Semantic Web rule-based language inference engine loads the standard ontology file. The engine then triggers pre-defined logical rule judgment instructions one by one. The system performs floating-point value comparison calculations in memory. The calculation process uses Boolean greater than instructions to complete the filtering operation.

[0030] The decision output module collects all tool instances that meet the Boolean greater than condition. It then presents the tool IDs and attribute lists of the eligible tools to the execution terminal. The entire data chain operates in an automated closed-loop manner.

[0031] This invention provides a tool fuzzy ontology analysis and selection system for machining, which may include: a data acquisition and input module, a fuzzy processing engine, an ontology knowledge base construction module, a semantic web rule language reasoning engine, and a decision output module.

[0032] The data acquisition and input module connects to the user's physical input terminal. It receives continuous performance index values ​​from the physical input terminal. These performance index values ​​include hot hardness, abrasion resistance, machining accuracy, and toughness. The data acquisition and input module converts the collected continuous performance index values ​​into a structured data format. Finally, it transmits the structured data format to the fuzzy processing engine.

[0033] The fuzzy processing engine receives structured data. Internally, it stores trapezoidal membership function formulas and modifying operators. The engine calls the trapezoidal membership function formulas to perform mapping operations on the structured data. The mapping operation outputs a membership degree value between zero and one. The engine further reads the power parameters of the modifying operators. It then performs a power product calculation on the membership degree values. Finally, the engine generates revised membership degree values ​​and sends them to the ontology knowledge base construction module.

[0034] The ontology knowledge base construction module incorporates a semantic hierarchical network for the mechanical processing domain. This network defines class nodes, subclass nodes, object attribute connections, and data attribute connections. The module reads the corrected dependency level values ​​sent by the fuzzy processing engine. It then binds these corrected dependency level values ​​to the data attribute connections. Based on these bindings, the module generates a machine-readable file containing nodes and connections. This machine-readable file includes a set of fact assertions. Finally, the module pushes this set of fact assertions to the semantic web rule-based language inference engine.

[0035] The Semantic Web rule-based language inference engine stores multiple basic derivation rules. Each basic derivation rule includes antecedents and consequents. The Semantic Web rule-based language inference engine receives a set of factual assertions. It substitutes the values ​​from the factual assertion set into the antecedents. The engine performs Boolean multiplication. It compares the output of the Boolean multiplication with a preset comparison threshold. When the output exceeds the threshold, the engine triggers the consequent. The engine generates a set of valid tool numbers. Finally, it passes this set of valid tool numbers to the decision output module.

[0036] The decision output module receives a set of valid tool IDs. It then queries the underlying database to retrieve the physical parameters of the entities corresponding to these valid tool IDs. Finally, the module displays the physical parameters along with the valid tool IDs on the physical terminal screen. The physical terminal screen then presents a suitable physical tool selection list. This entire process completes the cutting tool analysis and selection operation.

[0037] This invention provides a tool selection method for fuzzy ontology analysis in machining, which may include: Step S1: Receive physical parameter requirements; Step S2: Calculate the membership values ​​of continuous indicators; Step S3: Generate the ontology constraint assertion set; Step S4: Perform Boolean logic rule derivation; Step S5: Output the sequence of selected tool instances.

[0038] The data acquisition and input module executes step S1. The data acquisition and input module receives the basic material information of the workpiece. Simultaneously, the data acquisition and input module extracts the physical parameter requirements for the cutting process. These physical parameter requirements cover wear resistance, hot hardness, machining accuracy, and toughness. The data acquisition and input module converts the aforementioned basic material information and physical parameter requirements into a unified data structure unit. The data acquisition and input module then sends this data structure unit to the downstream node.

[0039] The fuzzy processing engine executes step S2. The fuzzy processing engine receives the data structure unit. The fuzzy processing engine calls its internally stored mathematical trapezoidal mapping function to transform the physical parameter requirements. After processing by the mathematical trapezoidal mapping function, the physical parameter requirements are transformed into initial membership degree values. The fuzzy processing engine obtains a preset modification operator. The fuzzy processing engine uses the modification operator as an exponent to perform a power operation on the initial membership degree values. The fuzzy processing engine obtains continuous index membership values. These continuous index membership values ​​represent the quantitative calculation results of the performance parameters.

[0040] The ontology knowledge base construction module executes step S3. The ontology knowledge base construction module extracts the membership values ​​of continuous indicators. Based on the established machining semantic hierarchy model, the ontology knowledge base construction module associates the membership values ​​of continuous indicators with the corresponding data attribute nodes. The ontology knowledge base construction module establishes a one-to-one mapping relationship between performance parameter entities and tool instances. Based on the mapping relationship, the ontology knowledge base construction module generates a set of ontology constraint assertions with a topological structure. The ontology knowledge base construction module compiles the set of ontology constraint assertions into a machine-readable file.

[0041] The Semantic Web rule-based language inference engine executes step S4. The engine loads the ontology constraint assertion set. The engine has a pre-built logical deduction instruction chain. This chain includes condition extraction terms and threshold comparison judgment terms. The engine extracts feature variables from the ontology constraint assertion set based on the logical deduction instruction chain. It then calls a Boolean multiplication operator to perform mathematical multiplication on the feature variables. Finally, the engine compares the final result of the multiplication with the threshold value in the logical deduction instruction chain. The engine then filters out matching tool number records that are greater than the threshold value.

[0042] The decision output module executes step S5. The decision output module obtains the matching tool number record. The decision output module retrieves the entity specification parameters corresponding to the matching tool number record from the database. The decision output module integrates the entity specification parameters with the matching tool number record to form a sequence of selected tool instances. The decision output module transmits the sequence of selected tool instances to the front-end display device. The front-end display device presents the determined applicable cutting tool selection scheme. The entire workflow cycle is completed.

