Block chain-based all-in-one fluorescent quantitative test strip management system and method

Abstract

The invention relates to the technical field of pesticide residue detection, and discloses an all-in-one fluorescent quantitative test strip management system and method based on a block chain, and the method comprises the steps: obtaining pesticide residue detection data of an all-in-one fluorescent quantitative test strip, carrying out the chaining storage of the detection data, and constructing a pesticide residue toxicity knowledge graph; comparing the detection data with the knowledge graph, and determining the original risk level of the pesticide; in combination with the region and time information, other detection data of the same region and the same time period in the block chain are obtained, clustering analysis is executed, and a pesticide risk reference of the region is determined; dynamically calibrating the original risk level based on the risk reference to generate a calibrated risk level; setting an intelligent contract on the block chain, and automatically triggering a corresponding execution strategy according to the calibrated risk level; the pesticide residue risk can be objectively and dynamically evaluated, intelligent management of pesticide residues is achieved, and the food safety guarantee level of agricultural products is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pesticide residue detection, and more specifically, to a management system and method for an all-in-one fluorescent quantitative test strip based on blockchain. Background Art

[0002] Although traditional pesticide residue detection technology can provide a certain degree of detection accuracy and efficiency, it still has major shortcomings in how to conduct intelligent analysis of detection data, establish a comprehensive risk assessment system, and achieve regional and dynamic pesticide residue risk management.

[0003] In the prior art, the relevant test data management methods and test strip information identification systems have solved the data collection and management problems in the detection process to a certain extent. For example, the Chinese patent with the authorization announcement number CN117554601B discloses a test data management method, device, equipment and storage medium, which analyzes the chemical reaction of the target test strip, associates the test results with the target identification code, and transmits the association relationship to the user device, thereby realizing the traceability and analysis of the test data. However, this technology is mainly aimed at urine detection scenarios, focusing on the correlation and traceability of detection data, and cannot dynamically analyze and risk assess the diverse data (such as environmental factors, regional information) in the pesticide residue detection scenario. In addition, this method lacks intelligent assessment of potential risks (such as the harm of excessive pesticide residues to the human body) and cannot provide a scientific basis for decision-making.

[0004] The Chinese patent application with publication number CN102509064A discloses a management system and its working method for obtaining test strip information based on video recording. By collecting one-dimensional code, two-dimensional code and other information of the test strip, the user is provided with test strip usage prompts, validity period management and calibration information transmission functions. This method has certain advantages in terms of the convenience of using test strip information, but its core functions are limited to the prompts and calibration of test strip product information, and do not involve in-depth analysis and risk assessment of test results. Therefore, the system cannot meet the needs of regional risk warning, historical data comparison and other functions in pesticide residue detection.

[0005] In summary, the existing technology cannot combine the correlation between test data and historical data in the management of test strip test data to dynamically analyze the risk level of pesticide residues; it fails to fully utilize the test data in the same region and time period for cluster analysis, and thus cannot provide a risk benchmark based on regional and time dynamics. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a management system and method for an all-in-one fluorescent quantitative test strip based on blockchain. By combining the decentralized and tamper-proof characteristics of blockchain with knowledge graphs and smart contract technology, historical transaction data of pesticide residues is extracted to construct a knowledge graph covering information such as pesticide types, toxicity index, and risk thresholds, thereby achieving a comprehensive analysis of pesticide residue risks. Real-time data on the blockchain is used to dynamically calibrate the risk level of pesticide residues through cluster analysis in combination with regional and time information, and an accurate risk assessment report is generated. By deploying smart contracts on the blockchain, corresponding execution strategies are automatically triggered according to the calibrated risk level, such as recall notices and regional warnings, thereby reducing manual intervention and improving regulatory efficiency.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] The management method of the all-in-one fluorescent quantitative test strips based on blockchain includes:

[0009] Obtain the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip, upload the first pesticide residue detection data to the blockchain for storage, and construct a pesticide residue toxicity knowledge graph;

[0010] The first pesticide residue detection data is compared with the pesticide residue toxicity knowledge graph to determine the original risk levels of n2 pesticide types; the first region information and the first time information of the first pesticide residue detection data are combined to obtain the second pesticide residue detection data of the same region and time period in the blockchain; cluster analysis is performed on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the region; based on the risk benchmark, the original risk level is dynamically calibrated to generate a calibrated risk level;

[0011] Set up smart contracts on the blockchain to automatically trigger corresponding execution strategies based on the calibrated risk level.

[0012] Further, the first pesticide residue detection data includes a first detection sample type, a first pesticide type, a first pesticide residue concentration, a first environmental factor, a first area information and a first time information; the first area information includes a first geographic coordinate, a first area name, a first area type and a first environmental feature; the first time information includes a first timestamp; the first pesticide type includes n2 pesticide types;

[0013] The step of obtaining the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip comprises:

[0014] Collect samples to be tested, use an all-in-one fluorescent quantitative test strip to test, and obtain a first pesticide residue concentration; record the type information of the samples to be tested as the first test sample type, and use the type of pesticide residue tested as the first pesticide type; record the geographical location information of the samples to be tested as the first regional information; record the test time as the first time information; and collect environmental parameters during the test as the first environmental factors.

[0015] Furthermore, the construction of the pesticide residue toxicity knowledge graph includes:

[0016] Obtain the historical transaction data of pesticide residues on the blockchain, and extract the first key indicator of pesticide residues from the historical transaction data of pesticide residues, wherein the first key indicator of pesticide residues includes a second pesticide type, a second test sample type, a second pesticide residue concentration, a second environmental factor, a second regional information, and a second time information; the second regional information includes a second geographic coordinate, a second regional name, a second regional type, and a second environmental feature; the second time information includes a second timestamp; and the second pesticide type includes n2 pesticide types;

[0017] Obtaining a second key indicator of pesticide residue, wherein the second key indicator of pesticide residue includes a pesticide residue threshold, a toxicity index, a hazard index, and a risk level label;

[0018] Based on the first key indicator of pesticide residue and the second key indicator of pesticide residue, a knowledge graph of pesticide residue toxicity is constructed.

