Order monitoring management system

Abstract

The invention belongs to the technical field of order monitoring and management, and provides an order monitoring and management system, which comprises a digital identity twinning verification module for generating a unique digital twinning ID, calculating credibility based on federated learning and generating a pre-authorization rule; the dynamic intelligent sandbox isolation module realizes field-level data slice encryption, quantum transmission protection and memory-level resource isolation; the block chain operation auditing module is used for atomizing an API calling behavior, uploading the behavior to a chain and storing the behavior, and verifying the compliance in real time through an intelligent contract; the AI risk decision module constructs a risk map by using a graph neural network, generates a dynamic score and executes a three-level response strategy, and the scheme realizes closed-loop management and control from identity authentication and data isolation to behavior auditing and intelligent decision, improves data interaction security and risk response efficiency, and improves the safety and reliability of the system. The method is suitable for an order monitoring scene requiring fine management and control of third party access.

Description

Technical Field

[0001] The invention belongs to the technical field of order monitoring and management, in particular to an order monitoring and management system. Background Art

[0002] The existing order monitoring system has the following defects when connected to a third-party system: It can only implement system-level permission control and cannot achieve detailed field-level data sharing, making it difficult to meet the needs of refined data management and control.

[0003] Third-party access risk monitoring relies on static rules, lacks real-time adaptive response capabilities to abnormal calling behaviors, and cannot respond to dynamically changing risks in a timely manner.

[0004] Operational audits lack a tamper-proof, full-link evidence storage mechanism, data traceability is inefficient, and it is difficult to quickly and accurately track and locate security issues when they arise.

[0005] To this end, the present invention provides an order monitoring and management system. Summary of the Invention

[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0007] The technical solution adopted by the present invention to solve its technical problem is: In a first aspect, the present invention provides an order monitoring and management system, comprising: Digital Identity Twin Verification Module: Generates a unique digital twin ID for the third-party platform, and uses federated learning to calculate the trust level of non-order data transmission. It verifies the security protection capabilities of third-party platforms that meet the trust level standards, uses smart contracts to parse cooperation agreements, and generates pre-authorization rules for third-party platforms that meet the security protection capability standards. Dynamic Intelligent Sandbox Isolation Module: This module sends order data to trusted third-party platforms based on pre-authorization rules. The module then encrypts and shards the order data at the field level. A quantum encryption transmission channel and a dynamic IP tunnel are established using the temporary keys of the encrypted shards. A memory-level micro-sandbox is created for each third-party platform, and the data transmission path for each order is switched in real time to enable order data transmission to the third-party platform. The micro-sandbox data is then destroyed upon completion of the transmission operation. Blockchain operation audit module: This module records third-party platform API call behavior, deploys smart contracts to verify third-party platform data access compliance in real time, builds a digital twin model of data flow, and visually displays sensitive data storage locations and third-party platform risk maps; AI risk decision module: Utilizes graph neural networks to access the risk map of third-party platforms, generates dynamic risk scores, and defines a three-level response strategy to achieve fully automated processing from risk detection to circuit breaker mechanism response.

[0008] As one of the embodiments, the specific process of generating a unique digital twin ID on a third-party platform is as follows: Convert the hardware fingerprint H, behavioral fingerprint B, and credit score C into feature vectors, calculate them through the hash encryption algorithm, and generate a digital twin ID. The digital twin ID generation formula is: ,in, represents the concatenation of feature vectors, It is a cryptographic hash function that uses the SHA-256 algorithm. is the hardware fingerprint feature vector, is the behavioral fingerprint feature vector, is the credit score feature vector.

[0009] As one embodiment, the specific process of calculating the trust level of order-free data transmission is as follows: Based on the multi-dimensional features (H, B, C) integrated into the digital twin ID, the order monitoring management system and the third-party platform further use federated learning technology to dynamically optimize feature weights and quantify the trust level of the third party without sharing the original data. The specific process is as follows: The order monitoring management system and the third-party platform calculate the trust of non-order data transmission through the federated learning method without sharing the original data. The specific formula for trust calculation is: ,in, The feature weights are optimized through federated learning, is the evaluation function of the i-th type of feature, n is the feature dimension, and i is the index of the feature.

