Intelligent contract automatic execution and supervision system for whole-process property right transaction

Abstract

The present application relates to the technical field of property transaction, in particular to an intelligent contract automatic execution and supervision system for the whole process of property transaction, comprising: an intelligent contract generation unit; a contract automatic execution unit, which maps the intellectual property right transfer and transaction settlement process agreed in the standardized intelligent contract into a multi-stage execution phase sequence of corresponding performance nodes, and each execution phase is subject to the necessary condition of performance risk assessment; a contract supervision and verification unit; and an intellectual property right ownership verification unit.The present application builds a whole-process intelligent contract automatic execution and supervision system, uses a time sequence attention mechanism and an improved lightweight residual time sequence convolution network to complete dynamic risk assessment, combines hierarchical execution and reversible rollback to realize abnormal disposal, forms an execution, supervision and ownership verification closed-loop architecture, can monitor and dispose transaction abnormalities in real time, ensures stable transaction execution, realizes whole-link ownership tracing and compliance supervision.

Description

Technical Field

[0001] This invention relates to the field of property rights transaction technology, and more specifically, to a smart contract automated execution and supervision system for the entire property rights transaction process. Background Technology

[0002] The entire process of intellectual property transactions covers key aspects such as the transfer of intellectual property rights, transaction settlement, compliance supervision and ownership verification. Blockchain smart contract technology has been widely used in this field to realize the online and automated circulation and data storage of intellectual property transactions, and is the mainstream technical solution to support the efficient conduct of intellectual property transactions.

[0003] In the existing technology, relevant patents have proposed corresponding technical solutions in the fields of property rights transactions and blockchain smart contract applications. For example, invention patent CN202310755243.5 discloses a two-stage data transaction method, device, equipment, and medium based on blockchain. This method completes key verification, phased settlement, and automatic transfer of data transactions through blockchain smart contracts, improving the efficiency and fairness of on-chain transactions and reducing transaction costs. Another example is invention patent CN202410096473.X, which discloses a method, device, equipment, and storage medium for processing rural property rights transaction information. This method stores encrypted property rights transaction data on the blockchain, completes asset delivery based on smart contracts, and records the results on the blockchain, improving data security and information transparency, reducing ownership disputes, and lowering transaction costs.

[0004] While the aforementioned technical solutions possess certain design advantages, they also suffer from the following technical shortcomings: Existing blockchain solutions for property rights transactions focus solely on optimizing the transaction execution process and ensuring secure data storage, neglecting to monitor and address abnormal states during transaction execution. For example, CN202310755243.5 only implements phased settlement and key verification of data transactions through smart contracts, without monitoring or handling abnormal states throughout the entire transaction execution process; CN202410096473.X only encrypts and uploads property rights transaction data to the blockchain and completes asset delivery storage, similarly lacking any abnormal monitoring and handling mechanisms for the transaction execution process, making it difficult to guarantee the stability and reliability of transaction execution. Therefore, we propose an automated smart contract execution and monitoring system covering the entire property rights transaction process. Summary of the Invention

[0005] The purpose of this invention is to provide an automated execution and supervision system for smart contracts covering the entire process of property rights transactions, in order to solve the problem that existing blockchain solutions for property rights transactions mentioned in the background only focus on optimizing the transaction execution process and ensuring data security, without monitoring and handling abnormal states during the transaction execution process.

[0006] To address the aforementioned technical problems, the present invention aims to provide an automated execution and monitoring system for smart contracts covering the entire process of property rights transactions, comprising: The smart contract generation unit is used to collect basic transaction information including information on the transaction subject, intellectual property rights information and transaction constraint clauses, and integrate preset contract execution technical rules including performance node rules, risk assessment rules, hierarchical execution conditions and reversible rollback rules to generate a standardized smart contract containing basic transaction information and contract execution technical rules. After completing the executability verification, it outputs a deployable standardized smart contract. The contract automated execution unit loads the standardized smart contract output by the smart contract generation unit, mapping the intellectual property ownership transfer and transaction settlement process stipulated in the standardized smart contract into a multi-stage execution sequence of corresponding performance nodes. Each execution stage requires performance risk assessment as a prerequisite. Before execution, multi-dimensional temporal feature collection is used to obtain real-time status temporal data of the intellectual property target and the transaction entity, historical performance risk data, and industry policy risk temporal data. A temporal attention mechanism and an improved lightweight residual temporal convolutional network are used to complete real-time risk assessment. Combined with historical risk data and policy risk coefficients, dynamic risk calibration is performed to output accurate risk assessment results. If the assessment meets the execution conditions, the corresponding action of the current stage is executed through atomic operations of the smart contract. If the assessment meets the suspension or termination conditions, the execution is suspended or the entire process is terminated, and reversible rollback processing is performed. Simultaneously, a contract execution record is generated and stored on the blockchain, and pushed to the contract supervision and verification unit and the intellectual property ownership verification unit. The contract supervision and verification unit is used to obtain the contract execution records pushed by the contract automated execution unit, verify the entire process of the phased execution of the contract according to the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements, generate regulatory certificates and put them on the blockchain for evidence storage, and then push them to the intellectual property ownership verification unit. The intellectual property ownership verification unit is used to obtain contract execution records, regulatory certificates and on-chain evidence data, to trace and verify the entire process of intellectual property ownership transfer in stages, and to output the ownership verification results and store them on the blockchain.

[0007] As a further improvement to this technical solution, the smart contract generation unit includes an information collection module, a rule integration module, a contract generation module, and a contract verification module, wherein: The information collection module is used to collect basic transaction information, including information on the parties involved in the property rights transaction, information on the ownership of intellectual property rights, and transaction constraint clauses. The rule integration module integrates preset contract execution technical rules, which include performance node rules, risk assessment rules, tiered execution conditions, and reversible rollback rules, wherein: The performance node rules refer to the technical rules used to divide the execution order, execution nodes, and corresponding execution content of the intellectual property ownership transfer and transaction settlement process agreed upon in standardized smart contracts. The risk assessment rules refer to the technical rules used to clarify the objects of risk assessment, the scope of information collection, and the assessment logic for each stage of contract execution. The tiered execution conditions refer to the technical criteria used to define the execution status corresponding to the risk assessment results of each execution stage of the contract. The reversible rollback rule refers to the technical execution rule used to restore the transaction state when the contract execution is triggered to stop or terminate. The contract generation module generates a standardized smart contract containing information on the property rights transaction subject, intellectual property rights ownership information, transaction constraint clauses, and contract execution technical rules based on the information collected by the information collection module and the contract execution technical rules integrated by the rule integration module. The contract verification module is used to perform executability verification on the standardized smart contract and output a deployable standardized smart contract.

[0008] As a further improvement to this technical solution, the automated contract execution unit includes an execution sequence decomposition module, a dynamic risk assessment module, a circuit breaker reversible handling module, and an execution record generation and push module, wherein: The execution sequence decomposition module, based on the standardized smart contract output by the smart contract generation unit, maps the intellectual property ownership transfer and transaction settlement process into a multi-stage execution sequence corresponding to the performance node. The dynamic risk assessment module, based on the execution phase sequence formed by the execution sequence decomposition module, collects multi-dimensional temporal status features of intellectual property targets, credit and temporal performance data of transaction entities, and temporal data of industry policy risks before each execution phase starts. It dynamically allocates features and time step weights through a temporal attention mechanism, extracts deep temporal correlation features by combining an improved lightweight residual temporal convolutional network, and introduces historical risk data and policy risk coefficients for dynamic risk calibration to complete a high-precision real-time risk assessment. The output includes risk values, attention weights, and calculation processes. The circuit breaker reversible handling module executes corresponding stage actions or triggers suspension, termination and reversible rollback processing based on the risk assessment results of the dynamic risk assessment module and the preset graded execution conditions. The execution record generation and push module generates on-chain evidence-based contract execution records based on the execution status, risk assessment results, and disposal information of each execution stage, and pushes them to the contract supervision and verification unit and the intellectual property ownership verification unit.

[0009] As a further improvement to this technical solution, the execution sequence decomposition module includes a performance node matching submodule, a stage sequence division submodule, and an execution connection triggering submodule, wherein: The performance node matching submodule is used to match the performance requirements stipulated in the standardized smart contract with the actual performance nodes of the property rights transaction one by one. The stage sequence division submodule is used to divide the intellectual property ownership transfer and transaction settlement process into multiple execution stage sequences based on the matched performance nodes; The execution connection triggering submodule is used to set the execution triggering conditions for adjacent execution phase sequences. The next execution phase sequence is triggered only after the previous execution phase sequence has been completed and the risk assessment is qualified.

[0010] As a further improvement to this technical solution, the risk assessment process of the dynamic risk assessment module includes the following steps: S22.1 Collect multi-dimensional temporal features and construct the original temporal feature matrix. The time-series characteristics include five categories of ownership status time-series characteristics of the intellectual property subject matter: pledge status, freeze status, invalidation status, objection status, and infringement litigation status, as well as the credit score of the transaction entity, performance compliance, and industry policy risk index. Bank of China corresponds to 8 types of features, and columns correspond to time steps. , For the first Class features in the first Feature values ​​of the time step; S22.2, On the original time series feature matrix Outlier truncation and standardization are performed to obtain the standardized time series feature matrix. Eliminate dimensional differences and data noise; S22.3 Calculate the attention weights for each feature and time step using a temporal attention mechanism. The standardized features are then weighted and fused to obtain an attention-weighted temporal feature vector. Attention weight The calculation is as follows: ; in: For the first Class features in the first Attention weights at time steps; For the first Weight coefficients of class features; For the first Class features in the first Standardized eigenvalues ​​of the time step; For the first Bias terms for class features; It is a linear rectification activation function; It is a natural exponential function; Traverse the index for feature categories; Index for time step traversal; This represents the total number of time steps for the temporal features. S22.4, Weighted time series feature vector Input an improved lightweight residual temporal convolutional network, and extract deep temporal fusion features through depthwise separable convolutions and residual connections. Deep temporal fusion features The extraction formula is as follows: ; in: This is a depthwise separable convolution operation; Attention-weighted temporal feature vectors; The kernel size for depth-separable convolutions; For residual join operations with dimension alignment; S22.5, combined with historical average performance risk Industry policy risk index The risk value is dynamically calibrated, and the real-time trading risk value is calculated using a normalized activation function. Real-time transaction risk value The calibration formula is as follows: ; in: This is the normalized real-time transaction risk value; This represents the Sigmoid normalized activation function; This represents the weight vector of deep temporal fusion features; For bias terms of fully connected layers; This is the historical performance risk weighting coefficient; This represents the average performance risk during the same historical period. This refers to the industry policy risk weighting coefficient. This is an industry policy risk index; S22.6 Output Trading Risk Value The corresponding core computing data is pushed to the circuit breaker reversible handling module to complete the entire risk assessment process.

