Supply chain financial article risk control method and system based on Internet of Things

Abstract

The invention discloses a supply chain financial article risk control method and system based on the Internet of Things, particularly relates to the technical field of financial risk control, and is used for solving the problems of existing dynamic property pledge monitoring data distortion and risk control lag. A dynamic correlation model of interference levels and equipment anti-interference capability is constructed by collecting cargo environment interference parameters, equipment type parameters and state data in real time, and the credibility weight of sensor data is self-adaptively corrected according to environment interference intensity; generating a dynamic pledge rate in combination with a market price fluctuation trend and a credibility segmentation weighting mechanism, synchronously introducing cross validation logic of pledge rate deviation degree and state change rate, and establishing an abnormal grade gradient judgment rule; a grading early warning mechanism is triggered through multi-dimensional risk parameter fusion analysis, so that accurate identification and grading management and control of abnormal movement, value deviation and market risks of pledges are realized; and the reliability of the multi-source heterogeneous data in a complex environment is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of financial risk control, and more specifically, to a method and system for risk control of supply chain finance items based on the Internet of Things. Background Art

[0002] In the field of supply chain finance, the application of Internet of Things technology can enhance the risk control ability of movable property pledge; the existing technology constructs a dynamic pledge valuation model by integrating warehouse environment monitoring equipment, goods status sensors and market price data to achieve real-time monitoring and risk warning of pledged goods; that is, it usually relies on the collaborative analysis of multi-source data (such as goods weight, location, environmental indicators and market fluctuation information), combines preset thresholds or linear rules to generate pledge rates and risk indicators, and then supports financing decisions and asset security management.

[0003] However, in the process of multi-source data fusion, due to environmental interference, device heterogeneity and differences in data collection scenarios, the credibility of different data sources changes dynamically and is difficult to quantify, resulting in the deviation of the pledge valuation result from the true asset value, further causing implicit deviations in risk control indicators (such as pledge rate, replenishment line), and increasing the asset security risks of the financing party and the capital party. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method and system for risk control of supply chain finance items based on the Internet of Things to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for risk control of supply chain finance items based on the Internet of Things includes the following steps:

[0007] S1. Collect environmental interference parameters, real-time status parameters and device type parameters of the target goods;

[0008] S2. Obtain the market price parameters and historical pledge rate parameters of the target goods;

[0009] S3. Determine the interference level according to the comparison result between the environmental interference parameters and the preset interference threshold interval, and generate an initial credibility parameter based on the device type parameters;

[0010] S4. Dynamically correct the initial credibility parameter according to the correlation relationship between the interference level and the device type parameters to generate a target credibility parameter;

[0011] S5. Segmentally weight the real-time status parameters based on the target credibility parameter, and generate a dynamic pledge rate through the segmentally weighted real-time status parameters and the market price parameters;

[0012] S6. Determine the abnormal level of the pledge rate according to the deviation degree between the dynamic pledge rate and the historical pledge rate parameter and the trend direction of the market price parameter;

[0013] S7. Generate a risk warning instruction based on the abnormal level of the pledge rate and the change rate of the real-time status parameter, and trigger a goods control operation instruction.

[0014] In a preferred embodiment, collect the environmental interference parameter, real-time status parameter and equipment type parameter of the target goods, including:

[0015] Collect the environmental interference parameter of the target goods through an environmental monitoring device, where the environmental monitoring device includes an electromagnetic sensor and a temperature and humidity sensor;

[0016] Collect the real-time status parameter of the target goods through a status sensor, where the status sensor includes a weight sensor and a positioning sensor;

[0017] When collecting the real-time status parameter, synchronously obtain the equipment type parameter of the status sensor, and the equipment type parameter includes classifications of electromagnetic sensitive type and non-electromagnetic sensitive type.

[0018] In a preferred embodiment, obtain the market price parameter and historical pledge rate parameter of the target goods, including:

[0019] Obtain the real-time market price parameter of the target goods through a financial data interface, where the financial data interface accesses the real-time quotation data of at least two independent bulk commodity trading platforms;

[0020] Obtain the historical pledge rate parameter of the target goods through a bank or supply chain finance platform interface, where the historical pledge rate parameter includes the pledge rate data of the target goods in different pledge cycles within the past twelve months;

[0021] Conduct a validity check on the real-time market price parameter. The validity check includes comparing whether the difference between the quotation data of at least two trading platforms is less than a preset difference threshold. If the difference is less than the threshold, take the average value as the final market price parameter.

[0022] In a preferred embodiment, determine the interference level according to the comparison result between the environmental interference parameter and the preset interference threshold interval, and generate an initial credibility parameter based on the equipment type parameter, including:

[0023] Compare the environmental interference parameter with the preset interference threshold interval, and the preset interference threshold interval includes a high interference interval, a medium interference interval and a low interference interval;

[0024] When the environmental interference parameter is in the high interference interval, generate a first interference level; when it is in the medium interference interval, generate a second interference level; when it is in the low interference interval, generate a third interference level;

[0025] Classify the credibility of real-time status parameters based on device type parameters. Electromagnetic sensitive devices generate a first credibility weight at the first interference level and a second credibility weight at the second interference level; non-electromagnetic sensitive devices generate a third credibility weight from the first interference level to the third interference level.

[0026] Multiply the real-time status parameters by the corresponding credibility weights according to the correspondence between the interference level and the device type parameters to generate initial credibility parameters.

[0027] In a preferred embodiment, dynamically correct the initial credibility parameters according to the association relationship between the interference level and the device type parameters to generate target credibility parameters, including:

[0028] Determine the dynamic correction coefficient based on the association relationship between the interference level and the device type parameters. The dynamic correction coefficient includes a downweighting coefficient for electromagnetic sensitive devices at high interference levels and a buffer coefficient for electromagnetic sensitive devices at medium interference levels.

[0029] Gradually attenuate the initial credibility parameters through the downweighting coefficient to generate preliminary corrected credibility parameters.

[0030] Smoothly adjust the preliminary corrected credibility parameters based on the buffer coefficient to generate target credibility parameters.

[0031] When there are unclassified sensors in the device type parameters, perform weighted processing on the initial credibility parameters according to the preset default correction rules to generate target credibility parameters.

[0032] In a preferred embodiment, perform segmented weighting on the real-time status parameters based on the target credibility parameters, and generate a dynamic pledge rate through the segmented weighted real-time status parameters and the market price parameters, including:

[0033] Divide the credibility interval according to the numerical range of the target credibility parameters. The credibility interval includes a high credibility interval, a medium credibility interval, and a low credibility interval.

[0034] Assign preset weight values to each credibility interval, where the high credibility interval corresponds to the first weight, the medium credibility interval corresponds to the second weight, and the low credibility interval corresponds to the third weight.

[0035] Multiply the real-time status parameters by the corresponding preset weight values according to the credibility interval to which their target credibility parameters belong to generate segmented weighted real-time status parameters.