[0010] As one embodiment, the specific process of verifying the security protection capability of a third-party platform that meets the trustworthiness standard is as follows: The assessment scope of security protection capabilities includes: vulnerability detection rate, encryption strength, and access control mechanism; The vulnerability detection rate, encryption strength, and access control mechanism are calculated using the security protection capability scoring formula to obtain the security protection capability score. The security protection capability scoring formula is as follows: ,in, 、 、 is the weight coefficient and , A is the vulnerability detection rate in vulnerability detection, and the specific calculation method is: , where B is the encryption strength coefficient in the encryption strength, which is determined by the complexity of the encryption algorithm. If the AES-256 encryption algorithm is used, the corresponding coefficient of the encryption strength coefficient is 1. If DES is used, the corresponding coefficient of the encryption strength coefficient is 0.5. C is the access control rate in the access control mechanism; If the security protection capability score is greater than or equal to the security protection capability score threshold, it means that the security protection capability meets the standard and the third-party platform is allowed to access; If the security protection capability score is less than the security protection capability score threshold, it means that the security protection capability meets the standard and third-party platform access is not allowed.

[0011] As one embodiment, the specific process of generating pre-authorization rules for a third-party platform that meets the security protection capability standards is as follows: Generate data access scope using parsing cooperation agreement; Based on allowing third-party platforms to access, the third-party platform can extract the data access scope from the smart contract. The data access scope includes: field collection , Time Limit Window and fuse mechanism.

[0012] As one embodiment, the specific process of encrypting and sharding the order data according to the field granularity is as follows: Split the original order data D into m independent slices according to the field dimension Each independent slice corresponds to a data field. A symmetric encryption algorithm is used to encrypt each independent slice at the field level and generate a temporary key. ; For each independent slice Use symmetric encryption algorithm (AES) to encrypt and generate encrypted data , using symmetric encryption algorithm to generate encrypted data The specific formula is: ,in, It is a temporary key generated by the symmetric encryption algorithm. Time t belongs to the pre-authorization time window. is the time window, For the i-th independent slice, the key validity period is determined by the time window in the pre-authorization rule To control, the third-party platform only obtains the temporary key of the authorized slice.

[0013] As one embodiment, the specific process of creating a memory-level micro-sandbox for each third-party platform is as follows: Create a separate Docker container for each third-party platform and limit the number of CPU cores and memory limits , prohibit data from being written to disk; The micro-sandbox only allows reading temporary keys for authorized slices , and decrypt the encrypted data in memory , and prohibit the decryption and persistence of encrypted data. Specifically, the prohibition of the persistence of encrypted data means that the encrypted data is only allowed to be decrypted within the time window. If the decryption time exceeds the end time specified in the time window, the third-party platform is prohibited from continuing to decrypt.

[0014] As one embodiment, the specific process of recording the third-party platform API call behavior is as follows: The evidence content includes the digital twin ID, timestamp, field-level access coordinates, and hash value comparison before and after the operation; The API call behavior is atomized and stored on the chain. The content of the evidence includes: digital twin ID, timestamp, field-level access coordinates, and hash values ​​before and after the operation.

[0015] As one embodiment, the specific process of real-time verification of third-party platform data access compliance is as follows: Deploy smart contracts to verify data access compliance in real time, including scope compliance, frequency compliance, and time compliance; Perform scope compliance check to verify whether the access field is within the field set of the pre-authorization rule; Perform frequency compliance check, count the call frequency per unit time, and compare it with the frequency threshold; Perform time compliance check to check whether the call time is within the time window; When any compliance check fails, the circuit breaker mechanism is triggered and the order monitoring and management system automatically activates the circuit breaker mechanism, revoking the third-party platform access permission.

[0016] As one embodiment, the specific process of defining the three-level response strategy is as follows: Execute corresponding measures according to the three-level response threshold. When the comprehensive risk score R is greater than or equal to the first score threshold and less than the second score threshold, a yellow warning notification is triggered; When the comprehensive risk score is greater than or equal to the second scoring threshold and less than the third scoring threshold, an orange quarantine warning is triggered, restricting third-party platforms from accessing order data; When the comprehensive risk score is greater than or equal to the third scoring threshold, a red circuit breaker warning is triggered and the circuit breaker mechanism is activated.

