Product analysis method and system based on life cycle-system level data chain model

Abstract

The application discloses a product analysis method and system based on a life cycle-system level data chain model, realizes transparent recording and tracing of product life cycle data through a decentralized data storage platform of a block chain, applies smart contract technology, accurately cooperates global and local decisions, and breaks the collaboration bottleneck of a traditional manufacturing system. Meanwhile, the problem of small sample data imbalance is solved by using a distributed sharing mechanism of the block chain, and the data credibility and the overall availability of the system are improved. On the basis of the decentralized data storage platform, a product life cycle and system level double digital main line block chain data model is established, an intelligent manufacturing data analysis and knowledge learning model is established in a method domain, the block chain is combined with edge computing, data processing is more efficient and intelligent, the intelligent level of the manufacturing system is further improved, and a comprehensive, efficient and intelligent solution is provided for digital transformation.

Description

Technical Field

[0001] The present invention relates to the technical field of product analysis, and in particular to a product analysis method and system based on a lifecycle-system hierarchical data chain model. Background Art

[0002] With the digital transformation of the manufacturing industry, product lifecycle and system-level management have become key concerns for manufacturers. However, traditional manufacturing digital systems face a series of challenges, including a lack of transparency in product lifecycle data management, making it difficult to fully track and monitor the production process; difficulties in local decision-making and global coordination, resulting in inflexible and inefficient production processes; imbalanced small sample data, which affects data reliability and overall usability; and limited intelligence, which hinders efficient big data processing and deep learning. These issues collectively hinder the manufacturing industry's goal of achieving comprehensive coordination, efficient intelligence, and digital transformation. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a product analysis method based on a life cycle-system level data chain model.

[0004] To achieve the above objectives, the technical solutions provided by the present invention are:

[0005] Product analysis method based on the lifecycle-system level data chain model, including:

[0006] Through a decentralized data storage platform, product lifecycle and system-level data can be uploaded to the chain;

[0007] Build a dual digital thread blockchain data model for the product lifecycle and system level. This dual digital thread blockchain data model includes the product lifecycle digital thread, the product multi-level digital thread, the blockchain information domain, the blockchain structure domain, the blockchain sample domain, and the blockchain method domain.

[0008] Through the coordination of blockchain information domain, blockchain structure domain, blockchain sample domain and blockchain method domain, the digital main line of the product life cycle and the multi-level digital main line of the product are combined to analyze the product.

[0009] Furthermore, in the blockchain information domain, the ontology of the manufacturing field is adopted, including the formal description of relevant concepts, attributes and relationships. The semantic web is used to realize the structured representation of manufacturing knowledge, and a unified description framework of manufacturing knowledge and engineering principles is constructed to facilitate the association and reasoning between data; smart contracts are designed to manage the creation, verification, update and sharing process of knowledge, and the authorization and access control of knowledge are automated through smart contracts; data mining and analysis technologies are implemented to identify and extract key data and information sources in physical and logical spaces, and map and manage complex relationships between data.

[0010] Furthermore, the specific working principles of the blockchain information domain include:

[0011] First, collect standard knowledge in the manufacturing field and build a standard manufacturing knowledge terminology database;

[0012] Then define the manufacturing domain ontology and use OWL language to formally describe the ontology, that is, the relevant concepts C, attributes O p , define the relationship R and get the OWL file;

[0013] Use RDF to annotate existing product lifecycle and system data, and correspond the data to the concepts, attributes and relationships in the ontology to obtain RDF = {c, o, r | c∈C, o∈O p ,r∈R};

[0014] Serialize OWL files and RDF data into JSON-LD format and upload them to the blockchain, using smart contracts to automatically manage the storage, update, and query of ontology and RDF data;

[0015] A graph database is linked to the blockchain to synchronize data regularly, and SPARQL is used for semantic queries to mine and analyze product and system information.

[0016] Related concepts C include parts, components, products, and equipment;

[0017] Attribute O p Including part size, weight, process, component shape, product packaging;

[0018] Relationship R includes the assembly of parts into components and the assembly of components into products;

[0019] Product information includes basic product information, product batch information, process requirements information, and technical parameter information;

[0020] System information includes equipment operating status, processing object description, processing quality information, and processing cycle plan.