[0043] See attached document Figure 2 The present invention provides a method for selecting fuzzy ontology analysis of cutting tools for machining, which may include: extracting objective physical continuous quantities, setting the performance domain space, dividing the fuzzy subset structure, establishing mapping correspondence and calculating the subset membership degree value.

[0044] The data acquisition and input module extracts objective physical continuous quantities. These quantities include wear resistance, hot hardness, machining accuracy, and toughness indices. The data acquisition and input module then transmits these objective physical continuous quantities to the fuzzy processing engine.

[0045] The fuzzy processing engine receives objective physical continuous quantities. It performs fuzzy concept quantization calculations on the boundaries of these objective physical continuous quantities. The engine constructs a performance domain based on fuzzy mathematics theory. This performance domain encompasses the entire range of values ​​for objective physical continuous quantities.

[0046] The fuzzy processing engine divides the performance domain into multiple fuzzy subsets. Each fuzzy subset represents a performance metric evaluation level. The engine establishes a mapping relationship between objective physical continuous quantities and these fuzzy subsets. This mapping relationship is expressed using mathematical functions.

[0047] The fuzzy processing engine calls the mapping correspondence to obtain the membership degree value. The membership degree value reflects the degree to which an objective physical continuous quantity belongs to a fuzzy subset. The mathematical model of the mapping correspondence stored internally by the fuzzy processing engine is expressed as follows: ; In the formula: Symbols representing mapping relationships; Represents the domain of performance; This represents the range of values ​​for the degree of subordination. Represents an objective physical continuous quantity element; The sign representing the mapping between variables; The constant representing the degree of belonging of an objective physical continuous quantity element to a fuzzy subset; Represents a fuzzy subset.

[0048] The fuzzy processing engine controls the distribution of the membership degree constant. The degree of association between objective physical continuous elements and fuzzy subsets increases as the membership degree constant increases. The fuzzy processing engine assigns values ​​in the range of zero to one to the membership degree constant.

[0049] The fuzzy processing engine outputs corresponding membership constants for the same objective physical continuous quantity element in different fuzzy subsets. The fuzzy processing engine reads the thermal hardness index values. The fuzzy processing engine determines that the thermal hardness index values ​​lie within the performance domain. The fuzzy processing engine calculates the independent membership constants for the thermal hardness index values ​​belonging to the first, second, and third evaluation levels, respectively.

[0050] The fuzzy processing engine integrates independent dependency constants to generate a dataset. It then packages and outputs the calculated dataset. Finally, the fuzzy processing engine transforms continuous performance indicators into numerical dependency variables.

[0051] This invention provides a method for selecting tool fuzzy ontology analysis for machining, which may include: receiving continuous performance index values, retrieving a trapezoidal membership function model, extracting boundary node parameter constraints, performing piecewise mapping algebra operations, and outputting basic membership variables.

[0052] The fuzzy processing engine receives continuous performance index values ​​from the data acquisition input module. These continuous performance index values ​​include physically quantified attribute values. The fuzzy processing engine then locates the coordinates of these continuous performance index values ​​within the performance domain.

[0053] The fuzzy processing engine reads the trapezoidal membership function model from internal storage. The trapezoidal membership function model establishes a mathematical transformation relationship between objective continuous variables and fuzzy evaluation level intervals. The trapezoidal membership function model sets four node constant parameters. These four node constant parameters divide the entire performance domain into five computational sub-intervals.

[0054] The formula for expressing the trapezoidal membership function model embedded in the fuzzy processing engine is as follows: ; In the formula: This represents the membership result value of the fuzzy evaluation level subset corresponding to the continuous performance parameter of the target; A continuous performance index value representing the input; Represents the constant parameter of the first node; Represents the constant parameter of the second node; Represents the constant parameter of the third node; This represents the constant parameter of the fourth node.

[0055] The fuzzy processing engine compares the continuous performance index value with the four node constant parameters. It then determines the specific calculation sub-interval the continuous performance index value falls into. Based on the matched sub-interval, the engine initiates the algebraic evaluation operation for the corresponding branch.

[0056] When the fuzzy processing engine determines that the continuous performance index value falls within the first and fifth calculation sub-intervals, it directly outputs a zero constant. When the fuzzy processing engine determines that the continuous performance index value falls within the third calculation sub-interval, it directly outputs a one-valued constant. When the fuzzy processing engine determines that the continuous performance index value falls within the second calculation sub-interval, it performs a first-order linearly increasing division operation. When the fuzzy processing engine determines that the continuous performance index value falls within the fourth calculation sub-interval, it performs a first-order linearly decreasing division operation.

[0057] The fuzzy processing engine derives a left-half trapezoidal membership function model based on the standard trapezoidal membership function model. The engine removes the constant parameter at the fourth node and retains the first-order linearly increasing division operation branch to construct the left-half trapezoidal membership function model. Similarly, the engine derives a right-half trapezoidal membership function model based on the standard trapezoidal membership function model. The engine removes the constant parameter at the first node and retains the first-order linearly decreasing division operation branch to construct the right-half trapezoidal membership function model.

[0058] The fuzzy processing engine uses a derived mathematical model to process attribute values ​​located at the boundary regions of the performance domain. The fuzzy processing engine completes the aforementioned piecewise mathematical mapping evaluation. The fuzzy processing engine generates precise values ​​for the basic dependency variables. The fuzzy processing engine then sends these basic dependency variable values ​​to the next stage for subsequent calculations.

[0059] The present invention provides a tool selection method for fuzzy ontology analysis of machining tools, which may include: receiving basic membership variables, extracting preset degree operators, determining the modification logic type, performing power algebra operations, and generating modified membership results.