[0017] In a second aspect, the present invention provides an order monitoring and management method, comprising: S1: Generate a unique digital twin ID for the third-party platform, and use federated learning to calculate the trust level of non-order data transmission. Verify the security protection capabilities of third-party platforms that meet the trust level standards. Use smart contracts to analyze cooperation agreements and generate pre-authorization rules for third-party platforms that meet the security protection capability standards. S2: Order data is sent to third-party platforms that meet the trust requirements according to pre-authorization rules. The order data is first encrypted and sharded according to field granularity. The temporary key of the encrypted shards is used to establish a quantum encryption transmission channel and a dynamic IP tunnel. A memory-level micro-sandbox is created for each third-party platform. The data transmission path for each order is switched in real time to realize order data transmission to the third-party platform. The micro-sandbox data is destroyed at the end of the transmission operation. S3: Evidence of third-party platform API call behavior is stored, smart contracts are deployed to verify the compliance of third-party platform data access in real time, a digital twin model of data flow is constructed, and a visual display of sensitive data storage locations and third-party platform risk maps is displayed; S4: Use graph neural networks to access the risk map of third-party platforms, generate dynamic risk scores, and define a three-level response strategy to achieve fully automatic processing from risk detection to circuit breaker mechanism response.

[0018] The beneficial effects of the present invention are as follows: 1. Through the dynamic intelligent sandbox isolation module, order data is encrypted and segmented according to field granularity. Third parties only obtain temporary keys for authorized slices. Compared with traditional system-level permission control, the risk of data leakage is reduced and data sharing is minimized. The temporary key is bound to the time window and automatically expires after timeout. After the decryption operation is completed, the micro-sandbox data is destroyed to avoid the risk of persistent data storage and strengthen the protection of data throughout its life cycle. The digital identity twin verification module calculates trust through federated learning. Combined with the graph neural network risk map and dynamic scoring of the AI ​​risk decision module, it realizes the upgrade from static rules to adaptive risk perception.

[0019] 2. Trigger yellow warnings, orange isolations, and red circuit breakers based on the comprehensive risk score (R), avoiding one-size-fits-all blocking and balancing security and business continuity. For example, minor anomalies only restrict resources without interrupting services. Smart contracts verify the scope, frequency, and time compliance of data access in real time, automatically triggering circuit breakers when violations occur, reducing manual intervention errors and improving system security.

[0020] 3. Use quantum communication protocols to protect transmission security, combined with high-frequency path switching of dynamic IP tunnels to effectively combat network tracking and man-in-the-middle attacks, enhance communication confidentiality, and integrate digital twin ID with hardware fingerprints, behavioral fingerprints, and credit scores. Trust calculation is combined with federated learning to optimize weights and improve system adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below with reference to the accompanying drawings.

[0022] Figure 1 This is a system module diagram of an order monitoring and management system of the present invention; Figure 2 It is a flowchart of the steps of an order monitoring and management method of the present invention. DETAILED DESCRIPTION