[0019] Furthermore, the construction of the pesticide residue toxicity knowledge graph based on the first key indicator of pesticide residue and the second key indicator of pesticide residue includes:

[0020] According to the first key indicator of pesticide residue, create nodes of the pesticide residue toxicity knowledge graph, wherein the nodes of the pesticide residue toxicity knowledge graph include pesticide nodes, test sample nodes, environment nodes, time nodes and region nodes;

[0021] According to the first key indicator of pesticide residue and the second key indicator of pesticide residue, the attributes of the nodes of the pesticide residue toxicity knowledge graph are expanded;

[0022] Add relationship edges between nodes in the pesticide residue toxicity knowledge graph;

[0023] Assign weights to the relationship edges between nodes in the pesticide residue toxicity knowledge graph.

[0024] Furthermore, the nodes for creating the pesticide residue toxicity knowledge graph include:

[0025] Create a pesticide node according to the second pesticide type, create a detection sample node according to the second detection sample type, create an environment node according to the second environmental factor, create a time node according to the second time information, and create a region node according to the second region information;

[0026] The attributes of the nodes of the expanded pesticide residue toxicity knowledge graph include:

[0027] The attributes of the pesticide node are expanded according to the pesticide residue threshold, toxicity index, hazard indicator and risk level label; the attributes of the detection sample node are expanded according to the second pesticide residue concentration; the attributes of the time node are expanded according to the second timestamp; the attributes of the region node are expanded according to the second geographic coordinates, the second region name, the second region type and the second environmental characteristics.

[0028] Furthermore, adding relationship edges between nodes of the pesticide residue toxicity knowledge graph includes:

[0029] Establish a detection relationship edge between the pesticide node and the detection sample node; establish an environmental impact relationship edge between the detection sample node and the environment node; establish a time relationship edge between the time node and the detection sample node; establish a region relationship edge between the region node and the detection sample node;

[0030] The step of assigning weights to the relationship edges between the pesticide residue toxicity knowledge graph nodes includes:

[0031] The ratio of the second pesticide residue concentration to the pesticide residue threshold is used as the weight of the detection relationship edge;

[0032] According to the second environmental factor and the second pesticide residue concentration, the weight of the environmental impact relationship edge is calculated;

[0033] Obtain the pesticide application time, and calculate the weight of the time relationship edge according to the second timestamp of the time node and the pesticide application time;

[0034] Obtain the frequency of regional pesticide application and calculate the weight of the regional relationship edge based on the frequency of regional pesticide application.

[0035] Further, the determining of the original risk levels of the n2 pesticide types includes:

[0036] Compare the first pesticide residue concentration with the pesticide residue threshold of the corresponding pesticide node in the pesticide residue toxicity knowledge graph to calculate a first excess concentration ratio; the first excess concentration ratio is equal to the first pesticide residue concentration divided by the pesticide residue threshold;

[0037] According to the toxicity index and hazard index of the corresponding pesticide node in the pesticide residue toxicity knowledge graph, combined with the first excessive concentration ratio, the first risk index is calculated ;

[0038] The first risk index The original risk levels of the n2 pesticide types are compared with the preset risk level thresholds and divided into original risk levels.

[0039] Furthermore, the risk benchmarks for determining n2 pesticide types in the area include:

[0040] Classifying the second pesticide residue detection data according to the second pesticide type to generate n2 second pesticide residue data sets;

[0041] Each second pesticide residue dataset was clustered and the residue concentrations were divided into three risk clusters: high, medium, and low;

[0042] Calculate the means of the three risk clusters of high, medium and low to generate a second pesticide residue concentration threshold; the second pesticide residue concentration threshold includes a high risk concentration threshold, a medium risk concentration threshold and a low risk concentration threshold;

[0043] The second pesticide residue concentration threshold is used as the risk benchmark for n2 pesticide types in the area.

[0044] Furthermore, the dynamically calibrating the original risk level to generate a calibrated risk level includes:

[0045] Obtain a second pesticide residue concentration threshold value of the same type as the first pesticide, and generate a first risk level adjustment coefficient based on a comparison between the first pesticide residue concentration and the second pesticide residue concentration threshold value. ; Obtain the first area type in the first area information, and obtain the risk level distribution probability of the area type from the pesticide residue toxicity knowledge graph ,in Safe, Mild, Moderate, Severe , generate the second risk level adjustment coefficient ; Combined with the first risk level adjustment coefficient and the second risk level adjustment factor , adjust the original risk level to generate a calibrated risk level;

[0046] Generating a first risk level adjustment coefficient include:

[0047] When the first pesticide residue concentration Less than low risk concentration threshold When the first risk level adjustment coefficient The value is 1; when the first pesticide residue concentration Greater than high risk concentration threshold When the first risk level adjustment coefficient The value is -1; when the first pesticide residue concentration Greater than or equal to the low risk concentration threshold And less than or equal to the high risk concentration threshold When the first risk level adjustment coefficient The value is 0.

[0048] The management system of the all-in-one fluorescent quantitative test strip based on blockchain is used to implement the management method of the all-in-one fluorescent quantitative test strip based on blockchain, and the system includes:

[0049] Data acquisition module: used to obtain the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip;

[0050] Data storage module: used to upload the first pesticide residue detection data to the blockchain for storage;

[0051] Knowledge graph construction module: used to construct the knowledge graph of pesticide residue toxicity;

[0052] Risk level classification module: used to compare the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk level of n2 pesticide types;

[0053] Calibration module: used to combine the first area information and the first time information of the first pesticide residue detection data to obtain the second pesticide residue detection data of the same area and time period in the blockchain; perform cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the area; based on the risk benchmark, dynamically calibrate the original risk level to generate a calibrated risk level;

[0054] Execution module: used to set up smart contracts on the blockchain and automatically trigger corresponding execution strategies based on the calibrated risk level.

[0055] An electronic device comprises a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit, wherein the central processing unit implements the above-mentioned blockchain-based all-in-one fluorescent quantitative test strip management method when executing the computer program.

