Scientific and technological achievement transfer and transformation certificate management system

Abstract

The invention discloses a scientific and technological achievement transfer and transformation certificate management system. The certificate management system comprises a structured data driven certificate automatic generation module, a collaborative approval chain module and a dynamic condition monitoring and automatic execution module. The structured certificate data generated by the automatic certificate generation module driven by the structured data is directly pushed to an approval process of the collaborative approval chain module, and a monitoring rule of the dynamic condition monitoring and automatic execution module is triggered; through structured field mapping and multi-source data real-time pulling, accurate pulling is realized, manual intervention is reduced, and data consistency is improved; through a dynamic flow chart and a material reuse mechanism, flexible configuration and execution efficiency improvement of a cross-department process are realized, and the problem of'process blockage 'is solved; through linkage of a rule engine and external data, real-time monitoring, automatic judgment and closed-loop execution of a certificate state are realized, and the risk of overdue abuse is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of certificate management, and specifically to a certificate management system for the transfer and transformation of scientific and technological achievements. Background Art

[0002] Under the background of accelerating the transformation of scientific and technological achievements, the certificate for the transfer and transformation of scientific and technological achievements, as the core voucher for technology transactions, ownership confirmation, and policy implementation, its management efficiency and security directly affect the efficiency of the market-oriented allocation of technological elements. However, the existing certificate management systems have significant defects in aspects such as process collaboration, risk prevention and control, and intelligent management. The specific problems are as follows:

[0003] The generation of traditional certificates relies on manual filling of templates, and the review process requires offline approvals from multiple departments, which takes up to several weeks. For example, when a certain technology is used as capital contribution by shares, it needs to go through links such as technology evaluation, legal compliance review, and approval by the state-owned assets management department in sequence. The systems of each department are not connected, and the progress cannot be tracked in real time, which is prone to problems such as repeated submission of materials or process jams; The technology transfer certificate usually stipulates a validity period or termination conditions (such as technology iteration, market substitution), but the existing system does not set up automatic monitoring rules. For example, a patent license certificate stipulates that "if the licensee fails to achieve industrialization within three years, it will be automatically terminated", but the system cannot be connected to the industrialization data source (such as enterprise production reports) in real time and still requires manual verification, resulting in the abuse of the certificate after it has expired and become invalid. Summary of the Invention

[0004] The purpose of the present invention is to provide a certificate management system for the transfer and transformation of scientific and technological achievements to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A certificate management system for the transfer and transformation of scientific and technological achievements, which includes a certificate automatic generation module driven by structured data, a collaborative approval chain module, and a dynamic condition monitoring and automatic execution module;

[0006] The certificate automatic generation module driven by structured data includes a configured template library, a multi-source data integration interface, and an intelligent error correction unit; The configured template library pre-sets a variety of certificate templates and supports dynamic field expansion; The multi-source data integration interface accesses the scientific research management system, the enterprise industrial and commercial information database, and the policy and regulation database through the API to automatically fill in the key fields of the certificate; The intelligent error correction unit verifies the data compliance based on the NLP algorithm and prompts the conflict items in real time;

[0007] The collaborative approval chain module includes a dynamic flowchart construction unit, a material intelligent reuse unit, and a blockchain evidence storage and tracking unit; the dynamic flowchart construction unit generates an approval path according to the approval type, and supports triggering a third-party audit when the technology valuation exceeds the threshold; the material intelligent reuse unit uses OCR+NLP technology to extract the technology valuation and patent numbers from the uploaded evaluation reports and qualification certification materials, and automatically reuses them in subsequent links; the blockchain evidence storage and tracking unit records the approval time, opinions, and operators through the consortium blockchain, generates a visual approval map, and automatically reminds the timeout nodes;

[0008] The dynamic condition monitoring and automatic execution module includes a data linkage interface, a rule engine, and a contract execution unit; the data linkage interface docks with the enterprise MES system and the patent database to obtain the industrial output value and the disclosure volume of alternative technologies in real time; the rule engine includes the output value compliance rate, technological iteration leading to substantial substitution, and dynamic determination based on the knowledge graph and GNN model; the contract execution unit automatically updates the certificate status to terminate or renew when the conditions are met, generates an electronic evidence chain containing timestamps and data sources, and stores it in the consortium blockchain;

[0009] The structured certificate data generated by the structured data-driven certificate automatic generation module is directly pushed to the approval process of the collaborative approval chain module, and triggers the monitoring rules of the dynamic condition monitoring and automatic execution module.