[0021] Furthermore, in the blockchain structure domain, the immutability and smart contracts of the blockchain are used to check data integrity, and the time window and process box granularity are dynamically adjusted in the system granularity control application to support the collaboration between local self-organization and global optimization, thereby creating an adaptive control framework and dynamically adjusting the control granularity according to the system scale and order acceptance rate.

[0022] Furthermore, the specific working principles of the blockchain domain include:

[0023] First, collect the time target points T={T1, T2,…, T5} in the product life cycle.

[0024] Where T1 is the time from demand analysis to process planning;

[0025] T2 is the manufacturing and assembly time;

[0026] T3 is the test and certification time;

[0027] T4 inventory logistics time;

[0028] T5 is the delivery and installation time.

[0029] And a complete set of product process sequences at the product system level, where the process sequence includes the following data:

[0030] N 子,j : The number of sub-processes in the process sequence of stage j;

[0031] R 需求,j : The amount of resources required for stage j;

[0032] C 复杂,j : complexity score of the process step in stage j;

[0033] T 需求,j : The time required to complete the process steps of stage j;

[0034] The acquired data is transmitted to the blockchain for on-chain operations to generate transactions, and the smart contract is automatically triggered to check the integrity of the data. After the check is completed and the data is stored in the block after consensus is reached;

[0035] The granular control application will then extract data from the blockchain at regular intervals. The data will be recorded and saved in the blockchain for each transaction generated with the system application; the granular control application will then analyze the current time target data and process sequence data.

[0036] Furthermore, the specific process of analyzing the current time target data and process sequence data by the granularity control application includes:

[0037] Analyze the system scale and order acceptance rate, among which,

[0038]

[0039] F 系统 is the system size adjustment factor, S 当前 is the current project scale, S 基准 is the benchmark project size;

[0040]

[0041] F 接单 is the order acceptance rate adjustment factor, R 当前 is the current order acceptance rate, R 基准 is the benchmark order acceptance rate;

[0042] Dynamically adjust the time window:

[0043] T 调整,i =T i ×F 系统 ×F 接单

[0044] Process sequence granularity calculation:

[0045] G j =α·N 子,j +β·R 需求,j +γ·C 复杂,j +δ·T 需求,j ,

[0046] Among them, G j is the process sequence granularity, α is the weight coefficient of the sub-process quantity, β is the resource demand weight coefficient, R 稀缺资源 Indicates the resource type with the highest demand among all steps; γ is the weight coefficient of complexity score, C 最高复杂 represents the complexity score of the step with the highest complexity score among all process steps; δ is the time requirement weight coefficient, T 最长需求 The time required for the longest step among all process steps;

[0047] Comprehensive control of process sequence and time window:

[0048] According to the calculated T 调整,i , i=1,…,5,G j ,j=1,2,…,n, respectively, sort and analyze them, and feedback the adjustment requirements to the production line; where n is the number of stages from the gradual processing of parts to the final production of products, T 调整,i As the maximum value that can be achieved at each time target point in the product life cycle according to the actual situation of the project; Gj It reflects the level of detail and complexity of the process steps as well as the resources and time required to complete these steps. j The larger the value, the more detailed the division of the process steps, and the value is adjusted according to the complexity and demand in actual production; thus dynamically implementing changes and coordinating consistency issues in the life cycle and system level.

[0049] Furthermore, in the blockchain sample domain, the fuzzy entropy method is used to quantify the uncertainty and complexity of small sample data, providing a quantitative metric for the equalization of small sample data; blockchain technology is used to record the product life cycle time target slice data and the process sequence slice data at the product system level to ensure the authenticity and non-tamperability of the data; the two types of small sample data are verified in real time through smart contracts to ensure the accuracy and integrity of the data; the fuzzy entropy method is executed in the sample data management application in the system to achieve equalization processing of small sample data.

[0050] Furthermore, the specific working principles of the blockchain sample domain include:

[0051] First, the time target slice and process sequence slice data are collected and preprocessed to ensure that the slice data is in a unified dimension. The slice data is then stored on the blockchain, and the data integrity is checked using smart contracts. The sample data management application in the system conducts data transactions with the blockchain through a data interface, and each transaction is recorded on the blockchain.