[0060] The fuzzy processing engine performs a degree of fuzzy constraint adjustment operation. The fuzzy processing engine receives the basic dependency variable output from the previous calculation stage.

[0061] The fuzzing engine reads the preset degree operator from the storage unit. The preset degree operator measures the strength level of the fuzzy concept modification. The fuzzing engine constructs a triplet structure variable containing the original fuzzy subset, the preset degree operator, and the entire performance universe.

[0062] The fuzzy processing engine internally establishes a mathematical model expression formula to limit and modify the degree of fuzziness: ; In the formula: Symbols representing the mapping relationship of the modified target fuzzy set; Represents an objective physical continuous quantity element; This represents the modified fuzzy target set; The constant representing the degree of belonging of an objective physical continuous quantity element to the original fuzzy subset; Represents the preset degree operator; This represents the algebraic operator for exponentiation.

[0063] The fuzzy processing engine extracts the preset degree operator value. It then performs a logical comparison between this preset degree operator value and a standard value. Based on the comparison result, the engine sets the modification logic type.

[0064] The fuzzing engine determines if the preset level operator value is greater than the standard value. The fuzzing engine then classifies the current preset level operator as an emphasis operator.

[0065] The fuzzy processing engine invokes an emphasis operator to perform a power-law algebra operation greater than one on the membership degree constant. This power-law algebra operation lowers the value of the membership degree constant. The fuzzy processing engine thus enhances and modifies the evaluation concept of the original fuzzy subset.

[0066] The fuzzing engine determines if the preset level operator value is less than the standard value. The fuzzing engine then classifies the current preset level operator as a weakened operator.

[0067] The fuzzy processing engine invokes a weakening operator to perform a power-law algebra operation less than one on the membership degree constant. This power-law algebra operation increases the value of the membership degree constant. The fuzzy processing engine thus implements weakening modification control on the evaluation concept of the original fuzzy subset.

[0068] The fuzzy processing engine performs the power algebra operation. The fuzzy processing engine obtains the corrected dependency result. The fuzzy processing engine sends the corrected dependency result to the ontology knowledge base construction module. The ontology knowledge base construction module expands the entity attribute binding configuration based on the corrected dependency result.

[0069] This invention provides a method for selecting fuzzy ontology analysis tools for machining, which includes: extracting domain keywords, defining core ontology class nodes, dividing performance factor subclass structures, establishing tool subordinate branch nodes, and generating semantic ontology network hierarchy.

[0070] The ontology knowledge base construction module performs the ontology framework building operation in the field of machining. This module receives basic data input instructions. It also organizes industrial technical terms relevant to machining application scenarios according to International Organization for Standardization (ISO) standards.

[0071] The ontology knowledge base construction module establishes entity class collection nodes. These nodes are used to group object entities with similar physical characteristics. The ontology knowledge base construction module abstracts domain knowledge into a machine-readable classification tree structure.

[0072] The ontology knowledge base construction module defines several first-level core class nodes. It creates part entity classes, which represent the physical workpiece objects to be processed. It also creates performance factor entity classes, which cover the various physical parameters and boundary conditions required for the cutting process. Finally, it creates material entity classes, representing the properties of the metal raw materials used to manufacture the workpiece. The module also creates tool entity classes, storing information about the physical tools involved in the cutting operation. Finally, it creates fuzzy function classes and fuzzy constraint classes, which handle the variables output from the mathematical quantization process.

[0073] The ontology knowledge base construction module creates second-level subclass nodes under the first-level core class nodes. The ontology knowledge base construction module adds a wear resistance requirement subclass to the performance factor entity class. The ontology knowledge base construction module adds a machining accuracy requirement subclass to the performance factor entity class. The ontology knowledge base construction module adds a toughness requirement subclass to the performance factor entity class. The ontology knowledge base construction module adds a thermohardness requirement subclass to the performance factor entity class.

[0074] The ontology knowledge base construction module derives a fuzziness level constraint subclass from the fuzzy constraint class. The fuzziness level constraint subclass records the modifier operators extracted from exponentiation algebra operations. The ontology knowledge base construction module further divides the fuzzy function class into trapezoidal subclasses, left-half trapezoidal subclasses, and right-half trapezoidal subclasses.

[0075] The ontology knowledge base construction module establishes multiple sub-material property branch nodes within the tool entity class. It assigns tool material subclasses, tool wear resistance information subclasses, tool hot hardness information subclasses, tool machining accuracy information subclasses, and tool toughness information subclasses to the tool entity class.

[0076] The ontology knowledge base construction module completes the assignment of node names and the binding of membership relationships within the classification tree structure. This module integrates scattered mechanical processing knowledge into a structured semantic model framework. It also sets up a basic container for subsequent entity attribute associations.

[0077] This invention provides a method for selecting fuzzy ontology analysis of cutting tools for machining, which includes: classifying ontology attribute types, setting object attribute node connections, setting data attribute node connections, defining the associated domain, and defining the associated value domain.

[0078] The ontology knowledge base construction module establishes attribute associations between different entity class nodes. Based on the differences between the objects at both ends of the attribute connection, the module divides the associations into object attribute connections and data attribute connections.

[0079] The ontology knowledge base construction module creates object attribute connections. These connections represent the semantic interaction states between individuals of different entity classes. The module also establishes connections for wear resistance requirements, part material constraints, machining accuracy requirements, toughness requirements, and thermal hardness requirements.

[0080] The ontology knowledge base construction module creates data attribute connections. These connections link entity classes to numerical variables. The module also establishes membership connections for wear resistance, machining accuracy, toughness, and thermal hardness.

