Industrial digital real-time risk control and decision support system based on block chain

Abstract

The invention relates to the technical field of block chains and data processing, in particular to an industrial digital real-time risk control and decision support system based on a block chain, which comprises a causal entropy chain construction module, a commercial plan simulation module, a multi-party credible cooperation module and a commercial credit assessment module. A causal entropy chain construction module is used for inferring a causal relationship of enterprise operation data by using a dynamic graph neural network, quantifying an entropy value of the enterprise operation data, packaging a result into a block, and recording the block on a block chain to form a causal relationship account book; based on the risk links with entropy values exceeding a preset threshold value in the causal relationship account book, the commercial plan simulation module performs adversarial risk simulation to generate a commercial decision plan; the multi-party credible cooperation module allows enterprises to cooperatively train a global risk prediction model through federated learning, zero-knowledge proof and other modes under the condition that private data is not exposed; and the commercial credit assessment module generates a dynamic credible two-dimensional code capable of representing the enterprise operation capability and the risk level, and the dynamic credible two-dimensional code is used for various commercial scenes.

Description

Technical Field

[0001] The present invention relates to the field of blockchain and data processing technology, and specifically to a blockchain-based industrial digital real-time risk control and decision support system. Background Art

[0002] In the era of industrial digitalization, the systematic and interconnected nature of business risks is becoming increasingly stronger, posing severe challenges to the traditional and backward business risk management model.

[0003] Existing business risk control systems often rely on historical data for correlation analysis, lacking insight into the causal transmission of risks and leading to delayed decision-making. While blockchain technology is being used for data authorization and storage, its application is relatively limited. Concerns about business data privacy and security hinder cross-party collaborative risk modeling, and data silos are widespread. Creditworthiness certificates issued by businesses are often static reports with poor timeliness and difficult verification. Therefore, a business risk decision support system that enables forward-looking, trusted collaboration, and dynamic verification is urgently needed.

[0004] To this end, a blockchain-based industrial digital real-time risk control and decision support system is proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a blockchain-based industrial digital real-time risk control and decision support system, which provides dynamic and verifiable business decision support by integrating deep causal insights, forward-looking risk simulation and multi-party trusted collaboration.

[0006] To achieve the above object, the present invention provides the following technical solutions: The blockchain-based industrial digital real-time risk control and decision support system includes: Causal Entropy Chain Building Module: Utilizing dynamic graph neural networks, enterprise operational data is modeled as dynamic graphs, inferring causal relationships between these data points. The information entropy principle is then used to quantify the entropy of these causal relationships, identifying risk links where the entropy exceeds a preset threshold. Causal relationships, entropy values, and model parameter hash values ​​are packaged into causal blocks and recorded on the blockchain, forming an immutable causal ledger.

[0007] Business plan simulation module: Based on the risk links in the causal relationship ledger whose entropy value exceeds the preset threshold, adversarial risk simulation is performed to actively discover business risks and generate business decision plans.

[0008] Multi-party trusted collaboration module: This module allows enterprises to maintain local digital twins based on their own private data. Without exposing private data, the module periodically collaborates on training and updating a shared global risk prediction model through federated learning. Each participant's submitted local model update is accompanied by a zero-knowledge proof and a hash of the updated model parameters.

[0009] Commercial credit assessment module: Generates a dynamic and credible QR code that can represent the business operation capability and risk level of the enterprise, for use in commercial business scenarios.

[0010] Preferably, the enterprise operation data includes industrial and commercial and judicial data, financial and tax declaration data, invoice data, intellectual property data and internal enterprise operation data.

[0011] Preferably, the causal entropy chain construction module includes a dynamic graph modeling unit, an entropy value quantification unit and a causal block generation unit. The dynamic graph modeling unit is used to model the entities in the enterprise operation as nodes of the dynamic graph, and model the materials and information between the entities as edges of the graph; and use the dynamic graph neural network to learn the risk propagation pattern and conditional probability between nodes to infer the causal relationship; the entropy value quantification unit is used to use the information entropy principle to quantify the entropy value of the causal relationship and identify the risk link whose entropy value exceeds a preset threshold; the causal block generation unit is used to package the inferred causal relationship, its corresponding entropy value and the hash value of the model parameter used for inference into a cryptographically signed causal relationship block to form an unalterable and auditable causal relationship ledger on the blockchain.

