A base liquor pledge supervision method and system based on multi-source credible data fusion

Abstract

This invention relates to the field of Industrial Internet of Things (IIoT) and provides a method and system for supervising the pledge of base liquor based on multi-source trusted data fusion. The method includes: associating and binding the basic attribute information of the pledged base liquor containers with multiple types of heterogeneous sensing devices to generate digital identity identifiers; collecting multi-dimensional raw data from multiple heterogeneous sensing devices to obtain a multi-source heterogeneous dataset; performing consistency cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset to obtain a multi-source consensus spatiotemporal stamp; digitally signing the multi-source consensus spatiotemporal stamp and associated data digest and submitting it to a consortium blockchain network to obtain an on-chain trusted state record; performing risk quantification processing through a multi-dimensional dynamic risk association model to obtain a comprehensive risk score; and triggering a level-based early warning and handling process based on the comparison result between the comprehensive risk score and a preset risk threshold. This invention achieves multi-source trusted data fusion and dynamic intelligent risk assessment of pledged base liquor assets.

Description

Technical Field

[0001] This invention belongs to the field of industrial Internet of Things technology, and in particular relates to a method and system for supervising base liquor pledge based on multi-source trusted data fusion. Background Technology

[0002] Base liquor, as a core semi-finished asset of liquor producers, is highly valuable and is an important collateral for movable asset pledge financing. In movable asset pledge financing, financial institutions need to continuously monitor the quantity, quality, and storage status of the pledged base liquor to ensure the safety of loan assets. Current monitoring models mainly rely on traditional methods such as regular manual inventory checks, physical seals, and on-site inspections by third-party monitoring companies. In recent years, preliminary explorations have emerged, including deploying IoT devices such as temperature and humidity sensors and liquid level sensors in liquor warehouses, or uploading the hash values ​​of collected data to the blockchain for tamper-proof evidence storage.

[0003] However, existing technical solutions suffer from key technical problems: data may be forged or interfered with at the source of collection, and existing solutions lack mechanisms for cross-verification and consensus-building of multi-source data at the data generation end. For example, attackers can tamper with location information using GPS spoofing devices or physically interfere with a single sensor to forge environmental data, resulting in the blockchain recording only the hash value of contaminated data, which cannot accurately reflect the actual state of pledged assets. Consequently, risk assessments and regulatory decisions are based on untrusted data. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of the invention is to provide a method and system for supervising base liquor pledge based on multi-source trusted data fusion.

[0005] This invention provides a method for supervising base liquor pledge based on multi-source trusted data fusion, comprising: S1. Associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container. S2. Using the digital identity identifier as an index, collect multi-dimensional raw data from multiple heterogeneous sensing devices to obtain a multi-source heterogeneous dataset. S3. Using the spatiotemporal consensus weight algorithm, perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset to obtain the multi-source consensus spatiotemporal stamp; S4. Digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record. S5. Using the on-chain trusted state records and external market data as input, a multi-dimensional dynamic risk correlation model is used to quantify the risks of multiple dimensions, including asset value, data credibility, operational compliance and environmental stability, to obtain a comprehensive risk score. S6. Based on the comparison result between the comprehensive risk score and the preset risk threshold, trigger the level warning and handling process to complete the supervision of base liquor pledge.

[0006] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S1 further includes: S11. Enter the basic attribute information of the pledged base liquor container to form an asset basic file; S12. Process multiple pledged base wine containers using a decentralized identity generation algorithm to obtain a globally unique decentralized identity. S13. Bind the device IDs of heterogeneous sensing devices, including liquid level sensors, temperature and humidity sensors, vibration sensors, GNSS positioning modules, smart electronic locks, and readers, to decentralized identity identifiers, and store the hash value of the binding relationship on the blockchain to complete the initial filing of the asset digital twin and obtain the corresponding digital identity identifier of the pledged base wine container.

[0007] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S2 further includes: S21. Container status data of multiple pledged base liquor containers are collected by multiple heterogeneous sensing devices. The container status data includes: liquid level, volume, multi-point average temperature, ambient temperature and humidity, and light intensity data. S22. Obtain current latitude and longitude coordinates, altitude and UTC timestamp from multiple heterogeneous sensing devices, and obtain network time to obtain spatiotemporal reference data; S23. The operator's employee ID is read by multiple heterogeneous sensing devices, and the opening and closing status and abnormal vibration detection results of multiple pledged base liquor containers are reported to obtain operation event data. S24. The container status data, the spatiotemporal reference data, and the operation event data are aggregated to the edge computing gateway and aligned with the digital identity of the corresponding pledged base wine container as the association key to obtain a multi-source heterogeneous dataset.

