A numerical control machine tool spindle service performance influencing factor intelligent analysis method and system based on a large language model

Abstract

The application discloses a kind of based on large language model's numerical control machine tool spindle service performance influence factor intelligent analysis method and system, comprising: spindle service performance influence factor analysis intelligent agent construction, spindle service performance knowledge base construction, three modules of spindle service performance cause intelligent analysis.Spindle service performance influence factor analysis intelligent agent construction, establish numerical control machine tool spindle influence factor correlation intelligent agent and field knowledge intelligent agent, realize the setting of service performance analysis theme;Spindle service performance knowledge base construction, from four dimensions of influence factor, evaluation index, mechanism model, data model, realize the construction of service performance multidimensional knowledge base;Spindle service performance cause intelligent analysis extracts service performance influence factor correlation and constructs its cause graph, realizes the intelligent analysis of service performance based on cause graph.The application can solve the problem of difficult analysis of service performance cause under the condition of multiple numerical control machine tool spindle service performance influence factors and complex coupling relationship.

Description

Technical Field

[0001] This invention belongs to the fields of electronic engineering and computer science, and specifically relates to an intelligent analysis method and system for the influencing factors of CNC machine tool spindle service performance based on a large language model. Background Technology

[0002] High-end CNC machine tools are crucial mother machines for key equipment in major fields such as aviation, aerospace, and shipbuilding. As the core power unit of high-end CNC machine tools, the service performance of the CNC machine tool spindle is a key indicator of its overall service capability and health status, directly affecting the machining capacity and service life of the high-end CNC machine tool. However, the evolution mechanism of the service performance of CNC machine tool spindles is complex due to the coupling of various errors such as thermal displacement, axial runout, and radial runout, as well as the intertwined influence of multiple explicit and implicit factors such as temperature rise, wear, and loosening. Accurate analysis of the factors affecting service performance faces severe challenges.

[0003] At present, preliminary research has been conducted on the analysis of factors affecting the service performance of CNC machine tool spindles. However, existing methods generally focus on single-dimensional or static indicators such as thermal error, accuracy retention, and radial runout, which makes it difficult to reveal the key factors affecting the service performance of CNC machine tool spindles under the coupling effect of multiple physical fields and multiple factors. The application threshold of service performance influencing factor analysis is high and it is difficult to implement.

[0004] To comprehensively analyze the service performance of high-end CNC machine tool spindles, it is urgent to deeply integrate knowledge in the field of CNC machine tools with next-generation generative artificial intelligence technology, and establish a complete knowledge base of influencing factors, rating indicators, and model components of CNC machine tool spindle service performance. This will provide methodological and technical support for the analysis of influencing factors and the maintenance and improvement of CNC machine tool spindle service performance. Summary of the Invention

[0005] To address the technical challenge of analyzing the causes of CNC machine tool spindle service performance issues, this invention constructs an intelligent agent for analyzing service performance influencing factors based on a generative large language model, and establishes a multi-dimensional knowledge base for CNC machine tool spindle service performance, thereby achieving intelligent analysis of the causes of CNC machine tool spindle service performance issues. To achieve the above objectives, this invention adopts the following technical solution:

[0006] This invention provides an intelligent analysis method for factors influencing the service performance of CNC machine tool spindles based on a large language model, comprising the following steps:

[0007] Step (1): Construct the intelligent agent for analyzing the influencing factors of spindle service performance, establish the intelligent agent related to the influencing factors of CNC machine tool spindle and the domain knowledge intelligent agent, and realize the setting of the service performance analysis topic;

[0008] Step (2): Construction of the spindle service performance knowledge base. From the four dimensions of influencing factors, evaluation indicators, mechanism model and data model, a multi-dimensional knowledge base for service performance is constructed.

[0009] Step (3): The intelligent analysis module for the causes of spindle service performance extracts the correlation between factors affecting service performance, constructs a service performance cause map, and realizes intelligent analysis of service performance based on the cause map.

[0010] This invention provides an intelligent analysis system for factors influencing the service performance of CNC machine tool spindles based on a large language model, comprising:

[0011] The module for constructing intelligent agents to analyze factors affecting the service performance of CNC machine tool spindles establishes intelligent agents that associate factors affecting CNC machine tool spindles with domain knowledge intelligent agents, and enables the setting of service performance analysis topics.