[0012] Preferably, the business plan simulation module includes a risk synthesis unit and a strategy optimization unit; the risk synthesis unit uses a generative artificial intelligence model to automatically generate a simulated risk scenario based on the risk link in the causal relationship ledger whose entropy value exceeds a preset threshold by combining multiple low-probability events; the strategy optimization unit uses a reinforcement learning method to find and output a business decision plan that can minimize operating losses through trial and error learning in the simulated risk scenario.

[0013] Preferably, the multi-party trusted collaboration module includes a local training and verification unit, a homomorphic encryption and security aggregation unit, and a multi-party collaborative decryption unit; the local training and verification unit is used for each enterprise to generate local model updates on its own private data through a federated learning algorithm, and generate a zero-knowledge proof for the updated calculation process to prove the compliance of the calculation without exposing private data; the homomorphic encryption and security aggregation unit is used to receive the encrypted local model updates and zero-knowledge proofs submitted by each enterprise, and after verifying that the zero-knowledge proof is valid, use homomorphic encryption technology to aggregate all valid local model updates in a ciphertext state to form an encrypted global model; the multi-party collaborative decryption unit organizes multiple enterprises to use their respective private key shards to collaboratively decrypt the encrypted global model through a secure multi-party computing protocol.

[0014] Preferably, the dynamic trusted QR code is a dynamic verifiable credential, comprising an access address pointing to a smart contract on the blockchain; the smart contract records an assessment result generated by the commercial credit assessment module, which characterizes the current operating capabilities and risk level of the enterprise; a zero-knowledge proof for proving that the assessment result is calculated by the authorization model and trusted data; and an index pointing to the causal relationship ledger and data authorization record.

[0015] Preferably, the dynamic update mechanism of the dynamic trusted QR code includes that when an immediate evaluation request is received from a third-party service agency and / or a preset update cycle is reached, the system will automatically trigger a new round of enterprise operation data collection and analysis process, refresh and update the evaluation results and corresponding zero-knowledge proof recorded in the smart contract.

[0016] Preferably, the business plan simulation module also includes an expert decision upgrade mechanism. When the strategy optimization unit cannot find a decision plan within the preset operating loss threshold, the simulation is automatically suspended; and the simulated risk scenario is packaged into an on-chain governance proposal and submitted to the decentralized expert unit for analysis and voting; finally, based on the voting results, the business decision formed by expert consensus is executed.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention builds a business causal relationship ledger to gain deep insight into the inherent logic of risk transmission. Combined with forward-looking adversarial simulation, it enables enterprises to proactively discover and formulate optimized plans for potential, complex, and unknown risks, fundamentally improving the foresight of business decisions and the anti-fragility of the system.

[0018] 2. This invention utilizes technologies such as federated learning, zero-knowledge proof, and homomorphic encryption to enable multiple enterprises within an industry chain or region to collaborate on modeling securely and pool ecological wisdom while fully protecting their respective trade secrets, thereby achieving accurate predictions of macro-market and systemic risks.

[0019] 3. This invention transforms a company's complex, real-time operating status into a concise, reliable, and readily verifiable digital credential by generating a dynamically verifiable, trusted QR code. This replaces traditional, delayed, and easily forged static reports, significantly reducing trust costs and business friction in scenarios such as finance and supply chains. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of the method for the blockchain-based industrial digital real-time risk control and decision support system proposed in an embodiment of the present invention; Figure 2 This is a system structure diagram of the blockchain-based industrial digital real-time risk control and decision support system proposed in an embodiment of the present invention; Figure 3 This is a system flow chart of the blockchain-based industrial digital real-time risk control and decision support system proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] Example 1 See also Figures 1 to 3 The present invention provides a blockchain-based industrial digital real-time risk control and decision support system. The technical solution is as follows: Industrial digital real-time risk control and decision support system based on blockchain, such as Figure 1-Figure 3 ,include: Causal Entropy Chain Building Module: Utilizing dynamic graph neural networks, enterprise operational data is modeled as dynamic graphs, inferring causal relationships between these data points. The information entropy principle is then used to quantify the entropy of these causal relationships, identifying risk links where the entropy exceeds a preset threshold. Causal relationships, entropy values, and model parameter hash values ​​are packaged into causal blocks and recorded on the blockchain, forming an immutable causal ledger.