[0052] Then the sample data management application reconstructs the multi-scale data based on the time series slice data t and the process sequence slice data g, and takes the average value of every s consecutive data points in the original sequence as the new data point t according to the scale s. s 、g s , if s = 2, the average of every two consecutive data points is used as the new data point;

[0053] According to the constructed new scale data, fuzzy entropy calculation is performed: the embedding dimension m and time delay τ are selected according to the dynamic characteristics of the data; a vector is constructed for each new data point, where for t s Data, construct space vector T s =[t s ,t s+τ ,…,t i+(m-1)τ ], and similarly for g s Data, construct space vector G s =[g s ,g s+τ ,…,g i+(m-1)τ ]; T s and G sRepresent the reconstructed vectors of time series slices and process sequence slices under the embedding dimension m respectively; then perform similarity measurement, for any two vectors in the space, calculate their Euclidean distance and obtain the maximum value in the calculation Where T s,i Indicates the i-th time slice construction space vector, G s,i Construct a space vector for the i-th process sequence slice and apply the fuzzy function after the calculation is completed The distance between each pair of vectors is converted into fuzzy similarity, where ε is the parameter that controls the steepness of the fuzzy function curve and r is the similarity threshold; finally, the average value of the fuzzy similarity under each embedding dimension m is calculated. And compare the average difference of fuzzy similarity when embedding dimension is m and m+1 Thus, the fuzzy entropy is obtained, where N is the total number of data points in the original time series t or process sequence g; according to the calculated fuzzy entropy, the complexity and uncertainty between the time slice and the process sequence slice are explained.

[0054] Furthermore, in the blockchain method domain, edge computing integration technology is used to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; blockchain technology is used to securely record and synchronize data on edge nodes, and smart contracts are used to manage edge computing tasks and data exchange; finally, the performance of edge computing and the status of the blockchain are continuously monitored to ensure the stability and efficiency of the system.

[0055] To achieve the above objectives, this technical solution also provides a product analysis system based on the lifecycle-system level data chain model, which is used to implement the above-mentioned product analysis method based on the lifecycle-system level data chain model. The system includes a decentralized data storage platform, which has a dual digital mainline blockchain data model of the product lifecycle and the system level.

[0056] The product lifecycle and system-level dual digital thread blockchain data model includes a product lifecycle digital thread, a product multi-level digital thread, a blockchain information domain, a blockchain structure domain, a blockchain sample domain, and a blockchain method domain.

[0057] in,

[0058] The decentralized data storage platform is used to store product lifecycle and system-level data on-chain;

[0059] The product lifecycle digital thread takes the process as the object, defines the key factors affecting product quality, cost, and completion time, and discovers the spatiotemporal coupling mechanism between processing activities and process systems;

[0060] The multi-level digital backbone of the product uses granular computing theory to model the relationships between multiple levels and their operating results;

[0061] The blockchain information domain adopts the ontology of the manufacturing field, including the formal description of relevant concepts, attributes and relationships, and uses the semantic web to achieve structured representation of manufacturing knowledge, build a unified description framework for manufacturing knowledge and engineering principles, and facilitate the association and reasoning between data; design smart contracts to manage the creation, verification, update and sharing of knowledge, and automate the authorization and access control of knowledge through smart contracts; implement data mining and analysis technologies to identify and extract key data and information sources in physical and logical spaces, and map and manage complex relationships between data;

[0062] The blockchain domain leverages the immutability of blockchain and smart contracts to check data integrity, dynamically adjusting time windows and process frame granularity in system-level control applications to support the synergy between local self-organization and global optimization, thereby creating an adaptive control framework that dynamically adjusts control granularity based on system scale and order acceptance rate.

[0063] The blockchain sample domain uses a fuzzy entropy method to quantify the uncertainty and complexity of small sample data, providing a quantitative metric for small sample data balancing. Blockchain technology is used to record product lifecycle time target slice data and process sequence slice data at the product system level to ensure the authenticity and immutability of the data. Smart contracts are used to verify the two types of small sample data in real time to ensure data accuracy and integrity. Fuzzy entropy methods are implemented in the sample data management application of the system to achieve balanced processing of small sample data.

[0064] The blockchain method domain utilizes edge computing integration technology to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; and uses blockchain technology to securely record and synchronize data on edge nodes, while using smart contracts to manage edge computing tasks and data exchange; finally, continuously monitors the performance of edge computing and the status of the blockchain to ensure the stability and efficiency of the system.