[0010] Preferably, the implementation method of the structured data-driven certificate automatic generation module includes the following steps:

[0011] S11: Configure the certificate template through the visualization interface, and define the text, numerical value, date field types, and data source mapping rules;

[0012] S12: Call the API interface to pull the patent status data in the scientific research management system, and perform matching verification with the registered capital and business scope in the enterprise industrial and commercial information database;

[0013] S13: Use the BERT model to perform semantic analysis on the certificate text, identify the constraint conditions of "patent validity period coverage" and "technology export compliance", and output conflict warnings.

[0014] Preferably, the implementation method of the collaborative approval chain module includes the following steps:

[0015] S21: Analyze the policy documents corresponding to the approval type, and generate an approval flowchart that triggers a third-party audit when the technology valuation exceeds the threshold;

[0016] S22: Identify the technology valuation and application scenario table data in the evaluation report through OCR, extract the key parameters, and store them in the shared database;

[0017] S23: Write the submission time and the hash value of the approval opinion into the Hyperledger Fabric consortium blockchain to generate an immutable evidence record.

[0018] Preferably, the implementation method of the dynamic condition monitoring and automatic execution module includes the following steps:

[0019] S31: Real-time collect the production report data of the enterprise MES system through the OPC UA protocol, and calculate the industrial output value achievement rate;

[0020] S32: Construct a technology substitution knowledge graph, associate substitute technology patents, market share, and technology life cycle curves, and use the GNN model to calculate the similarity weight;

[0021] S33: When the similarity weight exceeds the dynamic threshold of 0.65 and the output value is lower than 50% of the agreed value, trigger the intelligent contract termination certificate and generate an encrypted evidence package containing the data snapshot.

[0022] Preferably, the determination method for substantial substitution caused by technology iteration in the rule engine includes the following steps:

[0023] S41: Construct a technology substitution knowledge graph, the nodes include the target technology patent, substitute technology patent, and technology application scenario, and the edge relationships include patent citation relationships and technical efficacy similarities;

[0024] S42: Calculate the feature vector similarity weight between the target technology and the substitute technology through the GNN model, and the weight calculation formula is:

[0025]

[0026] where, V t ,V s are the semantic vectors of the target technology and the substitute technology respectively, C overlap is the number of overlapping claim items, and α = 0.6, β = 0.4 are dynamic adjustment coefficients;

[0027] S43: When the similarity weight is greater than 0.65 and the market share of the substitute technology is greater than 15%, it is determined that a substantial substitution is triggered.

[0028] Preferably, the implementation method of the blockchain evidence tracking unit includes the following steps:

[0029] S51: Sort the approval logs of the submission time, approval opinion, and operator ID by timestamp, generate a Merkle tree and calculate the root hash value;

[0030] S52: Encrypt the root hash value using the national cryptography SM3 algorithm, and write the encrypted result into the Hyperledger Fabric consortium blockchain after associating it with the certificate ID;

[0031] S53: When the approval node fails to be processed after the time limit, an early warning event is automatically triggered, the abnormal status is broadcast to the consortium blockchain, and the associated certificate process is frozen.