[0023] Business plan simulation module: Based on the risk links in the causal relationship ledger whose entropy value exceeds the preset threshold, adversarial risk simulation is performed to actively discover business risks and generate business decision plans.

[0024] Multi-party trusted collaboration module: This module allows enterprises to maintain local digital twins based on their own private data. Without exposing private data, the module periodically collaborates on training and updating a shared global risk prediction model through federated learning. Each participant's submitted local model update is accompanied by a zero-knowledge proof and a hash of the updated model parameters.

[0025] Commercial credit assessment module: Generates a dynamic and credible QR code that can represent the business operation capability and risk level of the enterprise, for use in commercial business scenarios.

[0026] Furthermore, the enterprise operation data includes industrial and commercial and judicial data, financial and tax declaration data, invoice data, intellectual property data and internal enterprise operation data.

[0027] For enterprise operational data, we use the predefined JSON Schema to perform structured cleaning on data from different sources; we use entity recognition technology to perform alignment; and we define a feature mapping table to map the enterprise's multi-dimensional operating indicators to the corresponding numerical attributes of the enterprise nodes in the graph model after normalization.

[0028] By integrating and standardizing multi-source data, we can effectively avoid the one-sidedness and fraud risks of a single data source, and lay a high-quality data foundation for building a comprehensive and objective enterprise risk profile.

[0029] Furthermore, the causal entropy chain construction module includes a dynamic graph modeling unit, an entropy value quantification unit and a causal block generation unit. The dynamic graph modeling unit is used to model the entities in the enterprise operation as nodes of the dynamic graph, and model the materials and information between the entities as edges of the graph; and use the dynamic graph neural network to learn the risk propagation pattern and conditional probability between nodes to infer the causal relationship; the entropy value quantification unit is used to use the information entropy principle to quantify the entropy value of the causal relationship and identify the risk link whose entropy value exceeds a preset threshold; the causal block generation unit is used to package the inferred causal relationship, its corresponding entropy value and the hash value of the model parameter used for inference into a cryptographically signed causal relationship block to form an unalterable and auditable causal relationship ledger on the blockchain.

[0030] The dynamic graph modeling unit specifically performs modeling in the following manner: Node Definition: Market entities such as enterprises, banks, suppliers, and customers are defined as entity nodes; invoices, contracts, lawsuits, and intellectual property are defined as event nodes. Each node has different attributes depending on its type.

[0031] Edge definition: If there is a direct business or legal relationship between two nodes, define an edge with direction and type. Dynamic graph construction: Take one month as the time window T, and construct all nodes and edges generated in the window as a graph snapshot. . Continuous time windows constitute a dynamic graph sequence .

[0032] Causal inference model: The dynamic graph neural network specifically uses a time-aware graph attention network. To achieve causal inference, the present invention combines the time-aware graph attention network with the Granger causality test. The specific steps are: First, use the time-aware graph attention network to learn the dynamic embedding representation of each node in the sequence graph. Then, for any two nodes i and j that have a link, extract the time series of their embedding representations and Finally, a vector autoregression model is implemented on these two time series, and a Granger causality test is performed to screen out node pairs with statistically significant predictive associations, identify potential risk transmission paths, and thus infer probabilistic causal relationships.

[0033] By modeling abstract operational data into a specific dynamic causal graph, a deep leap from "correlation" to "causality" is achieved, enabling the system to gain insight into the fundamental logic of risk transmission and providing a solid basis for accurate decision-making and risk auditing.

[0034] The entropy quantization unit specifically adopts the following logical steps to quantify the entropy of the causal relationship: For each causal relationship inferred by the dynamic graph neural network, the unit will obtain its corresponding probability score, which represents the likelihood of the result event occurring after the cause event occurs.