[0011] As a further improvement to this technical solution, the graded handling and reversible rollback process of the fuse reversible handling module includes the following steps: S23.1, Preset risk warning threshold With risk termination threshold The two thresholds satisfy the core numerical relationship: ; S23.2 Obtain the transaction risk value output by the dynamic risk assessment module. , to trade risk value The risk level is determined by comparing the execution values ​​with the two-level risk thresholds. S23.3, when When this occurs, a stop execution instruction is triggered, suspending the current execution phase sequence and pushing risk warning information to the trading entity and regulatory nodes; S23.4, when When this occurs, a full-process termination instruction is triggered, and the transaction assets are returned via the original payment method through atomic operations of the smart contract. The asset return relationship is as follows: ;in, To return the amount of assets; This represents the amount of assets that have already been transferred and traded. S23.5 Generate circuit breaker handling and reversible rollback records, and synchronize them to the execution record generation and push module to complete the entire process of hierarchical handling.

[0012] As a further improvement to this technical solution, the execution record generation and push module includes an execution information integration submodule, a hash encryption and evidence storage submodule, and an on-chain data push submodule, wherein: The execution information integration submodule is used to collect the execution status of each execution stage sequence and the transaction risk value output by the dynamic risk assessment module. Including core computing data, the actions of the circuit breaker reversible handling module, and reversible rollback information, forming a complete and traceable contract execution record; The hash encryption evidence storage submodule is used to perform hash encryption processing on the contract execution record, generate a hash value that uniquely corresponds to the contract execution record, and write the encrypted contract execution record and the corresponding hash value into the blockchain distributed node to complete the on-chain evidence storage, ensuring that the contract execution record cannot be tampered with. The on-chain data push submodule is used to synchronously push the contract execution records that have been stored on the blockchain to the contract supervision and verification unit and the intellectual property ownership verification unit through blockchain peer-to-peer transmission, and receive the receipt confirmation information from the corresponding units to ensure the integrity and reliability of data transmission.

[0013] As a further improvement to this technical solution, the contract supervision and verification unit includes an execution record collection module, a full-process compliance verification module, a supervision certificate generation module, and a certificate storage and push module, wherein: The execution record collection module receives and verifies the contract execution records of the entire process of phased execution of the contract based on the push instructions of the contract automated execution unit, and completes the standardized collection and integrity verification of the data. The full-process compliance verification module, based on the contract execution records verified by the execution record collection module, performs multi-dimensional compliance verification on the entire process of contract execution in stages according to preset contract execution technical rules, regulatory rules and intellectual property compliance requirements, and outputs a clear verification pass conclusion or a verification fail objection explanation. The regulatory certificate generation module generates a unique regulatory certificate corresponding to this property rights transaction based on the verification pass conclusion output by the full-process compliance verification module. The regulatory certificate includes transaction entity information, target information, full-process verification data, verification conclusion, and a unique compliance verification identifier. The certificate storage and push module writes the generated regulatory certificate into the blockchain distributed node to complete the on-chain immutable storage, and pushes the stored regulatory certificate to the intellectual property ownership verification unit simultaneously.

[0014] As a further improvement to this technical solution, the full-process compliance verification module includes an execution rule matching submodule, a phased execution verification submodule, a comprehensive compliance judgment submodule, and an objection data marking submodule, wherein: The execution rule matching submodule is used to accurately match the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements with the transaction type and intellectual property subject type corresponding to the contract execution record, and lock in the full set of rules applicable to this verification. The phased execution verification submodule is used to verify the execution status, risk assessment process, circuit breaker action and data link integrity of each phase of the contract's phased execution according to the matched verification rules, and output the individual verification results of each phase. The compliance comprehensive judgment submodule is used to complete the comprehensive compliance judgment of the entire contract execution process based on the individual verification results of each stage, combined with the consistency and closed-loop of the data throughout the process, and output the verification pass conclusion or verification fail conclusion. The objection data marking submodule is used to accurately locate and mark abnormal data and non-compliant items that fail verification, generate corresponding objection explanations, and synchronize them to the regulatory certificate generation module.

[0015] As a further improvement to this technical solution, the intellectual property ownership verification unit includes a verification data collection module, a full-process ownership traceability verification module, an ownership verification result generation module, and a verification result storage and synchronization module, wherein: The verification data collection module is used to receive contract execution records pushed by the contract automated execution unit and regulatory certificates pushed by the contract supervision and verification unit, and simultaneously retrieve the full amount of on-chain evidence data of the corresponding property rights transaction to complete the integrity verification, standardized collection and transaction link alignment of all data. The ownership traceability and verification module, based on the full amount of data collected by the verification data collection module and combined with the official ownership ledger data of intellectual property, conducts full-link traceability and verification of the ownership status of intellectual property targets before the transaction, the entire process of phased transfer, and the compliance of each transaction node. It identifies ownership defects, transfer gaps and compliance risks, and outputs the individual verification results of each link. The ownership verification result generation module, based on the single verification result output by the ownership full-process traceability verification module, completes the comprehensive judgment of the compliance of intellectual property ownership transfer and generates an ownership verification result uniquely corresponding to this transaction. The ownership verification result includes the target ownership status, the full-process traceability verification conclusion, the compliance judgment result and the unique verification identifier. The verification result storage and synchronization module is used to write the generated ownership verification result into the blockchain distributed node to complete the on-chain immutable storage, and to push the stored ownership verification result to the nodes of both parties to the transaction, the regulatory node and the intellectual property transaction publicity platform to complete the closed loop of the entire ownership verification process.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs an automated smart contract execution and supervision system for the entire property rights transaction process. It combines a temporal attention mechanism with an improved lightweight residual temporal convolutional network to achieve dynamic risk assessment and real-time calibration of the transaction performance process. It establishes a hierarchical execution, suspension, termination, and reversible rollback mechanism linked to the risk assessment results. At the same time, it builds a closed-loop architecture for the entire process of contract generation, automated execution, regulatory verification, and intellectual property ownership verification. This enables real-time monitoring, hierarchical handling, and reversible recovery of abnormal transaction execution states, ensuring the stability and reliability of the property rights transaction execution process. It also enables the linkage verification of transaction data throughout the entire process and the traceability of ownership across the entire chain, adapting to the compliance supervision and dynamic risk control needs of the entire property rights transaction process. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system framework of the present invention; The meanings of the labels in the diagram are as follows: 1. Smart contract generation unit; 11. Information collection module; 12. Rule integration module; 13. Contract generation module; 14. Contract verification module; 2. Contract Automated Execution Unit; 21. Execution Sequence Decomposition Module; 22. Dynamic Risk Assessment Module; 23. Circuit Breaker Reversible Handling Module; 24. Execution Record Generation and Push Module; 3. Contract supervision and verification unit; 31. Execution record collection module; 32. Full-process compliance verification module; 33. Supervision certificate generation module; 34. Certificate storage and push module; 4. Intellectual Property Ownership Verification Unit; 41. Verification Data Collection Module; 42. Ownership Full-Process Traceability Verification Module; 43. Ownership Verification Result Generation Module; 44. Verification Result Storage and Synchronization Module. Detailed Implementation

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

[0019] like Figure 1 As shown, this embodiment provides a smart contract automated execution and monitoring system for the entire property rights transaction process, including: The smart contract generation unit 1 is used to collect basic transaction information including information on the main body of the property rights transaction, information on the ownership of intellectual property rights and transaction constraints, and integrate preset contract execution technical rules including performance node rules, risk assessment rules, hierarchical execution conditions and reversible rollback rules to generate a standardized smart contract containing basic transaction information and contract execution technical rules. After completing the executability verification, it outputs a deployable standardized smart contract. In this embodiment, the smart contract generation unit 1 includes an information collection module 11, a rule integration module 12, a contract generation module 13, and a contract verification module 14, wherein: The information collection module 11 is used to collect basic transaction information, including information on the parties involved in the property rights transaction, information on the ownership of intellectual property rights, and transaction constraint clauses; Specifically, the information collection module 11 connects to the property rights transaction filing platform, the intellectual property rights ownership registration and publicity system, and the transaction entity qualification verification channel through a preset standardized data interface. It automatically collects the identity information, entity qualification, credit status, and other entity information of the property rights transaction entity; the type, registration number, ownership, validity period, and right restriction status of the intellectual property rights; and the transaction constraint clauses such as transaction consideration, payment method, ownership transfer triggering conditions, and liability for breach of contract. All collected information is standardized in terms of fields, format, and redundant data is removed to form standardized transaction basic information that meets the requirements for contract generation.

[0020] Furthermore, the information collection module 11 can also take confidentiality measures such as encrypted storage, access control isolation, and compliant transmission of the collected personal information, trade secret information, and intellectual property confidential information, strictly follow the relevant norms of information security and personal information protection, and ensure the confidentiality and security of information.

[0021] Rule integration module 12 integrates preset contract execution technical rules, including performance node rules, risk assessment rules, tiered execution conditions, and reversible rollback rules, among which: The performance node rules refer to the technical rules used to define the execution order, execution nodes, and corresponding execution content of the intellectual property ownership transfer and transaction settlement process agreed upon in standardized smart contracts. Risk assessment rules refer to the technical rules used to clarify the objects of risk assessment, the scope of information collection, and the assessment logic for each stage of contract execution. Tiered execution conditions refer to the technical criteria used to define the execution status of normal execution, suspended execution, and terminated execution corresponding to the risk assessment results of each execution stage of a contract. Reversible rollback rules refer to technical execution rules used to restore the transaction state when contract execution is triggered to a halt or termination. Specifically, the rule integration module 12 classifies, integrates, uniformly encapsulates, and adapts the preset performance node rules, risk assessment rules, tiered execution conditions, and reversible rollback rules. Based on the type of subject matter and transaction mode of this intellectual property transaction, it accurately matches and calls the corresponding rule entries, clarifies the execution order, execution node division, and execution content of each node in the performance node rules for the transfer of intellectual property rights and transaction settlement, clarifies the risk assessment objects, information collection scope, and core assessment logic of each execution stage in the risk assessment rules, clarifies the corresponding judgment criteria for the risk assessment results and the three execution states of normal execution, suspended execution, and terminated execution in the tiered execution conditions, and clarifies the restoration method and execution steps for the transaction funds, ownership, and process status when the contract execution is triggered to be suspended or terminated in the reversible rollback rules, forming a complete and usable set of contract execution technical rules.