[0032] Preferably, the implementation method of the contract execution unit includes the following steps:

[0033] S61: Define the chain code for the trigger condition and execution logic of the certificate status change, and deploy it to the consortium blockchain node;

[0034] S62: When the monitoring data meets the conditions, call the UpdateStatus method of the chain code to update the certificate status, and generate an evidence chain including the following elements: trigger condition data snapshot; rule engine judgment log; consortium blockchain transaction ID and timestamp;

[0035] S63: After the evidence chain is verified by the consensus node, store it in IPFS and return the CID, and bind the CID and the transaction ID to write into the blockchain.

[0036] Preferably, the specific method for OCR to recognize and evaluate the form data includes the following steps:

[0037] S71: Locate the table area in the evaluation report through the YOLOv5 model, and extract the cell text and values;

[0038] S72: Use a bidirectional LSTM model based on the attention mechanism to recognize the table row and column relationship and reconstruct the structured data;

[0039] S73: Perform a confidence score on the recognition result. If the score is less than 0.8, trigger the manual verification process.

[0040] Preferably, the method for generating the visual approval graph includes the following steps:

[0041] S81: Construct an approval flow chart with the approval node as the graph vertex and the approval flow direction as the directed edge;

[0042] S82: Use the Force Atlas algorithm to optimize the graph layout, and highlight the following features: the timeout node is highlighted in red; the completed node shows the percentile ranking of the processing duration; the critical path node adds a weighted border;

[0043] S83: Support clicking on the node to view the details of the approval opinion and the blockchain deposit hash value.

[0044] Preferably, the data cleaning method of the multi-source data integration interface includes the following steps:

[0045] S91: Perform anomaly detection on the output value data collected by the enterprise MES system. If the fluctuation of the output value data exceeds 3 times the standard deviation of the historical average, mark it as an anomaly and trigger manual review; use the ARIMA model to predict and fill in the missing data.

[0046] S92: For the technical similarity data in the patent database, use the DBSCAN clustering algorithm to remove outlier noise points.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] Through structured field mapping and real-time pulling of multi-source data, the present invention realizes accurate pulling, reduces manual intervention and improves data consistency; through the dynamic flowchart and material reuse mechanism, it realizes flexible configuration of cross-departmental processes and improves execution efficiency, solving the problem of "process jams"; through the rule engine and external data linkage, it realizes "real-time monitoring - automatic judgment - closed-loop execution" of the certificate status, avoiding the risk of overdue abuse. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a system block diagram of the technology transfer and transformation certificate management system of the present invention.

[0050] Figure 2 It is a flowchart of the certificate automatic generation module driven by structured data of the present invention.

[0051] Figure 3 It is a flowchart of the dynamic condition monitoring and automatic execution module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] Please refer to Figures 1 to 3 , the present invention provides a technical solution: a technology transfer and transformation certificate management system, which includes a certificate automatic generation module driven by structured data, a collaborative approval chain module, and a dynamic condition monitoring and automatic execution module;

[0054] The certificate automatic generation module driven by structured data includes a configuration template library, a multi-source data integration interface, and an intelligent error correction unit;

[0055] The configuration template library pre-sets a variety of certificate templates, supporting dynamic extension of fields; through the visual interface, certificate modules such as patent licensing and technology valuation for equity participation are pre-set, supporting dynamic extension of fields. For example, when adding a "Cross-border Technology Export Compliance Clause", users can customize the field types and data source mapping rules for text, numerical values, and dates; data mapping associates certificate fields with external data sources. For example, the "Patent Number" field is mapped to the API interface / api / patent / list of the scientific research management system to pull the list of valid patents in real time.

[0056] The multi-source data integration interface accesses the scientific research management system, enterprise industrial and commercial information database, and policy and regulation database through APIs, and automatically fills in the key fields of the certificate; the scientific research management system obtains the patent status of authorization or invalidation and the technology maturity through the RESTful API, and measures it by the TRL level; the enterprise industrial and commercial information database calls the Tianyancha API to verify the matching of the assignee's registered capital and business scope; the policy and regulation database subscribes to the Ministry of Commerce's technology export control list and automatically matches the technology classification code in the certificate.