[0036] Input the segmented weighted real-time status parameters and the market price parameters into the pledge rate calculation model, and generate a dynamic pledge rate through weighted average calculation.

[0037] When the target credibility parameter fails to match any credibility interval, the real-time status parameters are weighted based on a preset default weight value to generate a dynamic pledge rate.

[0038] In a preferred embodiment, the pledge rate anomaly level is determined according to the deviation degree between the dynamic pledge rate and the historical pledge rate parameter and the trend direction of the market price parameter, including:

[0039] Calculate the percentage deviation degree between the dynamic pledge rate and the historical pledge rate parameter, which is obtained by subtracting the historical pledge rate mean from the dynamic pledge rate and then dividing by the historical pledge rate mean;

[0040] Determine the trend direction of the market price parameter, where the trend direction includes an upward trend, a downward trend, and a flat trend;

[0041] Divide the anomaly level according to the comparison result between the percentage deviation degree and the preset deviation thresholds, and the preset deviation thresholds include a first-level deviation threshold, a second-level deviation threshold, and a third-level deviation threshold;

[0042] Determine the pledge rate anomaly level;

[0043] When the historical pledge rate parameter does not exist, the pledge rate anomaly level is determined according to the trend direction of the market price parameter and the industry average pledge rate.

[0044] In a preferred embodiment, the determination of the pledge rate anomaly level includes:

[0045] When the percentage deviation degree exceeds the first-level deviation threshold and the trend direction is a downward trend, generate a first-level pledge rate anomaly level; when the percentage deviation degree exceeds the second-level deviation threshold and the trend direction is a downward trend or a flat trend, generate a second-level pledge rate anomaly level; when the percentage deviation degree exceeds the third-level deviation threshold, generate a third-level pledge rate anomaly level.

[0046] In a preferred embodiment, a risk warning instruction is generated based on the pledge rate anomaly level and the change rate of the real-time status parameters, and a goods control operation instruction is triggered, including:

[0047] Calculate the change rate according to the change amount of the real-time status parameters within a preset time window;

[0048] Input the pledge rate anomaly level and the change rate into the risk warning rule library, and the risk warning rule library defines the warning instructions corresponding to different combinations of anomaly levels and change rate thresholds;

[0049] When the pledge rate anomaly level is the first level and the change rate exceeds the first rate threshold, generate a yellow risk warning instruction and trigger a goods location verification operation instruction;

[0050] When the abnormal level of the pledge rate is at the second level and the change rate exceeds the second rate threshold, an orange risk warning instruction is generated and a goods warehousing and outbound restriction operation instruction is triggered;

[0051] When the abnormal level of the pledge rate is at the third level or the change rate exceeds the third rate threshold, a red risk warning instruction is generated and a goods seizure operation instruction is triggered;

[0052] After the goods control operation instruction is generated, the input parameters of the pledge rate calculation model are synchronously updated.

[0053] On the other hand, the present invention provides an Internet of Things-based supply chain finance item risk control system, including:

[0054] Parameter acquisition module: acquiring environmental interference parameters, real-time status parameters and equipment type parameters of the target goods; obtaining the market price parameters and historical pledge rate parameters of the target goods;

[0055] Trust generation module: determining the interference level according to the comparison result between the environmental interference parameters and the preset interference threshold range, and generating initial credibility parameters based on the equipment type parameters;

[0056] Dynamic correction module: dynamically correcting the initial credibility parameters according to the correlation between the interference level and the equipment type parameters to generate target credibility parameters;

[0057] Pledge generation module: segmentally weighting the real-time status parameters based on the target credibility parameters, and generating a dynamic pledge rate through the segmentally weighted real-time status parameters and the market price parameters;

[0058] Abnormal grading module: determining the abnormal level of the pledge rate according to the deviation degree between the dynamic pledge rate and the historical pledge rate parameters and the trend direction of the market price parameters;

[0059] Risk warning module: generating a risk warning instruction based on the abnormal level of the pledge rate and the change rate of the real-time status parameters, and triggering a goods control operation instruction.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] 1. By introducing the correlation analysis of equipment type parameters and environmental interference parameters, a credibility dynamic correction model based on the interference level is constructed, which can adaptively adjust the data weight according to the anti-interference characteristics of different equipment in a specific environment, effectively solving the problem of data credibility distortion caused by equipment performance differences in traditional methods; at the same time, a dynamic pledge rate is generated through the combined determination of segmented weighting and trend direction, enabling the valuation model to reflect the correlation between the goods status and market fluctuations in real time, and avoiding the defect that the risk control index lags behind the market change under the static threshold rule;

[0062] 2. By integrating the multi-dimensional decision-making logic of the real-time state parameter change rate and the anomaly level, a gradient recognition and hierarchical response mechanism for risk events is achieved; the collaborative analysis of the pledge rate deviation and the market price trend can accurately identify the differences between market panic declines and normal fluctuation scenarios, avoiding false triggering of risk control instructions; while the cross-validation mechanism of the state parameter change rate and the anomaly level can quickly locate risks such as abnormal movement, deterioration, or human operation of goods, ensuring the accurate execution of control instructions; this multi-dimensional dynamic evaluation system significantly reduces the false alarm rate caused by data noise in traditional methods, and at the same time, through the real-time feedback and update mechanism of model parameters, the risk control strategy has the self-learning ability of continuous optimization, providing more reliable asset security protection for financial institutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is a flowchart of the method for controlling the risk of items in supply chain finance based on the Internet of Things according to the present invention;

[0064] Figure 2 is a schematic structural diagram of the system for controlling the risk of items in supply chain finance based on the Internet of Things according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0066] Embodiment 1: Figure 1 The method for controlling the risk of items in supply chain finance based on the Internet of Things according to the present invention is given, which includes the following steps:

[0067] S1. Collect the environmental interference parameters, real-time state parameters, and device type parameters of the target goods;

[0068] S2. Obtain the market price parameters and historical pledge rate parameters of the target goods;

[0069] S3. Determine the interference level according to the comparison result between the environmental interference parameters and the preset interference threshold interval, and generate an initial credibility parameter based on the device type parameters;

[0070] S4. Dynamically correct the initial credibility parameter according to the correlation relationship between the interference level and the device type parameters to generate a target credibility parameter;

[0071] S5. Segmentally weight the real-time state parameters based on the target credibility parameter, and generate a dynamic pledge rate through the segmentally weighted real-time state parameters and the market price parameters;

[0072] S6. Determine the abnormal level of the pledge rate based on the deviation degree between the dynamic pledge rate and the historical pledge rate parameter and the trend direction of the market price parameter;

[0073] S7. Generate a risk warning instruction based on the change rate of the abnormal level of the pledge rate and the real-time status parameter, and trigger a goods control operation instruction.