[0022] The contract generation module 13 generates a standardized smart contract containing information on the property rights transaction subject, intellectual property rights ownership information, transaction constraint clauses, and contract execution technical rules based on the information collected by the information collection module 11 and the contract execution technical rules integrated by the rule integration module 12. Specifically, the contract generation module 13 calls the system's built-in standardized smart contract template for intellectual property transactions, accurately fills in the standardized transaction basic information output by the information collection module 11 according to the corresponding fields of the template, embeds the contract execution technical rules output by the rule integration module 12 into the contract execution logic framework, completes the matching and fusion of the transaction basic information and the contract execution technical rules, and automatically generates a standardized smart contract with a complete structure, unified logic, and containing all necessary information and execution rules.

[0023] The contract verification module 14 is used to perform executability verification on standardized smart contracts and output deployable standardized smart contracts.

[0024] Specifically, the contract verification module 14 performs full-dimensional executability verification on the standardized smart contract generated by the contract generation module 13. The verification content includes verification of the integrity of basic transaction information, verification of the compliance of contract execution technical rules, verification of the closed-loop nature of contract execution logic, and verification of blockchain deployment format compatibility. It confirms that the contract has no missing information, rule conflicts, logical loopholes, or format errors. After completing all executability verifications and passing them, it outputs a standardized smart contract that can be directly deployed to a blockchain node.

[0025] Contract Automated Execution Unit 2 loads the standardized smart contract output by Smart Contract Generation Unit 1. It maps the intellectual property ownership transfer and transaction settlement process stipulated in the standardized smart contract into a multi-stage execution sequence of corresponding performance nodes. Each execution stage requires performance risk assessment as a prerequisite. Before execution, multi-dimensional temporal feature collection is used to obtain real-time status temporal data of the intellectual property target and transaction entity, historical performance risk data, and industry policy risk temporal data. A temporal attention mechanism and an improved lightweight residual temporal convolutional network are used to complete real-time risk assessment, combined with… Historical risk data and policy risk coefficients are dynamically calibrated to output accurate risk assessment results; if the assessment meets the execution conditions, the corresponding action for the current stage is executed through atomic operations of smart contracts; if the assessment meets the conditions for suspension or termination, the execution is suspended or the entire process is terminated accordingly, and reversible rollback processing is performed; contract execution records with on-chain evidence are generated simultaneously and pushed to the contract supervision and verification unit 3 and the intellectual property ownership verification unit 4; the contract automated execution unit 2 includes an execution sequence decomposition module 21, a dynamic risk assessment module 22, a circuit breaker reversible handling module 23, and an execution record generation and push module 24, wherein: In this embodiment, the execution sequence decomposition module 21 maps the intellectual property ownership transfer and transaction settlement process into a multi-stage execution sequence corresponding to the performance node based on the standardized smart contract output by the smart contract generation unit 1. The execution sequence decomposition module 21 is used to perform structured parsing of the transaction execution content in the standardized smart contract, and convert the performance arrangement that originally existed in the form of contract terms into a staged execution process that can be identified, arranged, and triggered by the system.

[0026] Specifically, the execution sequence decomposition module 21 can read the transaction entity information, intellectual property information, transaction consideration information, performance constraints, time limits, and corresponding performance node rules from the standardized smart contract, and extract key items related to automatic execution according to a preset mapping relationship. The preset mapping relationship is a pre-constructed one-to-one mapping rule between contract fields and execution fields, specifically including: First, the transaction entity information is mapped, which maps the names of Party A and Party B, their unified social credit codes, and the entity qualification document numbers in the contract to the unique identifier of the executing entity and the entity qualification verification field. Second, the information mapping of intellectual property rights involves mapping the patent / trademark / copyright name, application number / registration number, owner information, and rights status in the contract to the unique identifier of the target and the ownership status verification field. Third, the transaction consideration information is mapped, which maps the total consideration amount, payment node, payment amount, and payment conditions in the contract to consideration control fields and payment trigger condition fields; Fourth, the contractual obligations and time constraints are mapped, transforming the contractual obligations, breach clauses, and time requirements into execution trigger condition fields and timeout verification fields. Based on the above-mentioned preset mapping relationship, the execution sequence decomposition module 21 extracts and structures the contract fields, outputting a set of key execution items that the system can recognize.

[0027] In actual processing, the execution sequence decomposition module 21 does not simply list the contract terms, but transforms each performance requirement into a node-based execution unit with clear triggering conditions, execution objects, execution actions, and completion statuses. Then, based on the dependencies between nodes, it organizes these units into a sequence of interconnected execution stages. This approach transforms the intellectual property transaction process from a static description of terms into a dynamic execution chain, providing clear stage boundaries and a basis for subsequent dynamic risk assessment, circuit breaker-based reversible handling, and execution record generation.

[0028] The execution sequence decomposition module 21 includes a fulfillment node matching submodule, a stage sequence division submodule, and an execution connection triggering submodule, wherein: The performance node matching submodule is used to match the performance requirements stipulated in the standardized smart contract with the actual performance nodes of the property rights transaction. Specifically, the performance node matching submodule is used to extract various performance requirements from standardized smart contracts and map them to actual execution nodes in the property rights transaction process. These performance requirements may include, but are not limited to, confirmation of subject qualification, confirmation of the ownership status of the target, freezing or escrow of transaction funds, submission of ownership change application, confirmation of registration results, and settlement of final payment.

[0029] Furthermore, the performance node matching submodule can pre-establish a rule base corresponding to "contract clause type - performance node type". The rule base is built based on the BERT pre-trained language model and includes a built-in labeled dataset of intellectual property transaction contract clauses. The labeled dataset covers 32 subcategories of clause types in 5 major categories: subject qualification, ownership of the subject matter, consideration payment, ownership transfer, and compliance verification. Each clause type is bound to a unique corresponding performance node type. After reading the clause text of the standardized smart contract, the performance node matching submodule classifies the clauses using a text classification model, outputs clause type labels, and then completes the one-to-one matching of clause types with performance node types based on the rule base. For example, when a constraint clause related to transaction consideration payment is detected, it can be matched to the payment control node; when a clause related to intellectual property change registration is detected, it can be matched to the ownership transfer node. In this way, the complete transaction process can be divided into four core execution stage sequences: transaction preparation stage, consideration control stage, ownership transfer stage, and settlement completion stage. Each stage sequence has a built-in unique stage identifier, dependency identifier, and execution status identifier.

[0030] The phase sequence division submodule is used to divide the intellectual property ownership transfer and transaction settlement process into multiple execution phase sequences based on the matched performance nodes; Specifically, the phase sequence division submodule is used to group and sort the matched fulfillment nodes according to the dependencies between fulfillment nodes, business attributes and execution order, forming several interconnected execution phase sequences.

[0031] Specifically, if the completion result of a certain performance node constitutes a prerequisite for the start of the next node, the stage sequence division submodule configures the two as having a sequential execution relationship; if multiple nodes are in the same business stage and there is no strong dependency between them, they can be classified as parallel or combined tasks within the same execution stage. In this way, the complete transaction process can be divided into multiple stage sequences, such as the transaction preparation stage, the consideration control stage, the ownership transfer stage, and the settlement completion stage.

[0032] Each execution phase sequence preferably corresponds to at least one phase objective, at least one input condition, at least one output result, and a phase completion identifier. With this configuration, the system can identify and control the transaction execution status at the phase granularity, and it also facilitates the subsequent dynamic risk assessment module 22 to call the corresponding assessment logic before each phase starts.

[0033] The execution connection trigger submodule is used to set the execution trigger conditions for adjacent execution phase sequences. The next execution phase sequence will only be triggered after the previous execution phase sequence has been completed and the risk assessment is qualified.

[0034] Specifically, the execution transition trigger submodule can set the following trigger conditions for starting the next execution stage. All conditions must be verified in sequence, and the next stage can only be triggered after all verifications pass: First, all fulfillment nodes in the previous stage have reached the completion status. The criteria for determining the completion status of a fulfillment node are that all execution actions of the node have been executed, the node status identifier has been updated to "complete", and the hash verification value of the node execution result is consistent with the preset verification value. Second, the atomic operation of the smart contract corresponding to the previous stage is executed successfully. The criteria for judging the successful execution of the atomic operation are: the operation transaction is successfully submitted, the on-chain transaction receipt returns the "SUCCESS" status, there are no abnormal errors in the transaction execution status log, and the operation result is completely consistent with the execution content agreed upon in the contract. Third, the previous stage has generated a valid execution result. The criteria for determining a valid execution result are that a structured execution report containing the stage identifier, execution time, execution subject, execution content, execution status, and result hash value has been generated, and the report has been stored locally and temporarily stored on the blockchain. Fourth, the risk assessment results output by the dynamic risk assessment module 22 meet the normal execution conditions. The criterion for determining the normal execution conditions is that the real-time transaction risk value at the current stage is ≤ the preset risk warning threshold.

[0035] When all of the above conditions are met, the execution connection trigger submodule sends a start signal to the next stage; if any of the conditions are not met, the next stage remains in a waiting or locked state, and the specific reasons for the failure to meet the conditions and the verification details are pushed to the transaction entity.

[0036] This trigger control method ensures that the multi-stage execution sequence proceeds in a predetermined order, while making the risk assessment results a prerequisite for stage release, thereby enhancing the security, controllability, and consistency of the entire automated contract execution process.

[0037] In this embodiment, the dynamic risk assessment module 22, based on the execution phase sequence formed by the execution sequence decomposition module 21, collects multi-dimensional temporal status features of the intellectual property target, credit and temporal performance data of the transaction entity, and temporal data of industry policy risks before each execution phase is started. It dynamically allocates features and time step weights through a temporal attention mechanism, extracts deep temporal correlation features by combining an improved lightweight residual temporal convolutional network, and introduces historical risk data and policy risk coefficients for dynamic risk calibration to complete a high-precision real-time risk assessment and output a risk assessment result containing risk value, attention weight, and calculation process. Specifically, the dynamic risk assessment module 22 is used to uniformly model and quantitatively assess the status of the intellectual property target, the performance status of the transaction entity, and the external industry policy environment corresponding to the current stage before the formal start of each execution stage, and outputs the real-time transaction risk value corresponding to the current execution stage as a preliminary judgment basis for whether the stage can continue. The processing chain of the dynamic risk assessment module 22 strictly follows the sequence of "feature collection → preprocessing → feature fusion → deep extraction → risk calibration → output", thereby ensuring that the risk assessment process has clear data sources, clear calculation steps, and verifiable result output.