[0057] The data cleaning method of the multi-source data integration interface includes the following steps:

[0058] S111: Perform anomaly detection on the output value data collected by the enterprise MES system. If the fluctuation of the output value data exceeds 3 times the standard deviation of the historical average, mark it as abnormal and trigger manual review; use the ARIMA model to predict and fill in the missing data;

[0059] S112: For the technical similarity data in the patent database, use the DBSCAN clustering algorithm to remove outlier noise points.

[0060] The intelligent error correction unit verifies the data compliance based on the NLP algorithm and prompts conflict items in real time; the intelligent error correction unit uses the BERT model to parse the certificate text and identify key constraint conditions. For example, it detects whether the "Patent Validity Period" covers the agreed cycle of the certificate. If an expired patent is cited, a conflict warning is triggered; it verifies whether the "Technology Export Compliance" clause complies with the latest policy version; the conflict prompt highlights and pops up the conflict items through pop-up windows, supporting one-key jump for correction.

[0061] The implementation method of the structured data-driven certificate automatic generation module includes the following steps:

[0062] S211: Configure the certificate template through the visual interface, and define the field types and data source mapping rules for text, numerical values, and dates;

[0063] S212: Call the API interface to pull the patent status data in the scientific research management system and perform matching verification with the registered capital and business scope in the enterprise industrial and commercial information database;

[0064] S213: Use the BERT model to perform semantic analysis on the certificate text, identify the constraint conditions of "patent validity period coverage" and "technical export compliance", and output conflict warnings.

[0065] The collaborative approval chain module includes a dynamic flowchart construction unit, a material intelligent reuse unit, and a blockchain evidence storage and tracking unit;

[0066] The dynamic flowchart construction unit generates an approval path according to the approval type, and supports triggering a third-party audit when the technology valuation exceeds the threshold; the dynamic flowchart construction unit parses policy documents, such as the "Regulations on the Administration of State-owned Asset Appraisal", to generate an approval flowchart. For example, when the technology valuation exceeds 5 million yuan, a "third-party audit" node is automatically inserted; for cross-border technology transfer, a "national security review" branch needs to be added.

[0067] The material intelligent reuse unit uses OCR+NLP technology to extract the technology valuation and patent number from the uploaded evaluation report and qualification certificate materials, and automatically reuses them in subsequent links; the extracted patent number and technology valuation are stored in the shared database and automatically filled in during subsequent legal review.

[0068] The blockchain evidence storage and tracking unit records the approval time, opinions, and operators through the consortium chain, generates a visual approval map, and automatically reminds of overtime nodes; the implementation method of the blockchain evidence storage and tracking unit includes the following steps: sort the approval logs of the submission time, approval opinions, and operator IDs by timestamp, generate a Merkle tree and calculate the root hash value; encrypt the root hash value using the national secret SM3 algorithm, and associate the encrypted result with the certificate ID and write it into the Hyperledger Fabric consortium chain; when the approval node times out without being processed, an early warning event is automatically triggered, and the abnormal status is broadcast to the consortium chain and the associated certificate process is frozen.

[0069] The implementation method of the collaborative approval chain module includes the following steps:

[0070] S311: Parse the policy document corresponding to the approval type to generate an approval flowchart that triggers a third-party audit when the technology valuation exceeds the threshold;

[0071] S312: Identify the technology valuation and application scenario table data in the evaluation report through OCR, extract the key parameters and store them in the shared database;

[0072] S313: Write the submission time and the hash value of the approval opinion into the Hyperledger Fabric consortium chain to generate an immutable evidence record.

[0073] The tabular data in the OCR recognition evaluation report locates the table area in the evaluation report through the YOLOv5 model, extracts cell text and values; the bidirectional LSTM model based on the attention mechanism identifies the row and column relationships of the table and reconstructs the structured data; a confidence score is given to the recognition result, and if the score is less than 0.8, the manual verification process is triggered.