[0012] The spindle service performance knowledge base construction module realizes the construction of a multi-dimensional knowledge base for service performance from four dimensions: influencing factors, evaluation indicators, mechanism models, and data models.

[0013] The intelligent analysis module for the causes of spindle service performance extracts the correlation between factors affecting service performance, constructs a service performance cause map, and realizes intelligent analysis of service performance based on the cause map.

[0014] The present invention has the following beneficial effects:

[0015] To address the challenge of analyzing the causes of CNC machine tool spindle service performance issues, this paper proposes an intelligent analysis method based on a large language model to analyze the influencing factors of CNC machine tool spindle service performance. By deeply integrating knowledge from the CNC machine tool field with next-generation generative artificial intelligence technology, this method broadly correlates the influencing factors, rating indicators, and model components of CNC machine tool spindle service performance, thereby reducing the difficulty of analyzing and managing the influencing factors of service performance to a certain extent. Attached Figure Description

[0016] Figure 1 This invention presents a block diagram of the intelligent analysis method and system structure for the influencing factors of CNC machine tool spindle service performance based on a large language model. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other. To achieve the above objectives, this invention adopts the following technical solution.

[0018] This invention discloses an intelligent analysis method and system for the influencing factors of CNC machine tool spindle service performance based on a large language model. The system comprises three modules: construction of an intelligent agent for analyzing the influencing factors of spindle service performance, construction of a knowledge base for spindle service performance, and intelligent analysis of the causes of spindle service performance. The construction of the intelligent agent for analyzing the influencing factors of spindle service performance establishes an intelligent agent relating the influencing factors of the CNC machine tool spindle to domain knowledge, enabling the setting of service performance analysis topics. The construction of the knowledge base for spindle service performance constructs a multi-dimensional knowledge base for service performance from four dimensions: influencing factors, evaluation indicators, mechanism models, and data models. The intelligent analysis of the causes of spindle service performance extracts the relationships between the influencing factors and constructs a causal graph, enabling intelligent analysis of service performance based on the causal graph. This invention can solve the problem of difficult analysis of the causes of CNC machine tool spindle service performance when there are many influencing factors and complex coupling relationships.

[0019] The structural block diagram of the intelligent analysis method and system for the influencing factors of CNC machine tool spindle service performance based on a large language model is as follows: Figure 1 As shown, the process includes the construction of an intelligent agent for analyzing factors affecting spindle service performance, the construction of a knowledge base for spindle service performance, and an intelligent analysis module for the causes of spindle service performance. The specific steps are as follows:

[0020] Step (1), the intelligent agent for analyzing the factors affecting the service performance of CNC machine tool spindles is constructed, and its specific implementation is as follows:

[0021] Step (1.1), Construction of the Service Performance Influencing Factors Association Agent and Domain Knowledge Agent: Configure the calling key and basic URL of the large language model (e.g., Deepseek). Utilize the ReAct agent framework to construct an agent for analyzing the service performance influencing factors of CNC machine tool spindles based on the large language model. ,in: It is an intelligent agent that associates factors affecting the service performance of CNC machine tool spindles, covering a closed loop of factor processing: "factor analysis → relationship writing → relationship correction". It is a knowledge intelligence body for the service performance domain of CNC machine tool spindles, covering a closed loop of service performance domain knowledge processing: "knowledge extraction → knowledge identification → knowledge update";

[0022] Step (1.2), Setting the theme for CNC machine tool spindle service performance analysis: From the three dimensions of influencing factors, evaluation indicators, and model components, write the theme prompts for service performance analysis. ,in: Factors affecting the service performance of a spindle system or component, such as spindle speed and bearing heat generation power; The evaluation indicators for spindle service performance cover seven major indicators: thermal error, radial runout, axial runout, tilt oscillation, static stiffness, dynamic stiffness, and accuracy retention. This refers to the service performance mechanism model of the spindle, such as the Fourier heat transfer model, thermal expansion model, and convective heat transfer model. This refers to the service performance data model of the spindle system, such as convolutional neural networks and support vector machines.