[0038] Furthermore, the improvements to the dynamic risk assessment module 22 are mainly reflected in the following two aspects: First, a temporal attention mechanism is introduced in the feature fusion stage. Instead of treating all temporal features equally, attention weights are dynamically allocated according to the impact of different feature categories and different time steps on the risk of the current stage, thereby improving the ability to identify key risk signals. Secondly, an improved lightweight residual temporal convolutional network is used in the deep feature extraction stage. The model parameters and computational complexity are reduced by depthwise separable convolution, and the effective information in the original weighted features is preserved by residual connections, so as to improve the accuracy of temporal risk feature extraction while meeting the requirements of real-time processing.

[0039] Therefore, the dynamic risk assessment module 22 can not only adapt to multi-source heterogeneous and time-correlated risk data in property rights transaction scenarios, but also meet the requirements of automated execution scenarios for response speed and computational stability.

[0040] The risk assessment process of the dynamic risk assessment module 22 is executed sequentially according to the following steps, forming a complete algorithm chain of "feature acquisition → preprocessing → feature fusion → deep extraction → risk calibration → output", including the following steps: S22.1 Collect multi-dimensional temporal features and construct the original temporal feature matrix. The temporal characteristics include five categories of ownership status of intellectual property: pledge status, freeze status, invalidation status, objection status, and infringement litigation status, as well as the credit score of the transaction entity, performance compliance, and industry policy risk index. Bank of China corresponds to 8 types of features, and columns correspond to time steps. , For the first Class features in the first Feature values ​​of the time step; Specifically, before the current execution phase begins, the dynamic risk assessment module 22 simultaneously collects multi-dimensional time-series features from data sources including the status of intellectual property targets, the performance data of transaction entities, and industry policy data. These features are then organized into an original time-series feature matrix according to a unified time window. Among them, the characteristics related to intellectual property rights reflect the stability of the ownership status and potential disputes; the characteristics related to the transaction entities reflect their creditworthiness and performance reliability; and the characteristics related to industry policies reflect the impact of changes in the external institutional environment on the risks at the current stage. By organizing the above characteristics into the same time series matrix, a unified data input foundation can be provided for subsequent standardization processing, attention weighting, and deep time series extraction.

[0041] Furthermore, the original time-series feature matrix The rows correspond to 8 types of features, and the columns correspond to time steps. Among them, categories 1 to 5 correspond to pledge status, freeze status, invalidation status, objection status, and infringement litigation status, respectively; category 6 corresponds to the credit score of the transaction entity; category 7 corresponds to the compliance with performance obligations; and category 8 corresponds to the industry policy risk index. If the risk assessment window covers the most recent... The original time series feature matrix is ​​obtained at each time step. Each column in the table corresponds to one of the eight feature values ​​at a given time step, and each element... Corresponding to the Class features in the first The specific value at each time step.

[0042] The core quantization rules for the 8 types of features are as follows: The time series features of ownership status for categories 1 to 5 are as follows: using a 0-1 binary quantization method, the feature value is 1 if the corresponding state exists within the corresponding time step, and 0 if there is no corresponding state. Credit scores for Category 6 transaction entities: Normalization is performed using the 0-1 range. The original scores are sourced from the public credit information center or compliant third-party credit agencies, and normalization is performed by "original score / 100". Category 7 Compliance Performance: Calculated based on the entity's historical transaction defaults, with a value range of 0-1. The calculation formula is "Compliance Performance = 1 - (Number of Historical Defaults / Total Number of Historical Transactions)". The historical transaction data statistics window is the most recent 2 years. If there are no historical transaction records, the compliance performance score is set to 0.5 by default. Category 8 Industry Policy Risk Index: This index is quantified based on current intellectual property regulatory policies, compliance guidelines, and other environmental information, and its value ranges from 0 to 1.

[0043] This method allows risk information from different sources and with different meanings to be integrated into a unified time-series matrix representation, as follows: ; In the formula, Represents the feature category index, and ; Indicates the time step index.

[0044] S22.2, On the original time series feature matrix Outlier truncation and standardization are performed to obtain the standardized time series feature matrix. Eliminate dimensional differences and data noise; Specifically, step S22.2 is used to modify the original time-series feature matrix constructed in step S22.1. Preprocessing is performed to eliminate dimensional differences between different types of features and to reduce the interference of outliers and noisy data on subsequent risk assessment results. Because features related to intellectual property ownership status, transaction entity credit, and industry policy risk differ significantly in their value range, data type, and magnitude of variation, directly inputting them into subsequent calculations could easily lead to an abnormal amplification of the overall assessment results by some high-value features. Therefore, this step first truncates outliers and then standardizes the truncated features to obtain a standardized time-series feature matrix. .

[0045] Furthermore, outlier truncation can prune continuous features based on preset upper and lower thresholds, causing feature values ​​exceeding the threshold range to fall back to their corresponding boundaries. Standardization is used to map features of different dimensions and numerical ranges to a unified scale, thereby ensuring comparability of features in subsequent attention weight allocation and deep temporal extraction. After step S22.2, each element in the matrix is ​​derived from the original feature values. Convert to standardized eigenvalues And constitute a standardized time series feature matrix. After this step, all input data processed in subsequent steps are based on the elimination of dimensional differences and suppression of anomalous perturbations. Standardized time series feature matrix. The specific representation is as follows: .

[0046] S22.3 Calculate the attention weights for each feature and time step using a temporal attention mechanism. The standardized features are then weighted and fused to obtain an attention-weighted temporal feature vector. Attention weight The calculation is as follows: ; in: For the first Class features in the first Attention weights at time steps; For the first Weight coefficients of class features; For the first Class features in the first Standardized eigenvalues ​​of the time step; For the first Bias terms for class features; It is a linear rectification activation function; It is a natural exponential function; Traverse the index for feature categories; Index for time step traversal; This represents the total number of time steps for the temporal features. Specifically, step S22.3 is used to normalize the time-series feature matrix. Building upon this foundation, a temporal attention mechanism is used to calculate the relative importance of different feature categories and time steps to the risk outcome of the current execution stage, thereby highlighting key risk signals. Compared to applying equal weights to all temporal features, this step assigns different attention weights to each feature-time unit. This allows features and time steps with a higher degree of correlation to the current risk to occupy a larger proportion in the fusion result, while features and time steps with a lower degree of correlation are relatively suppressed, thereby enhancing the risk assessment's ability to identify key abnormal trends and key risk states.

[0047] Specifically, for any standardized eigenvalue First, through linear transformation Reflecting the Class features in the first The basic risk contribution at each time step, and then through Preserve the non-negative response, and then pass... Amplify the differences between different feature units, and finally analyze all of them. The exponential results of each feature-time unit are normalized to obtain the corresponding attention weights. Therefore, the sum of all attention weights is 1, and each weight represents the relative contribution of the corresponding feature unit to the risk assessment at the current stage.

[0048] Furthermore, after obtaining the attention weights Then, weighted fusion is performed on the standardized features to form an attention-weighted temporal feature vector. This vector is used to represent the temporal fusion result after introducing feature importance and temporal importance, and serves as the input basis for subsequent deep temporal convolutional networks. Considering the use of symbolic representation in step S22.4... As input to the convolutional network, in this embodiment, it can be Considered as the output representation of the attention-weighted temporal feature vector, its input form in step S22.4 is uniformly denoted as... Both refer to the same attention-weighted temporal feature result, namely: .

[0049] This embodiment also provides the following calculation example: Suppose that the standardized feature value of the sixth feature, "transaction entity credit score", at the fourth time step is And take the weight coefficient of this type of feature. Bias terms Then the activation value of this feature unit is: ; If the sum of the exponential activations of all feature-time units is 40.00, then the attention weight for that feature unit is: ; Then its weighted characteristic contribution is ; In the formula, This represents the attention weight of the 6th category feature at the 4th time step; This represents the standardized feature value of the 6th type feature at the 4th time step.

[0050] As can be seen from this example, step S22.3 does not simply superimpose all features, but rather uses a temporal attention mechanism to prioritize features that contribute more to the current risk assessment—the time unit—thereby improving the targeting of risk identification.

[0051] S22.4, Weighted time series feature vector Input an improved lightweight residual temporal convolutional network, and extract deep temporal fusion features through depthwise separable convolutions and residual connections. Deep temporal fusion features The extraction formula is as follows: ; in: This is a depthwise separable convolution operation; Attention-weighted temporal feature vectors; The kernel size for depth-separable convolutions; For dimension-aligned residual connection operations, the input features can be transformed into a feature matrix with the same dimensions as the convolution output, ensuring that the two items can be added element-wise. Specifically, step S22.4 is used to apply attention-weighted temporal feature vectors Building upon this foundation, higher-level temporal correlation features are further extracted to reflect the evolutionary trends and interactions of different risk signals across continuous time. Since risks in property rights transactions are typically not determined by a single feature or a single moment, but rather manifest as interconnected, cumulative, or abrupt changes in several features across multiple adjacent time steps, the weighted features obtained in step S22.3 alone are insufficient to fully characterize potential risks. This step improves upon these features by using a lightweight residual temporal convolutional network. Further modeling will be conducted to obtain deeper temporal fusion features that are more suitable for risk assessment. .

[0052] Furthermore, the improvements to the improved lightweight residual temporal convolutional network include: replacing conventional convolutions with depthwise separable convolutions to reduce the number of model parameters and computational complexity while maintaining the ability to extract local temporal patterns; and introducing residual connections to enable the input... The result can be directly added to the convolutional extraction result, thus preserving effective information from the original weighted features while extracting deep features. These improvements give the network both better temporal feature representation capabilities and higher computational efficiency and stability, making it suitable for deployment in smart contract execution scenarios requiring real-time responses. Specifically, the convolutional kernel size is fixed at [size missing]. It is used to extract local evolution features within a range of three adjacent time steps, enabling the system to effectively identify typical risk patterns such as short-term continuous anomalies, continuous risk accumulation, and local risk mutations.

[0053] This implementation also provides the following calculation example: If for a certain local time window... After performing a depthwise separable convolution, the output value is 0.48, and the output value corresponding to the residual connection is 0.12. Therefore, the deep temporal fusion feature at this local location is: ; As can be seen from this example, the output of this step is... It does not rely solely on the convolution result, but rather forms a result through the combined output of convolution and residual, thus balancing the representation of deep temporal patterns with the preservation of original effective information.