[0074] The dynamic condition monitoring and automatic execution module includes a data linkage interface, a rule engine, and a contract execution unit;

[0075] The data linkage interface connects to the enterprise MES system and the patent database to obtain the industrial output value and the disclosure volume of alternative technologies in real time; it connects to the enterprise MES system through the OPC UA protocol to collect the industrial output value in the production report hourly; an API interface is deployed to access the WIPO and USPTO patent databases to search for the abstracts and claims of alternative technology patents.

[0076] The rule engine includes the output value compliance rate, technological iteration leading to substantial substitution, and dynamic determination based on the knowledge graph and GNN model; the determination method for technological iteration leading to substantial substitution in the rule engine includes the following steps: constructing a technology substitution knowledge graph, with nodes including target technology patents, alternative technology patents, and technology application scenarios, and edge relationships including patent citation relationships and technological efficacy similarities; calculating the similarity weight of the feature vectors of the target technology and the alternative technology through the GNN model, and the weight calculation formula is:

[0077]

[0078] where, V t , V s are the semantic vectors of the target technology and the alternative technology respectively, C overlap is the number of overlapping claim items, and α = 0.6, β = 0.4 are dynamic adjustment coefficients;

[0079] When the similarity weight is greater than 0.65 and the market share of the alternative technology is greater than 15%, it is determined that a substantial substitution is triggered.

[0080] When the conditions are met, the contract execution unit automatically updates the certificate status to terminate or renew, generates an electronic evidence chain containing timestamps and data sources and stores it in the alliance chain; the implementation method of the contract execution unit includes the following steps: defining a chain code for the certificate status change trigger conditions and execution logic and deploying it to the alliance chain nodes; when the monitored data meets the conditions, calling the UpdateStatus method of the chain code to update the certificate status and generating an evidence chain containing the following elements: a data snapshot of the trigger conditions; the determination log of the rule engine; the alliance chain transaction ID and timestamp; after the evidence chain is verified by the consensus nodes, it is stored in IPFS and the CID is returned, and the CID is bound to the transaction ID and written into the blockchain.

[0081] The implementation method of the dynamic condition monitoring and automatic execution module includes the following steps:

[0082] S411: Real-time collect the production report data of the enterprise MES system through the OPC UA protocol, and calculate the industrial output value achievement rate;

[0083] S412: Construct a technology substitution knowledge graph, associate substitute technology patents, market share, and technology life cycle curves, and use the GNN model to calculate the similarity weight;

[0084] S413: When the similarity weight exceeds the dynamic threshold of 0.65 and the output value is lower than 50% of the agreed value, trigger the intelligent contract termination certificate and generate an encrypted evidence package containing a data snapshot.

[0085] The generation method of the visualization approval graph includes the following steps:

[0086] S511: Construct an approval flow chart with approval nodes as graph vertices and approval flows as directed edges;

[0087] S512: Optimize the graph layout using the Force Atlas algorithm to highlight the following features: timeout nodes are highlighted in red; completed nodes display the percentile ranking of processing duration; key path nodes have an increased weighted border;

[0088] S513: Support clicking on nodes to view the details of approval opinions and the blockchain deposit hash value.

[0089] The structured certificate data generated by the structured data-driven certificate automatic generation module is directly pushed to the approval process of the collaborative approval chain module, and triggers the monitoring rules of the dynamic condition monitoring and automatic execution module.

[0090] In actual use, taking the "lithography technology patent license certificate" as an example:

[0091] Pull the patent number CN202111476489 from the scientific research management system, match the assignee qualifications, and generate a preliminary certificate; when the technology valuation exceeds the threshold, trigger a third-party audit, OCR extract the valuation data in the audit report, and store it in the blockchain; real-time monitor the patent volume of the alternative technology EUV lithography, and if the similarity is greater than 0.65 and the output value does not meet the standard, automatically terminate the certificate.