[0023] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0024] Example 1 like Figure 1 As shown, an order monitoring and management system according to an embodiment of the present invention includes: Digital Identity Twin Verification Module: Generates a unique digital twin ID for the third-party platform, and uses federated learning to calculate the trust level of non-order data transmission. It verifies the security protection capabilities of third-party platforms that meet the trust level standards, uses smart contracts to parse cooperation agreements, and generates pre-authorization rules for third-party platforms that meet the security protection capability standards. The specific process of generating a unique digital twin ID on a third-party platform is as follows: Digital twin ID integrates hardware fingerprint (H), behavioral fingerprint (B) and credit score (C); The specific fusion and processing process is as follows: Hardware fingerprint (H): collects the hardware characteristics of the device, including MAC address and CPU serial number, and generates a unique identifier through a hash algorithm; Behavioral fingerprinting (B): Analyze the third-party platform's interface call frequency and timing pattern behavior characteristics and convert them into vector representations; Credit score (C): A quantitative score generated based on historical cooperation records and industry reputation data; Convert the hardware fingerprint (H), behavioral fingerprint (B), and credit score (C) into feature vectors, calculate them through the hash encryption algorithm, and generate a digital twin ID. The digital twin ID generation formula is: ,in, represents the concatenation of feature vectors, It is a cryptographic hash function that uses the SHA-256 algorithm. is the hardware fingerprint feature vector, is the behavioral fingerprint feature vector, is the credit score feature vector; In the present invention, the cryptographic hash function uses the SHA-256 algorithm to ensure that different feature combinations generate unique IDs; The specific process of calculating the trust level of orderless data transmission is as follows: Based on the multi-dimensional features (H, B, C) integrated into the digital twin ID, the order monitoring management system and the third-party platform further use federated learning technology to dynamically optimize feature weights and quantify the trust level of the third party without sharing the original data. The specific process is as follows: The order monitoring management system and the third-party platform calculate the trust of non-order data transmission through the federated learning method without sharing the original data. The specific formula for trust calculation is: ,in, The feature weights are optimized through federated learning, is the evaluation function of the i-th type of feature, n is the feature dimension, and i is the index of the feature; The aforementioned trust level focuses on the reliability of transmission between the third-party platform and the order monitoring and management system. However, the third-party platform's own security baseline still poses a risk of data leakage. Therefore, it is necessary to simultaneously evaluate the security protection capabilities of the third-party platform itself; Verify the security protection capabilities of third-party platforms that meet the trustworthiness standards; The assessment scope of security protection capabilities includes: vulnerability detection rate, encryption strength, and access control mechanism; The vulnerability detection rate, encryption strength, and access control mechanism are calculated using the security protection capability scoring formula to obtain the security protection capability score. The security protection capability scoring formula is as follows: ,in, 、 、 is the weight coefficient and , A is the vulnerability detection rate in vulnerability detection, and the specific calculation method is: , where B is the encryption strength coefficient in the encryption strength, which is determined by the complexity of the encryption algorithm. If the AES-256 encryption algorithm is used, the corresponding coefficient of the encryption strength coefficient is 1. If DES is used, the corresponding coefficient of the encryption strength coefficient is 0.5. C is the access control rate in the access control mechanism; Compare the calculated security protection capability score with the security protection capability score threshold; If the security protection capability score is greater than or equal to the security protection capability score threshold, it means that the security protection capability meets the standard and the third-party platform is allowed to access; If the security protection capability score is lower than the security protection capability score threshold, it means that the security protection capability meets the standard and third-party platform access is not allowed; It should be noted that the security protection capability scoring threshold is a reference value set by technical personnel in this industry and is used as an indicator to measure the security of third-party platforms; Generate pre-authorization rules for third-party platforms that meet security protection standards; Pre-authorization rules specifically include: data access scope, time window, and circuit breaker conditions; Generate data access scope using parsing cooperation agreement; Based on allowing third-party platforms to access, the third-party platform can extract the data access scope from the smart contract. The data access scope includes: field collection , Time Limit Window and circuit breaker mechanisms; The fuse mechanism is specifically as follows: When the third-party platform call behavior triggers the abnormal threshold, the circuit breaker mechanism is automatically activated; Abnormal thresholds specifically include: call frequency threshold and error rate threshold; If the third-party platform's call frequency is greater than or equal to the call frequency threshold, or the third-party platform's error rate is greater than or equal to the error rate threshold, or both of the above conditions are met, the order monitoring and management system will automatically activate the circuit breaker mechanism and cancel the third-party platform's access permission; The core function of the circuit breaker mechanism is to quickly cancel the third-party platform's access rights when the order monitoring management system detects abnormal risks during interaction between the third-party platform and the order monitoring management system, thereby preventing further losses. Through multi-dimensional feature fusion, federated learning trust calculation, security indicator quantification and smart contract analysis, the uniqueness verification of third-party identities and the automatic generation of pre-authorization rules are achieved. At the same time, the full process logic from ID generation to permission control is covered to ensure the security and compliance of data interaction.