[0065] Compared with the existing technology, the principles and advantages of this solution are as follows:

[0066] 1. Through blockchain's decentralized data storage platform, transparent recording and traceability of product lifecycle data are achieved. Smart contract technology is applied to precisely coordinate global and local decision-making, breaking the collaborative bottlenecks of traditional manufacturing systems. Furthermore, blockchain's distributed sharing mechanism addresses the imbalance of small sample data, improving data credibility and overall system availability.

[0067] 2. Build a dual digital thread blockchain data model for the product lifecycle and system level based on a decentralized data storage platform. By establishing an intelligent manufacturing data analysis and knowledge learning model in the method domain, blockchain and edge computing will be combined to make data processing more efficient and intelligent. This will further enhance the intelligence level of manufacturing systems and provide a comprehensive, efficient, and intelligent solution for digital transformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the services required for use in the embodiments or the prior art descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0069] Figure 1 This is a structural diagram of the product lifecycle and system-level dual digital mainline blockchain data model in the product analysis system based on the lifecycle-system-level data chain model of the present invention. DETAILED DESCRIPTION

[0070] The present invention will be further described below in conjunction with specific embodiments:

[0071] The product analysis system based on the lifecycle-system level data chain model described in this embodiment includes a decentralized data storage platform, which has a dual digital main line blockchain data model of product lifecycle and system level.

[0072] like Figure 1 As shown, the product lifecycle and system-level dual digital thread blockchain data model includes the product lifecycle digital thread, the product multi-level digital thread, the blockchain information domain, the blockchain structure domain, the blockchain sample domain, and the blockchain method domain;

[0073] Among them, the decentralized data storage platform is used to upload product life cycle and system-level data to the chain; the product life cycle digital backbone takes the process as the object, defines the key factors affecting product quality, cost, and completion time, and discovers the spatiotemporal coupling mechanism between processing activities and process systems; the product multi-level digital backbone uses granular computing theory to model the relationships at multiple levels and their operating results.

[0074] The blockchain information domain adopts the ontology of the manufacturing field, including the formal description of relevant concepts, attributes and relationships. It uses the semantic web to achieve structured representation of manufacturing knowledge and build a unified description framework for manufacturing knowledge and engineering principles to facilitate the association and reasoning between data. Smart contracts are designed to manage the creation, verification, update and sharing of knowledge, and the authorization and access control of knowledge are automated through smart contracts. Data mining and analysis technologies are implemented to identify and extract key data and information sources in physical and logical spaces, and to map and manage complex relationships between data.

[0075] The blockchain structure domain uses the immutability of blockchain and smart contracts to check data integrity, and dynamically adjusts the time window and process box granularity in system granularity control applications to support the collaboration between local self-organization and global optimization, thereby creating an adaptive control framework and dynamically adjusting the control granularity according to the system scale and order acceptance rate.

[0076] The blockchain sample domain uses the fuzzy entropy method to quantify the uncertainty and complexity of small sample data, providing a quantitative metric for the equalization of small sample data. Blockchain technology is used to record the product life cycle time target slice data and the process sequence slice data at the product system level to ensure the authenticity and non-tamperability of the data. The two types of small sample data are verified in real time through smart contracts to ensure the accuracy and integrity of the data. The fuzzy entropy method is executed in the sample data management application of the system to achieve the equalization of small sample data.

[0077] The blockchain method domain utilizes edge computing integration technology to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; and uses blockchain technology to securely record and synchronize data on edge nodes, while using smart contracts to manage edge computing tasks and data exchange; finally, continuously monitor the performance of edge computing and the status of the blockchain to ensure the stability and efficiency of the system.

[0078] The working principle of this embodiment is as follows:

[0079] Through a decentralized data storage platform, product lifecycle and system-level data can be uploaded to the chain;

[0080] Through the coordination of blockchain information domain, blockchain structure domain, blockchain sample domain and blockchain method domain, the digital main line of the product life cycle and the multi-level digital main line of the product are combined to analyze the product.