[0054] S22.5, combined with historical average performance risk Industry policy risk index The risk value is dynamically calibrated, and the real-time trading risk value is calculated using a normalized activation function. Real-time transaction risk value The calibration formula is as follows: ; in: This is the normalized real-time transaction risk value; This represents the Sigmoid normalized activation function; This represents the weight vector of deep temporal fusion features; For bias terms of fully connected layers; This is the historical performance risk weighting coefficient; This represents the average performance risk during the same historical period. This refers to the industry policy risk weighting coefficient. This is an industry policy risk index; Specifically, step S22.5 is used to fuse features in deep temporal sequences. Based on this, the risk results of the current execution phase are dynamically calibrated, and a real-time transaction risk value that can be directly used for subsequent threshold judgment is output. Unlike methods that directly output risk results based solely on deep characteristics, this step uses the historical average performance risk for the same period. Industry policy risk index By introducing a risk value calculation process, the final output risk value not only reflects the immediate risk characteristics inherent in the current time-series data, but also the combined impact of historical performance experience and changes in the external policy environment on the security of execution at the current stage. Therefore, this step achieves dynamic risk calibration based on multi-source information.

[0055] Specifically, The weighted mapping result representing the deep temporal fusion features is used to reflect the basic risk level determined by the internal temporal features at the current stage; Used to calibrate the impact of historical risk levels on current risk outcomes; This is used to calibrate the impact of changes in the industry policy environment on current risk outcomes. After summing the above terms, the activation function is then normalized using the Sigmoid function. Mapping the results to the 0-1 range yields the normalized real-time trading risk value. This allows for a unified standard for risk outcomes at different stages of implementation, facilitating subsequent comparisons of risk thresholds and tiered handling.

[0056] Understandably, this is to ensure real-time transaction risk levels. The calculation has a clear data source and a feasible, definite path, based on the historical average performance risk. Industry policy risk index All parameters are obtained and determined by the dynamic risk assessment module 22 according to preset rules, rather than by abstract given parameters.

[0057] Among them, the average historical performance risk This is used to characterize the average performance risk level in historical periods that are the same as or similar to the current execution stage. Specifically, the dynamic risk assessment module 22 can retrieve historical performance records corresponding to the current execution stage from the historical transaction sample library, prioritize selecting historical samples that are the same as or similar to the current transaction type, intellectual property subject type, execution stage type, and transaction constraints, extract the historical risk value of each historical sample in the corresponding stage, and calculate the average of the extracted historical risk values ​​to obtain the historical performance risk average. .

[0058] In one alternative implementation, if the number of historical samples is The historical risk values ​​of each historical sample at the corresponding stage are as follows: The historical average performance risk It can be determined by the following formula: ; In the formula, Indicates the first Historical risk values ​​of each historical sample at the corresponding execution stage; This represents the historical sample index.

[0059] In practice, the aforementioned historical risk value This can be the stage risk result output by a similar risk assessment model at the corresponding execution stage of a historical transaction, or it can be a risk characterization value obtained after normalizing the performance anomaly rate, default rate, suspension rate, or termination rate of a historical stage. Through these methods, It has clear data sources and statistical foundations, and can reflect the reference role of historical experience in risk assessment at the current stage.

[0060] Among them, the industry policy risk index This is used to characterize the impact of the current industry policy environment on the execution risk of intellectual property transactions. Specifically, the dynamic risk assessment module 22 can obtain policy texts, regulatory notices, compliance guidelines, or risk warnings related to the industry, type of intellectual property, and stage of the transaction from preset policy information sources, and quantify and score the obtained policy information to form an industry policy risk index. .

[0061] In one alternative implementation, factors such as the degree of transaction restriction, the intensity of compliance review, the complexity of approval, the sensitivity of regulation, and the frequency of institutional changes can be assigned corresponding scores based on policy information, and then determined by a weighted summation method. If each policy risk factor is denoted as... The corresponding weights are denoted as follows: The industry policy risk index It can be determined by the following formula: ; In the formula, Indicates the first The weighting coefficients of each policy risk factor; Indicates the first The score values ​​of each policy risk factor; This indicates the policy risk factor index.

[0062] In one specific implementation, the policy risk factor score The risk level can be determined using a pre-defined tiered scoring method, for example, assigning values ​​of 0.2, 0.5, and 0.8 to low, medium, and high risk respectively, or using a normalized score within the range of 0 to 1. Through these methods, It is not an abstract variable, but a risk index obtained by quantitative calculation based on the relevant policy environment information at the current stage. It can objectively reflect the risk impact of external policy changes on the current transaction execution stage.

[0063] Furthermore, to facilitate integration with real-time transaction risk values... Unified calculation, and Ideally, all values ​​should be processed into normalized values ​​within the range of 0 to 1. This allows... , The risk representations corresponding to the deep time series fusion features participate in the risk calibration calculation together on the same numerical scale, thereby improving the stability and comparability of the risk value calculation in step S22.5.

[0064] In addition, this embodiment also provides the following calculation example: Suppose that in a certain execution phase: ; ; The result of the linear combination is: ; If the Sigmoid function is used: ; In the formula, Indicates input variables; The real-time trading risk value is: ; This example demonstrates that when deep-seated time-series risk characteristics, historical risks from the same period, and industry policy risks all have a certain impact on the current stage, step S22.5 can output a real-time transaction risk value under a unified scale. This provides a quantitative basis for subsequent risk classification judgments.

[0065] S22.6 Output Trading Risk Value The corresponding core calculation data is pushed to the circuit breaker reversible handling module 23 to complete the entire risk assessment process.

[0066] Specifically, step S22.6 is used to encapsulate the result output of the dynamic risk assessment module 22. This is done after step S22.5 calculates the real-time transaction risk value for the current execution phase. Subsequently, the dynamic risk assessment module 22 outputs the risk value along with the corresponding core calculation data from this assessment process. The core calculation data includes at least the standardized time-series feature data and attention weights used in the assessment. Attention-weighted temporal feature vectors Or corresponding input representation Deep temporal fusion features Historical average performance risk Industry policy risk index In addition, the risk calibration calculation results are output simultaneously. By outputting the above data synchronously, the risk assessment process becomes traceable and interpretable, and provides complete input for subsequent modules to perform risk comparison and tiered handling.

[0067] Furthermore, step S22.6 is only used to standardize the output of the evaluation results of the dynamic risk assessment module 22, and uses this output as the input basis for subsequent circuit breaker reversible handling, without introducing any additional functions outside this module. Through the continuous execution of steps S22.1 to S22.6, the dynamic risk assessment module 22 forms a complete technical link from multi-dimensional temporal feature acquisition, standardization processing, temporal attention fusion, improved lightweight residual temporal convolution extraction, dynamic risk calibration to risk value output, making this module have a clear, complete and reproducible implementation path.

[0068] In this embodiment, the circuit breaker reversible handling module 23 executes corresponding stage actions or triggers suspension, termination, and reversible rollback processing based on the risk assessment results of the dynamic risk assessment module 22 and the preset hierarchical execution conditions. The hierarchical handling and reversible rollback process of the circuit breaker reversible handling module 23 is executed sequentially according to the following steps, forming a complete link of "threshold preset → risk comparison → hierarchical handling → state restoration → record push", including the following steps: S23.1, Preset risk warning threshold With risk termination threshold The two thresholds satisfy the core numerical relationship: ; Specifically, since the real-time transaction risk value RRR output by the dynamic risk assessment module 22 is a normalized continuous value, in order to transform this risk value into a clear decision-making process, at least two risk judgment thresholds need to be pre-set, namely risk warning thresholds. and risk termination threshold Among them, the risk warning threshold The lower bound used to define the transition of a transaction from a normal execution state to a risk-terminated state; risk termination threshold. This is used to define the upper bound of a transaction's transition from a suspended state to a fully terminated state with reversible rollback. By setting this dual-threshold structure, the risk status during transaction execution can be divided into low-risk, medium-risk, and high-risk zones.

[0069] In this embodiment, the reason for requiring The above-mentioned size relationship is satisfied because the real-time transaction risk value output by the dynamic risk assessment module 22... Located within the 0 to 1 range, the risk warning threshold and risk termination threshold are preferably set within the same range. Simultaneously, by ordering... Less than This ensures that the suspended execution state and the terminated execution state are distinguished from each other in terms of numerical range, avoiding system judgment conflicts. Therefore, step S23.1 provides a clear judgment benchmark for subsequent risk comparison and graded handling.

[0070] Furthermore, risk warning threshold and risk termination threshold The two thresholds satisfy the core numerical relationship: Risk warning threshold and risk termination threshold Based on a comprehensive consideration of historical transaction risk distribution, business tolerance requirements, and regulatory prudence requirements, the core setting rule is as follows: Based on the statistical distribution of historical risk values ​​for similar intellectual property transactions, corresponding quantile values ​​are selected as the baseline values ​​for the two thresholds, which are then adjusted according to the security requirements of the specific transaction scenario. For example, in scenarios with high transaction security requirements, the threshold can be adjusted accordingly. Set to a lower value, Set the threshold to a medium-low value to improve the system's sensitivity to risk; in scenarios with mature trading processes and high historical stability, the threshold value can be appropriately increased to reduce unnecessary suspension and termination triggers.

[0071] S23.2 Obtain the transaction risk value output by the dynamic risk assessment module 22 , to trade risk value The risk level is determined by comparing the execution values ​​with the two-level risk thresholds. Specifically, the circuit breaker reversible handling module 23 receives the real-time transaction risk value corresponding to the current execution stage. Then, the risk value is compared with the risk warning threshold preset in step S23.1. and risk termination threshold Numerical comparisons are performed, and the results are used to determine whether the current stage falls within the normal execution risk range, the suspended execution risk range, or the terminated execution risk range.

[0072] Specifically, step S23.2 only completes the risk level determination and does not directly implement the execution action. Its role is to establish a mapping relationship between continuous risk values ​​and discrete disposal states. Through this step, the quantitative risk results output by the dynamic risk assessment module 22 can be further transformed into decision control signals in the automatic contract execution process, providing a basis for judgment in the subsequent graded disposal in steps S23.3 and S23.4.

[0073] Furthermore, if the real-time transaction risk value Less than or equal to the risk warning threshold If so, it can be determined that the current stage is within the normal execution risk range; if the real-time transaction risk value Greater than the risk warning threshold And less than the risk termination threshold If so, it can be determined that the current stage is in the risk range of execution suspension; if the real-time transaction risk value Greater than or equal to the risk termination threshold If the current stage falls within the risk range for termination of execution, it can be determined that the risk level is clearly defined through the above-mentioned range division, providing clear triggering conditions for various actions. This normal execution range is naturally derived from the threshold relationship between steps S23.3 and S23.4, and is consistent with the hierarchical handling logic described above.