[0056] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the above-mentioned blockchain-based all-in-one fluorescent quantitative test strip management method.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] First, by uploading the pesticide residue detection data to the blockchain, the data is guaranteed to be tamper-proof and transparent, providing a reliable traceability basis for food safety supervision, and greatly improving the credibility and security of the data. Secondly, the present invention uses blockchain technology to construct a pesticide residue toxicity knowledge map. Combined with the historical transaction data and environmental factors of pesticide residues, it can comprehensively analyze the toxicity risk and regional distribution characteristics of pesticides, and provide a scientific basis for the risk assessment and dynamic calibration of pesticide residues. This analysis method based on knowledge maps can not only realize the accurate calculation of pesticide residue risks, but also dynamically adjust the risk level according to real-time data, thereby improving the accuracy and applicability of the test results. In addition, the invention realizes the automatic linkage between risk level and execution strategy through smart contracts, reduces manual intervention, improves the automation level and decision-making efficiency of the system, and can quickly respond to food safety risks. The combination of all-in-one fluorescent quantitative test strips makes the detection process convenient and efficient, suitable for a variety of scenarios, such as farmland, market and food processing plants, etc., realizes on-site rapid detection, and meets the diversified needs of food safety supervision. At the same time, the present invention can identify high-risk areas and high-risk periods through cluster analysis and environmental factor association, provide data support for regulatory authorities, and effectively prevent the further spread of pesticide residue problems. More importantly, the fluorescent immunochromatography technology of the test strip significantly improves the detection sensitivity and quantitative accuracy. Combined with the full-process management mechanism constructed by blockchain technology, a closed-loop system from data collection, storage, analysis to execution is formed, which improves the scientificity, effectiveness and intelligence level of food safety supervision. In summary, the present invention not only innovatively integrates blockchain and fluorescent quantitative detection technology at the technical level, but also demonstrates the ability to efficiently solve food safety problems in practical applications, providing a new technical path for pesticide residue management and food safety assurance. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 This is a principle flow chart of the management method of the all-in-one fluorescent quantitative test strip based on blockchain in the present invention;

[0061] Figure 2 A flowchart of a method for obtaining first pesticide residue detection data of an all-in-one fluorescent quantitative test strip in a blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0062] Figure 3A flowchart of a method for constructing a pesticide residue toxicity knowledge graph in a blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0063] Figure 4 A flowchart of a method for creating nodes of a pesticide residue toxicity knowledge graph in a blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0064] Figure 5 A flowchart of a method for expanding the attributes of nodes in a knowledge graph of pesticide residue toxicity in a blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0065] Figure 6 A flowchart of a method for adding relationship edges between nodes of a pesticide residue toxicity knowledge graph in a blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0066] Figure 7 A flowchart of a method for assigning weights to the relationship edges between nodes of the pesticide residue toxicity knowledge graph in the blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0067] Figure 8 A flowchart of a method for determining the original risk levels of n2 pesticide types in the blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0068] Fig. 9 A flowchart of a method for determining the risk benchmark of n2 types of pesticides in a region in the blockchain-based all-in-one fluorescent quantitative test strip management method of the present invention;

[0069] Fig.10 This is a functional module diagram of the management system of the all-in-one fluorescent quantitative test strips based on blockchain in the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Example 1

[0072] See also Figure 1 As shown, this embodiment provides a management method for an all-in-one fluorescent quantitative test strip based on blockchain, including:

[0073] Step S1000, obtaining first pesticide residue detection data of the all-in-one fluorescent quantitative test strip, uploading the first pesticide residue detection data to the blockchain for storage, and constructing a pesticide residue toxicity knowledge graph;

[0074] Furthermore, step S1000 includes:

[0075] Step S1100, obtaining first pesticide residue detection data of the all-in-one fluorescent quantitative test strip; the first pesticide residue detection data includes a first detection sample type, a first pesticide type, a first pesticide residue concentration, a first environmental factor, first area information and first time information; the first area information includes a first geographic coordinate, a first area name, a first area type and a first environmental feature; the first time information includes a first timestamp; the first pesticide type includes n2 pesticide types;

[0076] Furthermore, if Figure 2 As shown, step S1100 includes:

[0077] Step S1110, collecting a sample to be tested, using an all-in-one fluorescent quantitative test strip to test it, and obtaining a first pesticide residue concentration;

[0078] Step S1120, recording the type information of the sample to be tested as the first test sample type, and taking the type of pesticide residue to be tested as the first pesticide type;

[0079] Step S1130, recording the geographical location information of the sample to be detected as the first regional information; recording the detection time as the first time information;

[0080] Step S1140, collecting environmental parameters during detection as the first environmental factor.

[0081] Specifically, the all-in-one fluorescent quantitative test strip uses fluorescent immunochromatography technology (FICT) to fix specific pesticide residue antibodies labeled with fluorescent microspheres in the detection area. The pesticide residue antibodies in the sample combine to form an antigen-antibody complex, and the fluorescence signal intensity of the microspheres is proportional to the concentration of pesticide residues. The fluorescence signal is read by a time-resolved fluorescence detection device, and quantitative detection is achieved in combination with a built-in standard curve. The all-in-one fluorescent quantitative test strip combines time-resolved fluorescent immunochromatography technology and uses fluorescent microspheres to label specific pesticide antibodies, which significantly improves the detection sensitivity and quantitative accuracy. The test strip is designed as a portable structure with simple operation steps, which can meet the on-site detection needs of food production and processing companies, farmers' markets and home users.

[0082] The first test sample type refers to the specific type of the sample to be tested, such as vegetables (such as cabbage, spinach), fruits (such as apples, grapes), grains (such as wheat, rice), etc. The sample type is the basic information for the test, which affects the selection of the pesticide to be tested and its residue threshold. For example, different pesticides have different residue thresholds in different sample types, and the test results need to match the standards. Records of sample types can help trace the source of the test and facilitate classification processing. If the sample is spinach, the sample type is "vegetable". According to the standard, the pesticide residue limit in spinach may be different from that in apples (fruits), so classification information is an important basis for subsequent calculations.

[0083] The first pesticide type refers to the target pesticide residue type detected using the all-in-one fluorescent quantitative test strip, where n2 represents the number of pesticide types detected. By recording the specific name of the pesticide, the detection index can be clarified, which is convenient for comparison with the pesticide residue limit standard. The type of pesticide directly determines the target substance and the choice of detection method. For example, different types of pesticides have different chemical properties, and targeted antibodies are required for detection. The toxicity, degradation rate and residue threshold of different pesticides are different, so information on the type of pesticide is crucial for risk assessment. If the pesticide being detected is aldicarb, the aldicarb-specific antibody in the test strip needs to bind to the aldicarb in the sample to generate a fluorescent signal, and then analyze its concentration in combination with the standard curve.