[0025] Dynamic Intelligent Sandbox Isolation Module: This module sends order data to trusted third-party platforms based on pre-authorization rules. The module then encrypts and shards the order data at the field level. A quantum encryption transmission channel and a dynamic IP tunnel are established using the temporary keys of the encrypted shards. A memory-level micro-sandbox is created for each third-party platform, and the data transmission path for each order is switched in real time to enable order data transmission to the third-party platform. The micro-sandbox data is then destroyed upon completion of the transmission operation. The specific process of encrypting and sharding order data by field granularity is as follows: First, split the original order data by field granularity, and split the original order data D into m independent slices according to the field dimension. ,Each independent slice corresponds to a data field, and the original order data includes but is not limited to : order ID, amount, address, etc.; Use symmetric encryption algorithm to perform field-level encryption on each independent slice and generate temporary keys ; For each independent slice Use symmetric encryption algorithm (AES) to encrypt and generate encrypted data , using symmetric encryption algorithm to generate encrypted data The specific formula is: ,in, It is a temporary key generated by the symmetric encryption algorithm. Time t belongs to the pre-authorization time window. is the time window, For the i-th independent slice, the key validity period is determined by the time window in the pre-authorization rule To control, the third-party platform only obtains the temporary key of the authorized slice; The specific process of establishing a quantum encryption transmission channel and a dynamic IP tunnel using the temporary key of the encryption shard is as follows: Generate end-to-end encryption keys through quantum communication protocols , used to protect temporary keys during transmission and encrypted data transmission security; Establish a dynamic IP tunnel real-time switching strategy, and randomly select a set of IP nodes for the transmission path of each transaction The node combination n in is the total number of nodes, and the switching period is In the present invention, the switching cycle ; The specific path switching mechanism is: , is a random selection function; The specific process of creating a memory-level micro-sandbox for each third-party platform is as follows: Create a separate Docker container for each third-party platform and limit the number of CPU cores and memory limits , prohibit data from being written to disk; The micro-sandbox only allows reading temporary keys for authorized slices , and decrypt the encrypted data in memory , and prohibit the decryption and persistence of encrypted data. Specifically, the prohibition of encrypted data persistence is to only allow the decryption of encrypted data within the time window. If the decryption time exceeds the end time specified in the time window, the third-party platform is prohibited from continuing to decrypt; When the decryption operation is completed, the micro-sandbox data is destroyed; Through data slicing encryption, quantum transmission protection, sandbox resource isolation and automatic destruction mechanism, field-level data isolation and dynamic security protection are achieved, ensuring that third-party platforms can only temporarily access data within the authorized scope, avoiding data leakage and persistence risks.

[0026] Blockchain operation audit module: This module records third-party platform API call behavior, deploys smart contracts to verify third-party platform data access compliance in real time, builds a digital twin model of data flow, and visually displays sensitive data storage locations and third-party platform risk maps; The specific process of recording the third-party platform API call behavior is as follows: The evidence content includes the digital twin ID, timestamp, field-level access coordinates, and hash value comparison before and after the operation; The API call behavior is atomically stored on the chain, and the evidence content includes: digital twin ID, timestamp, field-level access coordinates, and hash values ​​before and after the operation; Generate hash values ​​for the order monitoring management system data before and after the operation to ensure that the order monitoring management system data cannot be tampered with. The hash calculation formula is: ,in, and They are the order data before and after the operation respectively; The specific process of real-time verification of third-party platform data access compliance is as follows: Deploy smart contracts to verify data access compliance in real time, including scope compliance, frequency compliance, and time compliance; 1. Perform scope compliance check to verify whether the access field is in the field set of the pre-authorization rule Inside; 2. Perform frequency compliance check and count the call frequency per unit time , and the frequency threshold contrast; 3. Perform time compliance check to check whether the call time is within the time window Inside; If any compliance check fails, the circuit breaker mechanism is triggered and the order monitoring and management system automatically activates the circuit breaker mechanism, revoking the third-party platform access permission. The specific process of building a digital twin model of data flow and visually displaying the sensitive data storage location and third-party platform risk map is as follows: Build a digital twin model of data flow and a risk map of third-party platforms; Using data fields as nodes and third-party platform access behaviors as edges, we construct a directed graph to quantitatively score the third-party platform risk map. We calculate the node risk score based on the node access frequency, sensitivity, and abnormal behavior in the directed graph. , the specific formula for calculating the node risk score is: ,in, is the weight coefficient. In the present invention, the sensitivity coefficient takes a value of 1 in the corresponding aspect of the amount field and a value of 0.3 in the corresponding aspect of other common fields; Through blockchain evidence storage, multi-dimensional compliance verification and graph model risk quantification, a full-link, tamper-proof audit of API call behavior is achieved, ensuring that data access behavior is traceable and verifiable. At the same time, the circuit breaker mechanism is automatically triggered through smart contracts to improve system security and compliance.