[0074] S23.3, when When this occurs, a stop execution instruction is triggered, suspending the current execution phase sequence and pushing risk warning information to the trading entity and regulatory nodes; Specifically, when step S23.2 determines that the real-time transaction risk value RRR in the current execution phase is greater than the risk warning threshold... And less than the risk termination threshold When this happens, the circuit breaker reversible handling module 23 does not immediately terminate the entire process, nor does it allow the current stage to continue execution. Instead, it triggers a stop execution command, putting the current execution stage into a pause state. This pause state is used to prevent the risk from spreading further if it is above the normal level but has not yet reached the termination condition, while reserving processing space for subsequent manual verification, supplementary materials, status restoration, or reassessment.

[0075] Furthermore, the so-called "pausing the current execution phase sequence" refers to freezing the unfinished smart contract execution actions of the current phase, preventing this phase from continuing to subsequent operations, and maintaining the subsequent execution phases in an unstartable state. Simultaneously, the circuit breaker reversible handling module 23 pushes risk warning information to the trading entity and regulatory nodes to indicate that there is an abnormal risk in the current execution phase, requiring further verification or handling. The risk warning information may include, but is not limited to, the current execution phase identifier, the current real-time trading risk value, the current risk range, the suspension time point, and the corresponding phase status information. In this way, the system can not only prevent the further propagation of risk but also provide clear warning basis for subsequent risk handling.

[0076] S23.4, when When this occurs, a full-process termination instruction is triggered, and the transaction assets are returned via the original payment method through atomic operations of the smart contract. The asset return relationship is as follows: ;in, To return the amount of assets; This represents the amount of assets that have already been transferred and traded. Specifically, when step S23.2 determines the real-time transaction risk value of the current execution stage... Greater than or equal to the risk termination threshold At this point, the circuit breaker reversible handling module 23 triggers a full-process termination instruction, switching the current transaction from automatic execution to termination, and executing the return of the transaction assets via the original path through atomic operations of the smart contract. The handling corresponding to this step is not simply stopping the current stage, but rather reversing the flow state of the transaction assets that has already occurred to restore the consistent state before the transaction execution, preventing incomplete, asymmetric, or irreversible ownership and settlement results under high-risk conditions.

[0077] Specifically, the "smart contract atomic operation execution transaction asset rollback" described above refers to integrating executed but still recoverable transaction asset transfer actions into a single atomic transaction. This transaction includes reverse rollback operations for all executed actions. This atomic transaction follows the rule of "either complete execution succeeds or the whole transaction fails." Only when all rollback operations are executed successfully is the transaction committed and effective, completing the full restoration of the transaction assets and state. If any rollback operation fails, the transaction is rolled back to the state before the commit, thus avoiding the abnormal situation where only some assets are rolled back and some states remain. Furthermore, the aforementioned asset return relationship indicates that upon triggering the termination of the entire process, the system will perform an equal-amount return based on the actual amount of transaction assets that have been transferred, without processing the return of any portion that has not yet been transferred. In other words, the scope of the return is strictly determined based on the actual transfer results completed prior to the current stage, ensuring that the return processing is consistent with the actual execution status. This avoids issues of insufficient or excessive returns and guarantees that the state restoration process has a clear quantitative basis.

[0078] S23.5 Generate circuit breaker handling and reversible rollback records, and synchronize them to the execution record generation and push module 24 to complete the entire process of hierarchical handling.

[0079] Specifically, after the circuit breaker reversible handling module 23 completes the suspension or termination and rollback process, it generates a circuit breaker handling record and a reversible rollback record corresponding to this handling, and synchronizes these records to the execution record generation and push module 24 so as to form a complete and traceable contract execution record. Through this step, the risk handling process is no longer just a momentary control action, but is transformed into execution result information with a clear chain of evidence and traceability attributes.

[0080] Specifically, the circuit breaker handling record may include at least the current execution stage identifier, the current real-time transaction risk value, the corresponding threshold comparison result, the triggered handling type, the time point of suspension or termination, and the corresponding stage status; the reversible rollback record may include at least the rollback trigger condition, the scope of rollback assets, the amount of rollback assets, the amount of originally transferred transaction assets, and the rollback completion status. By synchronizing the above records to the execution record generation and push module 24, a complete data foundation can be provided for subsequent on-chain evidence storage, regulatory verification, and ownership verification.

[0081] In addition, step S23.5 does not introduce new risk judgment logic. Its role is to standardize the recording and transmission of the aforementioned graded disposal results, so that the circuit breaker reversible disposal module 23 forms a complete closed loop from threshold setting, risk comparison, graded disposal, state restoration to result recording synchronization, and makes the entire risk control process have a traceable, verifiable and connectable implementation basis for subsequent processing procedures.

[0082] In this embodiment, the execution record generation and push module 24 generates on-chain evidence-based contract execution records based on the execution status, risk assessment results and disposal information of each execution stage, and pushes them to the contract supervision and verification unit 3 and the intellectual property ownership verification unit 4.

[0083] Specifically, the technical chain of the execution record generation and push module 24 can be executed sequentially in the order of "execution information integration → hash encryption and notarization → on-chain data push". The key technical point of the execution record generation and push module 24 is that it not only simply records the execution results, but also uniformly collects the execution status at each stage, the dynamic risk assessment results, and the circuit breaker reversible handling results. It uses hash encryption and blockchain writing to give the contract execution records unique identification and tamper-proof attributes, and then pushes them synchronously to the contract supervision and verification unit 3 and the intellectual property ownership verification unit 4 through on-chain peer-to-peer transmission, thereby ensuring that the subsequent verification process can call the same, complete, and reliable execution record data.

[0084] The execution record generation and push module 24 includes an execution information integration submodule, a hash encryption evidence storage submodule, and an on-chain data push submodule, wherein: The execution information integration submodule is used to collect the execution status of each execution stage sequence and the transaction risk value output by the dynamic risk assessment module 22. The core computing data, the processing actions of the circuit breaker reversible processing module 23, and the reversible rollback information form a complete and traceable contract execution record; Specifically, the execution information integration submodule is used to uniformly collect and structure the key information generated in each execution stage, and generate the original contract execution record. The execution information includes at least: the execution status information of each execution stage sequence, the transaction risk value and core calculation data output by the dynamic risk assessment module 22, and the handling action information and reversible rollback information output by the circuit breaker reversible handling module 23.

[0085] Furthermore, the execution status information includes stage identifiers, execution time, execution results, and stage status; core calculation data includes risk assessment results and calculation process data corresponding to the transaction risk value; disposal action information includes action identifiers for continuing execution, suspending execution, or terminating execution; reversible rollback information includes rollback trigger conditions, scope of rollback assets, amount of rollback assets, and rollback completion status. By integrating the above information, a complete contract execution record covering execution status, risk results, and disposal results can be formed.

[0086] The hash encryption evidence storage submodule is used to perform hash encryption processing on the contract execution record, generate a hash value that uniquely corresponds to the contract execution record, and write the encrypted contract execution record and the corresponding hash value into the blockchain distributed node to complete the on-chain evidence storage, ensuring that the contract execution record cannot be tampered with. Specifically, the hash encryption evidence storage submodule is used to perform hash encryption processing on the contract execution record formed by the execution information integration submodule, generate a hash value that uniquely corresponds to the contract execution record, and write the contract execution record and the corresponding hash value into the blockchain distributed node to complete the on-chain evidence storage.

[0087] ; In the formula, This represents a hash value that uniquely corresponds to a contract execution record; This indicates the SM3 hash encryption operation. This indicates the contract execution record.

[0088] Through the above processing, a one-to-one correspondence is established between the contract execution record content and the hash value. When the contract execution record content changes, the recalculated hash value will change, thus allowing verification of whether the record content has been tampered with. Therefore, the hash-encrypted evidence storage submodule implements the generation of unique identifiers for contract execution records and on-chain immutable evidence storage.

[0089] The on-chain data push submodule is used to synchronously push the contract execution records that have been stored on the blockchain to the contract supervision and verification unit 3 and the intellectual property ownership verification unit 4 through blockchain peer-to-peer transmission, and receive the receipt confirmation information from the corresponding units to ensure the integrity and reliability of data transmission.

[0090] Specifically, the on-chain data push submodule is used to synchronously push the contract execution record to the contract supervision and verification unit 3 and the intellectual property ownership verification unit 4 through blockchain peer-to-peer transmission after the contract execution record has been uploaded to the blockchain for notarization, and to receive the reception confirmation information returned by the corresponding units.

[0091] The confirmation message indicates whether the corresponding unit has successfully received the contract execution record. By recording and verifying the confirmation message, the integrity and reliability of the contract execution record during the push process can be guaranteed, and subsequent contract supervision verification and intellectual property ownership verification can be carried out based on the same execution record.

[0092] Furthermore, the core verification rule for receiving confirmation is as follows: after receiving the pushed data, the receiver verifies whether the data hash value is consistent with the hash value stored on the chain. If the verification is successful, a valid receipt confirmation receipt is returned. After receiving a valid receipt, the pusher completes the push process. If no valid receipt is received, exception handling is triggered.

[0093] Contract Supervision and Verification Unit 3 is used to obtain the contract execution records pushed by Contract Automated Execution Unit 2, verify the entire process of contract phased execution according to preset contract execution technical rules, regulatory rules and intellectual property compliance requirements, generate regulatory certificates and push them to the intellectual property ownership verification unit 4 after on-chain storage. In this embodiment, the contract supervision and verification unit 3 includes an execution record collection module 31, a full-process compliance verification module 32, a supervision certificate generation module 33, and a certificate storage and push module 34, wherein: The execution record collection module 31 receives and verifies the contract execution records of the entire process of the contract's phased execution based on the push instructions of the contract automated execution unit 2, and completes the standardized collection and integrity verification of the data. Specifically, the execution record collection module 31 performs full-dimensional collection and verification of the contract phased execution process records pushed by the contract automated execution unit 2. The verification content includes data format legality verification, core field integrity verification, digital signature validity verification, and confidential information compliance verification. It confirms that the records have no format errors, missing fields, data tampering, or confidentiality violations. After completing all collection and verification and passing the verification, it outputs a standardized regulatory collection dataset.