[0081] The specific working principles of the blockchain information domain include:

[0082] First, collect standard knowledge in the manufacturing field and build a standard manufacturing knowledge terminology database;

[0083] Then define the manufacturing domain ontology and use OWL language to formally describe the ontology, that is, the relevant concepts C, attributes O p , define the relationship R and get the OWL file;

[0084] Use RDF to annotate existing product lifecycle and system data, and correspond the data to the concepts, attributes, and relationships in the ontology to obtain RDF = {(c, o, r) | c∈C, o∈O p ,r∈R};

[0085] Serialize OWL files and RDF data into JSON-LD format and upload them to the blockchain, using smart contracts to automatically manage the storage, update, and query of ontology and RDF data;

[0086] A graph database is linked to the blockchain to synchronize data regularly, and SPARQL is used for semantic queries to mine and analyze product and system information.

[0087] In the above, the relevant concept C includes parts, components, products, and equipment;

[0088] Attribute O p Including part size, weight, process, component shape, product packaging;

[0089] Relationship R includes the assembly of parts into components and the assembly of components into products;

[0090] Product information includes basic product information, product batch information, process requirements information, and technical parameter information;

[0091] System information includes equipment operating status, processing object description, processing quality information, and processing cycle plan.

[0092] The specific working principles of the blockchain structure domain include:

[0093] First, collect the time target points T={T1, T2,…, T5} in the product life cycle.

[0094] Where T1 is the time from demand analysis to process planning;

[0095] T2 is the manufacturing and assembly time;

[0096] T3 is the test and certification time;

[0097] T4 inventory logistics time;

[0098] T5 is the delivery and installation time.

[0099] And a complete set of product process sequences at the product system level, where the process sequence includes the following data:

[0100] N子,j : The number of sub-processes in the process sequence of stage j;

[0101] R 需求,j : The amount of resources required for stage j;

[0102] C 复杂,j : complexity score of the process step in stage j;

[0103] T 需求,j : The time required to complete the process steps of stage j;

[0104] The acquired data is transmitted to the blockchain for on-chain operations to generate transactions, and the smart contract is automatically triggered to check the integrity of the data. After the check is completed and the data is stored in the block after consensus is reached;

[0105] The granular control application then extracts data from the blockchain at regular intervals. Each transaction with the system application is recorded and stored in the blockchain. The granular control application then analyzes the current target data and process sequence data. The specific analysis process includes:

[0106] Analyze the system scale and order acceptance rate, among which,

[0107]

[0108] F 系统 is the system size adjustment factor, S 当前 is the current project scale, S 基准 is the benchmark project size;

[0109]

[0110] F 接单 is the order acceptance rate adjustment factor, R 当前 is the current order acceptance rate, R 基准 is the benchmark order acceptance rate;

[0111] Dynamically adjust the time window:

[0112] T 调整,i =T i ×F 系统 ×F 接单

[0113] Process sequence granularity calculation:

[0114] G j =α·N 子,j +β·R 需求,j +γ·C 复杂,j +δ·T 需求,j ,

[0115] Among them, Gj is the process sequence granularity, α is the weight coefficient of the sub-process quantity, β is the resource demand weight coefficient, R 稀缺资源 Indicates the resource type with the highest demand among all steps; γ is the weight coefficient of complexity score, C 最高复杂 represents the complexity score of the step with the highest complexity score among all process steps; δ is the time requirement weight coefficient, T 最长需求 The time required for the longest step among all process steps;

[0116] Comprehensive control of process sequence and time window:

[0117] According to the calculated T 调整,i , i=1,…,5,G j ,j=1,2,…,n, respectively, sort and analyze them, and feedback the adjustment requirements to the production line; where n is the number of stages from the gradual processing of parts to the final production of products, T 调整,i As the maximum value that can be achieved at each time target point in the product life cycle according to the actual situation of the project; G j It reflects the level of detail and complexity of the process steps as well as the resources and time required to complete these steps. j The larger the value, the more detailed the division of the process steps, and the value is adjusted according to the complexity and demand in actual production; thus dynamically implementing changes and coordinating consistency issues in the life cycle and system level.

[0118] The specific working principles of the blockchain sample domain include:

[0119] First, the time target slice and process sequence slice data are collected and preprocessed to ensure that the slice data is in a unified dimension. The slice data is then stored on the blockchain, and the data integrity is checked using smart contracts. The sample data management application in the system conducts data transactions with the blockchain through a data interface, and each transaction is recorded on the blockchain.