[0027] AI risk decision module: This module uses graph neural networks to access the risk map of third-party platforms, generates dynamic risk scores, and defines a three-level response strategy to achieve fully automated processing from risk detection to circuit breaker response. The specific process of using graph neural networks to access the risk map of third-party platforms is as follows: The graph neural network is connected to the risk map of the third-party platform. The graph structure of the risk map is defined first, and the third-party digital twin identity is used. , data fields , interactive behavior As node V, with interaction relationship as edge E, construct a directed graph ,in, , , For third-party digital twin identities, is the data field, For interactive behavior, 、 The two nodes connected by an edge, is the interaction weight between two nodes; Vectorize node features, third-party platform node features Integration of digital twin ID, credit score, historical risk records; field node features Including sensitivity level, access frequency; behavioral node characteristics Contains call frequency and timestamp sequence; Graph convolutional networks (GCNs) are used to aggregate features, update node representations, and access the risk graph of third-party platforms using graph neural networks. The specific process of generating a dynamic risk score is as follows: Using the node embedding vector output by the graph neural network, the cosine similarity between the current operation and the historical behavior is calculated; Compare the current risk characteristics with the industry average and calculate the deviation ; Combining similarity, deviation and real-time compliance indicators, a comprehensive risk score R is generated. The risk score formula is: in, is a weight coefficient. Compliance deduction points are calculated based on the compliance verification results. For example, 5 points are deducted for each violation. The specific process of defining the three-level response strategy is as follows: Execute corresponding measures according to the three-level response threshold. When the comprehensive risk score R is greater than or equal to the first score threshold and less than the second score threshold, a yellow warning notification is triggered; When the comprehensive risk score is greater than or equal to the second scoring threshold and less than the third scoring threshold, an orange quarantine warning is triggered, restricting third-party platforms from accessing order data; When the comprehensive risk score is greater than or equal to the third scoring threshold, a red circuit breaker warning is triggered and the circuit breaker mechanism is activated; For example, based on the generated comprehensive risk score R, three-level response thresholds are defined. The three-level response thresholds are specifically divided into: yellow warning (60≤R<80), orange isolation (80≤R<90), and red circuit breaker (R≥90); Execute corresponding measures according to the third-level response threshold. When the comprehensive risk score R is greater than or equal to 60 and less than 80, a yellow warning notification is triggered; When the comprehensive risk score R is greater than or equal to 80 and less than 90, an orange quarantine warning is triggered, restricting third-party platforms from accessing order data; When the comprehensive risk score R is greater than or equal to 90, a red circuit breaker warning is triggered and the circuit breaker mechanism is activated; By building risk maps, dynamic scoring models, and adaptive response strategies through graph neural networks, we achieve automated intelligent decision-making from risk detection to response.

[0028] The technical solution of the embodiment of the present invention is: generating a unique digital twin ID for the third-party platform, and combining the federated learning method to calculate the trust of non-order data transmission, verifying the security protection capabilities of third-party platforms that meet the trust standards, using smart contracts to parse cooperation agreements, and generating pre-authorization rules for third-party platforms that meet the security protection capabilities; sending order data to third-party platforms that meet the trust standards according to the pre-authorization rules, first encrypting and sharding the order data according to field granularity, using the temporary key of the encrypted shards to establish a quantum encryption transmission channel and a dynamic IP tunnel, creating a memory-level micro-sandbox for each third-party platform, switching the data transmission path for each order in real time, realizing order data transmission to the third-party platform, and destroying the micro-sandbox data at the end of the transmission operation; recording the API call behavior of the third-party platform, deploying smart contracts to verify the data access compliance of the third-party platform in real time, building a digital twin model of data flow, and visually displaying the sensitive data storage location and the third-party platform risk map; using graph neural networks to access the third-party platform risk map, generating dynamic risk scores, and defining a three-level response strategy to achieve fully automatic processing from risk detection to circuit breaker mechanism response.

