Supply chain trusted data collaborative management method and system based on multi-mode block chain

Abstract

The invention relates to the technical field of data processing, and discloses a supply chain trusted data collaborative management method and system based on a multi-modal block chain, and the method comprises the steps: constructing a data modal sensing unit through the multi-source heterogeneous original data of a supply chain; generating a data classification evidence storage list in combination with the differential evidence storage rule base of the supply chain and the multi-source heterogeneous original data; according to service scene requirements and network operation states, consensus mechanism adaptation standards of the supply chain in different service scenes are defined; configuring a consensus strategy scheduler of the supply chain based on a consensus mechanism adaptation standard; and calculating a global performance state index of the supply chain to generate a collaborative optimization instruction of the data classification evidence storage list and the consensus strategy scheduler. According to the method, the credibility and the collaborative management efficiency of the multi-source heterogeneous data of the supply chain can be improved.

Description

Technical Field

[0001] This invention relates to a collaborative management method and system for trusted data in the supply chain based on multimodal blockchain, belonging to the field of data processing technology. Background Technology

[0002] Trusted data collaborative management of the supply chain refers to the systematic and standardized management of data flow and collaboration across the entire supply chain by integrating distributed ledger technology, data encryption algorithms, and cross-entity identity authentication mechanisms. It can collect and verify key data from each link of the supply chain in real time (such as material procurement information, production progress data, logistics and transportation trajectory, inventory changes, and transaction settlement records), and dynamically monitor the integrity of data transmission, the security of storage, and the compliance of use. It aims to break down data silos among various participants in the supply chain and ensure the trustworthiness of data in the process of sharing and collaboration.

[0003] However, existing implementation methods mostly adopt the single hash value on-chain method, which extracts the hash value of supply chain data and uploads it to the blockchain to complete the notarization, thereby realizing basic data traceability and integrity verification. Although this method can handle the needs of confirming ownership and tracing tampering of core data, it cannot deeply analyze the modalities and business value of multi-source heterogeneous raw data (such as structured transaction forms, unstructured production videos, etc.), and can only perform indiscriminate processing, thus having inherent defects in data management. Summary of the Invention

[0004] This invention provides a collaborative management method and system for trusted data in the supply chain based on multimodal blockchain, the main purpose of which is to improve the credibility and collaborative management efficiency of multi-source heterogeneous data in the supply chain.

[0005] To achieve the above objectives, this invention provides a supply chain trusted data collaborative management method based on multimodal blockchain, comprising: Acquire multi-source heterogeneous raw data and business process interaction data from the supply chain to construct a data modality perception unit for the supply chain; Based on the data modality sensing unit, a differentiated evidence storage rule base for the supply chain is established. Combining the differentiated evidence storage rule base and the multi-source heterogeneous original data, a data classification evidence storage list for the supply chain is generated. Identify the business scenario requirements and network operation status of the supply chain from the interaction data of the business process, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status. Based on the consensus mechanism adaptation standard, the consensus mechanism decision factor of the supply chain is determined, and the consensus strategy scheduler of the supply chain is configured according to the consensus mechanism decision factor. The data storage integrity index and consensus mechanism response efficiency value of the supply chain are monitored in real time to calculate the global performance status index of the supply chain. Based on the global performance status index, the collaborative optimization instructions of the data classification storage list and the consensus strategy scheduler are generated. By combining the consensus strategy scheduler, the data classification and storage list, and the collaborative optimization instructions, a data collaborative management scheme for the supply chain is output.

[0006] Optionally, the step of generating the data classification and storage list of the supply chain by combining the differentiated evidence storage rule base and the multi-source heterogeneous data features of the multi-source heterogeneous original data includes: The multi-source heterogeneous raw data is preprocessed to obtain a standardized dataset, wherein the preprocessing includes data cleaning, format standardization and metadata extraction; Identify the categorical data items in the standardized dataset and extract the data feature indicators corresponding to the categorical data items; Based on the differentiated evidence preservation rule base, calculate the matching weight between each data feature indicator and each evidence preservation rule; Based on the matching weights, the evidence preservation priority and evidence preservation timeliness level of the categorized data items are determined; Construct a business logic association model between the categorized data items to identify the data dependencies between them; By combining the evidence preservation priority, the evidence preservation timeliness level, and the data dependency relationship, an initial evidence preservation priority sequence for the supply chain is generated; Obtain the evidence storage node cluster corresponding to the supply chain, and collect the real-time evidence storage environment parameters of the evidence storage node cluster. Based on the real-time evidence storage environment parameters, dynamic priority optimization processing of the initial evidence storage priority sequence is performed to generate the data classification evidence storage list of the supply chain.

[0007] Optionally, configuring the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor includes: By using the dynamic evaluation data of the consensus mechanism decision factors, the strategy configuration node corresponding to the supply chain is located; Key performance indicators are extracted from the dynamic evaluation data, and the correlation weight between the key performance indicators and the strategy configuration node is calculated. Based on the association weights, the core configuration parameters of the strategy configuration node are selected; Identify the parameter adjustment path of the core configuration parameters in the policy configuration node; Based on the core configuration parameters, calculate the parameter sensitivity of the key performance indicators; Based on the parameter sensitivity, the adjustment priority of the core configuration parameters is set; Obtain a preset consensus strategy configuration scheme that matches the adjustment priority; Analyze the parameter adjustment boundary conditions of the core configuration parameters in the preset consensus strategy configuration scheme; By integrating the parameter adjustment boundary conditions and the parameter adjustment path, a two-level policy generation network for the consensus policy scheduler is constructed. The consensus policy scheduler for the supply chain is configured through the two-level policy generation network.

[0008] Optionally, the real-time monitoring of the data storage integrity index and consensus mechanism response efficiency value of the supply chain to calculate the global performance index of the supply chain includes: Identify the distributed nodes of the supply chain and obtain the consistency verification results of the ledger data of the distributed nodes; Calculate the data reliability index of the supply chain based on the data storage integrity index and the ledger data consistency verification results; Based on the consensus mechanism response efficiency value, the throughput variation coefficient and block propagation latency parameters of the supply chain are monitored in real time. The consensus stability coefficient of the supply chain is calculated based on the throughput variation coefficient and the block propagation delay parameter. Real-time collection of computing resource utilization, storage resource utilization, and network bandwidth utilization of the supply chain; By integrating the data reliability index, the consensus stability coefficient, the computing resource utilization rate, the storage resource utilization rate, and the network bandwidth utilization rate, the global performance status index of the supply chain is calculated using the following formula: ; in, This represents the overall performance status index. Indicates the data reliability index. This represents the consensus stability coefficient. The weighting coefficient represents the consensus stability coefficient. This indicates the utilization rate of computing resources. This represents the weighting coefficient used to calculate resource utilization. Indicates network bandwidth utilization. This represents the rated baseline value of network bandwidth corresponding to the network bandwidth utilization rate. This indicates the utilization rate of computing resources. This represents the rated baseline value of computing power corresponding to the computing resource utilization rate. Indicates storage resource utilization. This indicates the rated baseline value of storage space corresponding to the storage resource utilization rate. Indicates the system stability variance. This indicates a regulatory factor.

[0009] Optionally, the step of generating the collaborative optimization instructions for the data classification and storage list and the consensus policy scheduler based on the global performance state index includes: Obtain the performance evaluation benchmark corresponding to the global performance status index; Based on the performance evaluation benchmark, identify the system bottleneck type corresponding to the global performance status index, and set multi-level performance thresholds corresponding to the global performance status index; Based on the multi-level efficiency thresholds, the collaborative control levels corresponding to the global efficiency state index are divided. Based on the system bottleneck type, the multi-level performance threshold, and the collaborative control level, determine the joint optimization strategy for the data classification and storage list and the consensus strategy scheduler; The parameter adjustment step size of the joint optimization strategy is calculated based on the global performance state index and the multi-level performance threshold. Based on the parameter adjustment step size, construct the parameter coupling relationship matrix between the data classification and evidence storage list and the consensus strategy scheduler; The data classification and evidence storage list and the consensus strategy scheduler are generated through the parameter coupling relationship matrix.

[0010] Optionally, the step of acquiring multi-source heterogeneous raw data and business process interaction data of the supply chain to construct the data modality perception unit of the supply chain includes: Extract the data modal features and business source attributes of the multi-source heterogeneous raw data; Using the data modality features and the business source attributes, determine the set of structured data parsing strategies for the multi-source heterogeneous raw data; Based on the interaction data of the business processes, identify the business process nodes and data interaction relationships of the supply chain; The business logic rules and data permission levels for parsing the interactive data of the aforementioned business processes; Based on the business logic rules and the data permission levels, a value density evaluation standard for the multi-source heterogeneous raw data is set. By combining the structured data parsing strategy set, the business process nodes, and the value density assessment criteria, a data modality mapping model for the supply chain is constructed. Based on the data interaction relationship and the data modality mapping model, determine the type of multimodal fusion algorithm corresponding to the supply chain; By combining the structured data parsing strategy set, the value density assessment standard, and the multimodal fusion algorithm type, the data modality perception unit of the supply chain is constructed.

[0011] Optionally, establishing the differentiated evidence storage rule base for the supply chain based on the data modality sensing unit includes: The system acquires raw data from the entire supply chain and outputs the modality recognition results and data attribute information of the raw data through the data modality perception unit. The modal feature set of the original data of the entire link is extracted from the modal recognition result, wherein the modal feature set includes data format type, data generation frequency and data flow continuity characteristics; Based on the data attribute information, the business attribute set of the original data of the entire chain is extracted, wherein the business attribute set includes business link identifier, data sensitivity level and compliance evidence storage period; Based on the modal feature set, a selection strategy for evidence preservation technology corresponding to the supply chain is generated; Based on the aforementioned business attribute set, define the evidence preservation compliance constraint rules corresponding to the supply chain; By integrating the evidence preservation technology selection strategy and the evidence preservation compliance constraint rules, a differentiated evidence preservation strategy vector for the supply chain is output. The differentiated evidence preservation strategy vector is encapsulated into executable evidence preservation rules for the supply chain; Integrate the executable evidence storage rules to establish a differentiated evidence storage rule library for the supply chain.