[0074] S1. Collect the environmental interference parameter, real-time status parameter and equipment type parameter of the target goods, including:

[0075] Collect the environmental interference parameter of the target goods through an environmental monitoring device, where the environmental monitoring device includes an electromagnetic sensor and a temperature and humidity sensor;

[0076] Collect the real-time status parameter of the target goods through a status sensor, where the status sensor includes a weight sensor and a positioning sensor;

[0077] When collecting the real-time status parameter, synchronously obtain the equipment type parameter of the status sensor, and the equipment type parameter includes classifications of electromagnetic sensitivity type and non-electromagnetic sensitivity type.

[0078] The environmental monitoring device includes an electromagnetic sensor and a temperature and humidity sensor. The environmental interference parameter includes an electromagnetic field strength parameter, a temperature parameter and a humidity parameter. The electromagnetic sensor is used to collect the electromagnetic field strength parameter of the environment where the target goods are located. The electromagnetic field strength parameter is obtained by detecting the electromagnetic radiation intensity in a specific frequency range in the storage area, and the detection frequency range includes the common interference frequency bands of industrial equipment from 50 Hz to 2.4 GHz. The temperature and humidity sensor is installed on a fixed bracket in the storage area of the target goods and is used to collect the temperature parameter and humidity parameter of the environment where the target goods are located.

[0079] The status sensor includes a weight sensor and a positioning sensor. The real-time status parameter includes a weight parameter and a real-time position parameter. The weight sensor collects the weight parameter of the target goods through a pressure sensing unit installed at the bottom of the bearing device of the target goods. The pressure sensing unit generates a corresponding electrical signal according to the change of the goods weight and converts it into a weight value. The positioning sensor collects the real-time position parameter of the target goods through a passive radio frequency identification tag attached to the target goods or a device integrated with a global positioning system chip. The operating frequency of the passive radio frequency identification tag is ultra-high frequency from 860 MHz to 960 MHz.

[0080] When collecting real-time status parameters of the target goods through a status sensor, the device type parameter of the status sensor is synchronously obtained. The device type parameter includes classifications of electromagnetic-sensitive type and non-electromagnetic-sensitive type. An electromagnetic-sensitive type status sensor refers to a sensor with a measurement error exceeding 5% when the electromagnetic field intensity exceeds the preset interference threshold of 1 V / m. A non-electromagnetic-sensitive type status sensor refers to a sensor with a measurement error still less than or equal to 5% when the electromagnetic field intensity exceeds the preset interference threshold of 1 V / m. The classification basis of the device type parameter comes from the anti-interference ability test results of the sensor manufacturer according to the international electromagnetic compatibility standard IEC 61000-4-3.

[0081] The device type parameter is obtained by reading the model identification code built into the status sensor and matching the preset sensor parameter database. The sensor parameter database stores the anti-electromagnetic interference level classification information of different model sensors.

[0082] S2. Obtain the market price parameter and historical pledge rate parameter of the target goods, including:

[0083] Obtain the real-time market price parameter of the target goods through the financial data interface, where the financial data interface accesses the real-time quotation data of at least two independent bulk commodity trading platforms;

[0084] Obtain the historical pledge rate parameter of the target goods through the bank or supply chain finance platform interface, where the historical pledge rate parameter includes the pledge rate data of the target goods under different pledge periods in the past twelve months;

[0085] Conduct validity verification on the real-time market price parameter. The validity verification includes comparing whether the quotation data difference between at least two trading platforms is less than the preset difference threshold. If the difference is less than the threshold, the average value is taken as the final market price parameter.

[0086] The independent bulk commodity trading platforms accessed by the financial data interface include the Shanghai Futures Exchange, the London Metal Exchange, and the Chicago Mercantile Exchange. The real-time quotation data is obtained from the public market quotation interfaces of the above platforms in JSON format through the application programming interface. For example, when the target goods are electrolytic copper, the contract code of the Shanghai Futures Exchange is CU2312, and the contract code of the London Metal Exchange is MCUZ3. The real-time quotation data includes the latest bid price, ask price, and daily trading volume.

[0087] The bank or supply chain finance platform interface accesses the pledge business database of the supply chain finance system of the Industrial and Commercial Bank of China or Ping An Bank. The historical pledge rate parameter is obtained through the database query interface. The query conditions include the variety identification code of the target goods (such as the variety code CU-001 of electrolytic copper), the enterprise code of the pledgor (such as the unified social credit code of the enterprise), and the pledge period classification (such as three months, six months, twelve months).

[0088] For example, in the pledge cycle of a certain electrolytic copper supplier in the past twelve months, the average historical pledge rate for a three - month cycle was 70%, for a six - month cycle was 65%, and for a twelve - month cycle was 60%. The preset difference threshold is 5%. When the difference in real - time quotation data between two trading platforms exceeds 5%, the manual review process is triggered. For example, the quotation of electrolytic copper on the Shanghai Futures Exchange is 68,000 yuan per ton, and the quotation on the London Metal Exchange is 72,000 yuan per ton, with a difference of 5.8%. Then it is necessary to manually review whether the data is affected by exchange rate fluctuations or differences in delivery standards. If the data is confirmed to be valid, the middle value of 70,000 yuan is taken as the final market price parameter. If a certain platform returns an abnormal quotation due to a system failure (such as 100,000 yuan per ton), then this data is excluded and a valid quotation is obtained again.

[0089] After the validity check is completed, the final market price parameter and the historical pledge rate parameter are stored in the local database associated by the variety identification code. The associated fields include the data collection timestamp (such as 2023 - 10 - 01 10:00:00) and the check result identification code (such as the check - passed mark "VALID"). The historical pledge rate parameter is stored as a structured data table classified by the pledge cycle. For example, the data table for the three - month cycle contains fields such as "pledge start date", "pledge end date", "pledge rate value", and "pledgor enterprise code", where the pledge rate value is derived from the actual loan - disbursement ratio of the corresponding pledge contract in the bank system.

[0090] The basis for setting the preset difference threshold includes the standard deviation of the market price fluctuations of the target goods in the past year (such as the price standard deviation of electrolytic copper is 8%) and the industry - recognized price tolerance range (such as the maximum allowable price deviation stipulated by the China Nonferrous Metals Industry Association is 5%). For example, since the price standard deviation of electrolytic copper in the past year is 8%, the difference threshold is set to 5% to cover most normal fluctuation scenarios.

[0091] S3. Determine the interference level based on the comparison result between the environmental interference parameter and the preset interference threshold range, and generate an initial credibility parameter based on the equipment type parameter, including:

[0092] Compare the environmental interference parameter with the preset interference threshold range, which includes a high - interference range, a medium - interference range, and a low - interference range;

[0093] When the environmental interference parameter is in the high - interference range, generate the first interference level; when it is in the medium - interference range, generate the second interference level; when it is in the low - interference range, generate the third interference level;

[0094] Classify the credibility of real-time status parameters based on device type parameters. Electromagnetic-sensitive devices generate a first credibility weight at the first interference level and a second credibility weight at the second interference level; non-electromagnetic-sensitive devices generate a third credibility weight at the first to third interference levels.