[0023] Step (1.3), Construction of metadata file and field set for service performance cause map: Establish spindle service performance cause map G SPC The metadata file relation.json, based on the topic hints compiled in step (1.2). Construct a spindle service performance cause map G SPC field set G SPCFS ={'source_type', 'source_value', 'target_type', 'target_value', 'evidence'}. Where: source_type, source_value, target_type, target_value, and evidence are the spindle service performance cause graphs G... SPC The source node type field, source node value field, target node type field, target node value field, and service performance related semantic evidence field;

[0024] Step (2), the method for constructing a knowledge base for the service performance of CNC machine tool spindles, is implemented as follows:

[0025] Step (2.1) Construction of a knowledge base for factors influencing the service performance of CNC machine tool spindles: A knowledge base for factors influencing the service performance of CNC machine tool spindles is constructed based on five dimensions: Chinese translation, English translation, Chinese synonyms / near-synonyms, English synonyms / near-synonyms, and unit dimensions. and its knowledge entries ;in, For knowledge base Knowledge items related to service performance in group i. The Chinese translation of the knowledge item for the i-th group of service performance influencing factors is as follows: The English translation of the knowledge item for the i-th group of service performance influencing factors is as follows: For the knowledge item on the i-th group of service performance influencing factors, Chinese synonyms / near-synonyms are provided. The English synonyms / near-synonyms for the knowledge item on the i-th group of service performance influencing factors. The unit dimension is the knowledge item of the i-th group of service performance influencing factors;

[0026] Step (2.2) Construction of the CNC machine tool spindle service performance evaluation index knowledge base: Construct a CNC machine tool spindle service performance evaluation index knowledge base from five dimensions: Chinese translation, English translation, Chinese synonyms / near-synonyms, English synonyms / near-synonyms, and unit dimensions. and its knowledge entries .in, For knowledge base Knowledge items for the service performance evaluation indicators in group j of the Chinese language. The Chinese translation of the knowledge items for the service performance evaluation indicators in group j is as follows: The English translation of the knowledge items for the service performance evaluation indicators in group j is as follows: For the knowledge entries of the service performance evaluation indicators in group j, Chinese synonyms / near-synonyms For the knowledge items of the service performance evaluation indicators in group j, the English synonyms / near-synonyms are: The unit dimension is the knowledge item of the service performance evaluation index in the j-th group;

[0027] Step (2.3), Construction of the context parameter set for CNC machine tool spindle service performance analysis: Construct the context parameter set for CNC machine tool spindle service performance analysis from two dimensions: shape and size, and material properties. ,in This is a set of shape and dimension parameters for CNC machine tool spindles and their components. This is a set of material property parameters for CNC machine tool spindles and their components. These are physical quantity parameters that are not easily affected by factors influencing service performance;

[0028] Step (2.4), Construction of the knowledge base for the components of the CNC machine tool spindle service performance model: From the dimensions of mechanism model and data model, based on the input domain of the model components... and output domain Construct a knowledge base for the constituent elements of the service performance data model. Knowledge base of components of service performance mechanism model Based on this, a knowledge base of the components of the service performance model of CNC machine tool spindles is formed. ;in, and They are respectively and Knowledge entries. and It can include multiple parameters. and Only a single parameter;

[0029] Step (3), Intelligent Analysis Method for the Causes of Service Performance of CNC Machine Tool Spindles, is implemented as follows:

[0030] Step (3.1), Construction of the CNC machine tool spindle service performance analysis corpus: by crawling and keyword... Related Chinese and English corpus files Constructing a corpus for service performance analysis of CNC machine tool spindles , express The number of texts to be processed was determined. First, Chinese and English corpora were collected using CNKI and WOS respectively. For the Chinese corpus, using the keywords {Topic: CNC machine tools} AND {Topic: spindle} AND {Topic: error}, EI and Peking University core journals from the past ten years were retrieved from CNKI, exported as .es6 files, and then downloaded in batches as PDF files using CNKI Research software. For the English corpus, using the keyword combination {Topic: machine tools} AND {Topic: spindle} AND {Topic: error}, journal and conference articles from the past ten years were retrieved from the Web of Science core database, exported as .bib files, and then downloaded in batches as PDF files using Zotero software.