[0012] Optionally, defining the consensus mechanism adaptation standard for the supply chain under different business scenarios based on the business scenario requirements and the network operating status includes: Based on the business scenario requirements, the core consensus requirements of the supply chain under different business scenarios are extracted. Based on the business scenario requirements and the core consensus requirements, a candidate set of consensus mechanism types for the supply chain under different business scenarios is selected. Construct a dynamic adaptation matrix between the candidate set of consensus mechanism types and the network operating state, and calculate the consensus mechanism matching degree of the supply chain; The types of consensus nodes and network interfaces participating in the supply chain are clearly defined in order to determine the node admission conditions and communication protocol specifications of the candidate set of consensus mechanism types; Based on the dynamic adaptation matrix, the node admission conditions, and the communication protocol specifications, consensus parameter configuration rules for the supply chain under different business scenarios are generated. Based on the consensus parameter configuration rules, the consensus performance boundaries of the supply chain under different business scenarios are defined. The consensus mechanism matching degree, consensus parameter configuration rules, and consensus performance boundaries define the consensus mechanism adaptation standard for the supply chain in different business scenarios.

[0013] Optionally, determining the consensus mechanism decision factors for the supply chain based on the consensus mechanism adaptation standard includes: Real-time monitoring of the dynamic characteristics of node load and network communication quality in the supply chain; Calculate the load change rate corresponding to the node load dynamic characteristics and the network latency index corresponding to the network communication quality; Based on the load change rate and the network latency index, identify the real-time operating status mode of the supply chain; The candidate consensus mechanism type corresponding to the supply chain is determined by the state-mechanism mapping relationship between the real-time running state mode and the consensus mechanism adaptation standard. Based on the candidate consensus mechanism type, the transaction processing latency, data consistency index and system fault tolerance parameters of the supply chain in the real-time operating state mode are collected in real time. Based on the transaction processing latency, the data consistency index, and the system fault tolerance parameters, analyze the performance patterns of each candidate consensus mechanism type in the real-time operating state mode. Construct a matching degree evaluation matrix between the performance pattern and the consensus mechanism adaptation standard; Based on the matching degree evaluation matrix, the consensus mechanism decision factors of the supply chain are determined.

[0014] To address the aforementioned issues, this invention also provides a supply chain trusted data collaborative management system based on multimodal blockchain, the system comprising: The data modality perception module is used to acquire multi-source heterogeneous raw data and business process interaction data of the supply chain in order to construct the data modality perception unit of the supply chain. The data classification and evidence storage module is used to establish a differentiated evidence storage rule base for the supply chain based on the data modality perception unit, and generate a data classification and evidence storage list for the supply chain by combining the differentiated evidence storage rule base and the multi-source heterogeneous original data. The consensus mechanism intelligent adaptation module is used to identify the business scenario requirements and network operation status of the supply chain from the interaction data of the business process, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status. The consensus strategy dynamic scheduling module is used to determine the consensus mechanism decision factor of the supply chain based on the consensus mechanism adaptation standard, and configure the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor. The global performance collaborative optimization module is used to monitor the data storage integrity index and consensus mechanism response efficiency value of the supply chain in real time, so as to calculate the global performance status index of the supply chain, and generate the collaborative optimization instructions of the data classification storage list and the consensus strategy scheduler based on the global performance status index. The data collaboration management module is used to combine the consensus strategy scheduler, the data classification and storage list, and the collaboration optimization instructions to output the data collaboration management scheme of the supply chain.

[0015] Compared to the problems described in the background art, the embodiments of the present invention acquire multi-source heterogeneous raw data and business process interaction data from the supply chain to construct a data modality perception unit for the supply chain. This unit can transform the modal characteristics and business value associations of multi-source heterogeneous data in the supply chain into analyzable and usable structured perception results, improving the accuracy of supply chain collaborative management. Furthermore, the embodiments of the present invention establish a differentiated evidence storage rule base for the supply chain based on the data modality perception unit. This can accurately match the credible evidence storage needs of different modal data in the supply chain. By leveraging the hierarchical control capabilities of the rule base, it adapts to the storage characteristics and verification logic of multimodal blockchains, providing a basis for subsequent supply chain management. The trusted on-chain data and collaborative interaction provide a standardized operational framework, ensuring the authenticity and relevance of on-chain data from the data entry point, and guaranteeing the efficiency and reliability of trusted data collaborative management in the supply chain. This invention, by combining the differentiated evidence storage rule base and the multi-source heterogeneous raw data, generates a data classification and evidence storage list for the supply chain. This upgrades fragmented data governance to a systematic evidence storage framework, accurately mining the modal characteristics and business value of multi-source heterogeneous raw data (such as structured transaction forms, unstructured production videos, etc.), shifting from disordered data accumulation to ordered value extraction, and strengthening the full lifecycle management of supply chain data assets. Furthermore, this invention... This embodiment defines consensus mechanism adaptation standards for the supply chain under different business scenarios based on the business scenario requirements and network operation status. Adapted consensus mechanisms can enhance the accuracy of data modality recognition and the depth of business value association, improve the credibility and application efficiency of data parsing results, and enable synergy between multi-source heterogeneous raw data in modality parsing completeness and business value transformation efficiency, comprehensively enhancing the dynamic parsing and value mining capabilities of multi-source heterogeneous raw data in the supply chain. Furthermore, this embodiment configures the consensus strategy scheduler for the supply chain based on the consensus mechanism decision factors, allowing the consensus strategy scheduler to accurately match the supply chain's trusted data collaborative management. By understanding real-time business needs and network operation status, this invention enhances the targeting of consensus strategy scheduling and the security and efficiency of data collaboration, while providing core data support for the dynamic iteration and optimization of supply chain consensus strategies. Furthermore, by monitoring the data evidence integrity index and consensus mechanism response efficiency value of the supply chain in real time, this invention calculates the global performance status index of the supply chain. This enables the trusted data of the supply chain to accurately reflect the full-link operation status based on the trustworthiness of the supply chain data evidence and the dynamic changes in the consensus response, thereby improving the global control targeting of the supply chain data collaboration process and building a dynamic performance monitoring and optimization mechanism for trusted data collaboration management of the supply chain.This invention, through generating a data classification and evidence storage list and a consensus strategy scheduler based on the global performance status index, can optimize core elements such as data classification priority, evidence storage resource allocation ratio, and consensus strategy operation parameters in real time. This improves the accuracy of adapting to the multimodal data characteristics of the supply chain and network operation status, enhances the linkage between data evidence storage and consensus strategies, and strengthens the targeted and effective collaborative management of trusted data in the supply chain. It also builds a precise and dynamic technical management defense line for the trusted flow of data across the entire supply chain. Finally, this invention, by combining the consensus strategy scheduler... The algorithm, comprising the data classification and storage list and the collaborative optimization instructions, outputs a data collaborative management scheme for the supply chain. This scheme not only significantly overcomes the technical limitations of existing single-hash-value on-chain methods in deeply analyzing multi-source heterogeneous data modalities and business value, shifting supply chain data storage from "indiscriminate processing" to "value-based classification and control," but also adjusts the consensus strategy's operational logic, data storage resource allocation, and collaborative linkage rules in real time based on the dynamic changes of the global performance state vector. This effectively enhances the adaptability, accuracy, and intelligence of supply chain trusted data collaborative management based on multi-modal blockchain. Therefore, the supply chain trusted data collaborative management method and system based on multi-modal blockchain provided in this invention can improve the credibility and collaborative management efficiency of multi-source heterogeneous data in the supply chain. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a collaborative management method for trusted supply chain data based on multimodal blockchain, provided in one embodiment of the present invention. Figure 2 A flowchart illustrating the on-chain data integrity verification process in a supply chain trusted data collaborative management method based on multimodal blockchain, provided as an embodiment of the present invention; Figure 3 This is a schematic diagram of a module for implementing a supply chain trusted data collaborative management system based on multimodal blockchain, as provided in an embodiment of the present invention.

[0017] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for collaborative management of trusted supply chain data based on multimodal blockchain. The executing entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a collaborative management method for trusted supply chain data based on multimodal blockchain, according to an embodiment of the present invention. In this embodiment, the collaborative management method for trusted supply chain data based on multimodal blockchain includes: S1. Acquire multi-source heterogeneous raw data and business process interaction data of the supply chain to construct the data modality perception unit of the supply chain.

[0021] This invention, through acquiring multi-source heterogeneous raw data and business process interaction data from the supply chain, constructs a data modality perception unit for the supply chain. This unit can transform the modal characteristics and business value associations of multi-source heterogeneous data in the supply chain into analyzable and usable structured perception results, thereby improving the accuracy of supply chain collaborative management.

[0022] The supply chain refers to a complex network formed by multiple participants (such as suppliers, manufacturers, distributors, retailers, logistics service providers, financial institutions, etc.) through business interactions, aiming to realize the entire process of a product or service from its origin to the end user. The multi-source heterogeneous raw data refers to raw, unprocessed data of varying formats and types directly collected from various links in the supply chain, including raw data generated in the procurement, production, warehousing, logistics, and sales stages, such as raw material purchase orders, production line equipment operating parameters (temperature / speed), inventory ledgers, and goods in... Temperature and humidity during transit, etc.; the business process interaction data refers to the data generated when business transactions and interactions occur between upstream and downstream entities in the supply chain, including order interaction data, handover verification data, fund interaction data, abnormal collaboration data, etc., such as the goods handover signature records between warehouses and logistics providers, goods receipt vouchers between logistics providers and distributors, and quality inspection result confirmation information fed back to enterprises by quality inspection agencies; the data modality perception unit refers to a functional module composed of software algorithms, models, and computational logic that automatically identifies, parses, classifies, and extracts features from multi-source heterogeneous raw data and business process interaction data in the supply chain.

[0023] As an embodiment of the present invention, the step of acquiring multi-source heterogeneous raw data and business process interaction data of the supply chain to construct the data modality perception unit of the supply chain includes: Extract the data modal features and business source attributes of the multi-source heterogeneous raw data; Using the data modality features and the business source attributes, determine the set of structured data parsing strategies for the multi-source heterogeneous raw data; Based on the interaction data of the business processes, identify the business process nodes and data interaction relationships of the supply chain; The business logic rules and data permission levels for parsing the interactive data of the aforementioned business processes; Based on the business logic rules and the data permission levels, a value density evaluation standard for the multi-source heterogeneous raw data is set. By combining the structured data parsing strategy set, the business process nodes, and the value density assessment criteria, a data modality mapping model for the supply chain is constructed. Based on the data interaction relationship and the data modality mapping model, determine the type of multimodal fusion algorithm corresponding to the supply chain; By combining the structured data parsing strategy set, the value density assessment standard, and the multimodal fusion algorithm type, the data modality perception unit of the supply chain is constructed.