[0084] First, the pesticide residue concentration refers to the actual content of pesticides detected in the sample, usually in micrograms per kilogram (μg / kg) or milligrams per kilogram (mg / kg). The pesticide residue concentration is detected by fluorescent immunochromatography technology, and the intensity of the fluorescent signal on the test strip is proportional to the pesticide concentration. Detecting the pesticide residue concentration is an important basis for assessing food safety. The concentration directly affects the classification of risk levels. For example, if the test results show that the concentration of aldicarb in spinach is 0.8mg / kg, and the national standard stipulates that its maximum residue limit (MRL) is 0.5mg / kg, it means that the pesticide residue in the spinach sample exceeds the standard.

[0085] The first environmental factor refers to the relevant environmental parameters recorded when collecting test samples, such as temperature, air humidity, rainfall, soil pH, organic matter content, etc. Environmental factors may affect the residual level and degradation rate of pesticides in farmland and agricultural products. Including them in the test data will help to deeply analyze the behavior of pesticide residues. Temperature is an important environmental factor that affects the degradation rate of pesticides. High humidity may accelerate the degradation of certain pesticides, and the acidity and alkalinity of the soil have a significant effect on the degradation rate of pesticides. Environmental factors directly affect the degradation rate and residual characteristics of pesticides. For example, high temperature helps to quickly degrade certain organic pesticides, while low temperature may lead to prolonged residual time. For example, if the temperature during the test is 25°C, the humidity is 70%, and the soil pH is 6.5, the actual residual risk can be inferred based on the degradation characteristics of the pesticide.

[0086] The first regional information refers to the geographical location information of the area where the test samples are collected, including the following: First geographic coordinates: the longitude and latitude coordinates of the test area, such as "longitude: 120.123, latitude: 30.456", which clearly identifies the test location. First regional name: the name of the test area, such as "Shanghai Farmers' Market A", which intuitively describes the test site. First regional type: the type of the test area, such as farmland, orchard, vegetable greenhouse, farmers' market, supermarket, food processing plant, etc., reflecting the distribution of pesticide residue detection in agricultural production, circulation, processing and other links. For example, the regional type is marked as "farmland". First environmental characteristics: the description of the environmental characteristics of the test area, such as "wetland", "temperate monsoon climate", etc., reflecting the natural environmental background of the region. Regional information can help trace the source of pesticide residues, such as whether the residues are related to the pesticide application history of the test site. Environmental characteristics (such as wetlands) may affect the degradation rate and residual characteristics of pesticides. For example, if the detection location is "Shanghai Farmers' Market A" (latitude and longitude are 120.123, 30.456), the area type is "market", and the environmental characteristics are "temperate climate", it can provide node information for the subsequent establishment of a pesticide residue toxicity knowledge graph.

[0087] The first time information refers to the information related to the detection time, specifically the first timestamp, that is, the time point when the detection operation is performed, usually accurate to seconds, such as "2023-04-2014:30:00". By recording the timestamp, the detection behavior can be traced back and the time trend of the pesticide residue level can be analyzed. Time information can help determine whether the pesticide residue is close to the time of pesticide application. If the detection time is close to the application time, there may be a risk of excessive residue. Time information also plays an important role in dynamically calibrating the risk level. For example, if the detection time is "10:30:00 on December 4, 2024", and the pesticide application time is "December 3, 2024", it can be judged that the residue may belong to a high-risk period.

[0088] Step S1100 forms a complete data set by clearly recording the sample type, pesticide type, pesticide residue concentration, time, location and environmental factors, providing a solid foundation for subsequent analysis. By recording the first regional information, the source and detection location of the pesticide residue can be traced, which is convenient for regulatory authorities to conduct regional risk assessments. By combining environmental factors and time information, the actual risk of pesticide residues can be assessed more accurately. For example, a high temperature environment may accelerate degradation, thereby reducing risks. The all-in-one fluorescent quantitative test strip is simple and convenient to use, suitable for a variety of scenarios such as farmers' markets and food processing companies, and meets the needs of rapid on-site testing. By collecting comprehensive environmental parameters and regional information, a scientific basis can be provided for the establishment of a knowledge map of pesticide residue toxicity, thereby improving the scientificity and rationality of risk level division.

[0089] In summary, step S1100 lays a data foundation for the subsequent construction of the pesticide residue toxicity knowledge graph by collecting the first pesticide residue detection data in a standardized manner. Systematic and comprehensive pesticide residue detection data can fully describe the spatiotemporal distribution, environmental impact and other characteristics of pesticide residues, allowing multiple parties to understand the pesticide residue problem more clearly and intuitively, thereby effectively guiding agricultural production practices and ensuring the quality and safety of agricultural products.

[0090] Step S1200, uploading the first pesticide residue detection data to the blockchain for storage;

[0091] Further, step S1200 includes:

[0092] Step S1210, encoding the first pesticide residue detection data according to a predetermined data format, generating transaction data on the blockchain, and digitally signing the transaction data;

[0093] Step S1220, other blockchain nodes verify the digital signature of the transaction data;

[0094] Step S1230, through the consensus algorithm, the blockchain network reaches a consensus on the transaction data and packages it into a new block;

[0095] In step S1240, the new block is linked to the existing blockchain.

[0096] Specifically, first, the first pesticide residue detection data is encoded according to a predetermined data format to generate transaction data on the blockchain. The so-called predetermined data format refers to the standardized representation of fields such as the detection sample ID, detection time, detection location, first pesticide type, first pesticide residue concentration, and first environmental factor according to the data structure and protocol requirements of the blockchain network, such as using a common data exchange format such as JSON or XML. This process can be understood as the "translation" of data, converting heterogeneous raw data into standardized data that can be understood and processed by the blockchain system. The standardized data is called "transaction data", which is the basic unit of information exchange and storage on the blockchain.

[0097] In order to ensure the authenticity and non-repudiation of transaction data, it needs to be digitally signed. Digital signatures use asymmetric encryption algorithms, the most common of which are RSA and elliptic curve encryption algorithm (ECDSA). Asymmetric encryption algorithms are based on a pair of keys: public key and private key. The public key is public, while the private key is confidential and only held by the data provider (i.e. the uploader of the test data). The transaction data is signed with the private key. Anyone can verify the authenticity of the signature using the public key, but only the private key holder can generate a valid signature. This is like a digital "fingerprint" that ensures the reliability of the data source. The signed transaction data is broadcast to the entire blockchain network.