[0031] Step (3.2), Construction of the transformation model for service performance cue words-association mapping: from the spindle service performance analysis corpus Read Chinese and English corpus files Cleaning using regular expressions ,jump over In the internal figure captions, table captions, and literature review sections, all text is concatenated into a single string. Defines the number of characters per text block and the number of overlapping characters per text block for splitting long texts into shorter ones. Packed into chunk data blocks . use Reconstruction of service performance causes and correlation analysis prompts The system uses a hash algorithm to generate identifiers for text, sets the number of concurrent threads, request rate, and requests per second to improve text processing speed and the success rate of calling large language models, and constructs service performance indicator words. Correlation mapping with factors affecting service performance conversion model ;

[0032] Step (3.3), Construction of Service Performance Cause Map: Based on step (3.2), provide prompts for service performance cause correlation analysis. And step (1.3) provides the field set G SPCFS According to the corpus The internal file order calls the intelligent agent related to the influencing factors of spindle service performance constructed in step (1.1). And with key-value pairs {'source_type': ,'source_value': , 'targettype': , 'target_value': , 'evidence': Extraction in the form of association mapping of factors affecting service performance ,in , , , , These represent the correlation mappings of factors affecting service performance. The source node type value, source node value, target node type value, target node value, and service performance-related semantic evidence value are used to map the factors influencing service performance. The data was converted according to the JSON data specification format and integrated to form a service performance cause map. ,in , They are Fields within the service performance cause map;

[0033] Step (3.4), Establishment of the service performance cause map node set: by traversing the service performance cause map G SPC Found: 1) All source nodes are of type 1 The source node value and the target node type are The target node values ​​form a correlation mapping set of factors affecting service performance. Service performance influencing factors source node set Service performance influencing factors target node set ,in, Correlation mapping of factors affecting service performance, As the source node of factors affecting service performance, The set of target nodes for factors influencing service performance, together with the set of source nodes for factors influencing service performance, constitutes the set of nodes for factors influencing service performance. ;2) All source nodes are of type The source node value and target node type are The target node values ​​form the service performance mechanism model association mapping set. Service performance mechanism model source node set Service performance mechanism model target node set ,in, For service performance mechanism model correlation mapping, For the service performance mechanism model source node, The target node of the service performance mechanism model. The source node set and the target node set of the service performance mechanism model constitute the node set of the service performance mechanism model. 3) All source nodes are of type The source node value and target node type are The target node values ​​form a service performance data model association mapping set. Service performance data model source node set Service performance data model target node set ,in, For service performance data model association mapping, For service performance data model source node, The target node of the service performance data model. The source node set and the target node set of the service performance data model constitute the service performance data model node set. Service performance data model node set Service performance data model node set Together they constitute the service performance model node set .

[0034] Step (3.5), handling abnormal nodes within the service performance causation map: calling the service performance influencing factor association agent. Each node set of service performance influencing factors constructed in step (3.4) is traversed. Service performance mechanism model component node set and the node set of components of the service performance data model. ,accomplish:

[0035] 1) Clearing abnormal nodes affecting service performance: Set abnormal node detection prompts for factors affecting service performance. Invoke the intelligent agent that associates factors affecting service performance Traverse the source node set of factors affecting service performance Target node set of factors affecting service performance Based on the knowledge base of factors affecting service performance Determine the nodes affecting service performance from the following aspects. Are there any anomalies, among which The number of nodes in the service performance influencing factor node set is determined as follows: ① Nodes whose naming format does not conform to the 'object + factor' format are marked as anomalous nodes; ② 'Objects' do not belong to the spindle unit category, such as feed systems or rotary tables, are also marked as anomalous nodes. This yields the anomalous node set of service performance influencing factors. From the service performance cause diagram Remove the set of anomalous nodes that contain factors affecting service performance. Abnormal correlation mapping of factors affecting service performance .

[0036] 2) Determination of composite nodes of factors affecting service performance: Constructing discrimination prompts for composite nodes of factors affecting service performance. Invoke the intelligent agent that associates factors affecting service performance Determine the nodes that affect service performance If any punctuation marks or excessively long words are found, they should be broken down into multiple individual keywords related to service performance influencing factors. Using these decomposed service performance influencing factor nodes, a new service performance influencing factor association mapping is constructed, updating the service performance causal map. .

[0037] 3) Merging highly similar nodes of factors affecting service performance: Set similarity detection prompts for nodes affecting service performance. Invoke the intelligent agent that associates factors affecting service performance Search for knowledge entries related to factors affecting service performance. Chinese near / synonyms with semantic similarity and English near / synonyms And add it to the knowledge entry on factors affecting service performance. middle.