[0035] The entropy quantification unit evaluates the probability score based on the principle of information entropy. Its core logic is: if the probability score of a causal relationship approaches the two extremes of certainty, that is, the result is highly predictable, then it is assigned a low entropy value; conversely, if the probability score approaches the middle value, that is, the result is highly unpredictable, then it is assigned a high entropy value.

[0036] At the same time, a threshold for the entropy score is set. When the calculated entropy score of a causal relationship exceeds this preset threshold, the causal link is marked as a risk link with an entropy value exceeding the preset threshold and serves as the key input for the subsequent business plan simulation module.

[0037] Through the principle of information entropy, the present invention accurately mathematically quantifies the entropy value, enabling the system to automatically identify the core risk links that are most worthy of attention from massive relationships, providing key input for subsequent simulation decisions.

[0038] By modeling abstract operational data into concrete causal diagrams, we achieve a profound leap from correlation analysis to causal logic inference. Its quantification of risk entropy helps companies accurately identify and focus on core risk points. The resulting on-chain causal ledger provides an immutable basis for auditing and attributing decisions.

[0039] Furthermore, the business plan simulation module includes a risk synthesis unit and a strategy optimization unit; the risk synthesis unit uses a generative artificial intelligence model to automatically generate a simulated risk scenario based on the risk link in the causal relationship ledger whose entropy value exceeds a preset threshold by combining multiple low-probability events; the strategy optimization unit uses a reinforcement learning method to find and output a business decision plan that can minimize operating losses through trial and error learning in the simulated risk scenario.

[0040] The risk synthesis unit specifically uses a conditional variational autoencoder (CVAE) as a generative AI model. CVAE's training data consists of real-world risk events (e.g., a company's capital chain rupture). Each event consists of a set of precursor events and a label. The label is a risk link identified from the causal entropy chain ledger with an entropy value exceeding a preset threshold, such as "upstream core supplier A ceases production." During the simulation phase, the operator selects one or more high-entropy causal links from the causal entropy chain as input conditions. Based on these conditions, CVAE samples and decodes the latent space, generating a set of historically unconventional but logically plausible combinations of precursor events. For example, given the input condition "upstream core supplier A ceases production," the model might generate a combination of three low-probability events: "raw material prices rise 30%," "shipping disruptions," and "competitor B secures significant financing," forming a new, complex simulated risk scenario.

[0041] The strategy optimization unit specifically uses the proximal strategy optimization algorithm as a reinforcement learning method. Its core elements are defined as follows: State space: The currently simulated risk scenario is represented by a vector that contains the event combination generated by the risk synthesis unit and the current core operating indicators of the enterprise (such as cash flow, inventory, and order volume).

[0042] Action space: A discrete set of decisions representing the possible responses a company can take. Examples include: {“urgently purchase alternative raw materials,” “apply for a short-term loan of 1 million yuan,” “activate a backup production line,” “issue a delayed delivery notice to customers”}.

[0043] Reward function: It is defined as a weighted combination of changes in the enterprise's key performance indicators (KPIs). At each simulation time step t, the action is performed Afterwards, rewards The calculation is as follows .in, is the change in cash flow, Change in inventory holding costs, The amount of fines incurred due to delayed delivery, etc. , , are preset weight coefficients, representing the importance that the company attaches to cash, cost and reputation respectively.

[0044] Through continuous trial and error in thousands of simulations, the reinforcement learning model will gradually learn under what conditions and which actions to choose to obtain the highest long-term cumulative rewards. This learned optimal strategy will eventually be output as a "business decision plan."

[0045] Through adversarial simulation using generative AI and reinforcement learning, this invention achieves a strategic shift from passive risk response to active "immunity." By automatically reorganizing and deducing historical risk factors, the system generates complex risk scenarios that experts might intuitively overlook. In this process, it identifies optimal contingency plans, enabling stress testing and strategic preparation before a real crisis occurs.