[0081] The ontology knowledge base construction module applies boundary constraints to all attribute connections. Boundary constraints include domain constraints and range constraints. Domain constraints specify the range of class nodes to which the starting point of an attribute connection must belong. Range constraints specify the class node category and numeric format to which the ending point of an attribute connection must reach.

[0082] The ontology knowledge base construction module performs constraint assignment on object attribute connections. It binds the domain of the wear resistance requirement connection to the part entity class and the value domain of the wear resistance requirement connection to the wear resistance requirement subclass. It also binds the domain of the part material constraint connection to the part entity class and the value domain of the part material constraint connection to the material entity class. Furthermore, it binds the domain of the machining accuracy requirement connection to the part entity class and the value domain of the machining accuracy requirement connection to the machining accuracy requirement subclass. Finally, it binds the domain of the toughness requirement connection to the part entity class and the value domain of the toughness requirement connection to the toughness requirement subclass. Finally, it binds the domain of the hot hardness requirement connection to the part entity class and the value domain of the hot hardness requirement connection to the hot hardness requirement subclass.

[0083] The ontology knowledge base construction module performs constraint assignment on the data attribute connections. It receives continuous index membership values ​​output by the fuzzy processing engine. The module unifies the domain of the four membership connections established above to the tool entity class. It also sets the value range of the four membership connections established above to a floating-point data format.

[0084] The ontology knowledge base construction module completes the mapping configuration of each entity class individual within the classification tree structure to numerical variables. It integrates predecessor and successor instances to form a machine-readable file with attribute constraints. Finally, the module transmits this machine-readable file, containing underlying relational logic, to the semantic web rule-based language inference engine.

[0085] This invention provides a method for selecting tool fuzzy ontology analysis for machining, which includes: transforming mathematical calculation parameters, mapping membership function nodes, binding numerical data attributes, mapping modification operator nodes, and generating instance association networks.

[0086] The ontology knowledge base construction module performs the mapping operation from the mathematical model to the semantic ontology. This module receives mathematical quantization parameters transmitted from the fuzzy processing engine. These parameters include trapezoidal computation node variables and membership degree constants.

[0087] The ontology knowledge base construction module performs parameter mapping on the branches of fuzzy membership function nodes. It establishes a mapping and binding between mathematically quantified parameters and trapezoidal subclass nodes. The module allocates data attribute space for the trapezoidal subclass nodes. Based on the mathematical computational architecture, the module establishes mapping equations. ; In the formula: A container representing the collection of data attributes of trapezoidal subclass nodes; Represents the constant parameter of the first node; Represents the constant parameter of the second node; Represents the constant parameter of the third node; Represents the constant parameter of the fourth node; This represents the degree of subordination constant.

[0088] The ontology knowledge base construction module writes specific floating-point values ​​into the aforementioned data attribute set container. The ontology knowledge base construction module completes the loading of the first node constant parameters into the underlying data nodes of the ontology framework. Simultaneously, the ontology knowledge base construction module completes the loading of the dependency degree constants into the underlying data nodes of the ontology framework.

[0089] The ontology knowledge base construction module establishes the inclusion association between performance factor concepts and trapezoidal subclass nodes. This module transforms continuous performance attributes into concrete concept instances. These concept instances encompass both general hot-hardening instances and well-hardening instances. The module controls the connection between these concept instances and their corresponding trapezoidal subclass nodes through object attribute links.

[0090] The ontology knowledge base construction module executes the modification operator mapping at the fuzziness level constraint node branch. The ontology knowledge base construction module receives preset fuzziness level operator values ​​sent by the fuzzing engine. The ontology knowledge base construction module allocates fuzziness level constraint subclasses with fuzziness level constraints attribute space. The ontology knowledge base construction module establishes the modification operator mapping equation: ; In the formula: A container for a collection of data attributes that indicates the degree of ambiguity limiting the subclass; This represents the preset degree operator.

[0091] The ontology knowledge base construction module writes the preset degree operator values ​​into the data attribute collection container of the fuzziness degree-limited subclass. The ontology knowledge base construction module associates and binds individual concept instances with the fuzziness degree-limited subclass.

[0092] The ontology knowledge base construction module integrates the loaded numerical units with object connections. It completes the expansion and transformation from pure mathematical computational logic to a machine-readable semantic network. The module generates a set of fact assertions with clearly defined numerical boundaries. Finally, it submits this set of fact assertions to the semantic web rule-based language inference engine for the next step of inference.

[0093] This invention provides a tool selection method for fuzzy ontology analysis of machining tools, comprising: receiving entity machining features, generating a set of requirement assertions, generating a set of modal assertions, generating a set of constraint assertions, and constructing a fact derivation environment.

[0094] The ontology knowledge base construction module generates a dynamic assertion set. This module also establishes a data communication connection with the semantic web rule-based language inference engine.

[0095] The ontology knowledge base construction module receives part material elements and processing physical parameters from the data acquisition input module. The ontology knowledge base construction module establishes a set of requirement assertions between the part entity and its performance factors. This set of requirement assertions is denoted as... .

[0096] The ontology knowledge base construction module internally establishes a requirement assertion set expression model: ; In the formula: Represents a set of requirement assertions; This indicates the association of object attributes between a part entity and the performance requirements. Indicates the first Individual parts; Indicates the first Individual performance requirements.

[0097] The ontology knowledge base construction module fills the requirement assertion set with fact triples. The ontology knowledge base construction module establishes part material constraint triples. The ontology knowledge base construction module establishes wear resistance requirement triples. The ontology knowledge base construction module establishes thermohardness requirement triples. The ontology knowledge base construction module establishes machining accuracy requirement triples. The ontology knowledge base construction module establishes toughness requirement triples.

[0098] The ontology knowledge base construction module generates a set of modal assertions based on the modifier logic types output by the fuzzy processing engine. The set of modal assertions is denoted as... .