[0008] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S3 further includes: S31. Perform validity pre-verification on the multi-source heterogeneous dataset, remove abnormal sampling points whose values ​​exceed the preset reasonable range, and obtain the cleaned valid dataset. S32. For each data source in the valid dataset, calculate the positional deviation, time deviation, and reading deviation of similar sensors between the current data source and other data sources, and input the normalization function to obtain the current consistency score of multiple data sources. S33. For each data source in the effective dataset, the current consistency score, prior confidence and the weight of the previous period are weighted and fused to obtain a dynamic confidence weight vector of multiple data sources. S34. Based on the dynamic confidence weight vector, the location, time and environmental observations of the corresponding data source are weighted and summed to obtain the consensus location value, consensus time value and consensus environment value respectively. S35. Serialize the digital identity identifier, consensus location value, consensus time value, consensus environment value, operator ID, and dynamic confidence weight vector of the pledged base liquor container and calculate the hash value using the SHA-256 algorithm to obtain the multi-source consensus spatiotemporal stamp.

[0009] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S4 further includes: S41. The hardware security module HSM built into the edge computing gateway performs private key signing on the multi-source consensus time-space stamp to obtain signature data; S42. Construct a blockchain transaction from the data digest of digital identity, multi-source consensus time stamp, signature data and multi-dimensional raw data, and submit it to a consortium blockchain consisting of nodes composed of multiple regulatory agencies. S43. The signature data is verified through a smart contract on the consortium blockchain. When the verification is successful, the digital identity identifier is associated with the multi-source consensus time stamp and written into the state machine, triggering an on-chain state update event and obtaining a trusted on-chain state record.

[0010] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S5 further includes: S51. The consortium blockchain pulls the on-chain trusted state record, obtains the current market unit price of the base liquor through the oracle interface, and calculates the asset value risk sub-score by combining the current quantity and the pledged loan amount. S52. Extract the dynamic confidence weight vector from the multi-source consensus spatiotemporal stamp, calculate the information entropy, and count the frequency of data anomalies in the current monitoring period to obtain the data credibility risk sub-score. S53. Extract the number of unauthorized unlocks and the frequency of abnormal vibrations from the trusted state record on the chain, and calculate the operation compliance risk sub-score; S54. Calculate the environmental stability risk sub-score by calculating the cumulative duration and standard deviation of environmental parameters exceeding the limits of standard operating procedures (SOPs). S55. The asset value risk sub-score, the data credibility risk sub-score, the operational compliance risk sub-score, and the environmental stability risk sub-score are weighted and summed to obtain the comprehensive risk score.

[0011] According to the method for supervising base liquor pledge based on multi-source trusted data fusion provided by the present invention, step S6 further includes: S61. The comprehensive risk score is compared with the preset multi-level risk thresholds in sequence to determine the risk level to which the current score belongs. The risk levels include five risk levels: normal, attention, secondary, suspicious and loss. S62. For suspicious and loss-related risk levels, invoke the asset locking clause in the smart contract on the consortium blockchain to lock the digital twin of the base wine container associated with the corresponding digital identity to block the subsequent outbound approval process.

[0012] A second aspect of the present invention provides a base liquor pledge supervision system based on multi-source trusted data fusion, comprising: Binding module: Used to associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container; Receiving module: used to receive multi-dimensional raw data from multiple heterogeneous sensing devices, indexed by the digital identity, to obtain a multi-source heterogeneous dataset; The calculation module is used to perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset using a spatiotemporal consensus weight algorithm to obtain the multi-source consensus spatiotemporal stamp. The signature module is used to digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record. Quantification module: It is used to quantify risks in multiple dimensions, including asset value, data credibility, operational compliance and environmental stability, through a multi-dimensional dynamic risk correlation model, using the on-chain trusted state records and external market data as input, and obtain a comprehensive risk score. Comparison module: used to trigger a level warning and handling process based on the comparison result of the comprehensive risk score and the preset risk threshold, so as to complete the supervision of base liquor pledge.

[0013] A third aspect of this invention provides a base liquor pledge monitoring device based on multi-source trusted data fusion, comprising: A memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause a base liquor pledge supervision device based on multi-source trusted data fusion to execute a base liquor pledge supervision method based on multi-source trusted data fusion as described above.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement a base liquor pledge supervision method based on multi-source trusted data fusion as described in any of the preceding claims.

[0015] This invention first associates and binds decentralized identity identifiers (DIDs) with various heterogeneous sensing devices, giving each physical base wine container a unique and fully informed corresponding entity in the digital world. This fundamentally eliminates the ambiguity in the mapping between physical assets and digital records, laying a reliable data index foundation for subsequent fully automated supervision. Building upon this, the invention uses a spatiotemporal consensus weighting algorithm to cross-calculate the consistency deviations of multiple heterogeneous data sources and dynamically adjusts the confidence weight vectors of each data source based on historical reputation and prior credibility. This ensures that even if some data sources suffer from GPS spoofing or sensor interference, the invention can still output reliable location, time, and environmental consensus values ​​from multi-source consensus. The generated multi-source consensus spatiotemporal stamp is then signed by the hardware security module HSM and stored on the blockchain, demonstrating significantly superior anti-fraud capabilities and legal evidentiary validity compared to existing single-data-source storage schemes. Furthermore, the multidimensional dynamic risk association model of this invention integrates the risk sub-scores of four dimensions—asset value, data credibility, operational compliance, and environmental stability—into a comprehensive risk score. In particular, it introduces the information entropy of the weight vector into the calculation of the data credibility dimension, enabling the system to automatically identify the hidden risk of concentrated reliance on data sources and correct the score through a penalty coefficient. This effectively detects complex fraudulent activities such as collusion between internal and external parties and illegal transfer of assets under the cover of compliant operations, reducing the risk detection time from several weeks in traditional manual inventory checks to a response time of seconds. Moreover, it achieves automatic disposal of high-risk assets through smart contract locking clauses, significantly reducing the credit loss exposure and manual supervision costs of financial institutions. Attached Figure Description