[0024] The data modality features refer to the inherent attribute characteristics of multi-source heterogeneous raw data in terms of data type, format, structure, and expression form. For example, structured tabular data of purchase orders containing order numbers, amounts, and dates; and time-series GPS trajectory data of logistics transportation containing longitude, latitude, and timestamp sequences. The business source attributes refer to the attribute information of the specific business processes, responsible entities, and data generation scenarios from which the multi-source heterogeneous raw data originates. The structured data parsing strategy set refers to a set of rules and methods for converting unstructured / semi-structured data with different data modality features and business source attributes into a unified structured format. For example, using OCR to process unstructured PDF quality inspection reports. The system employs character recognition and key information extraction strategies to extract fields such as "product batch number" and "quality inspection results" and convert them into tabular data. For semi-structured logistics handover document images, an image segmentation and text localization strategy is used to extract structured information such as "handover time" and "recipient." The business process nodes refer to key links or operational steps with clear functional boundaries and specific business objectives throughout the entire supply chain. For example, the business process nodes of a home appliance supply chain include: "Raw Material Inbound Inspection" node (function: verifying raw material compliance), "Production Planning and Scheduling" node (function: allocating production resources), "Inter-warehouse Transfer" node (function: balancing regional inventory), and "Terminal Delivery Receipt" node (function: ...). Capable of: completing goods delivery); the data interaction relationship refers to the data transmission direction, dependency relationship, and association rules between different business process nodes or between different entities within the same node. For example, outbound orders and waybills between warehouse management nodes and logistics transportation nodes need to be bound one-to-one and associated through order numbers; the business logic rules refer to the constraints or operating guidelines that regulate data generation, flow, and processing in the supply chain business process to ensure that data interaction meets actual business needs; the data permission level refers to the hierarchical definition of data access, modification, and transmission permissions based on data sensitivity, business confidentiality requirements, and entity roles; the value density assessment standard refers to the standard used to measure the value of data to the supply chain. A quantitative indicator system for the usefulness of collaborative management and decision optimization can quantify data value from aspects such as data real-time performance (e.g., GPS trajectory data latency < 5 minutes is high value), data integrity (e.g., the completeness rate of key fields in quality inspection reports > 95% is high value), and data correlation (e.g., logistics data bound to order data is more valuable than isolated logistics data). The data modality mapping model refers to establishing a correspondence model between data modality characteristics, business source attributes, and business process nodes. The multimodal fusion algorithm type refers to the classification of algorithms that integrate data from different modalities (structured / unstructured / time-series, etc.) and different sources according to data interaction relationships to form a unified data view.

[0025] Optionally, the data interaction relationships in the supply chain can be identified using a directed graph path analysis algorithm; the business logic rules of the business process interaction data can be parsed using a decision tree logic extraction algorithm; the data modality mapping model of the supply chain can be constructed using a feature mapping algorithm, such as an attention mechanism model based on deep learning; and the type of multimodal fusion algorithm corresponding to the supply chain can be determined based on the Transformer cross-modal attention fusion algorithm.

[0026] S2. Based on the data modality perception unit, establish a differentiated evidence storage rule base for the supply chain, and combine the differentiated evidence storage rule base with the multi-source heterogeneous original data to generate a data classification evidence storage list for the supply chain.

[0027] This invention establishes a differentiated evidence storage rule base for the supply chain based on the data modality perception unit. This can accurately match the credible evidence storage needs of different modal data in the supply chain. By leveraging the hierarchical management capabilities of the rule base, it adapts to the storage characteristics and verification logic of multimodal blockchains, providing a standardized operational framework for the subsequent credible on-chain data and collaborative interaction of the supply chain. This ensures the authenticity and relevance of on-chain data from the data entry point, guaranteeing the efficiency and reliability of collaborative management of credible data in the supply chain.

[0028] The differentiated evidence preservation rule base refers to a set of personalized and hierarchical rules constructed based on the characteristics of multi-source supply chain data (such as data type, business attributes, value density, interaction frequency, etc.) parsed by the data modality perception unit, targeting data of different modalities, different business scenarios, and different trust requirements. It includes core elements such as evidence preservation methods, storage strategies, verification logic, update mechanisms, and access control.

[0029] As an embodiment of the present invention, the step of establishing a differentiated evidence storage rule base for the supply chain based on the data modality sensing unit includes: The system acquires raw data from the entire supply chain and outputs the modality recognition results and data attribute information of the raw data through the data modality perception unit. The modal feature set of the original data of the entire link is extracted from the modal recognition result, wherein the modal feature set includes data format type, data generation frequency and data flow continuity characteristics; Based on the data attribute information, the business attribute set of the original data of the entire chain is extracted, wherein the business attribute set includes business link identifier, data sensitivity level and compliance evidence storage period; Based on the modal feature set, a selection strategy for evidence preservation technology corresponding to the supply chain is generated; Based on the aforementioned business attribute set, define the evidence preservation compliance constraint rules corresponding to the supply chain; By integrating the evidence preservation technology selection strategy and the evidence preservation compliance constraint rules, a differentiated evidence preservation strategy vector for the supply chain is output. The differentiated evidence preservation strategy vector is encapsulated into executable evidence preservation rules for the supply chain; Integrate the executable evidence storage rules to establish a differentiated evidence storage rule library for the supply chain.

[0030] The term "end-to-end raw data" refers to the initial data set directly generated by participants, equipment, or systems in all business stages of the supply chain from start to finish. The "modal recognition result" refers to the classification conclusion regarding the data modality type output by the data modality perception unit after automatic analysis of the end-to-end raw data; for example, identifying "scanned purchase order" as an image modality, "production log" as a text modality, and "equipment vibration data" as a time-series modality. The "data attribute information" refers to metadata describing the inherent characteristics of the end-to-end raw data, including data source, generation time, data size, format version, etc. The "modal features" refer to... The modal feature set refers to the set of data physical features extracted from modal recognition results and data content to evaluate the suitability of evidence preservation technologies, including technical features such as data format type, generation frequency, and data flow continuity. The business attribute set refers to the set of business features extracted from data attribute information to evaluate the compliance and importance of evidence preservation, including business process identifiers, data sensitivity levels, and compliance evidence preservation periods. The evidence preservation technology selection strategy refers to the recommended scheme regarding which evidence preservation technology and its parameters to adopt after querying the evidence preservation technology cost library based on the modal feature set. For example, for "high-frequency continuous sensor data," by querying the evidence preservation technology... After reviewing the cost database, the selection strategy is "using cloud storage + scheduled blockchain uploading (daily summary hashing)" to balance real-time performance and cost. For "low-frequency discrete contract documents," after consulting the cost database of evidence storage technologies, the selection strategy is "directly using blockchain for evidence storage." The evidence storage compliance constraints refer to the mandatory evidence storage requirements that must be followed for specific data, derived from matching the business attribute set with the supply chain business rule database. These requirements include constraints on evidence storage format, retention period, and access permissions. For example, for "highly sensitive customer data," the constraint rule is "the evidence storage format must be encrypted (AES-256), and the retention period..." "≥3 years, access only by authorized personnel"; The differentiated evidence storage strategy vector refers to a mathematical vector representing the final evidence storage scheme, output by integrating technical selection strategies and compliance constraint rules through a multi-objective optimization algorithm. Each dimension of the vector represents a configurable technical parameter; The executable evidence storage rule refers to a specific instruction encapsulated by the differentiated evidence storage strategy vector that can be directly parsed and executed by blockchain evidence storage nodes or smart contracts. It can convert the differentiated evidence storage strategy vector into a string in a standard data exchange format (such as JSON, XML, Protocol Buffers) to facilitate network transmission and node parsing.

[0031] Optionally, the evidence preservation technology selection strategy corresponding to the supply chain can adopt a weighted decision algorithm, based on the mapping relationship between modal features and the cost of a preset evidence preservation technology cost library (such as hash table matching), and combined with the AHP hierarchical analysis method to calculate the optimal technology combination; the evidence preservation compliance constraint rules corresponding to the supply chain can be defined by a production rule engine (IF-THEN form); the differentiated evidence preservation strategy vector of the supply chain can be generated by a feature weighted fusion algorithm, which weights and sums the technology selection vector and the compliance constraint vector.

[0032] Furthermore, by combining the differentiated evidence preservation rule base and the multi-source heterogeneous raw data, this embodiment of the invention generates a data classification evidence preservation list for the supply chain. This upgrades fragmented data governance into a systematic evidence preservation framework, accurately mining the modal characteristics and business value of multi-source heterogeneous raw data (such as structured transaction forms, unstructured production videos, etc.), shifting from disordered data accumulation to ordered value extraction, and strengthening the full lifecycle management of supply chain data assets. The data classification evidence preservation list refers to a structured list that systematically classifies multi-source heterogeneous data across the entire supply chain based on differentiated evidence preservation strategies such as evidence preservation technology selection strategies and compliance constraint rules defined in the differentiated evidence preservation rule base, combined with the modal characteristics and business attributes of multi-source heterogeneous raw data. It also clarifies key information such as the evidence preservation method, evidence preservation period, storage medium, encryption requirements, access permissions, and compliance verification standards for each type of data.

[0033] As an embodiment of the present invention, the step of generating a data classification and evidence storage list for the supply chain by combining the differentiated evidence storage rule base and the multi-source heterogeneous data features of the multi-source heterogeneous original data includes: The multi-source heterogeneous raw data is preprocessed to obtain a standardized dataset, wherein the preprocessing includes data cleaning, format standardization and metadata extraction; Identify the categorical data items in the standardized dataset and extract the data feature indicators corresponding to the categorical data items; Based on the differentiated evidence preservation rule base, calculate the matching weight between each data feature indicator and each evidence preservation rule; Based on the matching weights, the evidence preservation priority and evidence preservation timeliness level of the categorized data items are determined; Construct a business logic association model between the categorized data items to identify the data dependencies between them; By combining the evidence preservation priority, the evidence preservation timeliness level, and the data dependency relationship, an initial evidence preservation priority sequence for the supply chain is generated; Obtain the evidence storage node cluster corresponding to the supply chain, and collect the real-time evidence storage environment parameters of the evidence storage node cluster. Based on the real-time evidence storage environment parameters, dynamic priority optimization processing of the initial evidence storage priority sequence is performed to generate the data classification evidence storage list of the supply chain.