[0046] Furthermore, the multi-party trusted collaboration module includes a local training and verification unit, a homomorphic encryption and security aggregation unit, and a multi-party collaborative decryption unit; the local training and verification unit is used for each enterprise to generate local model updates on its own private data through a federated learning algorithm, and generate a zero-knowledge proof for the updated calculation process to prove the compliance of the calculation without exposing private data; the homomorphic encryption and security aggregation unit is used to receive the encrypted local model updates and zero-knowledge proofs submitted by each enterprise, and after verifying that the zero-knowledge proof is valid, use homomorphic encryption technology to aggregate all valid local model updates in a ciphertext state to form an encrypted global model; the multi-party collaborative decryption unit organizes multiple enterprises to use their respective private key shards to collaboratively decrypt the encrypted global model through a secure multi-party computing protocol.

[0047] The specific implementation process of the multi-party trusted collaboration module is as follows, with each step aiming to ensure data privacy and the trustworthiness of the computing process: Participating enterprises adopt a common federated averaging strategy. First, each enterprise independently trains a global risk prediction model on its own private data and calculates local adjustments to the global model parameters, also known as a model update. This model update is the only data that needs to leave the enterprise's local environment. Before submitting a model update, each enterprise generates a zero-knowledge proof. This is a separate encrypted data package that proves to other parties in the system that the enterprise generated the model update strictly in accordance with pre-defined rules (for example, using the correct data format and standard training algorithm). Crucially, this proof process does not disclose any private data used for the calculation. To protect commercial confidentiality, enterprises encrypt their model updates using a homomorphic encryption scheme that supports addition operations before uploading them and their zero-knowledge proofs. Upon receiving the encrypted model updates and zero-knowledge proofs, the central aggregation server first verifies the validity of the proof to eliminate non-compliant submissions. Then, leveraging the unique properties of homomorphic encryption, the aggregator directly sums all valid model updates in an encrypted state, ultimately generating an encrypted, aggregated global model update.

[0048] The encrypted global model update cannot be decrypted by any single party (including aggregators). The full key required for decryption is split into multiple key shards, each of which is maintained by participating companies. Only when a predetermined number of companies (e.g., more than half) participate in a secure multi-party computation protocol and simultaneously provide their key shards can the decryption operation be collaboratively completed, recovering a clear and usable global model update. This mechanism ensures that the final result is the result of collective consensus, reducing the risk of single points of failure or malicious manipulation.

[0049] The global risk prediction model is a three-layer feedforward neural network. The input layer contains 128 neurons, corresponding to the 128 macroeconomic and industry risk factors jointly defined by the industry chain collaborators; the hidden layer contains 64 neurons and uses the ReLU activation function; the output layer contains 1 neuron and uses the Sigmoid activation function to output the probability of systemic risk occurring in the next quarter.

[0050] The local training and verification unit uses the zk-SNARKs protocol (such as the Groth16 algorithm) to generate zero-knowledge proof; the homomorphic encryption and security aggregation unit uses the Paillier cryptographic system to encrypt and aggregate model updates; and the multi-party collaborative decryption unit implements secure multi-party computing decryption based on the Shamir secret sharing scheme.

[0051] Through multi-layered cryptographic design, including zero-knowledge proofs and homomorphic encryption, this invention fundamentally resolves the trust paradox of data collaboration. While ensuring the integrity of all parties' computations, it completely removes reliance on a centralized trusted third party, enabling even competing entities to securely engage in deep data collaboration.

[0052] Furthermore, the dynamic trusted QR code is a dynamic verifiable credential, which includes an access address pointing to a smart contract on the blockchain; the smart contract records an assessment result generated by the commercial credit assessment module, which represents the company's current operating capabilities and risk level; a zero-knowledge proof used to prove that the assessment result is calculated by the authorization model and trusted data; and an index pointing to the causal relationship ledger and data authorization record.

[0053] Upgrading corporate creditworthiness to dynamic, verifiable digital credentials. Leveraging zero-knowledge proofs, third parties can verify the authenticity and compliance of assessment results without accessing private data, providing an audit basis for business decisions and significantly reducing the cost of mutual trust in business.