[0029] Example 2 like Figure 2 As shown, based on Example 1, the present invention provides an order monitoring and management method, including: S1: Generate a unique digital twin ID for the third-party platform, and use federated learning to calculate the trust level of non-order data transmission. Verify the security protection capabilities of third-party platforms that meet the trust level standards. Use smart contracts to analyze cooperation agreements and generate pre-authorization rules for third-party platforms that meet the security protection capability standards. S2: Order data is sent to third-party platforms that meet the trust requirements according to pre-authorization rules. The order data is first encrypted and sharded according to field granularity. The temporary key of the encrypted shards is used to establish a quantum encryption transmission channel and a dynamic IP tunnel. A memory-level micro-sandbox is created for each third-party platform. The data transmission path for each order is switched in real time to realize order data transmission to the third-party platform. The micro-sandbox data is destroyed at the end of the transmission operation. S3: Evidence of third-party platform API call behavior is stored, smart contracts are deployed to verify the compliance of third-party platform data access in real time, a digital twin model of data flow is constructed, and a visual display of sensitive data storage locations and third-party platform risk maps is displayed; S4: Use graph neural networks to access the risk map of third-party platforms, generate dynamic risk scores, and define a three-level response strategy to achieve fully automatic processing from risk detection to circuit breaker mechanism response.

[0030] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An order monitoring and management system, characterized by: include: Digital Identity Twin Verification Module: Generates a unique digital twin ID for the third-party platform, and uses federated learning to calculate the trust level of non-order data transmission. It verifies the security protection capabilities of third-party platforms that meet the trust level standards, uses smart contracts to parse cooperation agreements, and generates pre-authorization rules for third-party platforms that meet the security protection capability standards. Dynamic Intelligent Sandbox Isolation Module: This module sends order data to trusted third-party platforms based on pre-authorization rules. The module then encrypts and shards the order data at the field level. A quantum encryption transmission channel and a dynamic IP tunnel are established using the temporary keys of the encrypted shards. A memory-level micro-sandbox is created for each third-party platform, and the data transmission path for each order is switched in real time to enable order data transmission to the third-party platform. The micro-sandbox data is then destroyed upon completion of the transmission operation. Blockchain operation audit module: This module records third-party platform API call behavior, deploys smart contracts to verify third-party platform data access compliance in real time, builds a digital twin model of data flow, and visually displays sensitive data storage locations and third-party platform risk maps; AI risk decision module: Utilizes graph neural networks to access the risk map of third-party platforms, generates dynamic risk scores, and defines a three-level response strategy to achieve fully automated processing from risk detection to circuit breaker mechanism response.

2. The order monitoring and management system according to claim 1, characterized in that: The specific process of generating a unique digital twin ID on a third-party platform is as follows: Convert the hardware fingerprint H, behavioral fingerprint B, and credit score C into feature vectors, calculate them through the hash encryption algorithm, and generate a digital twin ID. The digital twin ID generation formula is: ,in, represents the concatenation of feature vectors, It is a cryptographic hash function that uses the SHA-256 algorithm. is the hardware fingerprint feature vector, is the behavioral fingerprint feature vector, is the credit score feature vector.

3. The order monitoring and management system according to claim 1, characterized in that: The specific process of calculating the trust level of orderless data transmission is as follows: The order monitoring management system and the third-party platform calculate the trust of non-order data transmission through the federated learning method without sharing the original data. The specific formula for trust calculation is: ,in, The feature weights are optimized through federated learning, is the evaluation function of the i-th type of feature, n is the feature dimension, and i is the index of the feature.