[0034] The standardized dataset refers to the dataset obtained after performing data cleaning, format standardization, and metadata extraction preprocessing on multi-source heterogeneous raw data from the supply chain; the evidence preservation rules refer to the instructions in the differentiated evidence preservation rule library, formulated based on supply chain industry regulatory requirements, enterprise business needs, and data attributes, used to regulate data evidence preservation behavior; the matching weight refers to the quantitative value of the degree of fit between the data feature indicators of the categorized data items and the applicable conditions of the evidence preservation rules; the evidence preservation priority refers to the urgency of data evidence preservation operations and the priority of resource allocation determined based on the matching weights of the categorized data items and each evidence preservation rule, combined with the degree of impact of the data on supply chain business. Level; the evidence retention timeliness level refers to the time level at which classified data items need to be retained in the evidence carrier based on the data retention period requirements and data lifecycle in the evidence retention rules (e.g., the lifecycle of temporary purchase orders is 1 year, and the lifecycle of core product design data is 10 years); the business logic association model refers to a mathematical model or graph structure that describes how different data items are related, referenced, or dependent on each other at the business level; the data dependency relationship refers to the relationship in the business logic association model where the integrity or validity of a data item depends on the existence or state of one or more other data items; the initial evidence retention priority sequence refers to the sequence of data items stored without considering the data retention period requirements and data lifecycle (e.g., the lifecycle of temporary purchase orders is 1 year, and the lifecycle of core product design data is 10 years); the business logic association model refers to the relationship ... Considering real-time resource constraints and costs, a preliminary list of data item storage order is generated solely based on the data's characteristics, rule matching degree, and business relevance through a sorting algorithm. The storage node cluster refers to a hardware / software cluster composed of various types of storage nodes (such as blockchain nodes, distributed storage nodes, edge computing nodes, and private cloud nodes) for distributed storage deployment of supply chain data. Examples include a blockchain storage cluster consisting of six servers across three regions, or a distributed storage cluster based on IPFS. The real-time storage environment parameters refer to parameters collected in real-time from the storage node cluster that reflect its current operation. The set of performance metrics for status and resource availability includes network transmission bandwidth, available capacity of distributed storage nodes, current computing resource load rate, and unit evidence storage operation cost coefficient. The dynamic priority optimization process refers to an optimization process that reorders and adjusts the initial evidence storage priority sequence based on real-time evidence storage environment parameters and cost constraints. For example, data item A has the highest priority in the initial sequence, but its evidence storage requires a lot of bandwidth. Since the current network bandwidth is very tight, the optimization algorithm may temporarily move a data item B with a slightly lower priority but smaller data volume and lower bandwidth requirement to the front for priority processing to avoid network congestion and improve overall throughput.

[0035] Optionally, the matching weights of each data feature index and each evidence preservation rule can be calculated using the cosine similarity formula; the business logic association model between the classified data items can be constructed using a graph database (such as Neo4j); the data dependencies between the classified data items can be identified using a topological sorting algorithm; and the dynamic priority optimization processing of the initial evidence preservation priority sequence can be implemented using a genetic algorithm (GA).

[0036] S3. Identify the business scenario requirements and network operation status of the supply chain from the interaction data of the business process, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status.

[0037] This invention, by identifying the business scenario requirements and network operation status of the supply chain from the interaction data of the business links, allows the modal analysis and business value mining of multi-source heterogeneous raw data to dynamically adjust the analysis strategy according to the business scenario requirements and network operation status, thereby optimizing the identification accuracy of data modalities and the correlation depth of business value in real time, and improving the parsing adaptability and value conversion efficiency of complex supply chain data.

[0038] Specifically, the business scenario requirements and network operation status of the supply chain are identified from the interaction data of the business links. The business scenario requirements refer to the functional, performance, and compliance requirements of data storage in different business scenarios of the supply chain, identified from the interaction data of the supply chain business links (such as purchase order confirmation, cross-border customs declaration collaboration, supplier reconciliation and settlement, logistics node signing, etc.). The network operation status refers to the set of real-time performance and health indicators of the underlying communication network supporting the operation of the supply chain business. Its core is to provide "real-time resource constraint basis" for the dynamic adjustment of the consensus mechanism (such as consensus parameter optimization and consensus node switching), ensuring that the consensus mechanism operates stably and efficiently under network resource constraints.

[0039] Optionally, business contract, SLA (Service Level Agreement), and compliance documents can be analyzed using Natural Language Processing (NLP) technology to identify the business scenario requirements corresponding to the supply chain; the network operation status corresponding to the supply chain can be identified by actively polling or passively collecting status information of network devices (switches, routers) using network monitoring protocols (such as SNMP, NetFlow).

[0040] Furthermore, this embodiment of the invention defines consensus mechanism adaptation standards for the supply chain under different business scenarios based on the business scenario requirements and the network operating status. Adapted consensus mechanisms can enhance the accuracy of data modality recognition and the depth of business value association, improve the credibility and application efficiency of data parsing results, and enable synergy between multi-source heterogeneous raw data in modality parsing integrity and business value transformation efficiency. This comprehensively enhances the dynamic parsing and value mining capabilities of multi-source heterogeneous raw data in the supply chain. The consensus mechanism adaptation standards refer to a standardized criterion system formulated based on the different business scenario requirements of the supply chain (such as the high security requirements of cross-border trade and the high timeliness requirements of emergency replenishment) and real-time network operating status (such as consensus node communication latency and computing resource load rate) for selecting the appropriate consensus mechanism type, configuring consensus parameters, and clarifying consensus execution boundaries.

[0041] As an embodiment of the present invention, defining the consensus mechanism adaptation standard of the supply chain under different business scenarios according to the business scenario requirements and the network operating status includes: Based on the business scenario requirements, the core consensus requirements of the supply chain under different business scenarios are extracted. Based on the business scenario requirements and the core consensus requirements, a candidate set of consensus mechanism types for the supply chain under different business scenarios is selected. Construct a dynamic adaptation matrix between the candidate set of consensus mechanism types and the network operating state, and calculate the consensus mechanism matching degree of the supply chain; The types of consensus nodes and network interfaces participating in the supply chain are clearly defined in order to determine the node admission conditions and communication protocol specifications of the candidate set of consensus mechanism types; Based on the dynamic adaptation matrix, the node admission conditions, and the communication protocol specifications, consensus parameter configuration rules for the supply chain under different business scenarios are generated. Based on the consensus parameter configuration rules, the consensus performance boundaries of the supply chain under different business scenarios are defined. The consensus mechanism matching degree, consensus parameter configuration rules, and consensus performance boundaries define the consensus mechanism adaptation standard for the supply chain in different business scenarios.

[0042] The core consensus requirements refer to the core functional needs and performance expectations of supply chain participants for the consensus mechanism in specific business scenarios. These requirements form the fundamental basis for selecting and designing consensus mechanisms. For example, in cross-border trade settlement scenarios, the core consensus requirements are "immutability" (ensuring transaction records cannot be tampered with), "high consistency" (consistent understanding of the transaction status among participating institutions), and "traceability" (supporting full-chain auditing of transactions). The candidate set of consensus mechanism types refers to a set of highly adaptable consensus mechanisms selected from existing mature consensus algorithms based on business scenario needs and core consensus requirements. For example, in fresh food supply chain traceability scenarios, the core consensus requirements are "high throughput" and "low energy consumption." The candidate set may include Practical Byzantine Fault Tolerance (PBFT) (suitable for high-consistency scenarios with small-scale nodes) and Delegated Proofof... Stake (DPoS)** (suitable for high-throughput scenarios); the dynamic adaptation matrix refers to a two-dimensional matrix constructed with network operating state parameters (such as the number of nodes, bandwidth, and latency) as rows and the consensus mechanism type candidate set as columns, used to quantify the adaptation degree of each candidate mechanism under different network conditions; the consensus mechanism matching degree refers to the comprehensive adaptation score of the candidate consensus mechanism and the current business scenario requirements and network state, calculated by weighting based on the dynamic adaptation matrix, used for the final decision on the consensus mechanism type; the participating consensus node type refers to the classification of node roles in the consensus process in the supply chain, based on the node's function, permissions, and trust level, determining its responsibilities in the consensus (such as proposal, verification, and accounting); the network interface refers to the standardized interface for data interaction between participating consensus nodes and between nodes and external systems (such as ERP and IoT platforms). This includes hardware interfaces (such as communication ports) and software interfaces (such as APIs); the node admission criteria refer to the qualification requirements set for nodes to join the consensus network to ensure the security and effectiveness of the consensus process, including constraints such as identity authentication, hardware performance, and credit rating; the communication protocol specification refers to a standardized protocol that specifies the format, timing, encryption method, and error handling rules for data transmission between nodes participating in the consensus, ensuring the consistency and security of information interaction; the consensus parameter configuration rules refer to the dynamic adjustment rules for the consensus mechanism's operating parameters (such as block size, consensus rounds, and fault tolerance thresholds) based on the dynamic adaptation matrix, node admission criteria, and communication protocol specifications; the consensus performance boundary refers to the performance indicator threshold range within which the consensus mechanism can operate stably under specific business scenarios. If this range is exceeded, the mechanism fails or needs to be switched. Indicators include throughput, latency, and fault tolerance rate.

[0043] Optionally, the core consensus requirements of the supply chain under different business scenarios can be extracted using the Kano model. For example, the business scenario requirements can be divided into basic requirements (such as anti-tampering), expected requirements (such as low latency), and attractive requirements (such as automatic execution of smart contracts) using a demand classification algorithm. The matching degree of the consensus mechanism of the supply chain can be calculated using TOPSIS (Approximation of Ideal Solution Ranking). The types of participating consensus nodes in the supply chain can be determined using a role mining algorithm (such as a cluster-based node classification algorithm). The node admission conditions of the candidate set of consensus mechanism types can be determined using a multi-attribute decision algorithm (such as the ELECTRE method).

[0044] S4. Based on the consensus mechanism adaptation standard, determine the consensus mechanism decision factor of the supply chain, and configure the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor.

[0045] This invention, through its embodiment, determines the consensus mechanism decision factors of the supply chain based on the aforementioned consensus mechanism adaptation standard. This enhances the accuracy and reliability of trusted data collaborative management in the supply chain and provides a core basis for subsequent optimization of multimodal data fusion rules and the formulation of trusted data flow strategies for the supply chain. It also strengthens the consensus mechanism's effective control over the collaborative process of multi-source data in the supply chain. The consensus mechanism decision factors refer to the set of core parameters used to quantify the operating logic, constraints, and dynamic adjustment rules of the decision consensus mechanism. Its core is to transform the abstract requirements in the adaptation standard (such as business scenario demands and network operation status adaptation rules) into calculable and executable specific indicators.

[0046] As an embodiment of the present invention, determining the consensus mechanism decision factor of the supply chain based on the consensus mechanism adaptation standard includes: Real-time monitoring of the dynamic characteristics of node load and network communication quality in the supply chain; Calculate the load change rate corresponding to the node load dynamic characteristics and the network latency index corresponding to the network communication quality; Based on the load change rate and the network latency index, identify the real-time operating status mode of the supply chain; The candidate consensus mechanism type corresponding to the supply chain is determined by the state-mechanism mapping relationship between the real-time running state mode and the consensus mechanism adaptation standard. Based on the candidate consensus mechanism type, the transaction processing latency, data consistency index and system fault tolerance parameters of the supply chain in the real-time operating state mode are collected in real time. Based on the transaction processing latency, the data consistency index, and the system fault tolerance parameters, analyze the performance patterns of each candidate consensus mechanism type in the real-time operating state mode. Construct a matching degree evaluation matrix between the performance pattern and the consensus mechanism adaptation standard; Based on the matching degree evaluation matrix, the consensus mechanism decision factors of the supply chain are determined.