[0095] According to the correspondence between the interference level and the device type parameters, multiply the real-time status parameters by the corresponding credibility weights to generate initial credibility parameters.

[0096] The preset interference threshold range is set according to the historical electromagnetic interference data and industry standards of the storage area where the target goods are located. For example, in an industrial storage scenario, the high-interference range may correspond to an electromagnetic field intensity exceeding a certain threshold (e.g., 5 V / m), and this threshold refers to the electromagnetic compatibility requirements for industrial environments in the international standard IEC 61000-4-3; the medium-interference range may be set as a medium-intensity range (e.g., 2 V / m to 5 V / m), and this range is determined based on the critical value of the sudden increase in equipment error in historical data; the low-interference range is a lower-intensity range (e.g., below 2 V / m), which is applicable to scenarios where the interference impact is negligible. The environmental interference parameters include the real-time electromagnetic field intensity value collected by the electromagnetic sensor and the temperature-humidity comprehensive interference index collected by the temperature-humidity sensor. The temperature-humidity comprehensive interference index is obtained through weighted calculation of temperature and humidity. For example, in food storage, humidity has a greater impact on the risk of goods spoilage, and the humidity weight may be set to 0.7 and the temperature weight to 0.3. In the storage of electronic components, the temperature weight may be increased to 0.8 to reflect its thermal sensitivity.

[0097] When the environmental interference parameters reach the high-interference range, generate the first interference level; when in the medium-interference range, generate the second interference level; when in the low-interference range, generate the third interference level. For example, in a metal storage scenario, if the electromagnetic sensor detects high-intensity electromagnetic interference (e.g., 6 V / m due to the operation of a nearby welding device), and at the same time the temperature-humidity comprehensive index rises to 75 (high temperature and high humidity) due to poor ventilation, it is determined as the first interference level; if the electromagnetic field intensity is 3 V / m and the temperature-humidity index is 50 (conventional operating environment), it is determined as the second interference level. The device type parameters include electromagnetic-sensitive and non-electromagnetic-sensitive types. For example, electromagnetic-sensitive devices may be low-cost general sensors (such as commercial electronic scales), while non-electromagnetic-sensitive devices are sensors that have passed industrial anti-interference certification (such as devices compliant with the IEC 61000-6-2 standard).

[0098] Electromagnetic sensitive devices generate differentiated credibility weights under different interference levels. For example, at the first interference level, the credibility weight of an electromagnetic sensitive device may be reduced to 0.5 to reflect its increased error rate in a high-interference environment; at the second interference level, the weight is adjusted to 0.7 to balance data availability under medium interference; non-electromagnetic sensitive devices maintain a high weight (e.g., 0.9) at all interference levels due to their stable anti-interference performance. The basis for setting the credibility weight includes the device's anti-interference performance test report and historical operation data. For example, if the error rate of an electromagnetic sensitive sensor exceeds the allowable range (e.g., error rate > 10%) in a laboratory-simulated high-interference environment, its weight at the high interference level is set to 0.5; while historical data shows that the error rate of this sensor is only 7% in a medium-interference scenario, so the weight is set to 0.7.

[0099] S4. Dynamically correct the initial credibility parameter according to the correlation between the interference level and the device type parameter to generate the target credibility parameter, including:

[0100] Determine the dynamic correction coefficient based on the correlation between the interference level and the device type parameter. The dynamic correction coefficient includes the weight reduction coefficient of electromagnetic sensitive devices at the high interference level and the buffer coefficient of electromagnetic sensitive devices at the medium interference level;

[0101] Gradually attenuate the initial credibility parameter through the weight reduction coefficient to generate the preliminary corrected credibility parameter;

[0102] Smoothly adjust the preliminary corrected credibility parameter based on the buffer coefficient to generate the target credibility parameter;

[0103] When there are unclassified sensors in the device type parameter, perform weighted processing on the initial credibility parameter according to the preset default correction rule to generate the target credibility parameter.

[0104] The dynamic correction factor is determined based on the correlation between the interference level and the device type parameters. For example, the downweighting factor for electromagnetic sensitive devices under a high interference level is set to 0.5. This factor is derived from the average error rate of sensors and safety redundancy adjustment in a laboratory-simulated high interference environment. For example, when the average error rate of a sensor in a high interference environment is 20%, the downweighting factor 0.5 is obtained by calculating (1 - error rate) and adding a 10% safety redundancy. The buffering factor for electromagnetic sensitive devices under a medium interference level is set to 0.8. This factor is based on the median error rate and the need for smoothing processing in historical data statistics. For example, when the median error rate of a sensor under medium interference is 12%, the buffering factor 0.8 is obtained by calculating (1 - error rate) and reserving a 5% smoothing factor. The correction factor for non-electromagnetic sensitive devices is fixed at 0.9 under all interference levels. This factor is based on the stability verification results in the anti-interference performance test report. For example, it is set to 0.9 when the error rate continuously remains below 5%.

[0105] The downweighting factor is used to gradually attenuate the initial credibility parameter. For example, when an electromagnetic sensitive weight sensor is under a high interference level, the initial credibility parameter (such as a weight value of 1000 tons) is multiplied by the downweighting factor 0.5 to generate a preliminary corrected credibility parameter of 500 tons.

[0106] The buffering factor is used to smoothly adjust the preliminary corrected credibility parameter. For example, when an electromagnetic sensitive temperature sensor is under a medium interference level, the preliminary corrected credibility parameter (such as a temperature value of 25°C) is multiplied by the buffering factor 0.8 to generate a target credibility parameter of 20°C.

[0107] If there are unclassified sensors in the device type parameters (such as newly connected devices without pre-set anti-interference levels), they are processed according to the pre-set default correction rules. For example, the default downweighting factor is set to 0.6 (referring to the industry's general minimum safety threshold). The initial credibility parameter (such as a humidity value of 60%) is multiplied by the default factor 0.6 to generate a target credibility parameter of 36%.

[0108] For example, in a metal sheet storage scenario, the target goods storage area is adjacent to high-frequency welding equipment, and the electromagnetic interference level is high. The initial credibility parameter of an electromagnetic sensitive weight sensor is a weight value of 800 tons. After being gradually attenuated by the downweighting factor 0.5, a preliminary corrected credibility parameter of 400 tons is generated. Since no additional smoothing processing is required in a high interference environment, the target credibility parameter is directly set to 400 tons. The initial credibility parameter of a non-electromagnetic sensitive positioning sensor is the location coordinate Area A, and the correction factor 0.9 generates the target credibility coordinate Area A.