[0054] Furthermore, the dynamic update mechanism of the dynamic trusted QR code includes that when an immediate evaluation request is received from a third-party service agency and / or a preset update cycle is reached, the system will automatically trigger a new round of enterprise operation data collection and analysis process, refresh and update the evaluation results and corresponding zero-knowledge proof recorded in the smart contract.

[0055] Through an on-demand or periodic automatic update mechanism, trusted credentials can reflect the company's latest operating conditions in real time, allowing third parties to have a more solid basis and control risks when making business decisions such as credit approval and cooperation review.

[0056] Furthermore, the business plan simulation module also includes an expert decision upgrade mechanism. When the strategy optimization unit cannot find a decision plan within the preset operating loss threshold, the simulation is automatically suspended; and the simulated risk scenario is packaged into an on-chain governance proposal and submitted to the decentralized expert unit for analysis and voting; finally, based on the voting results, the business decision formed by expert consensus is executed.

[0057] This human-machine governance model overcomes the rigid boundaries of purely automated decision-making. When AI faces extremely complex risks, it can seamlessly escalate to human expert consensus, balancing machine efficiency with expert wisdom. The on-chain governance process ensures transparency, fairness, and traceability in crisis management.

[0058] This system includes two core processes: real-time assessment and periodic collaborative modeling. The multi-party trusted collaboration module is executed periodically (e.g., weekly) to securely generate and update a global risk prediction model using federated learning and multiple cryptographic techniques. Each enterprise's causal entropy chain construction module and commercial credit assessment module utilize the latest locally stored global risk prediction model as environmental parameter input during real-time assessments, enabling immediate response to risks.

[0059] This invention provides a novel business risk decision support system. Through causal insights and forward-looking risk simulation, it achieves a strategic shift from passive response to proactive prediction. It leverages advanced cryptography to break down data silos, enabling ecosystem-level collaborative intelligence while protecting business privacy. Ultimately, it transforms complex analytical results into dynamically verifiable business credit credentials, significantly reducing the cost of trust in financial and supply chain scenarios and improving the operational efficiency and security of the entire industry.

[0060] Example 2 This example uses a high-end precision manufacturing industry chain as a specific scenario to further illustrate the implementation of the present invention. The core enterprise in this industry chain is automobile engine manufacturer A, which supplies dozens of key component suppliers (such as pistons, turbochargers, and electronic control units (ECUs)) upstream and multiple large automobile manufacturers downstream.

[0061] Step 1: Construction of causal entropy chain and risk identification Automaker A, as the chain leader, first utilizes the system. The system accesses A's internal operational data (such as production plans, inventory data, and order data from its ERP system) and, with authorization, accesses its invoice data, tax data, and relevant public legal proceedings and intellectual property data. Simultaneously, its core suppliers (such as German turbocharger supplier B and domestic ECU supplier C) also join the system as participants, authorizing the sharing of partially de-identified production and logistics data.

[0062] The system's dynamic graph modeling unit models enterprise A, supplier B, supplier C, and logistics company D as "entity nodes." It also models the purchase contract between A and B, the invoice from B to A, the payment from A to B, and the logistics documents handled by D as "event nodes" with time and amount attributes and "transaction" edges.

[0063] The system constructs graph snapshots on a weekly basis. By analyzing dynamic graph sequences spanning several consecutive weeks, the system's time-aware graph attention network and Granger causality test model discovered a potential causal relationship: a probabilistic causal link between the degree of port congestion in Supplier B's region and the risk of engine production line shutdown at Manufacturer A, with a confidence score of 60%. Because this score is close to 50%, the system's entropy quantification unit assigns it an entropy score of 0.7 based on the principle of information entropy. Because this score exceeds the system's preset risk threshold of 0.65 for this type of link, it is marked as a "risk link with entropy exceeding the preset threshold."