[0016] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. It is obvious that the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings.

[0017] Figure 1 This is a schematic diagram of a method for supervising base liquor pledge based on multi-source trusted data fusion, provided by an embodiment of the present invention. Figure 2 This is a schematic diagram of a base liquor pledge supervision system based on multi-source trusted data fusion, provided as an embodiment of the present invention. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions 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, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0019] Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts disclosed in this invention.

[0020] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The terms "installed," "connected," and "linked" should be interpreted broadly; for example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of methods and systems consistent with some aspects of the invention as detailed in the appended claims.

[0022] To better understand this invention, the technical terms appearing in the embodiments of this invention will be explained in detail below.

[0023] The Internet of Things (IoT) uses various information sensors, positioning technologies, radio frequency identification (RFID) technologies, and other devices and technologies to collect various information in real time about any object or process that needs to be monitored, connected, or interacted with. Through various network access methods, it enables ubiquitous connection between things and between things and people, achieving intelligent perception, identification, and management of items and processes.

[0024] Blockchain: A decentralized, multi-party maintained distributed ledger technology that uses cryptography to ensure secure transmission and access, enabling consistent, immutable, and traceable data storage.

[0025] Smart contract: A piece of code deployed on the blockchain that can automatically execute contract terms when preset conditions are met.

[0026] Spacetime stamp: A data structure that encrypts and binds specific time and space information to prove that an event occurred at a specific time and place.

[0027] Multi-Source Consensus Spatiotemporal Stamp (MCSS): This invention proposes an enhanced spatiotemporal stamp, which generates a highly reliable and fraud-resistant encrypted data digest by weighting and cross-validating data from multiple heterogeneous data sources (such as satellite positioning, network time, sensor arrays, operator identities, etc.) using the Spatiotemporal Consensus Weighted Algorithm (SCWA).

[0028] Spatiotemporal Consensus Weighting Algorithm (SCWA): This invention proposes a method for dynamically calculating the confidence weights of each data source and generating MCSS.

[0029] Multi-dimensional Dynamic Risk Correlation Model (MDRC-Model): The model proposed in this invention is used to continuously assess the risk of pledged base liquor assets. Its inputs include not only the physical state and market value of the assets, but also multiple dimensions such as the credibility of asset data, the stability of the regulatory environment, and the compliance of historical operations.

[0030] Blockchain oracles: Third-party services that securely and reliably provide data from the external world (off-chain) to smart contracts on the blockchain.

[0031] Edge computing gateway: A computing device deployed near the data source (such as a wine cellar) to collect sensor data in real time and run the SCWA algorithm locally to generate MCSS, thereby reducing the computing pressure on the cloud and improving response speed.

[0032] The embodiments of the present invention are described below with reference to the figures.

[0033] like Figure 1 As shown, this invention provides a method for supervising base liquor pledge based on multi-source trusted data fusion, including: S1. Associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container.

[0034] Step S1 further includes: S11. Enter the basic attribute information of the pledged base wine container to form the basic asset file.

[0035] Furthermore, in step S11, the present invention aims to digitally archive assets. In specific implementation, under the supervision of personnel from financial institutions, the asset administrator of the winery enters basic attribute information such as the wine jar or tank number, storage date, storage location information, aroma type, grade, initial liquid level, volume, density, and weight of each pledged base wine container into the system interface, forming the basic asset file for that container.

[0036] S12. Process multiple pledged base wine containers using a decentralized identity generation algorithm to obtain a globally unique decentralized identity.

[0037] In step S12, the present invention processes the asset basic file data of each container through a globally unique identifier, namely a decentralized identity identifier generation algorithm, and outputs a decentralized identity identifier that corresponds one-to-one with the physical container, serving as its unique identity credential in the digital world.

[0038] S13. Bind the device IDs of heterogeneous sensing devices, including liquid level sensors, temperature and humidity sensors, vibration sensors, GNSS positioning modules, smart electronic locks, and readers, to decentralized identity identifiers, and store the hash value of the binding relationship on the blockchain to complete the initial filing of the asset digital twin and obtain the corresponding digital identity identifier of the pledged base wine container.