[0038] 4) Error removal in the service performance mechanism model and data model: Write error correction prompts for the service performance model. Invoke the intelligent agent that associates factors affecting service performance Traverse the node set of the service performance mechanism model Service performance data model node set The criteria for identification are as follows: ① Non-model terms, belonging to engineering categories such as equipment / measurement / process / control strategy / hardware; ② Non-specific conceptual terms, such as deep learning / transfer learning, which cannot be identified as terms targeting a specific model. If any of the above situations occur, the node is marked as an abnormal node of the non-service performance model category. From the service performance cause diagram Delete the associated mapping containing the faulty node. .

[0039] 5) Correction of anomalies in the service performance mechanism model and data model: Write error correction prompts for service performance model nodes. Invoke the intelligent agent that associates factors affecting service performance Traverse the node set of the service performance mechanism model Service performance data model node set Initial identification of misclassified model nodes; final review by human experts of factors influencing service performance and their association with intelligent agents. To verify the accuracy of the analysis results, correct misclassified service performance models, and ultimately obtain a complete and usable service performance knowledge graph AG. SPC .

[0040] Step (3.6), Service Performance Knowledge Base Update: Utilize the domain knowledge intelligent agent of CNC machine tool spindle service performance already constructed in step (1.1) Combined with the available service performance knowledge graph AG SPC Update the knowledge base of factors affecting service performance. Knowledge base of components of service performance model A new knowledge base of factors affecting service performance was obtained by integrating these elements. Knowledge base of components of service performance model

[0041] 1) Update the knowledge base of components of the service performance model. This includes updating the service performance causation map. Internally, with service performance mechanism model nodes or service performance data model node Its connected non-service performance model class node set is defined as Where s is the number of nodes in the node set of the service performance mechanism model. The number of nodes in the node set of the service performance data model. Number of nodes in the non-service performance model node set: I. When the service performance model node Not appearing in the service performance mechanism model component library Or service performance data model component library In the middle, but non-service performance model class node set Includes a library of components for service performance mechanism models The inputs and outputs of any model, or the component library of service performance data models. Given the inputs and outputs of any model, construct new model component entries. II. When the service performance model node... Appears in the component library of service performance mechanism model In the middle, and service performance model nodes Knowledge entries that constitute any service performance mechanism model The names are the same. Among them, This represents the number of nodes within the service performance model node set. If the service performance model nodes... Connected non-model class node sets Knowledge items related to the components of the service performance mechanism model input field or output field Intersection with Existence And there is a service performance correlation mapping. Contains from set { } to set If the mapping relationship is true, then this mapping will be used. The source node is added to the knowledge entries of the service performance mechanism model. Middle; III. When the service performance model node Appears in the service performance data model component library In the process, its service performance model nodes Knowledge items related to the components of service performance data models The names are the same. If: ① Same service performance model node Connected non-model class node sets Knowledge items related to the components of service performance data models Output domain There is an intersection And there is a service performance correlation mapping. Contains from set { } to set The mapping relationship will then be the source node of the mapping relationship. Added to the knowledge entries of the components of the service performance mechanism model ② Same service performance model node Connected non-model class node sets Knowledge items related to the components of service performance data models input field There is an intersection And there is a service performance correlation mapping. Contains from set { } to set The mapping relationship is then found in the component library of the service performance data model. Add knowledge entries Its input and output entries are and .

[0042] 2) Updating the knowledge base of factors influencing service performance. New entries for factors influencing service performance are added to the knowledge base based on the identification of composite nodes. High-similarity nodes of factors influencing service performance are merged within the knowledge base. Knowledge items related to factors affecting service performance In the text, supplement the English and Chinese near / synonyms of factors affecting service performance, and associate these factors with a mapping set. The Chinese and English near / synonyms of factors affecting service performance were converted into standardized expressions, ultimately resulting in a new knowledge base of factors affecting service performance. Knowledge base of components of service performance model

[0043] In summary, this invention discloses an intelligent analysis method and system for factors influencing the service performance of CNC machine tool spindles based on a large language model, including: constructing an intelligent agent for analyzing factors influencing spindle service performance, constructing a knowledge base for spindle service performance, and intelligent analysis of the causes of spindle service performance. This invention can solve the problem of difficult analysis of the causes of service performance when there are many factors influencing the service performance of CNC machine tool spindles and complex coupling relationships.