[0094] The full-process compliance verification module 32, based on the contract execution records verified by the execution record collection module 31, performs multi-dimensional compliance verification of the entire phased execution process of the contract according to preset contract execution technical rules, regulatory rules, and intellectual property compliance requirements, and outputs a clear verification pass conclusion or a disclaimer of objection. The full-process compliance verification module 32 includes an execution rule matching submodule, a phased execution verification submodule, a comprehensive compliance judgment submodule, and an objection data marking submodule, among which: The execution rule matching submodule is used to accurately match the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements with the transaction type and intellectual property subject type corresponding to the contract execution record, and lock in the full set of rules applicable to this verification. Specifically, the execution rule matching submodule performs a full-dimensional matching and screening of the transaction type and intellectual property target type corresponding to the contract execution record with the preset rule base. The matching content includes matching of contract execution technical rules, regulatory rules, and intellectual property compliance requirements. After confirming that there are no mismatches, omissions, or redundancies in the rules, and after completing all rule matching and accurately locking, it outputs a full set of applicable rules specifically for this verification.

[0095] The phased execution verification submodule is used to verify the execution status, risk assessment process, circuit breaker action and data link integrity of each phase of the contract execution according to the matched verification rules, and output the individual verification results of each phase. Specifically, the phased execution verification submodule performs phased compliance verification on the execution status, risk assessment, circuit breaker handling, and data link of each performance stage. The verification content includes verification of the standardization of the execution status, verification of the compliance of the risk assessment process, verification of the legality of the circuit breaker handling action, and verification of the integrity of the data link. It confirms that there are no execution violations, risk control failures, improper handling, or data disconnection in each stage. After completing all phased verifications, it outputs the verification results of each stage as either passed or failed.

[0096] The compliance comprehensive judgment submodule is used to make a comprehensive judgment on the compliance of the entire contract execution process based on the individual verification results at each stage, combined with the consistency and closed-loop of the data throughout the process, and output a verification pass or verification fail conclusion. Specifically, the compliance comprehensive judgment submodule comprehensively judges the individual verification results of each stage, the consistency and closure of the data throughout the process. The judgment content includes stage result summary judgment, data consistency judgment, process closure judgment, and rule compliance judgment. It confirms that there are no stage non-compliance, data conflict, process breakpoints and rule violations throughout the process. After completing all comprehensive judgments, it outputs the final conclusion of verification pass or verification fail.

[0097] The objection data marking submodule is used to accurately locate and mark abnormal data and non-compliant items that fail verification, generate corresponding objection explanations, and synchronize them to the regulatory certificate generation module 33.

[0098] Specifically, the objection data marking submodule performs field-level location and hierarchical marking of abnormal data and non-compliant items that fail verification. The marking content includes abnormal stage location, violation field marking, rule violation annotation, and objection level classification. After confirming that the abnormal location is accurate, the marking is clear, and the basis is clear, a standardized objection description containing abnormal information and rectification suggestions is output after all objection marking is completed.

[0099] The regulatory certificate generation module 33 generates a unique regulatory certificate corresponding to this property rights transaction based on the verification conclusion output by the full-process compliance verification module 32. The regulatory certificate includes transaction entity information, target information, full-process verification data, verification conclusion, and a unique compliance verification identifier. Specifically, the regulatory certificate generation module 33 performs structured generation of the verified compliance conclusion and the full transaction data. The generated content includes transaction entity information, intellectual property target information, full-process verification data, verification conclusion, and a regulatory certificate with a unique compliance verification identifier. After confirming that the certificate information is complete, the identifier is unique, and the format is standardized, the module outputs a unique, verifiable regulatory certificate corresponding to this property rights transaction.

[0100] The certificate storage and push module 34 writes the generated regulatory certificate into the blockchain distributed node to complete the on-chain immutable certificate storage, and pushes the stored regulatory certificate to the intellectual property ownership verification unit 4.

[0101] Specifically, the certificate storage and push module 34 performs hash encryption and distributed on-chain storage on the generated standardized regulatory certificate. The stored content includes the original text of the regulatory certificate, the certificate hash value, the unique identifier for compliance verification, and the on-chain timestamp. It confirms that the stored information is tamper-proof and that it is synchronized and valid across multiple nodes. After completing all on-chain storage, the stored regulatory certificate is pushed to the intellectual property ownership verification unit 4.

[0102] Intellectual Property Ownership Verification Unit 4 is used to obtain contract execution records, regulatory certificates and on-chain evidence data, to trace and verify the entire process of intellectual property ownership transfer in stages, and to output the ownership verification results and store them on the chain.

[0103] In this embodiment, the intellectual property ownership verification unit 4 includes a verification data collection module 41, an ownership full-process traceability verification module 42, an ownership verification result generation module 43, and a verification result storage and synchronization module 44, wherein: The verification data collection module 41 is used to receive the contract execution records pushed by the contract automated execution unit 2 and the regulatory certificates pushed by the contract supervision verification unit 3, and simultaneously retrieve the full amount of on-chain evidence data of the corresponding property rights transaction to complete the integrity verification, standardized collection and transaction link alignment of all data. Specifically, the verification data collection module 41 performs full-dimensional collection processing on the received contract execution records, regulatory certificates, and retrieved on-chain evidence data. The processing includes data integrity verification, format standardization and regularization, transaction unique identifier matching, transaction link timing alignment, and on-chain data consistency verification. It confirms that the data is not missing, the format is not disordered, the identifier is not conflicting, the link is not misaligned, and there is no discrepancy with the on-chain data. After completing all collection and verification and passing the test, it outputs a standardized verification dataset that can be used for ownership traceability.

[0104] The ownership traceability and verification module 42, based on the full amount of data collected by the verification data collection module 41 and combined with the official ownership ledger data of intellectual property, conducts full-link traceability and verification of the ownership status of intellectual property targets before the transaction, the entire process of phased transfer, and the compliance of each transaction node. It identifies ownership defects, transfer gaps and compliance risks, and outputs the individual verification results of each link. Specifically, the ownership traceability and verification module 42 performs full-dimensional traceability and verification of the standardized verification dataset and the official intellectual property ownership ledger. The verification content includes verification of ownership status before the transaction, verification of the phased ownership transfer process, verification of compliance at each transaction node, verification of consistency of ownership ledger, and investigation of ownership defects and transfer gaps. It confirms that there are no ownership defects, transfer gaps, node violations, discrepancies in ledgers, or compliance risks. After completing the full-chain traceability and verification, it outputs the individual verification results of each ownership link.

[0105] The ownership verification result generation module 43, based on the single verification result output by the ownership full-process traceability verification module 42, completes the comprehensive judgment of the compliance of intellectual property ownership transfer and generates an ownership verification result uniquely corresponding to this transaction. The ownership verification result includes the target ownership status, the full-process traceability verification conclusion, the compliance judgment result and the unique verification identifier. Specifically, the ownership verification result generation module 43 performs a comprehensive and structured judgment on the individual verification results of each stage. The generated content includes the ownership status of the target, the conclusion of the full-process traceability verification, the compliance judgment result, the unique verification identifier and transaction association information. After confirming that the results are complete, the judgment is accurate, the identifier is unique and the format is compliant, after all the results are generated and qualified, the standardized ownership verification result that can be stored on the blockchain is output.

[0106] The verification result storage and synchronization module 44 is used to write the generated ownership verification result into the blockchain distributed node to complete the on-chain immutable storage, and to push the stored ownership verification result to the nodes of both parties to the transaction, the regulatory node and the intellectual property transaction publicity platform to complete the closed loop of the entire ownership verification process.

[0107] Specifically, the verification result storage and synchronization module 44 performs full-dimensional storage and synchronization processing on the standardized ownership verification results. The processing includes hash encryption storage, distributed node writing, multi-terminal synchronous push, and public information adaptation. It confirms that the stored evidence is tamper-proof, the synchronization is complete and reachable, and the public information is accurate. After all storage and synchronization are completed and qualified, the entire process of intellectual property ownership verification is closed-loop.

[0108] Those skilled in the art will understand that the process of implementing all or part of the steps of the above embodiments can be carried out by hardware or by a program instructing the relevant hardware.

[0109] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A smart contract automated execution and monitoring system for the entire property rights transaction process, characterized in that: include: The smart contract generation unit (1) is used to collect transaction basic information including property transaction subject information, intellectual property ownership information and transaction constraint clauses, and integrate preset contract execution technical rules including performance node rules, risk assessment rules, hierarchical execution conditions and reversible rollback rules, generate a standardized smart contract including transaction basic information and contract execution technical rules, and output a deployable standardized smart contract after completing the executability verification. Contract Automated Execution Unit (2) is used to load the standardized smart contract output by the smart contract generation unit (1), and map the intellectual property rights transfer and transaction settlement process agreed in the standardized smart contract into a multi-stage execution sequence of corresponding performance nodes. Each execution stage takes performance risk assessment as a prerequisite. Before execution, multi-dimensional temporal feature collection is used to obtain real-time status data of the intellectual property target and the transaction entity, historical performance risk data, and industry policy risk time-series data. A temporal attention mechanism and an improved lightweight residual temporal convolutional network are used to complete real-time risk assessment. Combined with historical risk data and policy risk coefficients, dynamic risk calibration is performed to output accurate risk assessment results. If the assessment meets the execution conditions, the corresponding action for the current stage is executed through atomic operations of smart contracts. If the assessment meets the conditions for suspension or termination, the execution of the entire process shall be suspended or terminated and a reversible rollback shall be performed accordingly; the contract execution record stored on the blockchain shall be generated simultaneously and pushed to the contract supervision and verification unit (3) and the intellectual property ownership verification unit (4). Contract supervision and verification unit (3) is used to obtain the contract execution record pushed by the contract automated execution unit (2), verify the entire process of contract phased execution according to the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements, generate regulatory certificates and push them to the intellectual property ownership verification unit (4) after they are stored on the blockchain. Intellectual property ownership verification unit (4) is used to obtain contract execution records, regulatory certificates and on-chain evidence data, to trace and verify the entire process of intellectual property ownership transfer in stages, and to output ownership verification results and store them on the chain.

2. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 1, characterized in that, The smart contract generation unit (1) includes an information collection module (11), a rule integration module (12), a contract generation module (13), and a contract verification module (14), wherein: The information collection module (11) is used to collect basic transaction information, including information on the parties involved in the property rights transaction, information on the ownership of intellectual property rights, and transaction constraint clauses; The rule integration module (12) integrates preset contract execution technical rules, which include performance node rules, risk assessment rules, tiered execution conditions, and reversible rollback rules, wherein: The performance node rules refer to the technical rules used to divide the execution order, execution nodes, and corresponding execution content of the intellectual property ownership transfer and transaction settlement process agreed upon in standardized smart contracts. The risk assessment rules refer to the technical rules used to clarify the objects of risk assessment, the scope of information collection, and the assessment logic for each stage of contract execution. The tiered execution conditions refer to the technical criteria used to define the execution status corresponding to the risk assessment results of each execution stage of the contract. The reversible rollback rule refers to the technical execution rule used to restore the transaction state when the contract execution is triggered to stop or terminate. The contract generation module (13) generates a standardized smart contract containing property rights transaction subject information, intellectual property rights ownership information, transaction constraint clauses and contract execution technical rules based on the information collected by the information collection module (11) and the contract execution technical rules integrated by the rule integration module (12). The contract verification module (14) is used to perform executability verification on the standardized smart contract and output a deployable standardized smart contract.

3. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 1, characterized in that, The contract automated execution unit (2) includes an execution sequence decomposition module (21), a dynamic risk assessment module (22), a circuit breaker reversible handling module (23), and an execution record generation and push module (24), wherein: The execution sequence decomposition module (21) maps the intellectual property ownership transfer and transaction settlement process into a multi-stage execution sequence corresponding to the performance node based on the standardized smart contract output by the smart contract generation unit (1). The dynamic risk assessment module (22) is based on the execution phase sequence formed by the execution sequence decomposition module (21). Before each execution phase starts, it collects multi-dimensional temporal status features of intellectual property targets, credit and temporal performance data of transaction entities, and temporal data of industry policy risks. It dynamically allocates features and time step weights through a temporal attention mechanism, extracts deep temporal correlation features by combining an improved lightweight residual temporal convolutional network, and introduces historical risk data and policy risk coefficients for dynamic risk calibration to complete high-precision real-time risk assessment and output risk assessment results containing risk values, attention weights, and calculation processes. The circuit breaker reversible handling module (23) performs corresponding stage actions or triggers suspension, termination and reversible rollback processing based on the risk assessment results of the dynamic risk assessment module (22) and the preset graded execution conditions. The execution record generation and push module (24) generates on-chain evidence-based contract execution records based on the execution status, risk assessment results and disposal information of each execution stage, and pushes them to the contract supervision and verification unit (3) and the intellectual property ownership verification unit (4).

4. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 3, characterized in that, The execution sequence decomposition module (21) includes a fulfillment node matching submodule, a stage sequence division submodule, and an execution connection triggering submodule, wherein: The performance node matching submodule is used to match the performance requirements stipulated in the standardized smart contract with the actual performance nodes of the property rights transaction one by one. The stage sequence division submodule is used to divide the intellectual property ownership transfer and transaction settlement process into multiple execution stage sequences based on the matched performance nodes; The execution connection triggering submodule is used to set the execution triggering conditions for adjacent execution phase sequences. The next execution phase sequence is triggered only after the previous execution phase sequence has been completed and the risk assessment is qualified.

5. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 3, characterized in that, The risk assessment process of the dynamic risk assessment module (22) includes the following steps: S22.1 Collect multi-dimensional temporal features and construct the original temporal feature matrix. The time-series characteristics include five categories of ownership status time-series characteristics of the intellectual property subject matter: pledge status, freeze status, invalidation status, objection status, and infringement litigation status, as well as the credit score of the transaction entity, performance compliance, and industry policy risk index. Bank of China corresponds to 8 types of features, and columns correspond to time steps. , For the first Class features in the first Feature values ​​of the time step; S22.2, On the original time series feature matrix Outlier truncation and standardization are performed to obtain the standardized time series feature matrix. Eliminate dimensional differences and data noise; S22.3 Calculate the attention weights for each feature and time step using a temporal attention mechanism. The standardized features are then weighted and fused to obtain an attention-weighted temporal feature vector. Attention weight The calculation is as follows: ； in: For the first Class features in the first Attention weights at time steps; For the first Weight coefficients of class features; For the first Class features in the first Standardized eigenvalues ​​of the time step; For the first Bias terms for class features; It is a linear rectification activation function; It is a natural exponential function; Traverse the index for feature categories; Index for time step traversal; This represents the total number of time steps for the temporal features. For the first Weight coefficients of class features; For the first Class features in the first Standardized eigenvalues ​​of the time step; For the first Bias terms for class features; S22.4, Weighted time series feature vector Input an improved lightweight residual temporal convolutional network, and extract deep temporal fusion features through depthwise separable convolutions and residual connections. Deep temporal fusion features The extraction formula is as follows: ； in: This is a depthwise separable convolution operation; Attention-weighted temporal feature vectors; The kernel size for depth-separable convolutions; For residual join operations with dimension alignment; S22.5, combined with historical average performance risk Industry policy risk index The risk value is dynamically calibrated, and the real-time trading risk value is calculated using a normalized activation function. Real-time transaction risk value The calibration formula is as follows: ； in: This is the normalized real-time transaction risk value; This represents the Sigmoid normalized activation function; This represents the weight vector of deep temporal fusion features; For bias terms of fully connected layers; This is the historical performance risk weighting coefficient; This represents the average performance risk during the same historical period. This refers to the industry policy risk weighting coefficient. This is an industry policy risk index; S22.6 Output Trading Risk Value The corresponding core calculation data is pushed to the circuit breaker reversible handling module (23) to complete the entire risk assessment process.

6. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 3, characterized in that, The graded handling and reversible rollback process of the reversible circuit breaker module (23) includes the following steps: S23.1, Preset risk warning threshold With risk termination threshold The two thresholds satisfy the core numerical relationship: ; S23.2 Obtain the transaction risk value output by the dynamic risk assessment module (22). , to trade risk value The risk level is determined by comparing the execution values ​​with the two-level risk thresholds. S23.3, when When this occurs, a stop execution instruction is triggered, suspending the current execution phase sequence and pushing risk warning information to the trading entity and regulatory nodes; S23.4, when When this occurs, a full-process termination instruction is triggered, and the transaction assets are returned via the original payment method through atomic operations of the smart contract. The asset return relationship is as follows: ;in, To return the amount of assets; This represents the amount of assets that have already been transferred and traded. S23.5 Generate circuit breaker handling and reversible rollback records, and synchronize them to the execution record generation and push module (24) to complete the entire process of hierarchical handling.

7. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 3, characterized in that, The execution record generation and push module (24) includes an execution information integration submodule, a hash encryption evidence storage submodule, and an on-chain data push submodule, wherein: The execution information integration submodule is used to collect the execution status of each execution stage sequence and the transaction risk value output by the dynamic risk assessment module (22). The core computing data, the handling actions of the circuit breaker reversible handling module (23), and the reversible rollback information form a complete and traceable contract execution record; The hash encryption evidence storage submodule is used to perform hash encryption processing on the contract execution record, generate a hash value that uniquely corresponds to the contract execution record, and write the encrypted contract execution record and the corresponding hash value into the blockchain distributed node to complete the on-chain evidence storage, ensuring that the contract execution record cannot be tampered with. The on-chain data push submodule is used to push the contract execution record that has been stored on the chain to the contract supervision and verification unit (3) and the intellectual property ownership verification unit (4) through blockchain peer-to-peer transmission, and receive the receiving confirmation information from the corresponding units to ensure the integrity and reliability of data transmission.

8. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 1, characterized in that, The contract supervision and verification unit (3) includes an execution record collection module (31), a full-process compliance verification module (32), a supervision certificate generation module (33), and a certificate storage and push module (34), wherein: The execution record collection module (31) receives and verifies the contract execution records of the entire process of the contract phased execution based on the push instruction of the contract automated execution unit (2), and completes the standardized collection and integrity verification of the data. The full-process compliance verification module (32) verifies the contract execution records completed by the execution record collection module (31), and performs multi-dimensional compliance verification on the full-process execution of the contract in stages according to the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements, and outputs a clear verification pass conclusion or verification fail objection statement. The regulatory certificate generation module (33) generates a unique regulatory certificate corresponding to this property rights transaction based on the verification pass conclusion output by the full-process compliance verification module (32). The regulatory certificate includes transaction entity information, target information, full-process verification data, verification conclusion and unique compliance verification identifier. The certificate storage and push module (34) writes the generated regulatory certificate into the blockchain distributed node to complete the on-chain immutable certificate storage, and pushes the stored regulatory certificate to the intellectual property ownership verification unit (4) simultaneously.

9. The smart contract automated execution and supervision system for the entire property rights transaction process as described in claim 8, characterized in that, The full-process compliance verification module (32) includes an execution rule matching submodule, a phased execution verification submodule, a comprehensive compliance judgment submodule, and an objection data marking submodule, wherein: The execution rule matching submodule is used to accurately match the preset contract execution technical rules, regulatory rules and intellectual property compliance requirements with the transaction type and intellectual property subject type corresponding to the contract execution record, and lock in the full set of rules applicable to this verification. The phased execution verification submodule is used to verify the execution status, risk assessment process, circuit breaker action and data link integrity of each phase of the contract's phased execution according to the matched verification rules, and output the individual verification results of each phase. The compliance comprehensive judgment submodule is used to complete the comprehensive compliance judgment of the entire contract execution process based on the individual verification results of each stage, combined with the consistency and closed-loop of the data throughout the process, and output the verification pass conclusion or verification fail conclusion. The objection data marking submodule is used to accurately locate and mark abnormal data and non-compliant items that fail verification, generate corresponding objection explanations, and synchronize them to the regulatory certificate generation module (33).

10. The smart contract automated execution and supervision system for the entire property rights transaction process according to claim 1, characterized in that, The intellectual property ownership verification unit (4) includes a verification data collection module (41), an ownership full-process traceability verification module (42), an ownership verification result generation module (43), and a verification result storage and synchronization module (44), wherein: The verification data collection module (41) is used to receive the contract execution record pushed by the contract automated execution unit (2) and the regulatory certificate pushed by the contract supervision verification unit (3), and simultaneously retrieve the full amount of on-chain evidence data of the corresponding property rights transaction to complete the integrity verification, standardized collection and transaction link alignment of all data. The ownership traceability and verification module (42) is based on the full amount of data collected by the verification data collection module (41) and combined with the official ownership ledger data of intellectual property rights. It performs full-link traceability and verification of the ownership status of the intellectual property target before the transaction, the full process of phased transfer, and the compliance of each transaction node. It investigates ownership defects, transfer gaps and compliance risks, and outputs the single verification results of each link. The ownership verification result generation module (43) completes the comprehensive judgment of the compliance of the transfer of intellectual property rights based on the single verification result output by the ownership full-process traceability verification module (42), and generates an ownership verification result that is uniquely corresponding to this transaction. The ownership verification result includes the target ownership status, the full-process traceability verification conclusion, the compliance judgment result and the unique verification identifier. The verification result storage and synchronization module (44) is used to write the generated ownership verification result into the blockchain distributed node to complete the on-chain immutable storage, and to push the stored ownership verification result to the nodes of both parties to the transaction, the regulatory node and the intellectual property transaction publicity platform to complete the closed loop of the entire ownership verification process.

Images