[0099] The ontology knowledge base construction module internally establishes a modal assertion set representation model: ; In the formula: Represents the set of modal assertions; This indicates the performance level requirement for individual data attributes that are associated with the modifier operator; Indicates the first Individual performance requirements; Indicates the first A modifier operator value.

[0100] The ontology knowledge base construction module fills the modal assertion set with modal fact triples. The ontology knowledge base construction module establishes a wear resistance modal triple. The ontology knowledge base construction module establishes a machining accuracy modal triple. The ontology knowledge base construction module establishes a toughness modal triple.

[0101] The ontology knowledge base construction module generates a set of constraint assertions based on the obtained modified membership results. .

[0102] The ontology knowledge base construction module internally establishes a set of constraint relation assertions to represent the model: ; In the formula: Represents the set of constraint assertions; This indicates a data attribute association where a tool instance points to a dependency value; Indicates the first Individual tool instances; Indicates the first Each degree of membership.

[0103] The ontology knowledge base construction module fills the constraint assertion set with tool performance value triples. The ontology knowledge base construction module establishes wear resistance membership triples. The ontology knowledge base construction module establishes hot hardness membership triples. The ontology knowledge base construction module establishes toughness membership triples.

[0104] The ontology knowledge base construction module integrates the requirement assertion set, modal assertion set, and contract constraint relation assertion set. This module then imports the integrated set of data into the Semantic Web rule-based language inference engine. The Semantic Web rule-based language inference engine then constructs a fact-reasoning environment in memory space.

[0105] This invention provides a tool selection method for fuzzy ontology analysis in machining, which includes: extracting fact assertion data, loading condition antecedent instructions, performing threshold comparison operations, triggering conclusion consequent generation, and establishing single-item matching associations.

[0106] The Semantic Web rule-based language inference engine executes the design and execution of single-performance-constraint inference rules. The engine reads the fact-reasoning environment built in memory. This environment is generated and imported by the ontology knowledge base construction module.

[0107] The Semantic Web rule-based language inference engine is internally configured with multiple single-dimensional basic inference rules. Each single-dimensional basic inference rule includes the antecedent of the inference condition and the consequent of the inference conclusion. The Semantic Web rule-based language inference engine extracts individual variables of parts and individual variables of tool instances from the factual inference environment.

[0108] The Semantic Web rule-based language inference engine internally establishes a single-dimensional basic inference rule logical operation model: ; In the formula: This represents a single-dimensional basic derivation rule; Represents individual variables of a part; Represents logical AND operators; Represents an individual variable of a tool instance; This indicates the association of object attributes between individual parts and performance requirements. Represents an individual's requirement for a single level of performance; This indicates a data attribute association where a tool instance points to a dependency value; Numerical variables representing the degree of dependence of a single performance metric; Represents the Boolean greater than criterion operator; This represents the preset comparison threshold value; This represents the derivation of the triggering operation operator; This indicates the object attribute association between an individual part and a matching tool instance.

[0109] The Semantic Web rule-based language inference engine loads the derivation condition antecedents. These antecedents encompass both logical AND operations and Boolean greater-than decisions. The Semantic Web rule-based language inference engine invokes the logical AND operations to perform concatenated decisions on the pre-assertional units. The pre-assertional units include object attribute association units and data attribute association units.

[0110] The Semantic Web rule-based language inference engine parses data attribute association units to obtain single performance dependency numerical variables. It extracts preset comparison threshold values ​​stored in memory. Finally, it calls a Boolean greater than operator to perform numerical comparison operations.

[0111] The Semantic Web rule-based language inference engine determines cases where a single performance dependency value exceeds a preset comparison threshold. The engine then establishes a state where all antecedents of the inference conditions are fully satisfied.

[0112] The Semantic Web rule-based language inference engine activates the derivation triggering operation operator. The Semantic Web rule-based language inference engine generates the derivation conclusion consequent. The derivation conclusion consequent establishes a one-to-one matching relationship between individual variables of the part and individual variables of the tool instance.

[0113] The Semantic Web rule-based language inference engine independently performs unidirectional performance derivation calculations based on wear resistance requirements. It also independently performs unidirectional performance derivation calculations based on machining accuracy requirements. The engine outputs a list of entities that meet each single constraint to the next-level cache space. Finally, the engine completes logical derivation tasks under a single metric dimension.

[0114] This invention provides a method for selecting tool fuzzy ontology analysis for machining, which includes: extracting multidimensional feature variables, loading comprehensive derivation conditions, performing Boolean multiplication operations, determining joint comparison thresholds, and outputting comprehensive matching results.

[0115] The Semantic Web rule-based language inference engine performs the design and execution of collaborative inference rules across all dimensions and with comprehensive performance. It reads multi-dimensional entity attribute data from the fact-reasoning environment.

[0116] The Semantic Web rule-based language inference engine is internally configured with multi-dimensional collaborative inference rules. These rules contain multiple antecedents of inference conditions and a single consequent of inference conclusion. The Semantic Web rule-based language inference engine extracts the degree of dependence values ​​of multiple cutting performance parameters in parallel from the factual inference environment.

[0117] Semantic Web rule-based language inference engine establishes a multi-dimensional collaborative inference rule logic operation model: ; In the formula: This represents a multi-dimensional collaborative derivation rule; Represents individual variables of a part; Represents logical AND operators; Represents an individual variable of a tool instance; This indicates a data attribute association between a tool instance and the wear resistance dependency. The numerical variable representing the degree of attribution of wear resistance; This indicates a data attribute association between a tool instance and the hot hardness dependency. Numerical variable representing the degree of dependence of thermal hardness; This indicates a data attribute association between a tool instance and the machining accuracy dependency. The numerical variable representing the degree of dependence of machining accuracy; This indicates a data attribute association between a tool instance and the toughness dependency. Table of resilience dependency numerical variables; Represents the Boolean multiplication operator; Numerical variables representing the degree of dependence in overall performance; Represents the Boolean greater than criterion operator; This represents the threshold value for comprehensive comparison; This represents the derivation of the triggering operation operator; This indicates the object attribute association between an individual part and a comprehensive matching tool instance.