4. The order monitoring and management system according to claim 1, characterized in that: The specific process of verifying the security protection capabilities of a third-party platform that meets the trustworthiness standards is as follows: The assessment scope of security protection capabilities includes: vulnerability detection rate, encryption strength, and access control mechanism; The vulnerability detection rate, encryption strength, and access control mechanism are calculated using the security protection capability scoring formula to obtain the security protection capability score. The security protection capability scoring formula is as follows: ,in, 、 、 is the weight coefficient and , A is the vulnerability detection rate in vulnerability detection, and the specific calculation method is: , where B is the encryption strength coefficient in the encryption strength, which is determined by the complexity of the encryption algorithm. If the AES-256 encryption algorithm is used, the corresponding coefficient of the encryption strength coefficient is 1. If DES is used, the corresponding coefficient of the encryption strength coefficient is 0.

5. C is the access control rate in the access control mechanism; If the security protection capability score is greater than or equal to the security protection capability score threshold, it means that the security protection capability meets the standard and the third-party platform is allowed to access; If the security protection capability score is less than the security protection capability score threshold, it means that the security protection capability meets the standard and third-party platform access is not allowed.

5. The order monitoring and management system according to claim 1, characterized in that: The specific process of generating pre-authorization rules for a third-party platform that meets the security protection capability standards is as follows: Generate data access scope using parsing cooperation agreement; By allowing third-party platforms to access, they can extract the data access scope from the smart contract. The data access scope includes: field set, time window and circuit breaker mechanism.

6. The order monitoring and management system according to claim 1, characterized in that: The specific process of encrypting and sharding the order data according to field granularity is as follows: Split the original order data D into m independent slices based on the field dimension. Each independent slice corresponds to a data field. Use a symmetric encryption algorithm to perform field-level encryption on each independent slice and generate a temporary key. Each independent slice is encrypted using the symmetric encryption algorithm AES to generate encrypted data. The specific formula is: ,in, It is a temporary key generated by the symmetric encryption algorithm. Time t belongs to the pre-authorization time window. is the time window, For the i-th independent slice, the key validity period is determined by the time window in the pre-authorization rule To control, the third-party platform only obtains the temporary key of the authorized slice.

7. The order monitoring and management system according to claim 1, characterized in that: The specific process of creating a memory-level micro-sandbox for each third-party platform is as follows: Create a separate Docker container for each third-party platform and limit the number of CPU cores and memory limits , prohibit data from being written to disk; The micro-sandbox only allows reading the temporary key of the authorized slice and decrypting the encrypted data in the memory, and prohibits the persistence of decrypted encrypted data. Specifically, the prohibition of persistence of encrypted data means that the encrypted data is only allowed to be decrypted within the time window. If the decryption time exceeds the end time specified in the time window, the third-party platform is prohibited from continuing to decrypt.

8. The order monitoring and management system according to claim 1, characterized in that: The specific process of recording the third-party platform API call behavior is as follows: The evidence content includes the digital twin ID, timestamp, field-level access coordinates, and hash value comparison before and after the operation; The API call behavior is atomized and stored on the chain. The content of the evidence includes: digital twin ID, timestamp, field-level access coordinates, and hash values ​​before and after the operation.

9. The order monitoring and management system according to claim 1, characterized in that: The specific process of real-time verification of third-party platform data access compliance is as follows: Deploy smart contracts to verify data access compliance in real time, including scope compliance, frequency compliance, and time compliance; Perform scope compliance check to verify whether the access field is within the field set of the pre-authorization rule; Perform frequency compliance check, count the call frequency per unit time, and compare it with the frequency threshold; Perform time compliance check to check whether the call time is within the time window; When any compliance check fails, the circuit breaker mechanism is triggered and the order monitoring and management system automatically activates the circuit breaker mechanism, revoking the third-party platform access permission.

10. The order monitoring and management system according to claim 1, characterized in that: The specific process of defining the three-level response strategy is as follows: Execute corresponding measures according to the three-level response threshold. When the comprehensive risk score R is greater than or equal to the first score threshold and less than the second score threshold, a yellow warning notification is triggered; When the comprehensive risk score is greater than or equal to the second scoring threshold and less than the third scoring threshold, an orange quarantine warning is triggered, restricting third-party platforms from accessing order data; When the comprehensive risk score is greater than or equal to the third scoring threshold, a red circuit breaker warning is triggered and the circuit breaker mechanism is activated.

Images