[0092] Through structured field mapping and real-time pulling of multi-source data, the present invention realizes accurate pulling, reduces manual intervention and improves data consistency; through dynamic flowcharts and material reuse mechanisms, it realizes flexible configuration of cross-departmental processes and improvement of execution efficiency, and solves the problem of "process jamming"; through the rule engine and external data linkage, it realizes "real-time monitoring - automatic judgment - closed-loop execution" of the certificate status, and avoids the risk of over-expiration abuse.

[0093] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A management system for scientific and technological achievement transfer and transformation certificates, characterized in that: The certificate management system includes a certificate automatic generation module driven by structured data, a collaborative approval chain module, and a dynamic condition monitoring and automatic execution module; The certificate automatic generation module driven by structured data includes a configured template library, a multi-source data integration interface, and an intelligent error correction unit; the configured template library pre-sets multiple certificate templates and supports dynamic field expansion; the multi-source data integration interface accesses the scientific research management system, the enterprise industrial and commercial information database, and the policy and regulation database through APIs to automatically fill in the key fields of the certificate; the intelligent error correction unit verifies the data compliance based on NLP algorithms and prompts conflict items in real time; The collaborative approval chain module includes a dynamic flowchart construction unit, a material intelligent reuse unit, and a blockchain evidence storage and tracking unit; the dynamic flowchart construction unit generates an approval path according to the approval type and supports triggering a third-party audit when the technology valuation exceeds the threshold; the material intelligent reuse unit uses OCR+NLP technology to extract the technology valuation and patent numbers in the uploaded evaluation reports and qualification certification materials and automatically reuses them in subsequent links; the blockchain evidence storage and tracking unit records the approval time, opinions, and operators through the consortium chain, generates a visual approval map, and automatically reminds of the overtime nodes; The dynamic condition monitoring and automatic execution module includes a data linkage interface, a rule engine, and a contract execution unit; the data linkage interface is connected to the enterprise MES system and the patent database to obtain the industrial output value and the disclosed quantity of alternative technologies in real time; the rule engine includes the output value compliance rate, the substantial replacement caused by technological iteration, and the dynamic determination based on the knowledge graph and the GNN model; when the conditions are met, the contract execution unit automatically updates the certificate status to terminate or renew, generates an electronic evidence chain containing the time stamp and data source, and stores it in the consortium chain; The structured certificate data generated by the certificate automatic generation module driven by structured data is directly pushed to the approval process of the collaborative approval chain module and triggers the monitoring rules of the dynamic condition monitoring and automatic execution module.

2. The management system for scientific and technological achievement transfer and transformation certificates according to claim 1, characterized in that: The implementation method of the certificate automatic generation module driven by structured data includes the following steps: S11: Configure the certificate template through the visual interface, define the text, numerical value, date field types, and data source mapping rules; S12: Call the API interface to pull the patent status data in the scientific research management system and perform matching verification with the registered capital and business scope in the enterprise industrial and commercial information database; S13: Use the BERT model to perform semantic analysis on the certificate text, identify the constraint conditions of "patent validity period coverage" and "technology export compliance", and output conflict warnings.

3. The technology achievement transfer and transformation certificate management system according to claim 1, wherein: The implementation method of the collaborative approval chain module includes the following steps: S21: Analyze the policy documents corresponding to the approval type and generate an approval flowchart that triggers a third-party audit when the technology valuation exceeds the threshold; S22: Identify the technology valuation and application scenario table data in the evaluation report through OCR, extract the key parameters, and store them in the shared database; S23: Write the submission time and the hash value of the approval opinion into the Hyperledger Fabric consortium chain to generate an immutable evidence record.