[0039] In step S13, the present invention binds the device serial numbers of the liquid level sensor, temperature and humidity sensor, vibration sensor, GNSS positioning module, smart electronic lock, and NFC / RFID reader deployed on or around the container with the aforementioned decentralized identity, forming a mapping relationship between one physical container and one decentralized identity, and one set of sensing devices. After binding, the present invention performs a hash operation on this mapping relationship and uploads the hash value to the consortium blockchain for notarization, ensuring that the binding relationship itself cannot be tampered with afterward. Ultimately, each physical base wine container has a digital twin in the system containing complete attribute information and bound to the complete set of sensing devices. All subsequent collected data is indexed and aggregated using this decentralized identity as the association key.

[0040] S2. Using the digital identity identifier as an index, collect multi-dimensional raw data from multiple heterogeneous sensing devices to obtain a multi-source heterogeneous dataset.

[0041] Step S2 further includes: S21. Multiple container status data of pledged base liquor containers are collected by various heterogeneous sensing devices. The container status data includes: liquid level, volume, multi-point average temperature, ambient temperature and humidity, and light intensity data.

[0042] In step S21, the present invention uses decentralized identity identifiers as indexes to drive various heterogeneous sensing devices to continuously collect data at a preset sampling frequency. Specifically, the liquid level sensor measures the liquid level height in the container through electromagnetic wave sensing and calculates the current volume by combining it with three-dimensional modeling; the multi-point temperature probe collects the temperature values ​​at multiple measurement locations in the container and takes the average, while collecting the ambient temperature, humidity and light intensity, and finally the above data together constitute the container status data.

[0043] S22. Obtain current latitude and longitude coordinates, altitude and UTC timestamp from multiple heterogeneous sensing devices, and obtain network time to obtain spatiotemporal reference data.

[0044] In step S22, the present invention obtains the current latitude and longitude coordinates and altitude based on the GPS, Beidou and GLONASS constellations supported by the GNSS positioning module, and calculates the UTC timestamp from the satellite signal; in addition, the NTP / PTP network time protocol obtains the network time from the Internet time server, and the two together serve as spatiotemporal reference data, forming a dual reference of time and location from different channels.

[0045] S23. The operator's employee ID is read by multiple heterogeneous sensing devices, and the opening and closing status and abnormal vibration detection results of multiple pledged base liquor containers are reported to obtain operation event data.

[0046] In step S23, the NFC / RFID reader reads the employee badge number when the operator approaches with the badge and records an access or operation event; at the same time, the smart electronic lock located in the base wine container reports the current switch status and whether abnormal vibration is detected in real time, which together constitute the operation event data.

[0047] S24. The container status data, the spatiotemporal reference data, and the operation event data are aggregated to the edge computing gateway and aligned with the digital identity of the corresponding pledged base wine container as the association key to obtain a multi-source heterogeneous dataset.

[0048] In step S24, the present invention aggregates the three types of data obtained in steps S21-S24 through LoRaWAN or LAN technology based on the edge computing gateway. Using the decentralized identity of the corresponding container as the association key, the container status data, spatiotemporal reference data, and operation event data of the same container in the same collection period are time-aligned and integrated to obtain a multi-source heterogeneous dataset. Each record in this dataset carries a decentralized identity and clearly points to the corresponding physical container.

[0049] S3. Using the spatiotemporal consensus weight algorithm, perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset to obtain the multi-source consensus spatiotemporal stamp.

[0050] Step S3 further includes: S31. Perform validity pre-verification on the multi-source heterogeneous dataset, remove abnormal sampling points whose values ​​exceed the preset reasonable range, and obtain the cleaned valid dataset.

[0051] After the multi-source heterogeneous dataset arrives at the edge computing gateway, in step S31, the present invention first performs a validity pre-verification, checking whether the values ​​of each data source fall within a preset reasonable range, such as the wine cellar temperature not exceeding 50℃ and the liquid level reading not lower than 0, etc. Abnormal sampling points that exceed the range are removed from the dataset to obtain a cleaned valid dataset.

[0052] S32. For each data source in the valid dataset, calculate the positional deviation, time deviation, and reading deviation of similar sensors between the current data source and other data sources, and input the normalization function to obtain the current consistency score of multiple data sources.

[0053] In step S32, this invention calculates three types of deviations for each data source in the valid dataset: location deviation, which is the spatial distance difference between GNSS coordinates and base station or Wi-Fi positioning coordinates; time deviation, which is the absolute difference between satellite UTC timestamps and NTP network time; and similar sensor reading deviation, which is the difference between readings from different sensors for the same physical quantity. After calculation, this invention inputs the three types of deviation values ​​into a normalization function, converting them into a consistency score between 0 and 1. The smaller the deviation, the higher the consistency score, reflecting the degree of agreement between this data source and other data sources.

[0054] S33. For each data source in the effective dataset, the current consistency score, prior confidence, and weight of the previous period are weighted and fused to obtain a dynamic confidence weight vector for multiple data sources.