[0098] When other blockchain nodes receive the signed transaction data, they will use the public key of the data provider to verify the validity of the digital signature. Only when the signature verification is passed will the transaction data be considered legitimate and enter the next processing flow. This verification process can prevent forgery or tampering of data, because if someone attempts to modify the signed transaction data, its digital signature will no longer match, resulting in verification failure. Through distributed node verification, the blockchain network can collectively maintain the authenticity of the data.

[0099] When most nodes reach a consensus on the validity of transaction data, the data will be packaged into a new block. Consensus algorithm is the core mechanism of blockchain, and common ones include proof of work (PoW) and proof of stake (PoS). Taking PoW as an example, miner nodes compete for the right to record accounts by solving a difficult mathematical problem (i.e. proof of work). The first miner to find the answer to the problem will get the right to package a new block. In addition to transaction data, the new block also contains metadata such as timestamp, hash value of the previous block, and the miner's digital signature. This additional information connects the new block to the existing blockchain to form an inseparable whole.

[0100] Finally, the new block is added to the end of the blockchain and becomes the latest block. Since each block records the hash value of the previous block, the entire chain is linked together. If someone tries to tamper with the data of a block in the middle, its hash value will change, causing the hash values ​​of all subsequent blocks to no longer match. This hash chain structure gives the blockchain extremely high data integrity and immutability. Each node saves a complete copy of the blockchain, and all copies are identical. Therefore, even if individual nodes fail or act maliciously, it will not affect the normal operation and data security of the entire network.

[0101] In summary, storing pesticide residue detection data on a blockchain can provide the following beneficial effects:

[0102] Data authenticity and credibility: Due to the use of digital signatures and consensus mechanisms, the test data uploaded to the chain can be collectively verified by the network, ensuring the authenticity and reliability of the data and eliminating fraud. This is crucial for the safe traceability of agricultural products.

[0103] Data cannot be tampered with: Once the data is written into the blockchain, it cannot be modified or deleted afterwards, because any tampering will cause the data hash value to change and be detected. This tamper-proof property protects the original appearance of the data, prevents the data from being forged or altered, and provides a reliable data basis for the supervision of pesticide residues.

[0104] Data is fully traceable: Blockchain records each piece of data in the form of timestamps and hash chains, forming a clear data footprint. With this feature, pesticide residue data can be tracked and traced throughout the process, and key information such as sample testing time, testing links, and circulation paths can be understood, thereby achieving traceability of agricultural product safety.

[0105] High data availability: The blockchain network uses distributed storage, and each node stores a complete copy of the data. This multi-copy fault-tolerant mechanism can prevent single point failures, and even if some nodes have problems, it will not affect the accessibility of the data. Distributed storage can also support the storage and high concurrent access of massive data, providing reliable protection for the long-term preservation and use of agricultural residue data.

[0106] Data privacy protection: Although blockchain is an open and transparent distributed ledger, not all data is directly exposed on the chain. Private data can be stored outside the chain, and only its hash value is saved on the chain as a fingerprint. In addition, some blockchain platforms also support privacy protection technologies such as multi-party secure computing and homomorphic encryption. Through these means, regulatory authorities can easily audit data and protect sensitive data from being leaked.

[0107] For example, suppose an agricultural product base uses an all-in-one fluorescent quantitative test strip to detect pesticide residues in vegetables. The test results show that a batch of spinach samples contained 0.08 mg / kg of aldicarb (a highly toxic insecticide). In order to ensure the authenticity and traceability of the data, the base stores this test data on the chain.

[0108] First, the base organizes the detection data into a standardized JSON format;

[0109] The base uses its own private key to digitally sign the JSON string and broadcast the signed transaction data to the blockchain network. Other nodes use the base's public key to verify the signature. After the consensus algorithm confirms that the transaction is valid, it is packaged into a block and linked to the blockchain. In this way, this test record is permanently and tamper-proof stored on the blockchain, becoming part of the pesticide residue safety traceability system.

[0110] Assuming that this batch of spinach flows into the farmers' market, the food safety regulatory department can retrieve the pesticide residue detection data of this batch of spinach from the blockchain by scanning the QR code on the spinach packaging. Since the data has been digitally signed and network consensus has been established, the regulatory department can be sure that the data is authentic and reliable. Once data falsification is discovered, the responsible party can be traced back.

[0111] The storage of pesticide residue detection data on the blockchain fully utilizes the advantages of blockchain technology, providing a reliable, traceable and verifiable data foundation for agricultural product safety. This is not only conducive to food supervision, but also provides a trust guarantee for food safety and promotes the healthy and sustainable development of agriculture.

[0112] Step S1300, constructing a pesticide residue toxicity knowledge graph;

[0113] Furthermore, step S1300 includes:

[0114] Step S1310, obtaining the pesticide residue historical transaction data on the blockchain, extracting the first key indicator of pesticide residue from the pesticide residue historical transaction data, wherein the first key indicator of pesticide residue includes a second pesticide type, a second test sample type, a second pesticide residue concentration, a second environmental factor, second regional information, and second time information; the second regional information includes a second geographic coordinate, a second regional name, a second regional type, and a second environmental feature; the second time information includes a second timestamp; and the second pesticide type includes n2 pesticide types;

[0115] Specifically, we first access the blockchain network to obtain historical transaction records of pesticide residue detection data. These transaction records are stored in blocks of the blockchain in chronological order, and each block contains all transactions within a certain time range. By parsing these transactions, we extract the key information elements required to build the knowledge graph, namely the first key indicator of pesticide residue:

[0116] (1) Second pesticide type: Extract the name of the pesticide detected in each transaction record as the attribute of the pesticide node. For example, chemical pesticides and organochlorine pesticides can be clearly identified by recording the specific type of pesticides.

[0117] (2) Second, the type of test sample: extract the specific type information of the test sample, such as wheat, rice, apple, etc., as the attribute of the test sample node. The type of test sample affects the evaluation standard of pesticide residues. For example, the maximum residue limits of pesticides in wheat, rice, and apples are not the same. By distinguishing the sample types, it can correspond to specific residue limit standards.