[0047] The node load dynamic characteristics refer to the resource occupancy status and changing trend of each participating consensus node in the supply chain multimodal blockchain network within a unit of time, including quantifiable indicators such as CPU utilization, memory usage, data storage increment, and consensus task processing volume. The network communication quality refers to the set of stability, effectiveness, and security indicators for data transmission between nodes in the supply chain multimodal blockchain network, mainly including data transmission rate, packet loss rate, transmission link stability, and data encryption compliance. The load change rate refers to the magnitude of change in node resource occupancy within a unit of time based on the node load dynamic characteristics, calculated using the formula: (current load value - previous load value). The network latency metric is calculated as: (1) / previous load value × 100%; The network latency metric refers to the total time elapsed in a supply chain multimodal blockchain network from the time data is initiated by the sending node to the time it is fully received and confirmed by the receiving node. This includes transmission latency (the time data travels through the link), processing latency (the time a node takes to parse and process the data after receiving it), and queuing latency (the time data waits for processing in the node's buffer). It is a core parameter for measuring network communication efficiency. For example, in a pharmaceutical supply chain, if a drug production node sends "drug batch traceability data" to a regulatory node, with a transmission latency of 20ms, a processing latency of 15ms, and a queuing latency of 10ms, the network latency metric is 45ms. If the latency exceeds 50ms, it may lead to… Consensus timeout occurred; the real-time operating state mode refers to the classification definition of the current operating state of the supply chain multimodal blockchain network based on node load change rate and network latency indicators. Each mode corresponds to a set of clear indicator threshold ranges, reflecting the overall load level and communication efficiency of the network; the state-mechanism mapping relationship refers to the set of corresponding rules between the "supply chain real-time operating state mode" and the "adapted consensus mechanism type" established based on the consensus mechanism adaptation standard. This can clarify the preferred consensus mechanism under different operating states. For example, according to the rule in the adaptation standard that "high throughput scenarios prioritize DPoS, and high consistency scenarios prioritize PBFT", the established mapping relationship is: "low load - low..." The consensus mechanism types are categorized as follows: "High-load-low-latency mode → PBFT / DPoS", "High-load-low-latency mode → DPoS", and "Low-load-high-latency mode → PBFT". When the network is in "high-load-low-latency mode", it directly maps to the DPoS mechanism. The candidate consensus mechanism type refers to the set of consensus mechanisms that are adapted to the current network state and supply chain business needs, selected through state-mechanism mapping based on the real-time operating state mode. For example, if an automotive parts supply chain is in "low-load-low-latency mode" and its core business requirement is "high consistency", PBFT (Practical Byzantine Fault Tolerance) and PoS (Proof of Stake) are selected through state-mechanism mapping, thus constituting the candidate consensus mechanism types.The transaction processing latency refers to the total time taken in the supply chain multimodal blockchain network from the submission of a consensus request by the initiating node to the achievement of consensus by all nodes in the network and the completion of the transaction on the chain. This includes transaction verification latency, consensus voting latency, and block generation latency, and is a key indicator for measuring the processing efficiency of the consensus mechanism. The data consistency indicator refers to the degree of consistency in the understanding of the same transaction data among different nodes in the supply chain multimodal blockchain network. Quantitative indicators include "node data synchronization rate" (number of nodes that have synchronized transaction data / total number of nodes × 100%) and "data tampering detection rate" (amount of tampered data detected / total amount of tampered data × 100%), which is a core standard for ensuring reliable data collaboration. The system fault tolerance parameter refers to the ability of the supply chain multimodal blockchain network to maintain normal operation and ensure data consistency even in the event of node failures, malicious attacks, data errors, or other anomalies. This mainly includes "fault tolerance..." The performance indicators are: "Node Ratio" (number of fault-tolerant nodes / total number of nodes × 100%), "Attack Resistance Rate" (number of attacks successfully resisted / total number of attacks × 100%), and "Error Recovery Time" (time taken from the occurrence of an anomaly to the system returning to normal). The performance pattern refers to the inherent correlation characteristics of the performance indicators of each candidate mechanism with changes in the network environment, extracted through multiple sets of experimental data collection and statistical analysis of transaction processing latency, data consistency indicators, and system fault tolerance parameters of candidate mechanisms under specific real-time operating conditions, as well as the comparative patterns of performance advantages and disadvantages among different candidate mechanisms. The matching degree evaluation matrix is ​​a two-dimensional numerical matrix constructed with "candidate consensus mechanism type" as the row and "core requirements of consensus mechanism adaptation standards" as the column, quantifying the degree of fit between "performance pattern" and "adaptation standards." The matrix element values ​​represent the degree of compliance or matching score of a specific consensus mechanism on a specific performance indicator.

[0048] Optionally, the state-mechanism mapping relationship between the real-time running state mode and the consensus mechanism adaptation standard can be generated by training historical "state-mechanism-performance" data using the C4.5 decision tree algorithm, and combined with the forward inference engine algorithm to achieve fast matching of real-time state to mechanism; the performance characteristics of the candidate consensus mechanism type in the real-time running state mode can be analyzed using linear / exponential regression algorithms.

[0049] Furthermore, by configuring the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factors, this embodiment of the invention enables the consensus strategy scheduler to accurately match the real-time business needs and network operation status of the supply chain's trusted data collaborative management, thereby improving the targeting of consensus strategy scheduling and the security and efficiency of data collaboration. At the same time, it provides core data support for the dynamic iteration and optimization of the supply chain consensus strategy. The consensus strategy scheduler refers to the core control module established in the supply chain multimodal blockchain network based on the consensus mechanism decision factors, which is responsible for dynamically selecting, switching, and executing consensus mechanism strategies.

[0050] As an embodiment of the present invention, configuring the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor includes: By using the dynamic evaluation data of the consensus mechanism decision factors, the strategy configuration node corresponding to the supply chain is located; Key performance indicators are extracted from the dynamic evaluation data, and the correlation weight between the key performance indicators and the strategy configuration node is calculated. Based on the association weights, the core configuration parameters of the strategy configuration node are selected; Identify the parameter adjustment path of the core configuration parameters in the policy configuration node; Based on the core configuration parameters, calculate the parameter sensitivity of the key performance indicators; Based on the parameter sensitivity, the adjustment priority of the core configuration parameters is set; Obtain a preset consensus strategy configuration scheme that matches the adjustment priority; Analyze the parameter adjustment boundary conditions of the core configuration parameters in the preset consensus strategy configuration scheme; By integrating the parameter adjustment boundary conditions and the parameter adjustment path, a two-level policy generation network for the consensus policy scheduler is constructed. The consensus policy scheduler for the supply chain is configured through the two-level policy generation network.

[0051] The dynamic evaluation data refers to a multi-dimensional time-series data set collected and updated in real time during system operation to evaluate the effectiveness of consensus mechanism decision factors. This includes, but is not limited to, time-varying sequence data such as real-time transaction throughput (TPS), transaction confirmation latency, block propagation time, node voting consensus rate, and network resource consumption (CPU, memory, bandwidth) under different consensus mechanisms. The strategy configuration node refers to a logical unit or physical component in the consensus strategy scheduler that performs specific configuration functions. The key performance indicators (KPIs) are core metrics used to quantitatively evaluate the performance of the consensus strategy scheduler, including but not limited to: transaction processing latency (≤500ms), data synchronization rate (≥98%), fault-tolerant node ratio (≥20%), and consensus success rate. The correlation weight is a quantitative value that measures the degree of influence or importance of a key performance indicator on a strategy configuration node. The core configuration parameters are those parameters with the greatest impact on key performance indicators, selected from all configurable parameters of the strategy configuration node based on correlation weights. These include algorithm-level parameters such as batch size and checkpoint interval, and network-level parameters such as heartbeat timeout. Timeout, message retransmission interval, and view-changetimeout; resource-level parameters: single block size limit and gas price cap; the parameter adjustment path refers to the operation process and data interaction link followed by adjusting the core configuration parameters from their current value to the target value in the strategy configuration node, including the generation, transmission, verification, execution, and feedback of parameter instructions; the parameter sensitivity refers to the degree of impact of changes in the value of core configuration parameters on key performance indicators; the adjustment priority refers to the hierarchical classification of the adjustment order of core configuration parameters based on parameter sensitivity and the business importance of key performance indicators (e.g., level 1 is the highest, level 3 is the lowest). The minimum threshold is set to ensure that parameters with high impact and importance are adjusted first. The preset consensus strategy configuration scheme refers to a consensus strategy configuration template pre-defined based on consensus mechanism adaptation standards and historical configuration data, matching different adjustment priorities. It includes recommended values ​​for core configuration parameters, adjustment steps, and applicable scenarios, and can be quickly invoked to shorten the configuration cycle. The parameter adjustment boundary conditions refer to the safe and effective value range set for the core configuration parameters. For example, the boundary conditions for "block generation time" parsed from the preset scheme are: "Value range 3-10 seconds, single adjustment amplitude ≤ 2 seconds, must meet 'block generation time × verification count ≤ 30 seconds·times' (to avoid excessive resource consumption)."The aforementioned two-tiered policy generation network refers to an intelligent decision-making network architecture comprising a macro-level policy generation layer and a micro-level parameter optimization layer. The macro-level layer selects or generates high-level policy directions from preset schemes based on decision factors, while the micro-level layer is responsible for fine-tuning core configuration parameters within the given policy direction to adapt them to real-time network conditions.

[0052] Optionally, the strategy configuration nodes corresponding to the supply chain can be located using a node importance assessment algorithm (such as the PageRank algorithm); key performance indicators can be extracted from the dynamic assessment data using a feature selection algorithm (such as the Relief-F algorithm); the association weight between the key performance indicators and the strategy configuration nodes can be calculated using a grey relational analysis algorithm; and the parameter sensitivity of the key performance indicators can be calculated using a sensitivity analysis algorithm (such as the Morris screening method).

[0053] S5. Monitor the data storage integrity index and consensus mechanism response efficiency value of the supply chain in real time to calculate the global performance status index of the supply chain. Based on the global performance status index, generate the data classification storage list and the collaborative optimization instructions of the consensus strategy scheduler.