4. The technology achievement transfer and transformation certificate management system according to claim 1, characterized in that: The implementation method of the dynamic condition monitoring and automatic execution module includes the following steps: S31: Real - time collect the production report data of the enterprise MES system through the OPC UA protocol, and calculate the achievement rate of industrial output value; S32: Construct a technology substitution knowledge graph, associate substitution technology patents, market share, and technology life - cycle curves, and use the GNN model to calculate the similarity weight; S33: When the similarity weight exceeds the dynamic threshold of 0.65 and the output value is lower than 50% of the agreed value, trigger the intelligent contract termination certificate and generate an encrypted evidence package containing a data snapshot.

5. The technology achievement transfer and transformation certificate management system according to claim 1, wherein: The determination method for substantial substitution caused by technology iteration in the rule engine includes the following steps: S41: Construct a technology substitution knowledge graph, where the nodes include target technology patents, substitution technology patents, and technology application scenarios, and the edge relationships include patent citation relationships and technology efficacy similarities; S42: Calculate the similarity weight of the feature vectors of the target technology and the substitution technology through the GNN model. The weight calculation formula is: Among them, V t , V s are the semantic vectors of the target technology and the alternative technology respectively, C overlap is the number of overlapping claim items, and α = 0.6, β = 0.4 are dynamic adjustment coefficients; S43: When the similarity weight is greater than 0.65 and the market share of the substitution technology is greater than 15%, it is determined that substantial substitution is triggered.

6. The management system for scientific and technological achievement transfer and transformation certificates according to claim 1, wherein: The implementation method of the blockchain evidence - storing and tracking unit includes the following steps: S51: Sort the approval logs of submission time, approval opinions, and operator ID by timestamp, generate a Merkle tree, and calculate the root hash value; S52: Encrypt the root hash value using the national cryptographic SM3 algorithm, associate the encryption result with the certificate ID, and write it into the Hyperledger Fabric consortium blockchain; S53: When the approval node fails to process within the timeout period, automatically trigger an early - warning event, broadcast the abnormal status to the consortium blockchain, and freeze the associated certificate process.

7. The technology achievement transfer and transformation certificate management system according to claim 1, wherein: The implementation method of the contract execution unit includes the following steps: S61: Define the chain code for the trigger conditions and execution logic of certificate status changes, and deploy it to the consortium blockchain nodes; S62: When the monitored data meets the conditions, call the UpdateStatus method of the chain code to update the certificate status, and generate an evidence chain containing the following elements: trigger condition data snapshot; rule engine determination log; consortium blockchain transaction ID and timestamp; S63: After the evidence chain is verified by the consensus node, store it in IPFS and return the CID, and bind the CID and the transaction ID and write them into the blockchain.

8. The technology achievement transfer and transformation certificate management system according to claim 3, characterized in that: The specific method for OCR - recognizing and evaluating the form data of the report includes the following steps: S71: Locate the table area in the evaluation report through the YOLOv5 model, and extract cell text and numerical values; S72: Use a bidirectional LSTM model based on the attention mechanism to recognize the table row - column relationship and reconstruct the structured data; S73: Perform a confidence score on the recognition result. If the score is less than 0.8, trigger the manual verification process.

9. The technology transfer and transformation certificate management system according to claim 1, wherein: The generation method of the visualized approval graph includes the following steps: S81: Construct an approval flow chart with approval nodes as graph vertices and approval flows as directed edges; S82: Use the Force Atlas algorithm to optimize the graph layout, highlighting the following features: timeout nodes are highlighted in red; completed nodes display the percentile ranking of processing duration; critical path nodes have increased weighted borders; S83: Support clicking on nodes to view the details of approval opinions and the blockchain evidence - storing hash value.

10. A management system for scientific and technological achievement transfer and transformation certificates according to claim 1, characterized in that: The data cleaning method of the multi-source data integration interface includes the following steps: S91: Perform anomaly detection on the output value data collected by the enterprise MES system. If the fluctuation of the output value data exceeds 3 times the standard deviation of the historical average, mark it as abnormal and trigger manual review; use the ARIMA model to predict and fill the missing data; S92: For the technical similarity data in the patent database, use the DBSCAN clustering algorithm to remove outlier noise points.

Images