[0120] Then the sample data management application reconstructs the multi-scale data based on the time series slice data t and the process sequence slice data g, and takes the average value of every s consecutive data points in the original sequence as the new data point t according to the scale s. s 、g s , if s = 2, the average of every two consecutive data points is used as the new data point;

[0121] According to the constructed new scale data, fuzzy entropy calculation is performed: the embedding dimension m and time delay τ are selected according to the dynamic characteristics of the data; a vector is constructed for each new data point, where for ts Data, construct space vector T s =[t s ,t s+τ ,…,t i+(m-1)τ ], and similarly for g s Data, construct space vector G s =[g s ,g s+τ ,…,g i+(m-1)τ ]; T s and G s Represent the reconstructed vectors of time series slices and process sequence slices under the embedding dimension m respectively; then perform similarity measurement, for any two vectors in the space, calculate their Euclidean distance and obtain the maximum value in the calculation Where T s,i Indicates the i-th time slice construction space vector, G s,i Construct a space vector for the i-th process sequence slice and apply the fuzzy function after the calculation is completed The distance between each pair of vectors is converted into fuzzy similarity, where ε is the parameter that controls the steepness of the fuzzy function curve and r is the similarity threshold; finally, the average value of the fuzzy similarity under each embedding dimension m is calculated. And compare the average difference of fuzzy similarity when embedding dimension is m and m+1 Thus, the fuzzy entropy is obtained, where N is the total number of data points in the original time series t or process sequence g; according to the calculated fuzzy entropy, the complexity and uncertainty between the time slice and the process sequence slice are explained.

[0122] In the blockchain method domain, edge computing integration technology is used to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; blockchain technology is used to securely record and synchronize data on edge nodes, and smart contracts are used to manage edge computing tasks and data exchange; finally, the performance of edge computing and the status of the blockchain are continuously monitored to ensure the stability and efficiency of the system.

[0123] The embodiments described above are only preferred embodiments of the present invention and are not intended to limit the scope of implementation of the present invention. Therefore, any changes made based on the shape and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A product analysis method based on the lifecycle-system level data chain model, characterized by: include: Through a decentralized data storage platform, product lifecycle and system-level data can be uploaded to the chain; Build a dual digital thread blockchain data model for the product lifecycle and system level. This dual digital thread blockchain data model includes the product lifecycle digital thread, the product multi-level digital thread, the blockchain information domain, the blockchain structure domain, the blockchain sample domain, and the blockchain method domain. By coordinating the blockchain information domain, blockchain structure domain, blockchain sample domain, and blockchain method domain, the product lifecycle digital thread and the product multi-level digital thread are combined to analyze the product. The product lifecycle digital thread takes the process as the object, defines the key factors affecting product quality, cost, and completion time, and discovers the spatiotemporal coupling mechanism between processing activities and process systems; The multi-level digital backbone of the product uses granular computing theory to model the relationships between multiple levels and their operating results; In the blockchain information domain, the ontology of the manufacturing field is adopted, including the formal description of relevant concepts, attributes and relationships. The semantic web is used to achieve the structured representation of manufacturing knowledge, and a unified description framework for manufacturing knowledge and engineering principles is constructed to facilitate the association and reasoning between data. Smart contracts are designed to manage the creation, verification, update and sharing of knowledge, and the authorization and access control of knowledge are automated through smart contracts. Data mining and analysis technologies are implemented to identify and extract key data and information sources in physical and logical spaces, and to map and manage complex relationships between data. In the blockchain domain, the immutability of blockchain and smart contracts are used to check data integrity. In system-level control applications, time windows and process frame granularity are dynamically adjusted to support the synergy between local self-organization and global optimization. This creates an adaptive control framework that dynamically adjusts control granularity based on system scale and order acceptance rate. In the blockchain sample domain, the fuzzy entropy method is used to quantify the uncertainty and complexity of small sample data, providing a quantitative metric for small sample data balancing. Blockchain technology is used to record product lifecycle time target slice data and process sequence slice data at the product system level to ensure data authenticity and immutability. Smart contracts are used to verify both small sample data in real time to ensure data accuracy and integrity. The fuzzy entropy method is implemented using the sample data management application in the system to achieve small sample data balancing. In the blockchain method domain, edge computing integration technology is used to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; blockchain technology is used to securely record and synchronize data on edge nodes, and smart contracts are used to manage edge computing tasks and data exchange; finally, the performance of edge computing and the status of the blockchain are continuously monitored to ensure the stability and efficiency of the system.