[0054] This invention, through real-time monitoring of the data evidence integrity index and consensus mechanism response efficiency value of the supply chain, calculates the global performance status index of the supply chain. This enables the trusted data of the supply chain to accurately reflect the entire chain's operational status based on the trustworthiness of the supply chain data evidence and the dynamic changes in the consensus response, thereby improving the global control and targeting of the supply chain data collaboration process and constructing a dynamic performance monitoring and optimization mechanism for trusted data collaborative management of the supply chain.

[0055] The data evidence integrity index is a composite indicator that quantitatively assesses whether supply chain data is recorded completely, accurately, and tamper-proofly during the evidence preservation process. Its calculation formula is: Data Evidence Integrity Index = Data upload success rate Data validation pass rate γ The data storage conflict rate, where α, β, and γ are weighting coefficients; the consensus mechanism response efficiency value is a comprehensive performance indicator that quantitatively evaluates the speed and resource consumption of the supply chain consensus network in processing transactions and reaching consensus. Its calculation formula is: response efficiency value = (transaction throughput * weight 1) / (average consensus latency * weight 2) - (unit transaction resource cost * weight 3); the global performance status index is a structured data set that integrates multiple key performance indicators such as the data storage integrity index and the consensus mechanism response efficiency value according to a specific dimension, and can comprehensively characterize the current operating health of the supply chain blockchain system.

[0056] As an embodiment of the present invention, the real-time monitoring of the data storage integrity index and consensus mechanism response efficiency value of the supply chain to calculate the global performance status index of the supply chain includes: Identify the distributed nodes of the supply chain and obtain the consistency verification results of the ledger data of the distributed nodes; Calculate the data reliability index of the supply chain based on the data storage integrity index and the ledger data consistency verification results; Based on the consensus mechanism response efficiency value, the throughput variation coefficient and block propagation latency parameters of the supply chain are monitored in real time. The consensus stability coefficient of the supply chain is calculated based on the throughput variation coefficient and the block propagation delay parameter. Real-time collection of computing resource utilization, storage resource utilization, and network bandwidth utilization of the supply chain; By integrating the data reliability index, the consensus stability coefficient, the computing resource utilization rate, the storage resource utilization rate, and the network bandwidth utilization rate, the global performance status index of the supply chain is calculated.

[0057] The distributed node refers to a computer node in the supply chain blockchain network that independently runs the blockchain protocol, stores complete or partial copies of ledger data, and participates in network communication and consensus processes. The ledger data consistency verification result refers to a binary or probabilistic judgment result regarding the consistency of data between nodes after comparing ledger data stored on different nodes using a specific algorithm. The data reliability index is a quantitative indicator reflecting the credibility of multimodal data (such as IoT device data and transaction voucher data) in the supply chain, calculated by combining the data storage integrity index and the ledger data consistency verification result. Its value ranges from 0 to 100 points; a higher score indicates more reliable data. The calculation logic is typically as follows: Data Reliability Index =α × Data Integrity Index + β × Ledger Data Consistency Pass Rate (α and β are weights, set according to business scenarios, α + β = 1); The throughput variation coefficient refers to the fluctuation of the number of transactions processed by the supply chain multimodal blockchain network per unit time (e.g., 1 minute) based on the consensus mechanism response efficiency value, and the formula is "throughput variation coefficient = throughput standard deviation / throughput average × 100%"; The block propagation delay parameter refers to the average time taken for a newly generated block to be transmitted from the producer node (e.g., the delegate node in DPoS, the master node in PBFT) to all distributed nodes in the entire network and to complete verification in the supply chain multimodal blockchain network, including the block transmission time (data transmission in the P2P network). The consensus stability coefficient is a quantitative indicator reflecting the operational stability of the supply chain consensus mechanism, calculated by combining the throughput variation coefficient and the block propagation delay parameter. Its value ranges from 0 to 1; the closer the coefficient is to 1, the more stable the consensus mechanism. The calculation logic is typically: Consensus Stability Coefficient = γ × (1 - Throughput Variation Coefficient) + δ × (1 - Block Propagation Delay Parameter / Threshold Delay), where γ and δ are weights, γ + δ = 1, and the threshold delay is the maximum allowable propagation delay specified in the consensus mechanism adaptation standard. The computing resource occupancy rate refers to the actual computing resources occupied by distributed nodes in the supply chain during consensus processing, data storage, and business interaction. The ratio of computing resources (such as CPU, GPU) to the total computing resources of the node; the storage resource utilization rate refers to the ratio of storage resources (such as hard drives, SSDs) used by the distributed nodes in the supply chain to the total storage resources of the node for storing blockchain ledger data and multimodal business data (such as images, videos, documents), and its calculation formula is: Storage resource utilization rate = Used storage capacity / Total storage capacity × 100%; the network bandwidth utilization rate refers to the ratio of the actual network bandwidth used by the distributed nodes in the supply chain to the maximum available network bandwidth of the node when interacting with other nodes (such as block transmission, consensus voting, business data synchronization), and its calculation formula is: Network bandwidth utilization rate = Actual bandwidth used / Maximum available bandwidth × 100%.

[0058] Optionally, the block propagation delay parameter of the supply chain can be determined by an average delay algorithm. For example, a timestamp recording algorithm can be used to record the time points when the supply chain blocks are generated, transmitted to each node, and verified, and the result can be calculated by the average delay algorithm. The storage resource utilization rate of the supply chain can be calculated by using a storage capacity monitoring algorithm (such as the df command or storage management API) to collect the used capacity and total capacity data, and combining them with the percentage.

[0059] To gain a clearer understanding of the ledger data consistency verification process in distributed nodes, please refer to [link / reference]. Figure 2 The diagram illustrates an on-chain data integrity verification process in a multimodal blockchain-based supply chain trusted data collaborative management method, as provided in an embodiment of the present invention. The diagram clearly shows that during real-time monitoring of the global supply chain performance index, distributed nodes (data users and publishers in the diagram) call smart contracts to compare the real-time SHA-256 hash of cloud data with the original fingerprint stored on the chain, directly generating a ledger data consistency verification result. This verification result, combined with the data storage integrity index, can calculate a data reliability index reflecting the authenticity and credibility of the data. Simultaneously, the entire verification process is submitted and packaged into blocks as transactions. Its processing speed and propagation efficiency are directly constrained by the underlying consensus mechanism. Therefore, the efficiency of this verification process itself is an important practical source for monitoring the consensus mechanism response efficiency value, calculating the throughput variation coefficient and block propagation latency parameters, and ultimately solving for the consensus stability coefficient.

[0060] In another embodiment of the present invention, the global performance index of the supply chain is calculated using the following formula: ; in, This represents the overall performance status index. Indicates the data reliability index. This represents the consensus stability coefficient. The weighting coefficient represents the consensus stability coefficient. This indicates the utilization rate of computing resources. This represents the weighting coefficient used to calculate resource utilization. Indicates network bandwidth utilization. This represents the rated baseline value of network bandwidth corresponding to the network bandwidth utilization rate. This indicates the utilization rate of computing resources. This represents the rated baseline value of computing power corresponding to the computing resource utilization rate. Indicates storage resource utilization. This indicates the rated baseline value of storage space corresponding to the storage resource utilization rate. Indicates the system stability variance. This indicates a regulatory factor.

[0061] It should be noted that, in this application, the above formula breaks through the limitation of traditional performance evaluation focusing only on a single-dimensional indicator, and constructs a three-dimensional collaborative evaluation system of "data reliability - consensus stability - resource utilization efficiency". This formula uses the data reliability index With consensus stability coefficient The core performance of blockchain systems in terms of data trustworthiness and consensus efficiency is quantified separately, and weighted by coefficients. , Integration is the basic performance evaluation term; a geometric mean term for multi-dimensional resource utilization is introduced. It reflects the efficiency of coordinated utilization of computing, storage, and network resources, avoiding evaluation distortion caused by single resource bottlenecks; and introduces system stability variance. ,pass This is to mitigate the negative impact of system volatility and weight allocation bias on overall efficiency, thereby enabling the supply chain to shift from single-point monitoring to global collaborative governance.

[0062] Furthermore, this embodiment of the invention generates collaborative optimization instructions for the data classification and evidence storage list and the consensus strategy scheduler based on the global performance status index. This allows for real-time optimization of core elements such as the priority of data classification, the allocation ratio of evidence storage resources, and the operating parameters of the consensus strategy. It improves the accuracy of adapting to the characteristics of multimodal data in the supply chain and the network operating status, enhances the linkage between data evidence storage and the consensus strategy, and strengthens the targeted and effective collaborative management of trusted data in the supply chain. This builds a precise and dynamic technical management defense line for the trusted flow of data across the entire supply chain. The collaborative optimization instructions refer to a set of structured and executable coordination control commands automatically generated by the system based on the analysis results of the global performance status index. The core of these instructions is to synchronously adjust the operating parameters of the data evidence storage strategy and the consensus strategy scheduler, enabling them to cooperate to achieve the goal of optimal global performance of the supply chain blockchain system.

[0063] As an embodiment of the present invention, the step of generating the collaborative optimization instructions for the data classification and storage list and the consensus policy scheduler based on the global performance state index includes: Obtain the performance evaluation benchmark corresponding to the global performance status index; Based on the performance evaluation benchmark, identify the system bottleneck type corresponding to the global performance status index, and set multi-level performance thresholds corresponding to the global performance status index; Based on the multi-level efficiency thresholds, the collaborative control levels corresponding to the global efficiency state index are divided. Based on the system bottleneck type, the multi-level performance threshold, and the collaborative control level, determine the joint optimization strategy for the data classification and storage list and the consensus strategy scheduler; The parameter adjustment step size of the joint optimization strategy is calculated based on the global performance state index and the multi-level performance threshold. Based on the parameter adjustment step size, construct the parameter coupling relationship matrix between the data classification and evidence storage list and the consensus strategy scheduler; The data classification and evidence storage list and the consensus strategy scheduler are generated through the parameter coupling relationship matrix.