[0109] In the monitoring scenario of chemical raw material storage tanks, the electromagnetic interference level is medium. The initial credibility parameter of the electromagnetic sensitive humidity sensor is a humidity value of 70%. After smooth adjustment through a buffer coefficient of 0.8, the target credibility parameter of 56% is generated. The initial credibility parameter of the non-electromagnetic sensitive pressure sensor is a pressure value of 1.0 MPa. A correction coefficient of 0.9 generates a target credibility parameter of 0.9 MPa. If the newly added temperature sensor does not have the device type parameter preset, the system processes the initial credibility parameter of 30°C according to the default correction rule (downgrading coefficient of 0.6) to generate a target credibility parameter of 18°C. Subsequently, if the sensor is classified as non-electromagnetic sensitive, the correction coefficient is switched to 0.9 and the target credibility parameter is adjusted to 27°C.

[0110] The setting basis of the downgrading coefficient of 0.5 includes the average error rate (such as 20%) of electromagnetic sensitive devices under high interference in laboratory tests and safety redundancy adjustment (such as 10%). The final weight calculation is (1 - 0.2) × (1 - 0.1) = 0.72, which is rounded and simplified to 0.5.

[0111] The setting basis of the buffer coefficient of 0.8 is the median error rate (such as 12%) of electromagnetic sensitive devices under medium interference in historical data and the smoothing factor (such as 5%). The calculation is (1 - 0.12) × (1 - 0.05) = 0.836, which is approximated to 0.8. The coefficient of 0.6 in the default correction rule can refer to the minimum credibility requirement for unknown devices in the Internet of Things financial data credibility specification.

[0112] When the correction results of the weight sensor (electromagnetic sensitive type) and the positioning sensor (non-electromagnetic sensitive type) for the same target goods conflict, for example, the weight credible value of 500 tons conflicts with the positioning coordinate of Warehouse Area B, the system triggers an artificial review process, preferentially uses the positioning coordinate of the non-electromagnetic sensitive device as the benchmark, and recalculates the weight credible value. If a sensor has long performed better than its classification (for example, the error rate of an electromagnetic sensitive device continuously remains below 5%), the system statistically calculates the error rate monthly and gradually increases its correction coefficient (for example, from 0.5 to 0.7).

[0113] After the target credibility parameter is generated, it is bound and stored together with the initial credibility parameter, interference level code (such as G1), device type code (such as EM_SENSITIVE), and correction coefficient. For example, the weight target credible value of 400 tons is associated with the initial value of 800 tons, interference level code G1, device type code EM_SENSITIVE, and downgrading coefficient of 0.5, and is stored in the local database for subsequent calculation of the pledge rate.

[0114] S5. Segmentally weight the real-time status parameters based on the target credibility parameter, and generate a dynamic pledge rate through the segmentally weighted real-time status parameters and market price parameters, including:

[0115] Divide the credibility interval according to the numerical range of the target credibility parameter. The credibility interval includes a high credibility interval, a medium credibility interval, and a low credibility interval;

[0116] Assign a preset weight value to each credibility interval, where the high credibility interval corresponds to a first weight, the medium credibility interval corresponds to a second weight, and the low credibility interval corresponds to a third weight;

[0117] Multiply the real-time status parameter by the corresponding preset weight value according to the credibility interval to which its target credibility parameter belongs, and generate a segmented weighted real-time status parameter;

[0118] Input the segmented weighted real-time status parameter and the market price parameter into the pledge rate calculation model, and generate a dynamic pledge rate through weighted average calculation;

[0119] When the target credibility parameter cannot match any credibility interval, weight the real-time status parameter based on a preset default weight value to generate a dynamic pledge rate.

[0120] The credibility interval is divided according to the numerical range of the target credibility parameter. For example, the high credibility interval is set to the target credibility parameter being greater than or equal to 0.8, and this threshold is based on the verification result that the error rate is less than 5% during the long-term operation test of the device in a non-interference environment. The medium credibility interval is set to be greater than or equal to 0.5 and less than 0.8, based on the weight allocation suggestions for medium-credibility data in the industry risk control operation specifications. The low credibility interval is set to be less than 0.5, which is applicable to scenarios where the device is affected by environmental interference resulting in a significant increase in the error rate. A preset weight value is assigned to each credibility interval. For example, the high credibility interval corresponds to a weight of 0.9, the medium credibility interval corresponds to a weight of 0.7, and the low credibility interval corresponds to a weight of 0.5.

[0121] The real-time status parameter is multiplied by the corresponding weight according to the interval to which its target credibility parameter belongs to generate a segmented weighted real-time status parameter. For example, when the target credibility parameter is 0.85, the weight value of the real-time status parameter is 1000 tons multiplied by the weight of 0.9, resulting in a segmented weighted weight value of 900 tons. If the target credibility parameter is 0.6, the weight value of the real-time status parameter is 1000 tons multiplied by the weight of 0.7, resulting in a segmented weighted weight value of 700 tons.

[0122] Input the segmented weighted real-time status parameters and market price parameters into the pledge rate calculation model, and generate a dynamic pledge rate through weighted average calculation. For example, after multiplying the segmented weighted value of 900 tons of the goods weight by the market price parameter of 5000 yuan per ton, a dynamic pledge rate is generated in combination with the historical average pledge rate. When the target credibility parameter cannot match any credibility interval, the real-time status parameters are weighted based on the preset default weight value of 0.4. For example, when the target credibility parameter is 0.3, the weight value of 1000 tons of the real-time status parameter weight is multiplied by the default weight of 0.4 to generate a segmented weighted weight value of 400 tons.

[0123] In the metal warehousing scenario, due to the low electromagnetic interference level and strong anti-interference performance of the equipment, the target credibility parameter belongs to the high credibility interval. The weight value of 800 tons of the real-time status parameter is multiplied by the weight of 0.9 to generate a segmented weighted weight value of 720 tons. Combining the current market price parameter of 6000 yuan per ton, the pledge rate calculation model calculates and generates a dynamic pledge rate by associating the weighted weight with the price. If a sudden strong electromagnetic interference occurs in the warehousing area, resulting in the target credibility parameter dropping to 0.6 (medium credibility interval), the weight value of 800 tons is multiplied by the weight of 0.7 to generate a segmented weighted weight value of 560 tons, and the system synchronously triggers an artificial review process to verify the data reliability.

[0124] In the cold chain logistics scenario, due to the influence of condensate water on the temperature and humidity sensor, the target credibility parameter is temporarily in the low credibility interval (0.4). The temperature value of -18°C of the real-time status parameter is multiplied by the weight of 0.5 to generate a segmented weighted temperature value of -9°C. The system starts the backup sensor to collect data to replace the abnormal parameters to ensure the accuracy of the pledge rate calculation.

[0125] The basis for setting the credibility interval threshold includes the statistical data of the long-term operation of the equipment and the industry risk tolerance index. For example, the high credibility threshold of 0.8 is set based on the test results that the error rate of the equipment continuously remains below 5% in the interference-free environment. In the weight allocation rule, the weight of 0.9 corresponds to the reliability verification result of high credibility data, the weight of 0.7 is based on the median of the error rate in historical data and the buffer requirement in the risk control strategy, and the weight of 0.5 refers to the conservative processing principle of the industry for high-risk data. The default weight value of 0.4 is determined based on the minimum risk control guarantee requirement in extreme scenarios.