[0118] (3) Second pesticide residue concentration: Extract the pesticide residue concentration value of the test sample in mg / kg as the attribute of the edge from the pesticide node to the test sample node. Pesticide residue concentration is a key indicator for evaluating the quality and safety of agricultural products. By comparing the residue concentration with the residue limit threshold, it can be determined whether it exceeds the standard. For example, if 0.5 mg / kg of aldicarb is detected in wheat, and the maximum residue limit of aldicarb in wheat is 0.1 mg / kg, it is considered that the pesticide residue in this batch of wheat exceeds the standard.

[0119] (4) Second environmental factor: Extract environmental parameters recorded during testing, such as temperature, humidity, soil pH, etc., as attributes of environmental nodes. Environmental factors affect the degradation of pesticides in the environment. For example, high temperature may accelerate the degradation of pesticides, while acidic soil slows down the degradation. By recording environmental factors, the dynamic changes of pesticide residues can be analyzed.

[0120] (5) Secondary regional information: Extract the location information of the detection area, including geographic coordinates (latitude and longitude), regional name (a certain town, a certain village), regional type (farmland, vegetable field, orchard), and environmental characteristics (soil, climate conditions) as attributes of the regional node. Regional information can trace the geographical distribution of pesticide residue detection. Environmental differences in different regions may affect the level of pesticide residues.

[0121] (6) Second time information: extract the timestamp of the detection operation, accurate to seconds, as the attribute of the time node. Time information can describe the dynamic trend of pesticide residues. For example, higher pesticide residues are detected before harvest, while pesticide residues decrease after harvest, indicating that pesticide residues have a certain timeliness.

[0122] Through the above steps, a wealth of pesticide residue detection data was obtained from the blockchain network, covering multi-dimensional data such as pesticide types, test samples, residue concentrations, environmental conditions, and spatiotemporal information, laying a solid data foundation for building a knowledge graph. Storing data on the chain can ensure the authenticity and immutability of the data and improve data quality.

[0123] Step S1320, obtaining a second key indicator of pesticide residue, wherein the second key indicator of pesticide residue includes a pesticide residue threshold, a toxicity index, a hazard index, and a risk level label;

[0124] Specifically, after extracting the first key indicator of pesticide residues in the blockchain network, it is also necessary to obtain the second key indicator as the attribute of the edge and node in the knowledge graph, giving the pesticide residue data clear toxicology and risk assessment semantics, so that the knowledge graph has the ability to analyze pesticide residue risks. The second key indicator includes:

[0125] (1) Pesticide residue threshold: refers to the maximum residue level (MRL) of pesticides allowed in agricultural products. MRL is a food safety standard formulated based on factors such as the toxicological properties of pesticides, application methods, and environmental behavior. By comparing the pesticide residue detection value with the MRL, it can be determined whether the agricultural products exceed the standard. MRL data can be obtained from authoritative standard databases, such as the CODEX standard established by the WHO and China's GB 2763-2019 "National Food Safety Standard Maximum Residue Limits of Pesticides in Food". Access these databases, match the detected pesticide types and test samples, and extract the corresponding MRL values ​​as the attributes of the pesticide node. For example, the MRL of aldicarb in wheat is 0.01 mg / kg. By comparing the detection concentration of the aldicarb node with 0.1 mg / kg, it can be found whether it exceeds the standard.

[0126] (2) Toxicity index: It is a quantitative indicator to measure the toxic effect of pesticides on organisms. Common toxicity indices include acute reference dose (ARfD) and acceptable daily intake (ADI). ARfD is the estimated dose of pesticide that can be tolerated for a single or short-term intake, and ADI is the tolerable dose of pesticide for long-term intake. Toxicity index data can be obtained from toxicology databases, such as the IRIS database of the US EPA and the WHO Pesticide Residue Toxicology Assessment Report. These databases are searched to extract the ARfD and ADI values ​​of different pesticides and mark them on the pesticide nodes. The toxicity index directly reflects the hazard level of the pesticide. The smaller the index, the more toxic the pesticide. For example, the ADI of aldicarb is 0.003 mg / kg, which is lower than that of most pesticides, indicating that it is highly toxic.

[0127] (3) Hazard indicators: These are qualitative descriptors of the degree of hazard that are summarized and refined based on cases of acute and chronic poisoning caused by pesticide residues. Common hazard indicators include “mild,” “moderate,” and “severe.” Relevant cases are searched through search engines and academic databases using keywords such as “pesticide residues,” “poisoning,” and “exceeding the standard.” Qualitative analysis is performed on the search results to summarize the symptoms of poisoning caused by pesticides, and the hazard level is summarized and marked on the pesticide node. For example, in the case of “exceeding the standard for aldicarb residues causing 10 people to be poisoned and hospitalized,” the hazard indicator of aldicarb can be set as “severe.”

[0128] (4) Risk level label: It is a graded evaluation label for pesticide residue risk by comprehensively considering factors such as residue concentration, toxicity index, and hazard index. The grades are from low to high, and common grade labels are "low risk", "medium risk", and "high risk". The core idea of ​​risk level evaluation is to multiply the ratio of pesticide residue concentration to residue threshold (i.e., exposure parameter) by the toxicity index (toxicity parameter), divide the risk level according to the size of the product, and then add the hazard index correction. By automatically calculating the exposure parameters and toxicity parameters, generating risk level labels, and marking the risk level labels on the knowledge graph, users can be intuitively warned of pesticide residue risks.

[0129] By extracting and calculating the above-mentioned second key indicator of pesticide residues and associating it with the first key indicator, a semantically rich knowledge graph is constructed. The knowledge graph not only describes the current spatiotemporal distribution of pesticide residues, but also mines the toxicological characteristics and risk information contained in the pesticide residue data, which can support the rapid assessment and early warning of pesticide residue risks.

[0130] Step S1330, constructing a pesticide residue toxicity knowledge graph based on the first key indicator of pesticide residue and the second key indicator of pesticide residue.

[0131] Furthermore, if Figure 3 As shown, step S1330 includes:

[0132] Step S1331, creating nodes of a pesticide residue toxicity knowledge graph according to the first key indicator of pesticide residue, wherein the nodes of the pesticide residue toxicity knowledge graph include pesticide nodes, test sample nodes, environment nodes, time nodes and region nodes;

[0133] Furthermore, if Figure 4 As shown, step S1331 includes:

[0134] Step S13311, creating a pesticide node according to the second pesticide type;

[0135] Step S13312, creating a detection sample node according to the second detection sample type;

[0136] Step S13313, creating an environment node according to the second environment factor;

[0137] Step S13314, creating a time node according to the second time information;

[0138] Step S13315: Create a regional node according to the second regional information.