[0064] Step 2: Simulation and optimization of business plans The business plan simulation module was triggered, and the risk synthesis unit used the high-entropy link "Supplier B's port congestion" as its foundational condition. Its built-in conditional variational autoencoder (CVAE) model, combined with other unrelated, low-probability historical events, generated a complex simulated risk scenario: "Due to geopolitical factors, Supplier B's port is severely congested, resulting in a delay of at least three weeks in the ocean shipment of turbochargers. Simultaneously, a domestic competitor releases a new engine and announces a 15% price cut to seize market share."

[0065] Faced with this simulated scenario (state space), the proximal policy optimization algorithm of the policy optimization unit begins trial-and-error learning.

[0066] Action 1: Execute the decision to "airfreight some urgent supplies." Simulations show that while this action can partially resolve the immediate need, the high air freight costs will severely erode product profits, resulting in a low negative feedback score from the reward function.

[0067] Action 2: Execute "Initiate the certification and procurement process for backup domestic turbocharger supplier E." Simulations show that while supplier E's product performance is slightly lower than supplier B's, it ensures supply chain continuity and offers lower costs, maintaining profitability despite competitors' price cuts. The reward function, based on a comprehensive assessment of financial and operational impact, assigns a high positive feedback score.

[0068] After thousands of rounds of simulation, the system ultimately output the optimal business decision plan: "Immediately initiate emergency certification of supplier E, inform major customers of potential delivery risks to manage expectations, and use part of the risk reserves to hedge short-term profit fluctuations."

[0069] Step 3: Multi-party trusted collaboration and macro-risk prediction Manufacturer A, supplier C, and dozens of other companies in the industry chain jointly participate in a multi-party trusted collaboration network to jointly train a "macro-risk prediction model for the automotive industry chain."

[0070] Each enterprise uses local data to independently calculate model updates and uploads them after encrypting them using a homomorphic encryption scheme.

[0071] The central aggregator aggregates all encrypted updates using the properties of homomorphic encryption without touching the plaintext data.

[0072] The decryption process requires more than half of the companies to collaborate using their own private key shards to ensure the fairness of the results and the confidentiality of the process.

[0073] Ultimately, this global model can predict macro risks such as "the overall upward trend of chip prices in the next six months", providing all participating companies with a broader decision-making perspective.

[0074] Step 4: Application of dynamic trusted QR code A commercial bank, F, plans to provide supply chain financing to manufacturer A. Traditionally, due diligence would take weeks. Now, a loan officer at Bank F simply scans a "dynamic trusted QR code" provided by company A.

[0075] The QR code contains an access address pointing to the smart contract on the blockchain.

[0076] By accessing this address, loan officers can instantly view the assessment results of Company A's operating capacity and risk level, generated by the commercial credit assessment module, as of that day. Bank F's system can also automatically verify the zero-knowledge proof accompanying the assessment results, confirming that the results were indeed calculated by the authorized model based on Company A's authentic, trusted data (such as the latest invoices and tax data), and that the entire calculation process was compliant. Because the QR code automatically refreshes when Manufacturer A completes a new large-value order or reaches a preset update cycle (e.g., weekly), Bank F can make credit decisions based on near-real-time, trusted data, significantly improving approval efficiency and reducing risk.

[0077] This invention provides core enterprises in the industrial chain and their partners with all-round support from micro-risk plans to macro-risk insights and trusted financing, demonstrating its great value in improving the stability and efficiency of the entire industrial digital ecosystem.

[0078] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The blockchain-based industrial digital real-time risk control and decision support system is characterized by: include: Causal entropy chain building module: Utilizes dynamic graph neural networks to model enterprise operational data into dynamic graphs, inferring causal relationships between enterprise operational data, and uses the information entropy principle to quantify the entropy value of causal relationships, identifying risk links whose entropy values ​​exceed a preset threshold. The causal relationship, entropy value and model parameter hash value are packaged into a causal relationship block and recorded on the blockchain to form an unalterable causal relationship ledger; Business plan simulation module: Based on the risk links in the causal relationship ledger whose entropy value exceeds the preset threshold, it conducts adversarial risk simulation to proactively identify operational risks and generate business decision plans; Multi-party trusted collaboration module: This module allows enterprises to maintain local digital twins on their own private data. Without exposing private data, it periodically collaborates to train and update a shared global risk prediction model through federated learning. Each participant's submitted local model update is accompanied by a zero-knowledge proof and a hash of the updated model parameters. Commercial credit assessment module: Generates a dynamic and credible QR code that can represent the business operation capability and risk level of the enterprise, for use in commercial business scenarios.