[0055] Based on the current consistency scores of multiple data sources calculated in step S32, this invention performs weighted fusion calculation on the current consistency score of each data source, the prior confidence of the data source (e.g., GNSS prior confidence is 0.9, base station positioning is 0.7, both are static configuration values), and the historical weight of the data source in the previous period, according to the smoothing factor and the adjustment factor, to obtain the dynamic confidence weight of the data source in the current period. The weights of all data sources together constitute the dynamic confidence weight vector.

[0056] Furthermore, weight The update formula is: in, Indicates time, For data source index value, For the first Data sources Weight of time, For the first Data sources Weight of time, For the first Data sources Consistency score at any given moment For the first Prior confidence of each data source The smoothing factor is set to 0.7. As the first regulating factor, It is the second regulatory factor.

[0057] S34. Based on the dynamic confidence weight vector, the location, time and environment observations of the corresponding data source are weighted and summed to obtain the consensus location value, consensus time value and consensus environment value respectively.

[0058] Furthermore, in step S34, the present invention uses the weight values ​​corresponding to each data source in the dynamic confidence weight vector as coefficients to perform weighted summation and normalization on the location observation values, time observation values ​​and environmental observation values ​​provided by each data source, respectively, to obtain consensus location value, consensus time value and consensus environment value. The obtained calculated values ​​represent the most credible location, time and environmental state after multi-source consensus within the current collection period.

[0059] S35. Serialize the digital identity identifier, consensus location value, consensus time value, consensus environment value, operator ID, and dynamic confidence weight vector of the pledged base liquor container and calculate the hash value using the SHA-256 algorithm to obtain the multi-source consensus spatiotemporal stamp.

[0060] Finally, this invention serializes and concatenates the decentralized identity identifier, consensus location value, consensus time value, consensus environment value, operator ID number, and dynamic confidence weight vector of the current container, inputs them into the SHA-256 hash algorithm for calculation, and outputs a fixed-length hash digest, namely the multi-source consensus spatiotemporal stamp. This multi-source consensus spatiotemporal stamp uniquely corresponds to the complete data state verified by multi-source consensus within this collection period.

[0061] S4. Digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record.

[0062] Step S4 further includes: S41. The hardware security module HSM built into the edge computing gateway performs private key signing processing on the multi-source consensus spatiotemporal stamp to obtain signature data.

[0063] After generating the multi-source consensus time stamp, in step S41, the present invention aims to solidify it into the consortium blockchain in an immutable form. The hardware security module is a dedicated security chip built into the edge computing gateway. Its core function is to store the private key and perform signature operations in a physically isolated secure environment. The private key never leaves the chip in plaintext form from the time it is generated.

[0064] Specifically, the present invention uses a hardware security module built into the edge computing gateway to perform asymmetric cryptographic signature operation on the multi-source consensus spatiotemporal stamp with a private key and outputs signature data. This signature data forms a unique correspondence with the multi-source consensus spatiotemporal stamp, and any subsequent tampering with the multi-source consensus spatiotemporal stamp will result in signature verification failure.

[0065] S42. Construct a blockchain transaction from the data digest of digital identity, multi-source consensus time stamp, signature data, and multi-dimensional original data, and submit it to a consortium blockchain consisting of nodes composed of multiple regulatory agencies.

[0066] In step S42, the present invention constructs a standard-format blockchain transaction by combining decentralized identity identifiers, multi-source consensus spatiotemporal stamps, signature data, and a data digest obtained by hashing and compressing the multi-dimensional raw data collected in step S2. This transaction is then submitted to the consortium blockchain through the gateway's network interface. The consortium blockchain consists of nodes comprised of financial institutions, wineries, regulatory agencies, and auditing firms. Each node maintains a complete copy of the ledger, and any data anomaly in any node will be detected by the other nodes.

[0067] S43. The signature data is verified through a smart contract on the consortium blockchain. When the verification is successful, the digital identity identifier is associated with the multi-source consensus time stamp and written into the state machine, triggering an on-chain state update event and obtaining a trusted on-chain state record.

[0068] In step S43, after the transaction is submitted in step S42, the smart contract deployed on the consortium blockchain automatically receives the transaction, extracts its signature data, and performs a signature verification operation using the public key corresponding to the edge computing gateway. The verification process involves decrypting the signature data with the public key and comparing it with the multi-source consensus spatiotemporal stamp. If the two match, the verification is successful. After successful verification, the smart contract writes the association between the decentralized identity identifier and the multi-source consensus spatiotemporal stamp into the on-chain state machine. The state machine is a key-value storage structure maintained by the smart contract, updated with the decentralized identity identifier as the key and the latest multi-source consensus spatiotemporal stamp as the value. It also broadcasts an on-chain state update event to off-chain applications, thus obtaining a trusted on-chain state record.

[0069] S5. Using the on-chain trusted state records and external market data as input, a multi-dimensional dynamic risk correlation model is used to quantify the risks of multiple dimensions, including asset value, data credibility, operational compliance, and environmental stability, to obtain a comprehensive risk score.