[0139] Step S1332, expanding the attributes of the nodes of the pesticide residue toxicity knowledge graph according to the first key indicator of pesticide residue and the second key indicator of pesticide residue;

[0140] Furthermore, if Figure 5 As shown, step S1332 includes:

[0141] Step S13321, expanding the attributes of the pesticide node according to the pesticide residue threshold, toxicity index, hazard index and risk level label;

[0142] Step S13322, expanding the attributes of the detection sample node according to the second pesticide residue concentration;

[0143] Step S13323, expanding the attributes of the time node according to the second timestamp;

[0144] Step S13324, expanding the attributes of the area node according to the second geographic coordinates, the second area name, the second area type and the second environmental characteristics.

[0145] Specifically, step S1331 constructs the basic framework of the pesticide residue toxicity knowledge graph through the information extracted from the first key indicator of pesticide residue. The framework consists of five types of nodes: pesticide nodes, test sample nodes, environment nodes, time nodes, and region nodes. Each type of node represents an important dimension of pesticide residue detection data.

[0146] The pesticide node corresponds to the second pesticide type, that is, the specific name of the detected pesticide. Each pesticide will be an independent node in the knowledge graph. The pesticide node is the core of the entire graph, and other nodes will be associated around the pesticide node. Through the pesticide node, the residual status of different pesticides under various conditions can be intuitively displayed.

[0147] The detection sample node corresponds to the second detection sample type, that is, the sample type for pesticide residue detection, such as vegetables, fruits, grains, etc. The sample type affects the risk assessment of pesticide residues because different agricultural products have different absorption and enrichment capabilities for pesticides. For example, leafy vegetables are more likely to absorb and retain pesticide residues than root vegetables. Therefore, using the detection sample type as a node in the knowledge graph can reveal the association between pesticide residues and agricultural product types.

[0148] The environmental node corresponds to the second environmental factor, which is the environmental parameters recorded during pesticide residue detection, such as temperature, humidity, rainfall, etc. Environmental conditions significantly affect the degradation rate and residual degree of pesticides. For example, high temperature and high humidity environments may accelerate the degradation of certain pesticides. Incorporating environmental factors into the knowledge graph will help to deeply understand the environmental drivers of pesticide residue behavior and provide more clues for pesticide residue risk prediction.

[0149] The time node corresponds to the second time information, that is, the timestamp attribute of the pesticide residue detection. The time information can be a specific date or time, such as "April 20, 2023 14:30:00". Through the time node, the dynamic trend of the pesticide residue level can be portrayed. For example, by connecting the pesticide residue data at multiple time points in a year, the seasonal pattern of pesticide residues can be analyzed, providing time guidance for pesticide residue monitoring.

[0150] The regional node corresponds to the second regional information, including multiple attributes such as geographic coordinates, regional name, regional type, and environmental characteristics. Regional nodes can represent different geographical locations such as farmland, orchards, and farmers' markets. Through regional nodes, the spatial distribution pattern of pesticide residues can be displayed, and high-incidence areas and low-risk areas of pesticide residues can be identified to achieve precise governance.

[0151] Step S1332 further expands the attribute information of each node based on the nodes constructed in S1331 according to the first key indicator and the second key indicator of pesticide residue. The attribute information is a detailed description of the node, which can comprehensively characterize the characteristics of pesticide residue.

[0152] For the pesticide node, step S13321 expands its attributes according to the pesticide residue threshold, toxicity index, hazard index and risk level label. The pesticide residue threshold reflects the maximum residue limit standard of the pesticide in agricultural products and is the basis for judging whether the pesticide residue exceeds the standard. The toxicity index quantifies the degree of toxicity of pesticides to organisms, such as acute reference doses. The hazard index assesses the hazard level based on poisoning cases caused by pesticides. The risk level label combines the above information to classify and control pesticides. The expanded pesticide node not only includes the threshold specifications for residues, but also covers toxicological parameters, while taking into account actual hazard cases, and characterizing pesticides in a multi-dimensional and comprehensive manner, which is convenient for comprehensive risk assessment.

[0153] For the detection sample node, step S13322 expands its attributes according to the second pesticide residue concentration. The second pesticide residue concentration reflects the actual amount of pesticide residue detected in the agricultural product sample. By using the residue concentration as an attribute of the sample node, the detection result can be directly associated with the type of agricultural product. This is of great significance for understanding the residual patterns of pesticides in different agricultural products. For example, by comparing the pesticide residue concentration attributes of different vegetables or fruits, the accumulation characteristics of pesticides in specific agricultural products can be discovered.

[0154] For the time node, step S13323 is expanded according to the second timestamp attribute. The second timestamp is accurate to a specific date and time, such as "2023-04-20 14:30:00". Using the timestamp as an attribute of the time node can accurately locate and trace the pesticide residue detection event. For example, the time interval between pesticide application and sampling detection can be analyzed according to the timestamp to assess the risk of excessive pesticide residues. The introduction of the time attribute allows the knowledge graph to depict the dynamic evolution of pesticide residues over time.

[0155] For the regional node, step S13324 expands its attributes according to the second geographic coordinates, the second area name, the second area type, and the second environmental characteristics. The geographic coordinates clearly identify the location of the detection area; the area name such as "Shanghai Farmers' Market A" intuitively describes the detection site; the area type such as farmland, orchard, etc. reflects the agricultural production links; environmental characteristics reflect the natural background of the region. Through these attributes, the regional node can comprehensively record the spatiotemporal background of the detection behavior and show the regional characteristics of the distribution of agricultural residues. For example, agricultural residues in the farmland area may come from the application of pesticides, while agricultural residues in the farmers' market may be brought in through transportation. Attribute information allows regional nodes to contain richer semantics and enhance the expressive power of the knowledge graph.