[0126] When the weight credibility parameter of the same goods is 0.6 and the positioning credibility parameter is 0.9, the pledge rate calculation model preferentially adopts the calculation result of the weight corresponding to the high credibility parameter, and the low credibility data is used as an auxiliary reference. If the long-term operation error rate of a certain type of equipment is significantly better than its initial classification (for example, the error rate is stably below 3%), the system dynamically increases its weight value according to the monthly statistical results (for example, from 0.7 to 0.8).

[0127] The real-time status parameters after segmented weighting are associated and stored with the target credibility parameters, weight values, and acquisition timestamps. For example, the weight segmented weighting value of 720 tons is associated with the target credibility parameter of 0.85, a weight of 0.9, and the timestamp "October 2023" for subsequent data traceability and auditing. The calculation result of the dynamic pledge rate is bound and recorded with the unique code of the goods, market price parameters, and operator identifier.

[0128] S6. Determine the pledge rate anomaly level based on the deviation degree between the dynamic pledge rate and the historical pledge rate parameter and the trend direction of the market price parameter, including:

[0129] Calculate the percentage deviation degree between the dynamic pledge rate and the historical pledge rate parameter. The percentage deviation degree is obtained by subtracting the historical pledge rate mean from the dynamic pledge rate and then dividing by the historical pledge rate mean;

[0130] Determine the trend direction of the market price parameter. The trend direction includes an upward trend, a downward trend, and a flat trend;

[0131] Divide the anomaly level according to the comparison result between the percentage deviation degree and the preset deviation thresholds. The preset deviation thresholds include a first-level deviation threshold, a second-level deviation threshold, and a third-level deviation threshold;

[0132] When the percentage deviation degree exceeds the first-level deviation threshold and the trend direction is a downward trend, generate a first-level pledge rate anomaly level; when the percentage deviation degree exceeds the second-level deviation threshold and the trend direction is a downward trend or a flat trend, generate a second-level pledge rate anomaly level; when the percentage deviation degree exceeds the third-level deviation threshold, generate a third-level pledge rate anomaly level;

[0133] When the historical pledge rate parameter does not exist, determine the pledge rate anomaly level based on the trend direction of the market price parameter and the industry average pledge rate.

[0134] The percentage deviation degree is calculated by subtracting the historical pledge rate mean from the dynamic pledge rate and then dividing by the historical pledge rate mean. For example, if the dynamic pledge rate is 3.6 million yuan and the historical pledge rate mean is 3 million yuan, then the percentage deviation degree is (3.6 - 3) / 3 = 20%.

[0135] The trend direction of the market price parameter is determined by comparing the current price with the average price of the past three trading days. For example, if the current price is 5% higher than the historical mean, it is an upward trend; if it is 5% lower, it is a downward trend; and if the fluctuation range is within ±5%, it is a flat trend. The preset deviation thresholds are set according to the industry risk control guidelines. The first-level deviation threshold is 15%, the second-level deviation threshold is 25%, and the third-level deviation threshold is 40%.

[0136] When the percentage deviation exceeds the first-level deviation threshold and the trend direction is a downward trend, a first-level pledge rate anomaly level is generated. For example, if the deviation of the dynamic pledge rate is 18% and the market price trend is a downward trend, it is determined as a first-level anomaly. When the percentage deviation exceeds the second-level deviation threshold and the trend direction is a downward or flat trend, a second-level pledge rate anomaly level is generated. For example, if the deviation is 28% and the market price trend is a flat trend, it is determined as a second-level anomaly. When the percentage deviation exceeds the third-level deviation threshold, a third-level pledge rate anomaly level is generated regardless of the trend direction. For example, if the deviation is 45%, it is determined as a third-level anomaly regardless of whether the market price rises or falls. When the historical pledge rate parameter does not exist (such as for the first-time pledged goods), the anomaly level is determined based on the trend direction of the market price parameter and the industry average pledge rate. For example, if the market price trend is a downward trend and the current pledge rate is 20% higher than the industry average, a second-level anomaly level is generated.

[0137] In the metal warehousing scenario, due to the decreased credibility of the goods weight data, the deviation of the dynamic pledge rate reaches 20% (exceeding the first-level threshold of 15%). With the market price trend being a downward trend, the system generates a first-level pledge rate anomaly level and triggers the manual review process.

[0138] In the food warehousing scenario, the deviation of the dynamic pledge rate is 30% (exceeding the second-level threshold of 25%). With the market price trend being a flat trend, a second-level anomaly level is generated and the pledge rate adjustment authority is restricted.

[0139] In the chemical transportation scenario, due to a sudden sharp price drop, the deviation of the dynamic pledge rate reaches 50% (exceeding the third-level threshold of 40%), generating a third-level anomaly level and immediately initiating the forced liquidation mechanism.

[0140] For the first-time pledged electronic component goods, since there is no historical pledge rate parameter, the system generates a second-level anomaly level and freezes part of the financing quota based on the comparison result of the market price downward trend and the industry average pledge rate (the current pledge rate is 25% higher than the industry average).

[0141] The basis for setting the preset deviation threshold includes the statistical fluctuations of historical pledge rates and the industry risk control consensus. For example, the first-level threshold of 15% is set based on the maximum allowable deviation value of the pledge rate during normal market fluctuations in the past year. The second-level threshold of 25% corresponds to the industry's high-risk warning line, and the third-level threshold of 40% refers to the extreme value of the pledge rate decline during extreme market crash events.

[0142] In the trend direction judgment rule, the up and down threshold of 5% is set according to the short-term trend judgment standard in the "Commodity Price Fluctuation Management Specification".

[0143] When historical data is missing, the industry average pledge rate is calculated by querying the public pledge data of the same category of goods in major financial institutions.

[0144] After the abnormal level is generated, it is associated and stored with the dynamic pledge rate value, the deviation calculation result, and the timestamp. For example, the first-level abnormal level is associated with a deviation of 18%, a dynamic pledge rate of 3.6 million yuan, and a timestamp of "October 2023" for subsequent risk audit traceability.

[0145] S7. Generate a risk warning instruction based on the change rate of the abnormal level of the pledge rate and the real-time status parameters, and trigger the cargo control operation instruction, including:

[0146] Calculate the change rate according to the change amount of the real-time status parameters within a preset time window;

[0147] Input the abnormal level of the pledge rate and the change rate into the risk warning rule library, and the risk warning rule library defines the warning instructions corresponding to different combinations of abnormal levels and change rate thresholds;

[0148] When the abnormal level of the pledge rate is at the first level and the change rate exceeds the first rate threshold, generate a yellow risk warning instruction and trigger the cargo location verification operation instruction;

[0149] When the abnormal level of the pledge rate is at the second level and the change rate exceeds the second rate threshold, generate an orange risk warning instruction and trigger the cargo inbound and outbound restriction operation instruction;

[0150] When the abnormal level of the pledge rate is at the third level or the change rate exceeds the third rate threshold, generate a red risk warning instruction and trigger the cargo seizure operation instruction;

[0151] After the cargo control operation instruction is generated, synchronously update the input parameters of the pledge rate calculation model.