[0118] The Semantic Web rule-based language inference engine loads multi-dimensional collaborative inference rules. The engine invokes logic and operational operators to concatenate four independent data attribute association instructions.

[0119] The Semantic Web rule language inference engine parses four independent data attribute association instructions to obtain all independent dependency numerical variables. The Semantic Web rule language inference engine activates the Boolean multiplication operator. The Semantic Web rule language inference engine performs algebraic multiplication on the extracted independent dependency numerical variables.

[0120] The Semantic Web rule-based language inference engine assigns the result of the multiplication algebra calculation to the comprehensive performance membership degree numerical variable. The Semantic Web rule-based language inference engine also extracts the system's built-in comprehensive comparison threshold value.

[0121] The Semantic Web rule-based language inference engine invokes the Boolean greater than decision operator to evaluate numerical values. The engine determines that the overall performance dependency value of the variable is greater than the overall comparison threshold. Finally, the engine confirms that the current input variable satisfies the multi-dimensional collaborative derivation rules.

[0122] The Semantic Web rule-based language inference engine triggers inference-triggered operators. The Semantic Web rule-based language inference engine generates a single inference conclusion consequent. This single inference conclusion consequent establishes a global mapping relationship between individual part variables and individual variables of the optimal tool instance within the system.

[0123] The Semantic Web rule-based language inference engine outputs global mapping relationship data. This data is then passed to the decision output module. The decision output module performs a tool selection and configuration display operation based on the global mapping relationship data.

[0124] This invention provides a method for selecting fuzzy ontology analysis tools for machining, which includes: receiving scene physical parameters, parsing basic material elements, extracting machining performance limitations, establishing entity object constraints, and generating scene data packets.

[0125] The data acquisition and input module performs scene input setting operations. It receives scene physical parameters, which originate from specific machining task instructions. The data acquisition and input module then breaks down these scene physical parameters into independent data dimension variables.

[0126] The data acquisition input module parses the basic material elements. These elements include the material attribute identifier of the workpiece to be processed. The material attribute identifier indicates the specific metal grade data. The data acquisition input module extracts the hardness index constant contained within the corresponding metal grade data.

[0127] The data acquisition and input module extracts the machining performance constraints. These constraints include the target surface roughness value and the dimensional tolerance value. The data acquisition and input module then converts these values ​​into a comprehensive accuracy requirement value.

[0128] The data acquisition and input module establishes cutting environment constraint variables. These variables include spindle speed limit constants and feed rate limit constants. Based on these constraints, the module extracts continuous quantities for hot hardness and wear resistance requirements. Finally, the module transmits these extracted numerical variables to the ontology knowledge base construction module.

[0129] The ontology knowledge base construction module receives basic material elements and processing performance constraint variables. Within the semantic hierarchical network, the ontology knowledge base construction module instantiates individual scene parts.

[0130] The ontology knowledge base construction module establishes constraint associations for entity objects. It binds individual scene parts to their material attribute identifiers. It also binds individual scene parts to continuous quantities related to thermal hardness requirements. Finally, it binds individual scene parts to numerical values ​​related to overall accuracy requirements.

[0131] The ontology knowledge base construction module internally sets up a scenario constraint expression model: ; In the formula: Represents the set of scenario constraints; This indicates an object attribute association that links individual parts to material properties; Represents the individual part currently being processed; Represents a specific metal grade; Represents logical AND operators; This indicates the association of object attributes between individual parts and performance requirements. This represents an individual's requirement for a single level of performance.

[0132] The ontology knowledge base construction module invokes logic and operation operators to concatenate multiple independent object attribute binding instructions. It then packages and integrates all independent constraint entries. Finally, it generates a complete scene data package. This package is then sent to the fuzzy processing engine for quantization calculations.

[0133] This invention provides a method for selecting tool fuzzy ontology analysis for machining, which includes: extracting scene physical parameters, performing operator mapping algebra operations, establishing ontology concept instances, loading floating-point values, and generating parameterized fact assertions.

[0134] The fuzzing engine receives scene data packets. It then extracts the continuous physical variables within these packets. These continuous physical variables include continuous quantities related to thermal hardness requirements and wear resistance requirements.

[0135] The fuzzy processing engine reads the matching trapezoidal membership function model. It then substitutes the continuous quantity of the thermal hardness requirement into the model and performs piecewise mapping algebraic operations. The engine then derives the values ​​of the basic membership variables.

[0136] The fuzzy processing engine extracts preset degree operator values. It then uses these preset degree operator values ​​to perform exponential algebra operations on the base dependency variable values. Finally, the engine generates a corrected dependency result. This corrected dependency result represents the quantitative rating under the current processing task environment.

[0137] The ontology knowledge base construction module receives the corrected membership degree results. It performs the conversion operation from numerical values ​​to the ontology knowledge level. The module instantiates fuzzy concept individuals within the semantic hierarchical network. These fuzzy concept individuals correspond to the quantitative ratings obtained from the aforementioned calculations.

[0138] The ontology knowledge base construction module establishes an assertion triple loading model: ; In the formula: Represents a set of instance assertions; Indicates data attribute association; The subject represents an instance of the ontology concept; Represents a floating-point numerical object variable.

[0139] The ontology knowledge base construction module assigns fuzzy concept individuals to ontology concept instance subjects. The ontology knowledge base construction module also assigns the modified membership result to floating-point numeric object variables.