[0070] In step S5, after the on-chain trusted state record is generated, the cloud-based AI platform detects the on-chain state update event and then initiates the calculation process of the multi-dimensional dynamic risk association model. Specifically, the multi-dimensional dynamic risk association model is a calculation framework in this invention that weights and integrates quantitative indicators of four dimensions—asset value, data credibility, operational compliance, and environmental stability—into a single risk score. Each dimension calculates its sub-score independently and then integrates them into the comprehensive scoring module.

[0071] Step S5 further includes: S51. The consortium blockchain pulls the on-chain trusted state record, obtains the current market price of the base liquor through the oracle interface, and calculates the asset value risk sub-score by combining the current quantity and the pledged loan amount.

[0072] In step S51, the present invention first pulls the on-chain trusted state record corresponding to the current container from the consortium blockchain to obtain the current quantity of base liquor; simultaneously, it obtains the current market price per unit of base liquor from an external data source through an oracle interface. An oracle is a middleware service that reliably transmits external blockchain data to on-chain or off-chain smart systems; the present invention uses it to obtain off-chain market price data. After obtaining the data, the present invention multiplies the current market price per unit by the current quantity to obtain the current asset market value, and then performs a ratio calculation with the pledged loan amount. When the ratio is too low or the market price fluctuates drastically, the asset value risk sub-score increases accordingly.

[0073] S52. Extract the dynamic confidence weight vector from the multi-source consensus spatiotemporal stamp, calculate the information entropy, and count the frequency of data anomalies in the current monitoring period to obtain the data credibility risk sub-score.

[0074] In step S52, this invention extracts a dynamic confidence weight vector from the multi-source consensus spatiotemporal stamp and calculates the information entropy of this vector. Information entropy is an indicator that measures the uniformity of a set of numerical distributions. When the weights in the dynamic confidence weight vector are concentrated in a few sensors for a long period, the information entropy decreases, indicating a decline in data source diversity and a risk of single-point dependence. Simultaneously, the proportion of sampling points marked as abnormal in the valid dataset within the current monitoring period is calculated as the frequency of data anomalies. Finally, this invention inputs both the information entropy and the frequency of data anomalies into a sub-score calculation function to obtain the data credibility risk sub-score.

[0075] S53. Extract the number of unauthorized unlocks and the frequency of abnormal vibrations from the trusted state record on the chain, and calculate the operation compliance risk sub-score.

[0076] Next, the present invention extracts operation event data from the on-chain trusted state record, counts the number of unlocking events triggered by unauthorized identities within the current monitoring period, and the frequency of abnormal vibration events reported by the smart electronic lock. The two indicators are input into the sub-score calculation function to obtain the operation compliance risk sub-score. An abnormal increase in any indicator will directly raise the sub-score.

[0077] S54. Calculate the environmental stability risk sub-score by calculating the cumulative duration and standard deviation of environmental parameters exceeding the limits of standard operating procedures (SOPs).

[0078] In step S54, in terms of environmental stability, the present invention extracts historical time-series data of environmental parameters such as temperature, humidity and light intensity from the container status data, calculates the cumulative duration of each parameter exceeding the preset limit of the standard operating procedure, and calculates the standard deviation of the time-series data of each parameter relative to the benchmark value of the standard operating procedure. The two indicators are input into the sub-score calculation function to obtain the environmental stability risk sub-score.

[0079] S55. The asset value risk sub-score, the data credibility risk sub-score, the operational compliance risk sub-score, and the environmental stability risk sub-score are weighted and summed to obtain the comprehensive risk score.

[0080] In step S55, the present invention multiplies the asset value risk sub-score, data credibility risk sub-score, operational compliance risk sub-score, and environmental stability risk sub-score by the risk weights of the corresponding dimensions and then sums them. The risk weights of each dimension are pre-set by domain experts or obtained by machine learning training on historical risk event data. The weighted sum is then used to output a comprehensive risk score.

[0081] Furthermore, the expression for the comprehensive risk score is as follows: in, To calculate the comprehensive risk score, This is a sub-score for asset value risk. The weights corresponding to the asset value risk sub-scores. For data credibility risk sub-score, The weights corresponding to the data credibility risk sub-scores. For operational compliance risk sub-scores, The weights corresponding to the sub-scores of operational compliance risk. For environmental stability risk sub-scores, The weights are the sub-scores corresponding to environmental stability risk.

[0082] S6. Based on the comparison result between the comprehensive risk score and the preset risk threshold, trigger the level warning and handling process to complete the supervision of base liquor pledge.

[0083] Step S6 further includes: S61. The comprehensive risk score is compared with the preset multi-level risk thresholds in sequence to determine the risk level to which the current score belongs. The risk levels include five risk levels: normal, attention, secondary, suspicious and loss.