2. The product analysis method based on the life cycle-system level data chain model according to claim 1 is characterized in that: The specific working principles of the blockchain information domain include: First, collect standard knowledge in the manufacturing field and build a standard manufacturing knowledge terminology database; Then define the manufacturing domain ontology and use OWL language to formally describe the ontology, that is, the relevant concepts C, attributes O p , define the relationship R and get the OWL file; Use RDF to annotate existing product lifecycle and system data, and correspond the data to the concepts, attributes, and relationships in the ontology to obtain RDF = {(c, o, r) | c∈C, o∈O p ,r∈R}; Serialize OWL files and RDF data into JSON-LD format and upload them to the blockchain, using smart contracts to automatically manage the storage, update, and query of ontology and RDF data; A graph database is linked to the blockchain to synchronize data regularly, and SPARQL is used for semantic queries to mine and analyze product and system information.

3. The product analysis method based on the life cycle-system level data chain model according to claim 1 is characterized in that: The specific working principles of the blockchain structure domain include: First, collect the time target points T={T1, T2,…, T5} in the product life cycle. Where T1 is the time from demand analysis to process planning; T2 is the manufacturing and assembly time; T3 is the test and certification time; T4 inventory logistics time; T5 is the delivery and installation time; And a complete set of product process sequences at the product system level, where the process sequence includes the following data: N 子,j : The number of sub-processes in the process sequence of stage j; R 需求,j : The amount of resources required for stage j; C 复杂,j : complexity score of the process step in stage j; T 需求,j : The time required to complete the process steps of stage j; The acquired data is transmitted to the blockchain for on-chain operations to generate transactions, and the smart contract is automatically triggered to check the integrity of the data. After the check is completed and the data is stored in the block after consensus is reached; The granular control application will then extract data from the blockchain at regular intervals. The data will be recorded and saved in the blockchain for each transaction generated with the system application; the granular control application will then analyze the current time target data and process sequence data.

4. The product analysis method based on the lifecycle-system level data chain model according to claim 3 is characterized in that: The specific process of analyzing the current time target data and process sequence data by the particle size control application includes: Analyze the system scale and order acceptance rate, among which, F 系统 is the system size adjustment factor, S 当前 is the current project scale, S 基准 is the benchmark project size; F 接单 is the order acceptance rate adjustment factor, R 当前 is the current order acceptance rate, R 基准 is the benchmark order acceptance rate; Dynamically adjust the time window: T 调整,i =T i ×F 系统 ×F 接单 Process sequence granularity calculation: G j =α·N 子,j +β·β 需求,j +γ·C 复杂,j +δ·T 需求,j , Among them, G j is the process sequence granularity, α is the weight coefficient of the sub-process quantity, β is the resource demand weight coefficient, R 稀缺资源 Indicates the resource type with the highest demand among all steps; γ is the weight coefficient of complexity score, C 最高复杂 represents the complexity score of the step with the highest complexity score among all process steps; δ is the time requirement weight coefficient, T 最长需求 The time required for the longest step among all process steps; Comprehensive control of process sequence and time window: According to the calculated T 调整,i , i=1,…,5,G j ,j=1,2,…,n, respectively, sort and analyze them, and feedback the adjustment requirements to the production line; where n is the number of stages from the gradual processing of parts to the final production of products, T 调整,i As the maximum value that can be achieved at each time target point in the product life cycle according to the actual situation of the project; G j It reflects the level of detail and complexity of the process steps as well as the resources and time required to complete these steps. j The larger the value, the more detailed the division of the process steps, and the value is adjusted according to the complexity and demand in actual production; thus dynamically implementing changes and coordinating consistency issues in the life cycle and system level.