[0140] The ontology knowledge base construction module calls data attributes to associate and link ontology concept instances, subjects, and floating-point numerical object variables. The ontology knowledge base construction module completes attribute binding for a single physical dimension. The ontology knowledge base construction module generates specific parameterized fact assertions.

[0141] The ontology knowledge base construction module synchronously performs the same instance transformation operation for continuous quantities with wear resistance requirements. Within the semantic hierarchical network, the ontology knowledge base construction module generates multiple parameterized fact assertions.

[0142] The ontology knowledge base construction module integrates all parameterized fact assertions to generate an instance assertion set. This module then outputs the instance assertion set to the Semantic Web rule-based language inference engine. The Semantic Web rule-based language inference engine uses this instance assertion set to perform subsequent multi-dimensional rule comparisons.

[0143] This invention provides a method for selecting fuzzy ontology analysis tools for machining, comprising: receiving a set of instance assertions, performing logical rule matching, establishing entity mapping associations, calculating a comprehensive evaluation score, and outputting a final decision list.

[0144] The Semantic Web rule-based language inference engine receives the set of instance assertions transmitted by the ontology knowledge base construction module. It then loads this set of instance assertions into the fact-reasoning environment. Next, the engine activates multi-dimensional collaborative inference rules. Finally, it performs a step-by-step comparison between the set of instance assertions and the preconditions within these multi-dimensional collaborative inference rules.

[0145] The Semantic Web rule-based language inference engine determines whether the set of instance assertions satisfies the preconditions. It then triggers the consequent of the derivation. Next, it generates mapping connections between scene part individuals and candidate tool entities. Finally, it aggregates candidate tool entities that meet the logical constraints to form an initial matching pool.

[0146] The decision output module connects to the Semantic Web rule-based language inference engine. It extracts candidate tool entities from the initial matching pool. The module then performs a comprehensive evaluation score calculation for each candidate tool entity. Finally, it establishes an algebraic model for the comprehensive evaluation score. ; In the formula: This represents the overall evaluation score variable; Represents the first weighting constant; Table of wear resistance subordination numerical variables; Represents the second weighting constant; Numerical variable representing the degree of dependence of thermal hardness; Represents the third weighting constant; The numerical variable representing the degree of dependence of machining accuracy; Represents the fourth weighting constant; The numerical variable representing the degree of resilience dependency.

[0147] The decision output module inputs specific numerical variables and weighting constants into the comprehensive evaluation score algebraic model. The decision output module calculates the comprehensive evaluation score variable value for each candidate tool entity. The decision output module then sorts the candidate tool entities in descending order based on the magnitude of their comprehensive evaluation score variable values.

[0148] The decision output module extracts the top-ranked candidate tool entity. This top-ranked candidate tool entity is then designated as the recommended selection. The final decision list is generated, containing both the tool number and the overall evaluation score.

[0149] The decision output module sends the final decision list to the human-computer interaction interface for display. The decision output module completes the closed-loop process of tool selection under the machining scenario conditions.

Claims

1. A method for selecting cutting tools based on fuzzy ontology analysis in machining, characterized in that, include: Receive physical parameter requirements and extract basic material information of the workpiece and objective physical continuity of the cutting process; Calculate the membership value, call the mathematical trapezoidal mapping function to transform the objective physical continuous quantity to obtain the initial membership value, obtain the preset modification operator, and use the modification operator as an exponent to perform a power operation on the initial membership value to obtain the continuous index membership value. Generate ontology assertions, associate continuous index values ​​with data attribute nodes according to the semantic hierarchy model of machining, establish mapping associations between performance parameter entities and tool instances, and generate a set of ontology constraint assertions with topological structure based on the mapping associations. Execute Boolean logic deduction, load the ontology constraint assertion set, extract the feature variables in the ontology constraint assertion set according to the logic deduction instruction chain, perform mathematical multiplication calculation on the feature variables, compare the multiplication calculation result with the threshold value, and filter out the matching tool number records that are greater than the threshold value. Obtain the entity specification parameters corresponding to the matching tool number record, integrate the entity specification parameters with the matching tool number record to generate a sequence of selected tool instances for output display, and complete a tool selection method based on fuzzy ontology analysis for machining.

2. The method for selecting fuzzy body analysis tools for machining according to claim 1, characterized in that, Objective physical continuous quantities include thermal hardness values, wear resistance values, machining accuracy values, and toughness values; Calling the mathematical trapezoidal mapping function to transform objective physical continuous quantities to obtain initial membership values ​​includes: Read the trapezoidal membership function model with four node constant parameters, which divide the performance domain into five computational subintervals. The objective physical continuous quantity is compared with the four node constant parameters to determine the specific calculation subinterval in which the objective physical continuous quantity currently falls. The algebraic evaluation operation of the corresponding branch is initiated according to the specific computational sub-interval of the match, and the initial membership degree value is output.

3. The method for selecting fuzzy body analysis tools for machining according to claim 2, characterized in that, Calling the mathematical trapezoidal mapping function to transform objective physical continuous quantities to obtain initial membership values ​​includes: When an objective physical continuous quantity falls into the first and fifth calculation sub-intervals, the zero constant is output directly. When an objective physical continuous quantity falls into the third calculation sub-interval, a single constant is directly output. When it is determined that an objective physical continuous quantity falls into the second calculation sub-interval, a first-order linear incremental division operation is performed. When it is determined that an objective physical continuous quantity falls into the fourth calculation sub-interval, a first-order linear decreasing division operation is performed.