[0084] In step S61, the present invention compares the comprehensive risk score calculated in step S5 with the multi-level risk thresholds preset by the system. The multi-level risk thresholds divide the value space of the comprehensive risk score into five intervals, corresponding to five risk levels: normal, attention, secondary, suspicious, and loss. The comprehensive risk score of the normal category is lower than the first threshold; the comprehensive risk score of the attention category is between the first threshold T1 and the second threshold T1; the comprehensive risk score of the secondary category is between the second threshold T2 and the third threshold T3; the comprehensive risk score of the suspicious category is between the third threshold T3 and the fourth threshold T4; and the comprehensive risk score of the loss category is not lower than the fourth threshold T4. After the comparison is completed, the system determines the level of the current comprehensive risk score and pushes a graded early warning notification to the corresponding responsible party. At the same time, it generates a visualized risk analysis report containing risk sub-scores of each dimension and historical trend comparisons.

[0085] S62. For suspicious and loss-related risk levels, invoke the asset locking clause in the smart contract on the consortium blockchain to lock the digital twin of the base wine container associated with the corresponding digital identity to block the subsequent outbound approval process.

[0086] For suspicious and loss-related risk levels, this invention further invokes pre-deployed asset locking clauses in the smart contract on the consortium blockchain. Upon receiving the invocation instruction, this clause uses the decentralized identity of the current container as the search key to locate the corresponding digital twin record in the on-chain state machine, updates its state field from normal to locked, and broadcasts the locking event to the outbound approval module after the state machine completes the write. Upon receiving this event, the outbound approval module rejects all subsequent outbound applications for that container until the risk is resolved and the authorized party manually unlocks it.

[0087] Specifically, the risk threshold classification and corresponding handling measures in step S6 are shown in Table 1.

[0088] Table 1 Risk Thresholds and Classifications like Figure 2 As shown, the present invention also provides a base liquor pledge supervision system based on multi-source trusted data fusion, comprising: Binding module 100: Used to associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container; Receiving module 200: Used to receive multi-dimensional raw data from multiple heterogeneous sensing devices collected by the digital identity identifier as an index, and obtain a multi-source heterogeneous dataset; Calculation module 300: used to perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset through a spatiotemporal consensus weight algorithm to obtain a multi-source consensus spatiotemporal stamp; Signature module 400: Used to digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record; Quantification Module 500: This module takes the on-chain trusted state records and external market data as inputs, and uses a multi-dimensional dynamic risk correlation model to perform risk quantification on multiple dimensions, including asset value, data credibility, operational compliance, and environmental stability, to obtain a comprehensive risk score. Comparison module 600: Used to trigger a level warning and handling process based on the comparison result between the comprehensive risk score and the preset risk threshold, so as to complete the supervision of base liquor pledge.

[0089] The present invention also provides a base liquor pledge supervision device based on multi-source trusted data fusion, comprising: a memory and at least one processor, wherein the memory stores instructions; at least one processor invokes the instructions in the memory to cause the base liquor pledge supervision device based on multi-source trusted data fusion to perform a base liquor pledge supervision method based on multi-source trusted data fusion as described above.

[0090] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement a base liquor pledge supervision method based on multi-source trusted data fusion as described in any of the above claims.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A base liquor pledge supervision method based on multi-source trusted data fusion, characterized in that, include: S1. Associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container. S2. Using the digital identity identifier as an index, collect multi-dimensional raw data from multiple heterogeneous sensing devices to obtain a multi-source heterogeneous dataset. S3. Using the spatiotemporal consensus weight algorithm, perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset to obtain the multi-source consensus spatiotemporal stamp; S4. Digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record. S5. Using the on-chain trusted state records and external market data as input, a multi-dimensional dynamic risk correlation model is used to quantify the risks of multiple dimensions, including asset value, data credibility, operational compliance and environmental stability, to obtain a comprehensive risk score. S6. Based on the comparison result between the comprehensive risk score and the preset risk threshold, trigger the level warning and handling process to complete the supervision of base liquor pledge.

2. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S1 further includes: S11. Enter the basic attribute information of the pledged base liquor container to form an asset basic file; S12. Process multiple pledged base wine containers using a decentralized identity generation algorithm to obtain a globally unique decentralized identity. S13. Bind the device IDs of heterogeneous sensing devices, including liquid level sensors, temperature and humidity sensors, vibration sensors, GNSS positioning modules, smart electronic locks, and readers, to decentralized identity identifiers, and store the hash value of the binding relationship on the blockchain to complete the initial filing of the asset digital twin and obtain the corresponding digital identity identifier of the pledged base wine container.

3. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S2 further includes: S21. Container status data of multiple pledged base liquor containers are collected by multiple heterogeneous sensing devices. The container status data includes: liquid level, volume, multi-point average temperature, ambient temperature and humidity, and light intensity data. S22. Obtain current latitude and longitude coordinates, altitude and UTC timestamp from multiple heterogeneous sensing devices, and obtain network time to obtain spatiotemporal reference data; S23. The operator's employee ID is read by multiple heterogeneous sensing devices, and the opening and closing status and abnormal vibration detection results of multiple pledged base liquor containers are reported to obtain operation event data. S24. The container status data, the spatiotemporal reference data, and the operation event data are aggregated to the edge computing gateway and aligned with the digital identity of the corresponding pledged base wine container as the association key to obtain a multi-source heterogeneous dataset.

4. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S3 further includes: S31. Perform validity pre-verification on the multi-source heterogeneous dataset, remove abnormal sampling points whose values ​​exceed the preset reasonable range, and obtain the cleaned valid dataset. S32. For each data source in the valid dataset, calculate the positional deviation, time deviation, and reading deviation of similar sensors between the current data source and other data sources, and input the normalization function to obtain the current consistency score of multiple data sources. S33. For each data source in the effective dataset, the current consistency score, prior confidence and the weight of the previous period are weighted and fused to obtain a dynamic confidence weight vector of multiple data sources. S34. Based on the dynamic confidence weight vector, the location, time and environmental observations of the corresponding data source are weighted and summed to obtain the consensus location value, consensus time value and consensus environment value respectively. S35. Serialize the digital identity identifier, consensus location value, consensus time value, consensus environment value, operator ID, and dynamic confidence weight vector of the pledged base liquor container and calculate the hash value using the SHA-256 algorithm to obtain the multi-source consensus spatiotemporal stamp.

5. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S4 further includes: S41. The hardware security module HSM built into the edge computing gateway performs private key signing on the multi-source consensus time-space stamp to obtain signature data; S42. Construct a blockchain transaction from the data digest of digital identity, multi-source consensus time stamp, signature data and multi-dimensional raw data, and submit it to a consortium blockchain consisting of nodes composed of multiple regulatory agencies. S43. The signature data is verified through a smart contract on the consortium blockchain. When the verification is successful, the digital identity identifier is associated with the multi-source consensus time stamp and written into the state machine, triggering an on-chain state update event and obtaining a trusted on-chain state record.

6. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S5 further includes: S51. The consortium blockchain pulls the on-chain trusted state record, obtains the current market unit price of the base liquor through the oracle interface, and calculates the asset value risk sub-score by combining the current quantity and the pledged loan amount. S52. Extract the dynamic confidence weight vector from the multi-source consensus spatiotemporal stamp, calculate the information entropy, and count the frequency of data anomalies in the current monitoring period to obtain the data credibility risk sub-score. S53. Extract the number of unauthorized unlocks and the frequency of abnormal vibrations from the trusted state record on the chain, and calculate the operation compliance risk sub-score; S54. Calculate the environmental stability risk sub-score by calculating the cumulative duration and standard deviation of environmental parameters exceeding the limits of standard operating procedures (SOPs). S55. The asset value risk sub-score, the data credibility risk sub-score, the operational compliance risk sub-score, and the environmental stability risk sub-score are weighted and summed to obtain the comprehensive risk score.

7. The method for supervising base liquor pledge based on multi-source trusted data fusion according to claim 1, characterized in that, Step S6 further includes: S61. The comprehensive risk score is compared with the preset multi-level risk thresholds in sequence to determine the risk level to which the current score belongs. The risk levels include five risk levels: normal, attention, secondary, suspicious and loss. S62. For suspicious and loss-related risk levels, invoke the asset locking clause in the smart contract on the consortium blockchain to lock the digital twin of the base wine container associated with the corresponding digital identity to block the subsequent outbound approval process.

8. A base liquor pledge supervision system based on multi-source trusted data fusion, characterized in that, include: Binding module: Used to associate and bind the basic attribute information of the pledged base liquor container with multiple heterogeneous sensing devices to generate a corresponding digital identity identifier for the pledged base liquor container; Receiving module: used to receive multi-dimensional raw data from multiple heterogeneous sensing devices, indexed by the digital identity, to obtain a multi-source heterogeneous dataset; The calculation module is used to perform consensus cross-validation and dynamic weight calculation on the multi-source heterogeneous dataset using a spatiotemporal consensus weight algorithm to obtain the multi-source consensus spatiotemporal stamp. The signature module is used to digitally sign the multi-source consensus spatiotemporal stamp and the corresponding associated data digest and submit them to the consortium blockchain network to obtain an on-chain trusted state record. Quantification module: It is used to quantify risks in multiple dimensions, including asset value, data credibility, operational compliance and environmental stability, through a multi-dimensional dynamic risk correlation model, using the on-chain trusted state records and external market data as input, and obtain a comprehensive risk score. Comparison module: used to trigger a level warning and handling process based on the comparison result of the comprehensive risk score and the preset risk threshold, so as to complete the supervision of base liquor pledge.

9. A base liquor pledge monitoring device based on multi-source trusted data fusion, characterized in that, include: A memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause a base liquor pledge supervision device based on multi-source trusted data fusion to perform a base liquor pledge supervision method based on multi-source trusted data fusion as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed by a processor, implement a base liquor pledge supervision method based on multi-source trusted data fusion as described in any one of claims 1 to 7.

Images