5. The product analysis method based on the life cycle-system level data chain model according to claim 1 is characterized in that: The specific working principles of the blockchain sample domain include: First, the time target slice and process sequence slice data are collected and preprocessed to ensure that the slice data is in a unified dimension. The slice data is then stored on the blockchain, and the data integrity is checked using smart contracts. The sample data management application in the system conducts data transactions with the blockchain through a data interface, and each transaction is recorded on the blockchain. Then the sample data management application reconstructs the multi-scale data based on the time series slice data t and the process sequence slice data g, and takes the average value of every s consecutive data points in the original sequence as the new data point t according to the scale s. s 、g s , if s = 2, the average of every two consecutive data points is used as the new data point; According to the constructed new scale data, fuzzy entropy calculation is performed: the embedding dimension m and time delay τ are selected according to the dynamic characteristics of the data; a vector is constructed for each new data point, where for t s Data, construct space vector T s =[t s ,t s+τ ,…,t i+(m-1)τ ], for g s Data, construct space vector G s =[g s ,g s+τ ,…,g i+(m-1)τ ]; T s and G s Represent the reconstructed vectors of time series slices and process sequence slices under the embedding dimension m respectively; then perform similarity measurement, for any two vectors in the space, calculate their Euclidean distance and obtain the maximum value in the calculation Where T s,i Indicates the i-th time slice construction space vector, G s,i Construct a space vector for the i-th process sequence slice and apply the fuzzy function after the calculation is completed The distance between each pair of vectors is converted into fuzzy similarity, where ε is the parameter that controls the steepness of the fuzzy function curve and r is the similarity threshold; finally, the average value of the fuzzy similarity under each embedding dimension m is calculated. And compare the average difference of fuzzy similarity when embedding dimension is m and m+1 Thus, the fuzzy entropy is obtained, where N is the total number of data points in the original time series t or process sequence g; according to the calculated fuzzy entropy, the complexity and uncertainty between the time slice and the process sequence slice are explained.

6. A product analysis system based on a lifecycle-system level data chain model, for implementing the product analysis method based on a lifecycle-system level data chain model according to any one of claims 1 to 5, characterized in that: Includes a decentralized data storage platform with dual digital thread blockchain data models for product lifecycle and system level; The product lifecycle and system-level dual digital thread blockchain data model includes a product lifecycle digital thread, a product multi-level digital thread, a blockchain information domain, a blockchain structure domain, a blockchain sample domain, and a blockchain method domain. in, The decentralized data storage platform is used to store product lifecycle and system-level data on-chain; The product lifecycle digital thread takes the process as the object, defines the key factors affecting product quality, cost, and completion time, and discovers the spatiotemporal coupling mechanism between processing activities and process systems; The multi-level digital backbone of the product uses granular computing theory to model the relationships between multiple levels and their operating results; The blockchain information domain adopts the ontology of the manufacturing field, including the formal description of relevant concepts, attributes and relationships, and uses the semantic web to achieve structured representation of manufacturing knowledge, build a unified description framework for manufacturing knowledge and engineering principles, and facilitate the association and reasoning between data; design smart contracts to manage the creation, verification, update and sharing of knowledge, and automate the authorization and access control of knowledge through smart contracts; implement data mining and analysis technologies to identify and extract key data and information sources in physical and logical spaces, and map and manage complex relationships between data; The blockchain domain leverages the immutability of blockchain and smart contracts to check data integrity, dynamically adjusting time windows and process frame granularity in system-level control applications to support the synergy between local self-organization and global optimization, thereby creating an adaptive control framework that dynamically adjusts control granularity based on system scale and order acceptance rate. The blockchain sample domain uses a fuzzy entropy method to quantify the uncertainty and complexity of small sample data, providing a quantitative metric for small sample data balancing. Blockchain technology is used to record product lifecycle time target slice data and process sequence slice data at the product system level to ensure the authenticity and immutability of the data. Smart contracts are used to verify the two types of small sample data in real time to ensure data accuracy and integrity. The fuzzy entropy method is executed using the sample data management application in the system to achieve small sample data balancing. The blockchain method domain utilizes edge computing integration technology to deploy edge computing resources on edge nodes of manufacturing equipment or near-equipment to achieve rapid data processing and real-time analysis; and uses blockchain technology to securely record and synchronize data on edge nodes, while using smart contracts to manage edge computing tasks and data exchange; finally, continuously monitors the performance of edge computing and the status of the blockchain to ensure the stability and efficiency of the system.

Images