2. The blockchain-based real-time risk control and decision support system for industrial digitization according to claim 1 is characterized by: The enterprise operation data includes industrial and commercial and judicial data, financial and tax declaration data, invoice data, intellectual property data and internal enterprise operation data.

3. The blockchain-based industrial digital real-time risk control and decision support system according to claim 1 is characterized by: The causal entropy chain construction module includes a dynamic graph modeling unit, an entropy value quantification unit, and a causal block generation unit; The dynamic graph modeling unit is used to model entities in enterprise operations as nodes of a dynamic graph, and to model materials and information between entities as edges of the graph; and to use a dynamic graph neural network to learn the risk propagation pattern and conditional probability between nodes and infer causal relationships; The entropy value quantification unit is used to quantify the entropy value of the causal relationship using the information entropy principle and identify risk links whose entropy value exceeds a preset threshold; The causal block generation unit is used to package the inferred causal relationship, its corresponding entropy value, and the hash value of the model parameter used for inference into a cryptographically signed causal relationship block, forming an unalterable and auditable causal relationship ledger on the blockchain.

4. The blockchain-based industrial digital real-time risk control and decision support system according to claim 1 is characterized by: The business plan simulation module includes a risk synthesis unit and a strategy optimization unit; The risk synthesis unit uses a generative artificial intelligence model to automatically generate simulated risk scenarios by combining multiple low-probability events based on risk links in the causal relationship ledger whose entropy values ​​exceed a preset threshold; The strategy optimization unit uses a reinforcement learning method to find and output a business decision plan that can minimize operating losses through trial and error learning in the simulated risk scenario.

5. The blockchain-based industrial digital real-time risk control and decision support system according to claim 1 is characterized by: The multi-party trusted collaboration module includes a local training and verification unit, a homomorphic encryption and security aggregation unit, and a multi-party collaborative decryption unit; The local training and verification unit is used by each enterprise to generate local model updates using a federated learning algorithm on its own private data, and to generate a zero-knowledge proof for the updated calculation process, proving the compliance of the calculation without exposing private data; The homomorphic encryption and security aggregation unit is used to receive the encrypted local model updates and zero-knowledge proofs submitted by each enterprise. After verifying the validity of the zero-knowledge proofs, it uses homomorphic encryption technology to aggregate all valid local model updates in the ciphertext state to form an encrypted global model. The multi-party collaborative decryption unit organizes multiple enterprises to use their respective private key shards to collaboratively decrypt the encrypted global model through a secure multi-party computing protocol.

6. The blockchain-based industrial digital real-time risk control and decision support system according to claim 1 is characterized by: The dynamic trusted QR code is a dynamic and verifiable credential that contains an access address pointing to a smart contract on the blockchain; the smart contract records an assessment result generated by the business credit assessment module that represents the company's current operating capabilities and risk level; A zero-knowledge proof that the evaluation result is calculated using an authorized model and trusted data; A reference to the causal relationship ledger and data authorization records.

7. The blockchain-based industrial digital real-time risk control and decision support system according to claim 6 is characterized in that: The dynamic update mechanism of the dynamic trusted QR code includes: When an immediate evaluation request is received from a third-party service agency and / or the preset update cycle is reached, the system will automatically trigger a new round of enterprise operation data collection and analysis process, refresh and update the evaluation results and corresponding zero-knowledge proof recorded in the smart contract.

8. The blockchain-based industrial digital real-time risk control and decision support system according to claim 4 is characterized by: The business plan simulation module also includes an expert decision upgrade mechanism: When the strategy optimization unit is unable to find a decision plan within the preset operating loss threshold, the simulation is automatically suspended; the simulated risk scenario is packaged into an on-chain governance proposal and submitted to the decentralized expert unit for analysis and voting; finally, based on the voting results, the business decision formed by expert consensus is executed.

Images