[0064] The performance evaluation benchmark refers to a pre-established multi-dimensional reference model used to measure the health of the global performance status index of a supply chain blockchain system. For example, for a food traceability supply chain, the performance evaluation benchmark can be defined as follows: during peak business periods, the global performance status index should be maintained above 7.0, the data reliability index should be greater than 0.95, and the consensus latency should be less than 500ms. The system bottleneck type refers to the category of the root cause that causes the global performance status index to fail to reach the evaluation benchmark. It identifies the main resources or components in the system with limited performance, including computing bottlenecks, storage I / O bottlenecks, network bandwidth bottlenecks, and consensus algorithm bottlenecks. The multi-level performance threshold refers to multiple different levels of critical values ​​set for the global performance status index, used to trigger optimization actions of different intensities and priorities. The collaborative control level refers to the level of aggression and impact of the optimization strategy defined according to the multi-level performance threshold range in which the global performance status index is located. The higher the level, the more drastic the control action and the more modules involved. The joint optimization... The optimization strategy refers to a set of optimization actions formulated to address identified system bottlenecks, requiring the coordinated execution of data storage and consensus strategy schedulers. For example, a joint optimization strategy for "network bandwidth bottlenecks" could be: "reducing the frequency of storing non-critical data (storage module action) and simultaneously switching the consensus message broadcast protocol from flooding to the more bandwidth-efficient Gossip protocol (consensus module action)." The parameter adjustment step size refers to the magnitude or increment of a single adjustment to a system parameter when executing the joint optimization strategy. For example, if the current block size is 1MB and the joint optimization strategy decides to increase the block size to improve throughput, the parameter adjustment step size can be set to 0.5MB. The first adjustment would be to 1.5MB, and if the effect is not satisfactory, the next adjustment would be to 2.0MB. The parameter coupling matrix is ​​a mathematical matrix describing the mutual influence between the parameters of the data storage module and the parameters of the consensus strategy scheduler. Its matrix elements can quantitatively represent the degree of indirect influence of a change in one parameter on another.

[0065] Optionally, the multi-level performance threshold corresponding to the global performance state index can be set by a clustering algorithm (such as K-Means); the parameter adjustment step size of the joint optimization strategy can be dynamically calculated based on the deviation between the current performance index and the benchmark using an adaptive control algorithm; the parameter coupling relationship matrix between the data classification and evidence storage list and the consensus strategy scheduler can be constructed by a system identification method, such as the least squares method or a neural network, by inputting parameter perturbations into the system and observing the output changes to fit the dynamic relationship between parameters.

[0066] S6. Combining the consensus strategy scheduler, the data classification and storage list, and the collaborative optimization instructions, output the data collaborative management scheme for the supply chain.

[0067] This invention, by combining the consensus strategy scheduler, the data classification and storage list, and the collaborative optimization instructions, outputs a data collaborative management scheme for the supply chain. This not only significantly overcomes the technical limitations of existing single hash value on-chain methods, which cannot deeply analyze multi-source heterogeneous data modalities and business values, allowing supply chain data storage to shift from "indiscriminate processing" to "value-based classification and control," but also adjusts the consensus strategy's operating logic, data storage resource allocation, and collaborative linkage rules in real time based on the dynamic changes of the global performance state vector. This effectively enhances the adaptability, accuracy, and intelligence of supply chain trusted data collaborative management based on multimodal blockchain.

[0068] The data collaboration management scheme refers to a complete and dynamically adjustable supply chain multimodal data trusted collaboration execution strategy formed through the collaborative adaptation of a consensus strategy scheduler, a data classification and evidence storage list, and collaborative optimization instructions. The data classification and evidence storage list provides the basic framework for data control, clarifying the evidence storage priorities, resource allocation rules, and security protection criteria for data of different values ​​and modalities. The consensus strategy scheduler provides the core basis for consensus mechanism regulation, defining the type selection, parameter configuration range, and node collaboration relationships of the consensus mechanism under different operating states, ensuring the efficiency and reliability of the data consensus process. The collaborative optimization instructions endow the scheme with real-time linkage adjustment capabilities, optimizing the collaborative rules of evidence storage and consensus based on the dynamic changes of the global performance state vector, ensuring that supply chain data collaboration always meets business needs and network operating status, and avoiding insufficient data credibility, low collaboration efficiency, or resource waste caused by indiscriminate data processing, rigid consensus strategies, or resource mismatch.

[0069] Compared to the problems described in the background art, the embodiments of the present invention acquire multi-source heterogeneous raw data and business process interaction data from the supply chain to construct a data modality perception unit for the supply chain. This unit can transform the modal characteristics and business value associations of multi-source heterogeneous data in the supply chain into analyzable and usable structured perception results, improving the accuracy of supply chain collaborative management. Furthermore, the embodiments of the present invention establish a differentiated evidence storage rule base for the supply chain based on the data modality perception unit. This can accurately match the credible evidence storage needs of different modal data in the supply chain. By leveraging the hierarchical control capabilities of the rule base, it adapts to the storage characteristics and verification logic of multimodal blockchains, providing a basis for subsequent supply chain management. The trusted on-chain data and collaborative interaction provide a standardized operational framework, ensuring the authenticity and relevance of on-chain data from the data entry point, and guaranteeing the efficiency and reliability of trusted data collaborative management in the supply chain. This invention, by combining the differentiated evidence storage rule base and the multi-source heterogeneous raw data, generates a data classification and evidence storage list for the supply chain. This upgrades fragmented data governance to a systematic evidence storage framework, accurately mining the modal characteristics and business value of multi-source heterogeneous raw data (such as structured transaction forms, unstructured production videos, etc.), shifting from disordered data accumulation to ordered value extraction, and strengthening the full lifecycle management of supply chain data assets. Furthermore, this invention... This embodiment defines consensus mechanism adaptation standards for the supply chain under different business scenarios based on the business scenario requirements and network operation status. Adapted consensus mechanisms can enhance the accuracy of data modality recognition and the depth of business value association, improve the credibility and application efficiency of data parsing results, and enable synergy between multi-source heterogeneous raw data in modality parsing completeness and business value transformation efficiency, comprehensively enhancing the dynamic parsing and value mining capabilities of multi-source heterogeneous raw data in the supply chain. Furthermore, this embodiment configures the consensus strategy scheduler for the supply chain based on the consensus mechanism decision factors, allowing the consensus strategy scheduler to accurately match the supply chain's trusted data collaborative management. By understanding real-time business needs and network operation status, this invention enhances the targeting of consensus strategy scheduling and the security and efficiency of data collaboration, while providing core data support for the dynamic iteration and optimization of supply chain consensus strategies. Furthermore, by monitoring the data evidence integrity index and consensus mechanism response efficiency value of the supply chain in real time, this invention calculates the global performance status index of the supply chain. This enables the trusted data of the supply chain to accurately reflect the full-link operation status based on the trustworthiness of the supply chain data evidence and the dynamic changes in the consensus response, thereby improving the global control targeting of the supply chain data collaboration process and building a dynamic performance monitoring and optimization mechanism for trusted data collaboration management of the supply chain.This invention, through generating a data classification and evidence storage list and a consensus strategy scheduler based on the global performance status index, can optimize core elements such as data classification priority, evidence storage resource allocation ratio, and consensus strategy operation parameters in real time. This improves the accuracy of adapting to the multimodal data characteristics of the supply chain and network operation status, enhances the linkage between data evidence storage and consensus strategies, and strengthens the targeted and effective collaborative management of trusted data in the supply chain. It also builds a precise and dynamic technical management defense line for the trusted flow of data across the entire supply chain. Finally, this invention, by combining the consensus strategy scheduler... The algorithm, comprising the data classification and storage list and the collaborative optimization instructions, outputs a data collaborative management scheme for the supply chain. This scheme not only significantly overcomes the technical limitations of existing single-hash-value on-chain methods in deeply analyzing multi-source heterogeneous data modalities and business value, shifting supply chain data storage from "indiscriminate processing" to "value-based classification and control," but also adjusts the consensus strategy's operational logic, data storage resource allocation, and collaborative linkage rules in real time based on the dynamic changes of the global performance state vector. This effectively enhances the adaptability, accuracy, and intelligence of supply chain trusted data collaborative management based on multi-modal blockchain. Therefore, the supply chain trusted data collaborative management method and system based on multi-modal blockchain provided in this invention can improve the credibility and collaborative management efficiency of multi-source heterogeneous data in the supply chain.

[0070] like Figure 3 The diagram shown is a functional block diagram of a supply chain trusted data collaborative management system based on multimodal blockchain according to the present invention.

[0071] The supply chain trusted data collaborative management system 200 based on multimodal blockchain described in this invention can be installed in an electronic device. Depending on the functions implemented, the supply chain trusted data collaborative management system based on multimodal blockchain may include a data modality perception module 201, a data classification and storage module 202, a consensus mechanism intelligent adaptation module 203, a consensus strategy dynamic scheduling module 204, a global performance collaborative optimization module 205, and a data collaborative management module 206. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0072] In this embodiment of the invention, the functions of each module / unit are as follows: The data modality perception module 201 is used to acquire multi-source heterogeneous raw data and business process interaction data of the supply chain in order to construct the data modality perception unit of the supply chain. The data classification and evidence storage module 202 is used to establish a differentiated evidence storage rule base for the supply chain based on the data modality perception unit, and generate a data classification and evidence storage list for the supply chain by combining the differentiated evidence storage rule base and the multi-source heterogeneous original data. The consensus mechanism intelligent adaptation module 203 is used to identify the business scenario requirements and network operation status of the supply chain from the business process interaction data, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status. The consensus strategy dynamic scheduling module 204 is used to determine the consensus mechanism decision factor of the supply chain based on the consensus mechanism adaptation standard, and configure the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor. The global performance collaborative optimization module 205 is used to monitor the data storage integrity index and consensus mechanism response efficiency value of the supply chain in real time, so as to calculate the global performance status index of the supply chain, and generate the collaborative optimization instructions of the data classification storage list and the consensus strategy scheduler based on the global performance status index. The data collaboration management module 206 is used to combine the consensus strategy scheduler, the data classification and evidence storage list, and the collaboration optimization instructions to output the data collaboration management scheme of the supply chain.

[0073] In detail, the modules in the supply chain trusted data collaborative management system 200 based on multimodal blockchain described in this embodiment of the invention adopt the same approach as described above when in use. Figure 1 The method uses the same technical means as the multimodal blockchain-based supply chain trusted data collaborative management method described in the article, and can produce the same technical effect, so it will not be elaborated here.

[0074] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0075] Finally, it should be noted that deleting any one of the above embodiments does not affect the technical solutions of other embodiments. The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A collaborative management method for trusted supply chain data based on multimodal blockchain, characterized in that, The method includes: Acquire multi-source heterogeneous raw data and business process interaction data from the supply chain to construct a data modality perception unit for the supply chain; Based on the data modality sensing unit, a differentiated evidence storage rule base for the supply chain is established. Combining the differentiated evidence storage rule base and the multi-source heterogeneous original data, a data classification evidence storage list for the supply chain is generated. Identify the business scenario requirements and network operation status of the supply chain from the interaction data of the business process, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status. Based on the consensus mechanism adaptation standard, the consensus mechanism decision factor of the supply chain is determined, and the consensus strategy scheduler of the supply chain is configured according to the consensus mechanism decision factor. The data storage integrity index and consensus mechanism response efficiency value of the supply chain are monitored in real time to calculate the global performance status index of the supply chain. Based on the global performance status index, the collaborative optimization instructions of the data classification storage list and the consensus strategy scheduler are generated. By combining the consensus strategy scheduler, the data classification and storage list, and the collaborative optimization instructions, a data collaborative management scheme for the supply chain is output.

2. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The process of combining the differentiated evidence preservation rule base and the multi-source heterogeneous data characteristics of the multi-source heterogeneous raw data to generate the data classification and evidence preservation list of the supply chain includes: The multi-source heterogeneous raw data is preprocessed to obtain a standardized dataset, wherein the preprocessing includes data cleaning, format standardization and metadata extraction; Identify the categorical data items in the standardized dataset and extract the data feature indicators corresponding to the categorical data items; Based on the differentiated evidence preservation rule base, calculate the matching weight between each data feature indicator and each evidence preservation rule; Based on the matching weights, the evidence preservation priority and evidence preservation timeliness level of the categorized data items are determined; Construct a business logic association model between the categorized data items to identify the data dependencies between them; By combining the evidence preservation priority, the evidence preservation timeliness level, and the data dependency relationship, an initial evidence preservation priority sequence for the supply chain is generated; Obtain the evidence storage node cluster corresponding to the supply chain, and collect the real-time evidence storage environment parameters of the evidence storage node cluster. Based on the real-time evidence storage environment parameters, dynamic priority optimization processing of the initial evidence storage priority sequence is performed to generate the data classification evidence storage list of the supply chain.

3. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The step of configuring the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor includes: By using the dynamic evaluation data of the consensus mechanism decision factors, the strategy configuration node corresponding to the supply chain is located; Key performance indicators are extracted from the dynamic evaluation data, and the correlation weight between the key performance indicators and the strategy configuration node is calculated. Based on the association weights, the core configuration parameters of the strategy configuration node are selected; Identify the parameter adjustment path of the core configuration parameters in the policy configuration node; Based on the core configuration parameters, calculate the parameter sensitivity of the key performance indicators; Based on the parameter sensitivity, the adjustment priority of the core configuration parameters is set; Obtain a preset consensus strategy configuration scheme that matches the adjustment priority; Analyze the parameter adjustment boundary conditions of the core configuration parameters in the preset consensus strategy configuration scheme; By integrating the parameter adjustment boundary conditions and the parameter adjustment path, a two-level policy generation network for the consensus policy scheduler is constructed. The consensus policy scheduler for the supply chain is configured through the two-level policy generation network.

4. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The real-time monitoring of the data storage integrity index and consensus mechanism response efficiency value of the supply chain, in order to calculate the global performance status index of the supply chain, includes: Identify the distributed nodes of the supply chain and obtain the consistency verification results of the ledger data of the distributed nodes; Calculate the data reliability index of the supply chain based on the data storage integrity index and the ledger data consistency verification results; Based on the consensus mechanism response efficiency value, the throughput variation coefficient and block propagation latency parameters of the supply chain are monitored in real time. The consensus stability coefficient of the supply chain is calculated based on the throughput variation coefficient and the block propagation delay parameter. Real-time collection of computing resource utilization, storage resource utilization, and network bandwidth utilization of the supply chain; By integrating the data reliability index, the consensus stability coefficient, the computing resource utilization rate, the storage resource utilization rate, and the network bandwidth utilization rate, the global performance status index of the supply chain is calculated using the following formula: ； in, This represents the overall performance status index. Indicates the data reliability index. This represents the consensus stability coefficient. The weighting coefficient represents the consensus stability coefficient. This indicates the utilization rate of computing resources. This represents the weighting coefficient used to calculate resource utilization. Indicates network bandwidth utilization. This represents the rated baseline value of network bandwidth corresponding to the network bandwidth utilization rate. This indicates the utilization rate of computing resources. This represents the rated baseline value of computing power corresponding to the computing resource utilization rate. Indicates storage resource utilization. This indicates the rated baseline value of storage space corresponding to the storage resource utilization rate. Indicates the system stability variance. This indicates a regulatory factor.

5. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The step of generating the data classification and evidence storage list and the consensus policy scheduler collaborative optimization instructions based on the global performance state index includes: Obtain the performance evaluation benchmark corresponding to the global performance status index; Based on the performance evaluation benchmark, identify the system bottleneck type corresponding to the global performance status index, and set multi-level performance thresholds corresponding to the global performance status index; Based on the multi-level efficiency thresholds, the collaborative control levels corresponding to the global efficiency state index are divided. Based on the system bottleneck type, the multi-level performance threshold, and the collaborative control level, determine the joint optimization strategy for the data classification and storage list and the consensus strategy scheduler; The parameter adjustment step size of the joint optimization strategy is calculated based on the global performance state index and the multi-level performance threshold. Based on the parameter adjustment step size, construct the parameter coupling relationship matrix between the data classification and evidence storage list and the consensus strategy scheduler; The data classification and evidence storage list and the consensus strategy scheduler are generated through the parameter coupling relationship matrix.

6. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The acquisition of multi-source heterogeneous raw data and business process interaction data from the supply chain to construct the data modality perception unit of the supply chain includes: Extract the data modal features and business source attributes of the multi-source heterogeneous raw data; Using the data modality features and the business source attributes, determine the set of structured data parsing strategies for the multi-source heterogeneous raw data; Based on the interaction data of the business processes, identify the business process nodes and data interaction relationships of the supply chain; The business logic rules and data permission levels for parsing the interactive data of the aforementioned business processes; Based on the business logic rules and the data permission levels, a value density evaluation standard for the multi-source heterogeneous raw data is set. By combining the structured data parsing strategy set, the business process nodes, and the value density assessment criteria, a data modality mapping model for the supply chain is constructed. Based on the data interaction relationship and the data modality mapping model, determine the type of multimodal fusion algorithm corresponding to the supply chain; By combining the structured data parsing strategy set, the value density assessment standard, and the multimodal fusion algorithm type, the data modality perception unit of the supply chain is constructed.

7. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The establishment of the differentiated evidence storage rule base for the supply chain based on the data modality sensing unit includes: The system acquires raw data from the entire supply chain and outputs the modality recognition results and data attribute information of the raw data through the data modality perception unit. The modal feature set of the original data of the entire link is extracted from the modal recognition result, wherein the modal feature set includes data format type, data generation frequency and data flow continuity characteristics; Based on the data attribute information, the business attribute set of the original data of the entire chain is extracted, wherein the business attribute set includes business link identifier, data sensitivity level and compliance evidence storage period; Based on the modal feature set, a selection strategy for evidence preservation technology corresponding to the supply chain is generated; Based on the aforementioned business attribute set, define the evidence preservation compliance constraint rules corresponding to the supply chain; By integrating the evidence preservation technology selection strategy and the evidence preservation compliance constraint rules, a differentiated evidence preservation strategy vector for the supply chain is output. The differentiated evidence preservation strategy vector is encapsulated into executable evidence preservation rules for the supply chain; Integrate the executable evidence storage rules to establish a differentiated evidence storage rule library for the supply chain.

8. The supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The definition of consensus mechanism adaptation standards for the supply chain under different business scenarios, based on the business scenario requirements and the network operating status, includes: Based on the business scenario requirements, the core consensus requirements of the supply chain under different business scenarios are extracted. Based on the business scenario requirements and the core consensus requirements, a candidate set of consensus mechanism types for the supply chain under different business scenarios is selected. Construct a dynamic adaptation matrix between the candidate set of consensus mechanism types and the network operating state, and calculate the consensus mechanism matching degree of the supply chain; The types of consensus nodes and network interfaces participating in the supply chain are clearly defined in order to determine the node admission conditions and communication protocol specifications of the candidate set of consensus mechanism types; Based on the dynamic adaptation matrix, the node admission conditions, and the communication protocol specifications, consensus parameter configuration rules for the supply chain under different business scenarios are generated. Based on the consensus parameter configuration rules, the consensus performance boundaries of the supply chain under different business scenarios are defined. The consensus mechanism matching degree, consensus parameter configuration rules, and consensus performance boundaries define the consensus mechanism adaptation standard for the supply chain in different business scenarios.

9. A supply chain trusted data collaborative management method based on multimodal blockchain as described in claim 1, characterized in that, The determination of consensus mechanism decision factors for the supply chain based on the consensus mechanism adaptation standard includes: Real-time monitoring of the dynamic characteristics of node load and network communication quality in the supply chain; Calculate the load change rate corresponding to the node load dynamic characteristics and the network latency index corresponding to the network communication quality; Based on the load change rate and the network latency index, identify the real-time operating status mode of the supply chain; The candidate consensus mechanism type corresponding to the supply chain is determined by the state-mechanism mapping relationship between the real-time running state mode and the consensus mechanism adaptation standard. Based on the candidate consensus mechanism type, the transaction processing latency, data consistency index and system fault tolerance parameters of the supply chain in the real-time operating state mode are collected in real time. Based on the transaction processing latency, the data consistency index, and the system fault tolerance parameters, analyze the performance patterns of each candidate consensus mechanism type in the real-time operating state mode. Construct a matching degree evaluation matrix between the performance pattern and the consensus mechanism adaptation standard; Based on the matching degree evaluation matrix, the consensus mechanism decision factors of the supply chain are determined.

10. A supply chain trusted data collaborative management system based on multimodal blockchain, characterized in that, The system includes: The data modality perception module is used to acquire multi-source heterogeneous raw data and business process interaction data of the supply chain in order to construct the data modality perception unit of the supply chain. The data classification and evidence storage module is used to establish a differentiated evidence storage rule base for the supply chain based on the data modality perception unit, and generate a data classification and evidence storage list for the supply chain by combining the differentiated evidence storage rule base and the multi-source heterogeneous original data. The consensus mechanism intelligent adaptation module is used to identify the business scenario requirements and network operation status of the supply chain from the interaction data of the business process, and define the consensus mechanism adaptation standard of the supply chain under different business scenarios based on the business scenario requirements and the network operation status. The consensus strategy dynamic scheduling module is used to determine the consensus mechanism decision factor of the supply chain based on the consensus mechanism adaptation standard, and configure the consensus strategy scheduler of the supply chain according to the consensus mechanism decision factor. The global performance collaborative optimization module is used to monitor the data storage integrity index and consensus mechanism response efficiency value of the supply chain in real time, so as to calculate the global performance status index of the supply chain, and generate the collaborative optimization instructions of the data classification storage list and the consensus strategy scheduler based on the global performance status index. The data collaboration management module is used to combine the consensus strategy scheduler, the data classification and storage list, and the collaboration optimization instructions to output the data collaboration management scheme of the supply chain.

Images