[0044] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

[0045] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent analysis of factors influencing the service performance of CNC machine tool spindles based on a large language model, characterized in that, Includes the following steps: Step (1): Construct the intelligent agent for analyzing the influencing factors of spindle service performance, establish the intelligent agent related to the influencing factors of CNC machine tool spindle and the domain knowledge intelligent agent, and realize the setting of the service performance analysis topic; Step (2): Construction of the spindle service performance knowledge base. From the four dimensions of influencing factors, evaluation indicators, mechanism model and data model, a multi-dimensional knowledge base for service performance is constructed. Step (3): Intelligent analysis of the causes of spindle service performance, extracting the correlation of factors affecting service performance, constructing a service performance cause map, and realizing intelligent analysis of service performance based on the cause map.

2. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 1, characterized in that, Step (1) is implemented as follows: Step (1.1), Construction of the intelligent agent for associating factors affecting service performance and the intelligent agent for domain knowledge: Construct an intelligent agent for analyzing factors affecting the service performance of CNC machine tool spindles based on a large language model using the ReAct intelligent agent framework. ,in: It is an intelligent agent that associates factors affecting the service performance of CNC machine tool spindles, covering a closed loop of factor processing: "factor analysis → relationship writing → relationship correction". It is a knowledge intelligence agent for the service performance domain of CNC machine tool spindles, covering a closed loop of service performance domain knowledge processing: "knowledge extraction → knowledge identification → knowledge update"; Step (1.2), Setting the topic for CNC machine tool spindle service performance analysis: From the four dimensions of influencing factors, evaluation indicators, mechanism model, and data model, write the topic prompts for CNC machine tool spindle service performance analysis. ,in: Factors affecting the service performance of a spindle system or component; The evaluation indicators refer to the service performance of the spindle; Refers to the service performance mechanism model of the spindle; Refers to the service performance data model of the spindle; Step (1.3), Construction of metadata file and field set for service performance cause map: Establish spindle service performance cause map G SPC The metadata file relation.json contains topic hints compiled based on step (1.2). Construct a spindle service performance cause map G SPC field set G SPCFS ={'source_type', 'source_value', 'target_type', 'target_value', 'evidence'}, where: source_type, source_value, target_type, target_value, and evidence are the spindle service performance cause map G. SPC The source node type field, source node value field, target node type field, target node value field, and service performance related semantic evidence field.

3. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 1, characterized in that, Step (2) is implemented as follows: Step (2.1) Construction of a knowledge base for factors influencing the service performance of CNC machine tool spindles: A knowledge base for factors influencing the service performance of CNC machine tool spindles is constructed based on five dimensions: Chinese translation, English translation, Chinese synonyms / near-synonyms, English synonyms / near-synonyms, and unit dimensions. and its knowledge entries ;in, For knowledge base Knowledge items related to service performance in group i. The Chinese translation of the knowledge item for the i-th group of service performance influencing factors is as follows: The English translation of the knowledge item for the i-th group of service performance influencing factors is as follows: For the knowledge item on the i-th group of service performance influencing factors, Chinese synonyms / near-synonyms are provided. The English synonyms / near-synonyms for the knowledge item on the i-th group of service performance influencing factors. The unit dimension is the knowledge item of the i-th group of service performance influencing factors; Step (2.2) Construction of the CNC machine tool spindle service performance evaluation index knowledge base: Construct a CNC machine tool spindle service performance evaluation index knowledge base from five dimensions: Chinese translation, English translation, Chinese synonyms / near-synonyms, English synonyms / near-synonyms, and unit dimensions. and its knowledge entries ,in, For knowledge base Knowledge items for the service performance evaluation indicators in group j of the Chinese language. The Chinese translation of the knowledge items for the service performance evaluation indicators in group j is as follows: The English translation of the knowledge items for the service performance evaluation indicators in group j is as follows: For the knowledge entries of the service performance evaluation indicators in group j, Chinese synonyms / near-synonyms For the knowledge items of the service performance evaluation indicators in group j, the English synonyms / near-synonyms are: The unit dimension is the knowledge item of the service performance evaluation index in the j-th group; Step (2.3), Construction of the context parameter set for CNC machine tool spindle service performance analysis: Construct the context parameter set for CNC machine tool spindle service performance analysis from two dimensions: shape and size, and material properties. ,in This is a set of shape and dimension parameters for CNC machine tool spindles and their components. This is a set of material property parameters for CNC machine tool spindles and their components. These are physical quantity parameters that are not easily affected by factors influencing service performance; Step (2.4), Construction of the knowledge base for the components of the CNC machine tool spindle service performance model: From the dimensions of mechanism model and data model, based on the input domain of the model components... and output domain Construct a knowledge base for the constituent elements of the service performance data model. Knowledge base of components of service performance mechanism model Based on this, a knowledge base of the components of the service performance model of CNC machine tool spindles is formed. ;in, and They are respectively and Knowledge entries.

4. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 1, characterized in that, Step (3) is implemented as follows: Step (3.1), Construction of the CNC machine tool spindle service performance analysis corpus: by crawling and keyword... Related Chinese and English corpus files Constructing a corpus for service performance analysis of CNC machine tool spindles , express The number of texts to be processed; Step (3.2), Construction of the transformation model for service performance cue words-association mapping: from the CNC machine tool spindle service performance analysis corpus Reading Chinese and English corpus files ; Cleaning Chinese and English text files using regular expressions Package it into chunk data blocks Reconstructing service performance causes and correlation analysis prompts Construct service performance prompts Mapping related to service performance conversion model ; Step (3.3), Construction of Service Performance Cause Map: Based on step (3.2), provide prompts for service performance cause correlation analysis. And step (1.3) provides the field set G SPCFS According to the corpus The internal file order calls the intelligent agent related to the influencing factors of spindle service performance constructed in step (1.1). And with key-value pairs {'source_type': ,'source_value': , 'target_type': , 'target_value': , 'evidence': Extraction in the form of association mapping of factors affecting service performance ,in , , , , These represent the correlation mappings of factors affecting service performance. The source node type value, source node value, target node type value, target node value, and service performance-related semantic evidence value are used to map the factors influencing service performance. The data was converted according to the JSON data specification format and integrated to form a service performance cause map. ,in , They are Fields within the service performance cause map; Step (3.4), Establishment of the service performance cause map node set: by traversing the service performance cause map G SPC Found: ① All source nodes of type The source node value and the target node type are The target node values ​​form a set of nodes affecting service performance. ② All source nodes are of type The source node value and the target node type are The target node values ​​form the node set of the service performance mechanism model. ③ All source nodes are of type The source node value and target node type are The target node values ​​form the node set of the service performance data model. ; Step (3.5), handling abnormal nodes within the service performance causation map: calling the service performance influencing factor association agent. ① Traverse the set of service performance influencing factors constructed in step (3.4) ① Abnormal nodes affecting service performance are removed by comparing and analyzing naming rules and topic categories, and highly similar nodes affecting service performance are merged based on latent semantic analysis; ② The node set of constituent elements of the service performance mechanism model constructed in step (3.4) is traversed. and the node set of components of the service performance data model Based on the topic categories, abnormal nodes in the service performance model are labeled and removed, and misclassified service performance model nodes are corrected. Based on sub-steps ① and ②, the usable service performance causal map AG is obtained. SPC ; Step (3.6), Service Performance Knowledge Base Update: Utilize the domain knowledge agent of CNC machine tool spindle service performance already constructed in step (1.1) Combined with the available service performance cause map AG SPC Update the knowledge base of factors affecting service performance. Knowledge base of components of service performance model A new knowledge base of factors affecting service performance was obtained by integrating these elements. Knowledge base of components of service performance model .

5. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 2, characterized in that, This includes spindle speed and bearing heating power.

6. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 2, characterized in that, It covers seven major indicators: thermal error, radial runout, axial runout, tilting oscillation, static stiffness, dynamic stiffness, and accuracy retention.

7. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model according to claim 2, characterized in that, This includes heat transfer models, thermal expansion models, and convective heat transfer models.

8. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model, as described in claim 2. This includes convolutional neural networks and support vector machines.

9. The intelligent analysis method for factors affecting the service performance of CNC machine tool spindles based on a large language model, as described in claim 3. and Includes multiple parameters, and It is a single parameter only.

10. An intelligent analysis system for factors influencing the service performance of CNC machine tool spindles based on a large language model, characterized in that, include: The module for constructing intelligent agents to analyze factors affecting the service performance of CNC machine tool spindles establishes intelligent agents that associate factors affecting CNC machine tool spindles with domain knowledge intelligent agents, and enables the setting of service performance analysis topics. The spindle service performance knowledge base construction module realizes the construction of a multi-dimensional knowledge base for service performance from four dimensions: influencing factors, evaluation indicators, mechanism models, and data models. The intelligent analysis module for the causes of spindle service performance extracts the correlation between factors affecting service performance, constructs a service performance cause map, and realizes intelligent analysis of service performance based on the cause map.

Images