[0152] The change rate of the real-time status parameters is obtained by dividing the change amount within a preset time window by the time window length. For example, the time window is set to one hour, and the cargo weight decreases from 1000 tons to 950 tons, and the change rate is (1000 - 950) / 1 = 50 tons / hour. The risk warning rule library defines the combination rules of different abnormal levels and change rate thresholds. For example, the first-level abnormal level of the pledge rate corresponds to a change rate threshold of 5%, the second level corresponds to 10%, and the third level corresponds to 20%.

[0153] When the abnormal level of the pledge rate is at the first level and the change rate exceeds 5%, generate a yellow risk warning instruction and trigger the cargo location verification operation instruction. For example, the abnormal level of the dynamic pledge rate is at the first level, and the change rate of the cargo weight is 6%. The system generates a yellow warning and notifies the warehouse administrator to verify whether the cargo location is abnormal.

[0154] When the abnormal level of the pledge rate is level two and the change rate exceeds 10%, an orange risk warning instruction is generated and a goods inbound and outbound restriction operation instruction is triggered. For example, when the abnormal level of the dynamic pledge rate is level two and the change rate of the goods temperature is 12%, the system generates an orange warning and suspends the inbound and outbound permissions of the cold storage goods.

[0155] When the abnormal level of the pledge rate is level three or the change rate exceeds 20%, a red risk warning instruction is generated and a goods seizure operation instruction is triggered. For example, when the abnormal level of the dynamic pledge rate is level three and the change rate of the goods pressure is 25%, the system generates a red warning and requests the regulatory agency to seal the goods storage area.

[0156] After the goods control operation instruction is generated, the input parameters of the pledge rate calculation model are updated synchronously. For example, after the goods seizure instruction is triggered, the system marks the abnormal goods data as invalid, and the data is excluded from the subsequent calculation of the pledge rate calculation model.

[0157] In the metal warehousing scenario, the abnormal level of the dynamic pledge rate is level one, and the change rate of the goods weight is 6% (exceeding the threshold of 5%). The system generates a yellow warning and triggers a location verification instruction. After verifying that the goods location is normal, the warning is lifted and marked as a false alarm. In the cold chain logistics scenario, the abnormal level of the dynamic pledge rate is level two, and the change rate of the goods temperature is 12% (exceeding the threshold of 10%). The system generates an orange warning and restricts the inbound and outbound operations of the cold storage until the temperature stabilizes. In the chemical transportation scenario, the abnormal level of the dynamic pledge rate is level three, and the change rate of the goods pressure is 25% (exceeding the threshold of 20%). The system generates a red warning and forces the transport vehicle to dock at the regulatory area. For the first-time pledged electronic component goods, due to no historical data, the system determines that the abnormal level is level two according to the industry average pledge rate and the market price trend, and the change rate of the weight is 15% (exceeding the threshold of 10%). An orange warning is generated and the goods transfer permission is frozen.

[0158] The basis for setting the change rate threshold includes historical data statistics and industry risk control specifications. For example, the level one threshold of 5% is based on the upper limit of the normal fluctuation rate of goods in the past year, the level two threshold of 10% corresponds to the typical rate of abnormal transfer of high-risk goods, and the level three threshold of 20% refers to the rate extreme value in extreme events (such as goods theft). The mapping relationship of the risk warning rule library is set according to the "abnormal level and disposal measures" clause in the industry operation guidelines. After the goods control operation instruction is generated, the update rules for the input parameters of the pledge rate calculation model include: marking abnormal data but retaining the calculation for yellow warnings, reducing the weight of abnormal data for orange warnings, and removing abnormal data for red warnings. For example, after the goods are seized, the relevant real-time status parameters are removed from the pledge rate calculation model to avoid interfering with subsequent risk control.

[0159] Example 2: Figure 2The structure diagram of the supply chain finance item risk control system based on the Internet of Things according to the present invention is given. The supply chain finance item risk control system based on the Internet of Things includes:

[0160] Parameter acquisition module: acquiring the environmental interference parameters, real-time status parameters and device type parameters of the target goods; obtaining the market price parameters and historical pledge rate parameters of the target goods;

[0161] Trusted generation module: determining the interference level according to the comparison result between the environmental interference parameters and the preset interference threshold range, and generating initial credibility parameters based on the device type parameters;

[0162] Dynamic correction module: dynamically correcting the initial credibility parameters according to the correlation between the interference level and the device type parameters to generate target credibility parameters;

[0163] Pledge generation module: segmentally weighting the real-time status parameters based on the target credibility parameters, and generating a dynamic pledge rate through the segmentally weighted real-time status parameters and the market price parameters;

[0164] Abnormal grading module: determining the abnormal grade of the pledge rate according to the deviation degree between the dynamic pledge rate and the historical pledge rate parameters and the trend direction of the market price parameters;

[0165] Risk warning module: generating a risk warning instruction based on the abnormal grade of the pledge rate and the change rate of the real-time status parameters, and triggering a goods control operation instruction.

[0166] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by software simulation of a large amount of collected data to obtain a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0167] It should be noted that the present invention can be deployed on the device itself to achieve embedded applications, or can also run on a PC or other terminal with a user interface, so as to meet various hardware environments and usage requirements.

[0168] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0169] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0170] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0171] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules. They can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0172] In addition, in each embodiment of the present application, each functional module can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0173] If the above-mentioned function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0174] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0175] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A supply chain finance item risk control method based on the Internet of Things, characterized in that: The steps include: S1. Collect environmental interference parameters, real-time status parameters and equipment type parameters of target goods; S2. Obtain market price parameters and historical pledge rate parameters of target goods; S3. Determine the interference level according to the comparison result between the environmental interference parameter and the preset interference threshold interval, and generate an initial credibility parameter based on the device type parameter; S4. Dynamically modify the initial credibility parameters according to the correlation between the interference level and the device type parameters to generate target credibility parameters; S5. Based on the target credibility parameter, the real-time status parameter is weighted in sections, and a dynamic pledge rate is generated through the weighted real-time status parameter and the market price parameter; S6. Determine the abnormal level of the pledge rate based on the deviation between the dynamic pledge rate and the historical pledge rate parameters and the trend direction of the market price parameters; S7. Generate risk warning instructions based on the abnormal level of pledge rate and the change rate of real-time status parameters, and trigger cargo control operation instructions.

2. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: Collect environmental interference parameters, real-time status parameters and equipment type parameters of the target goods, including: Collecting environmental interference parameters of the target goods through environmental monitoring equipment, where the environmental monitoring equipment includes electromagnetic sensors and temperature and humidity sensors; The real-time status parameters of the target goods are collected through status sensors, wherein the status sensors include weight sensors and positioning sensors; When collecting real-time status parameters, the device type parameters of the status sensor are obtained synchronously, and the device type parameters include electromagnetic sensitive type and non-electromagnetic sensitive type classification.

3. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: Obtain the market price parameters and historical pledge rate parameters of the target goods, including: Obtaining real-time market price parameters of target goods through a financial data interface, wherein the financial data interface is connected to real-time quotation data of at least two independent commodity trading platforms; Obtain historical pledge rate parameters of the target goods through the bank or supply chain finance platform interface, where the historical pledge rate parameters include pledge rate data of the target goods under different pledge periods in the past twelve months; The real-time market price parameters are checked for validity. The validity check includes comparing whether the difference in quotation data of at least two trading platforms is less than a preset difference threshold. If the difference is less than the threshold, the average is taken as the final market price parameter.

4. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: The interference level is determined based on the comparison result between the environmental interference parameters and the preset interference threshold interval, and the initial credibility parameters are generated based on the device type parameters, including: Comparing the environmental interference parameter with a preset interference threshold interval, where the preset interference threshold interval includes a high interference interval, a medium interference interval, and a low interference interval; When the environmental interference parameter is in a high interference range, a first interference level is generated; when it is in a medium interference range, a second interference level is generated; when it is in a low interference range, a third interference level is generated; Based on the device type parameter, the real-time status parameter is classified according to credibility. The electromagnetic sensitive device generates a first credibility weight at the first interference level and a second credibility weight at the second interference level; the electromagnetic non-sensitive device generates a third credibility weight at the first interference level to the third interference level; According to the corresponding relationship between the interference level and the device type parameter, the real-time status parameter is multiplied by the corresponding credibility weight to generate an initial credibility parameter.

5. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: The initial credibility parameters are dynamically modified according to the correlation between the interference level and the equipment type parameters to generate the target credibility parameters, including: Determine a dynamic correction coefficient based on the correlation between the interference level and the equipment type parameter, the dynamic correction coefficient including a weight reduction coefficient for electromagnetic sensitive equipment at a high interference level and a buffer coefficient for electromagnetic sensitive equipment at a medium interference level; The initial credibility parameters are gradually attenuated by the weight reduction coefficient to generate preliminary revised credibility parameters; Based on the buffer coefficient, the initial revised credibility parameter is smoothly adjusted to generate the target credibility parameter; When there are unclassified sensors in the device type parameters, the initial credibility parameters are weighted according to the preset default correction rules to generate target credibility parameters.

6. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: The real-time status parameters are weighted in segments based on the target credibility parameters, and a dynamic pledge rate is generated through the weighted real-time status parameters and market price parameters, including: The credibility interval is divided according to the numerical range of the target credibility parameter, and the credibility interval includes a high credibility interval, a medium credibility interval and a low credibility interval; Assigning a preset weight value to each credibility interval, wherein a high credibility interval corresponds to a first weight, a medium credibility interval corresponds to a second weight, and a low credibility interval corresponds to a third weight; The real-time state parameter is multiplied by the corresponding preset weight value according to the credibility interval to which the target credibility parameter belongs, to generate a real-time state parameter after segment weighting; Input the real-time status parameters and market price parameters after segment weighting into the pledge rate calculation model, and generate the dynamic pledge rate through weighted average calculation; When the target credibility parameter cannot match any credibility interval, the real-time status parameter is weighted based on the preset default weight value to generate a dynamic pledge rate.

7. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1 is characterized in that: The abnormal level of pledge rate is determined based on the deviation between the dynamic pledge rate and the historical pledge rate parameters and the trend direction of the market price parameters, including: Calculate the percentage deviation between the dynamic pledge rate and the historical pledge rate parameters. The percentage deviation is obtained by subtracting the average historical pledge rate from the dynamic pledge rate and dividing it by the average historical pledge rate. Determine the trend direction of market price parameters, including upward trend, downward trend and flat trend; The abnormality level is divided according to the comparison result of the percentage deviation and the preset deviation threshold, and the preset deviation threshold includes a first-level deviation threshold, a second-level deviation threshold and a third-level deviation threshold; Determine the abnormal level of pledge rate; When the historical pledge rate parameter does not exist, the pledge rate abnormality level is determined based on the trend direction of the market price parameter and the industry average pledge rate.

8. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 7 is characterized in that: The determination of abnormal pledge rate level includes: When the percentage deviation exceeds the first-level deviation threshold and the trend direction is a downward trend, a first-level pledge rate abnormality level is generated; when the percentage deviation exceeds the second-level deviation threshold and the trend direction is a downward trend or a flat trend, a second-level pledge rate abnormality level is generated; when the percentage deviation exceeds the third-level deviation threshold, a third-level pledge rate abnormality level is generated.

9. The method for controlling supply chain financial goods risks based on the Internet of Things according to claim 1, characterized in that: Generate risk warning instructions based on the abnormal level of pledge rate and the rate of change of real-time status parameters, and trigger cargo control operation instructions, including: Calculate the change rate based on the change amount of the real-time state parameter within a preset time window; The abnormal level and change rate of the pledge rate are input into the risk warning rule base, which defines the warning instructions corresponding to different combinations of abnormal levels and change rate thresholds; When the pledge rate abnormality level is level one and the change rate exceeds the first rate threshold, a yellow risk warning instruction is generated and a cargo location verification operation instruction is triggered; When the pledge rate abnormality level is level 2 and the change rate exceeds the second rate threshold, an orange risk warning instruction is generated and a cargo in and out restriction operation instruction is triggered; When the pledge rate abnormality level reaches level three or the change rate exceeds the third rate threshold, a red risk warning instruction is generated and a cargo seizure operation instruction is triggered; After the cargo control operation instructions are generated, the input parameters of the pledge rate calculation model are updated synchronously.

10. A supply chain financial item risk control system based on the Internet of Things, used to implement the supply chain financial item risk control method based on the Internet of Things as described in any one of claims 1 to 9, characterized in that: include: Parameter collection module: collects environmental interference parameters, real-time status parameters and equipment type parameters of target goods; obtains market price parameters and historical pledge rate parameters of target goods; Credibility generation module: determines the interference level based on the comparison result between the environmental interference parameter and the preset interference threshold interval, and generates the initial credibility parameter based on the device type parameter; Dynamic correction module: dynamically corrects the initial credibility parameters according to the correlation between the interference level and the equipment type parameters to generate the target credibility parameters; Pledge generation module: based on the target credibility parameter, the real-time status parameter is segmented and weighted, and a dynamic pledge rate is generated through the segmented weighted real-time status parameter and the market price parameter; Abnormal grading module: Determines the abnormal level of pledge rate according to the deviation between the dynamic pledge rate and the historical pledge rate parameters and the trend direction of the market price parameters; Risk warning module: Generates risk warning instructions based on the abnormal level of pledge rate and the change rate of real-time status parameters, and triggers cargo control operation instructions.

Images