4. The method for selecting fuzzy ontology analysis tools for machining according to claim 1, characterized in that, By exponentiating the initial membership values ​​with the modifier as an exponent, continuous membership values ​​are obtained, including: Extract the modifier operator, compare the modifier operator with the standard value, and set the modifier logic type based on the comparison result. When the modifier operator is determined to be greater than the standard value of one, the current modifier operator is identified as an emphasis operator, and the emphasis operator is called to perform a power algebra operation greater than one on the initial membership degree value; When the modifier operator is determined to be less than the standard value of one, the current modifier operator is identified as a weakened operator, and the weakened operator is called to perform a power algebra operation less than one on the initial membership degree value.

5. The method for selecting cutting tools based on fuzzy ontology analysis for machining according to claim 1, characterized in that, According to the semantic hierarchy model of machining, the continuous index values ​​are associated with data attribute nodes, including: A semantic hierarchy model for machining is established. The semantic hierarchy model for machining includes part entity classes that represent physical workpiece objects to be machined, performance parameter entity classes that cover physical parameters and boundary conditions required for the cutting process, material entity classes that represent the properties of metal raw materials, tool instance classes that store physical tool information, and fuzzy function classes and fuzzy constraint classes that accept mathematical quantification variables. Establish object attribute connections between different entity class nodes and data attribute connections between data attribute nodes, and apply boundary constraints containing domain restriction clauses and value domain restriction clauses to all attribute connections; The continuous indicator's value is bound to the data attribute by connecting the data. Based on the binding, a machine-readable file containing nodes and connecting lines is generated. The machine-readable file contains a set of fact assertions.

6. The method for selecting a tool based on fuzzy ontology analysis for machining according to claim 1, characterized in that, Generate a set of ontology constraint assertions with topological structure based on mapping associations, including: Establish a set of requirement assertions between the part entity and performance factors, and fill the requirement assertion set with fact triples; Generate a set of modal assertions and fill the set of modal assertions with modal fact triples; Generate a set of constraint relationship assertions based on the membership values ​​of continuous indices, and fill the set of constraint relationship assertions with triplets of tool performance values; The set of integrated demand assertions, modal assertions, and contractual constraint assertions constitutes the fact deduction environment.

7. The method for selecting a tool based on fuzzy ontology analysis for machining according to claim 6, characterized in that, Performing Boolean logic derivations includes: Read the fact derivation environment and extract the individual variables of the part and the individual variables of the tool instance from the fact derivation environment; Loading a single-dimensional basic derivation rule that includes the antecedent of the derivation condition and the consequent of the derivation conclusion; The logical AND operation operator is invoked to perform a chained determination on the pre-assertion unit containing the object attribute association unit and the data attribute association unit; The data attribute association unit is parsed to obtain the single performance dependency value variable. The Boolean greater than judgment operator is called to perform a numerical comparison operation between the single performance dependency value variable and the preset comparison threshold value. When the numerical value of a single performance dependency is greater than the preset benchmark threshold, the antecedent of the derivation condition is fully satisfied, the derivation triggering operation operator is activated, and the consequent of the derivation conclusion is generated to establish the one-to-one matching relationship between the individual variables of the part and the individual variables of the tool instance.

8. The method for selecting a tool based on fuzzy ontology analysis in machining according to claim 6, characterized in that, Performing Boolean logic derivations also includes: Independent dependency variables of multiple cutting performance characteristics are extracted in parallel from the factual environment. Load a multi-dimensional collaborative derivation rule that includes multiple derivation conditions antecedents and a single derivation conclusion consequent, and link multiple independent data attribute association instructions with calling logic and operation operators. Activate the Boolean multiplication operator, perform a series algebraic calculation on the extracted independent membership degree numerical variables, and assign the result of the series algebraic calculation to the comprehensive performance membership degree numerical variable; Extract the built-in comprehensive comparison threshold value of the system, call the Boolean greater than judgment operator to determine when the comprehensive performance membership degree numerical variable is greater than the comprehensive comparison threshold value, trigger the derivation trigger operation operator, and generate a single derivation conclusion consequent that establishes a global mapping relationship between the individual variables of the part and the individual variables of the optimal tool instance.

9. The method for selecting cutting tools based on fuzzy ontology analysis for machining according to claim 1, characterized in that, Receive physical parameter requirements, extract the basic material information of the workpiece and the objective physical continuity of the cutting process, including: Receive scene physical parameters and decompose the scene physical parameters into independent data dimension variables; Parse the basic material information containing material attribute identifiers and extract the hardness index constant contained within the specific metal grade data; Extract the machining performance constraints that include the target surface roughness value and the dimensional tolerance value, and convert the target surface roughness value and the dimensional tolerance value into a comprehensive accuracy requirement value; Establish cutting environment constraint variables that cover both the spindle speed limit constant and the feed rate limit constant. Based on the cutting environment constraint variables, extract the continuous quantities of hot hardness requirement and wear resistance requirement as objective physical continuous quantities.

10. The method for selecting a tool based on fuzzy ontology analysis for machining according to claim 1, characterized in that, Retrieve the entity specification parameters corresponding to the matching tool number record, integrate the entity specification parameters with the matching tool number record to generate a sequence of optional tool instances for output display, including: Extract individual candidate tool entities from the initial matching pool formed by the matching conditions; For each candidate tool entity, a comprehensive evaluation score is calculated. An algebraic model of the comprehensive evaluation score is established. Specific numerical variables and weight constants are input into the algebraic model of the comprehensive evaluation score to obtain the comprehensive evaluation score variable value for each candidate tool entity. The candidate tool entities are sorted in descending order according to the value of the comprehensive evaluation score variable. The candidate tool entity with the highest ranking is selected as the recommended selection scheme. Generate a final decision list containing tool number characters and comprehensive evaluation score variables, and send the final decision list to the human-computer interaction interface for display.

Images