[0156] In summary, steps S1331 and S1332 construct a pesticide residue toxicity knowledge graph skeleton containing five types of nodes: pesticide, test sample, environment, time and region, and extract attribute information from multiple key indicators to enrich the graph, forming a comprehensive and systematic pesticide residue feature library. The graph skeleton is clearly classified and covers all the core elements of pesticide residue detection, while the attribute information deeply depicts the characteristics of pesticide residues from multiple angles such as pesticide toxicity, residue concentration, and spatiotemporal distribution. The constructed knowledge graph is like an information panorama, which can intuitively present the overall picture of pesticide residues and reveal their internal laws. This provides a decision-making basis for pesticide residue management, such as identifying high-risk pesticides and key prevention and control areas, and giving full play to the effectiveness of digital means in agricultural ecological management. At the same time, organizing knowledge into the form of a graph is also conducive to the correlation analysis and intelligent search of information, making pesticide residue supervision more efficient and convenient. The construction of the knowledge graph makes the scattered pesticide residue data clear and organized, so that pesticide residue prevention and control can be done with a clear mind, improve the accuracy and effectiveness of management, and promote the green and sustainable development of agriculture.

[0157] Step S1333, adding relationship edges between nodes of the pesticide residue toxicity knowledge graph;

[0158] Furthermore, if Figure 6 As shown, step S1333 includes:

[0159] Step S13331, establishing a detection relationship edge between the pesticide node and the detection sample node;

[0160] Step S13332, establishing an environmental impact relationship edge between the detection sample node and the environment node;

[0161] Step S13333, establishing a time relationship edge between the time node and the detection sample node;

[0162] Step S13334, establishing a regional relationship edge between the regional node and the detection sample node.

[0163] Step S1334, assigning weights to the relationship edges between the nodes of the pesticide residue toxicity knowledge graph.

[0164] Furthermore, if Figure 7 As shown, step S1334 includes:

[0165] Step S13341, using the ratio of the second pesticide residue concentration to the pesticide residue threshold as the weight of the detection relationship edge;

[0166] Step S13342, calculating the weight of the environmental impact relationship edge according to the second environmental factor and the second pesticide residue concentration;

[0167]

[0168] in:

[0169] : Environmental factor weight coefficient, indicating the The relative importance of environmental factors (such as temperature, humidity, etc.) on pesticide degradation. Obtained through experimental data or literature research: For example, the effect of temperature on degradation is usually greater than humidity, and humidity is greater than wind speed, so different weight values ​​can be assigned. Value range: , and satisfies .

[0170] n: total number of environmental factors.

[0171] : No. The actual value of an environmental factor is obtained through detection equipment, such as temperature (unit: ), humidity (unit: %), rainfall (unit: mm), soil pH value, etc.

[0172] : No. The optimal value of an environmental factor (optimal degradation condition) indicates the value when the degradation rate of the pesticide reaches its peak under this environmental factor. It is determined through experiments. For example, the optimal degradation temperature of some pesticides may be , humidity is .

[0173] : Environmental factor sensitivity coefficient, used to describe the sensitivity of the environmental factor to the degradation rate of pesticides when it deviates from the optimal value. Obtained through experiments or modeling, for example: temperature is more sensitive to degradation ( smaller), while the sensitivity to wind speed is lower ( The larger the sensitivity, the higher the The smaller it is), the faster the function value decreases.

[0174] : No. The degree of influence of an environmental factor on pesticide degradation indicates whether the environmental factor is conducive to degradation. The value range is: . Acquisition method: through fitting of experimental data, or based on degradation rate models in literature.

[0175] : Residual concentration adjustment factor, used to adjust the influence of the second pesticide residue concentration on the weight. Determined by model optimization, usually in the range of .if The higher the value, the more obvious the inhibitory effect of concentration on weight.

[0176] : The second pesticide residue concentration.

[0177] : An exponential function with the base e of the natural logarithm as its base.

[0178] : Indicates the deviation between the actual value of the environmental factor and the optimal value. The smaller the deviation, the closer the current environmental conditions are to the optimal degradation conditions. An exponential function is used to describe the impact of the deviation. The larger the deviation, the smaller the exponential function value, and the weaker the degradation effect.

[0179] :The second regulating factor of the impact of pesticide residue concentration on the environment. The higher the concentration, the stronger the inhibitory effect on the weight. The use of logarithmic function can avoid excessive nonlinear effects on the weight when the concentration is too high.

[0180] This formula is obtained by and , the influence weight and degree of each environmental factor can be flexibly set to adapt to the degradation characteristics of different pesticides. The calculation deviation can dynamically reflect the impact of environmental conditions on the degradation effect. and , to avoid nonlinear distortion of weights when the concentration is too high or too low. Use By controlling the sensitivity of environmental factor deviations, the model can be optimized for specific pesticides.

[0181] when When (the environment is close to the optimal value):

[0182]

[0183] The environmental impact weights tend to the maximum value, indicating that the degradation conditions are optimal.

[0184] when Deviation When larger:

[0185]

[0186] The environmental impact weight decreased significantly, indicating that the degradation conditions worsened.

[0187] when (Second pesticide residue concentration is low):

[0188]

[0189] The environmental impact weight is entirely determined by environmental factors.

[0190] when (Second pesticide residue concentration is too high):

[0191]

[0192] This indicates that high residual concentration has a significant inhibitory effect on the degradation effect.

[0193] By comprehensively considering environmental factors, pesticide residue concentration and degradation characteristics, this formula can not only accurately describe the dynamic impact of the environment on the degradation rate of pesticides, but also provide a quantitative basis for pesticide residue management. The complexity and flexibility of the formula make it applicable to a variety of scenarios, especially in the fields of food safety and environmental protection.

[0194] Step S13343, obtaining the pesticide application time, and calculating the weight of the time relationship edge according to the second timestamp of the time node and the pesticide application time;

[0195]

[0196] in:

[0197] : Time relationship edge weight, reflecting the importance of time to pesticide residue concentration.

[0198] : Weight adjustment factor, set according to experience or experiment, usually takes a value range of 0<α≤1, controls the maximum value of the time weight, and reflects the influence of the time factor on the overall weight.

[0199] : Time sensitivity coefficient, obtained by fitting the pesticide degradation experimental data, usually in the range of 0<κ≤10, the sensitivity of pesticides to time factors. The larger the value, the more sensitive the time weight is to the change of time deviation.

[0200] : Pesticide degradation characteristic time constant (such as half-life), found in literature or experiments, in hours or days. For example, the degradation half-life of aldicarb is about 1 day.

[0201] : Detection time, obtained from the second timestamp of the time node, in hours or days.

[0202] : Pesticide application time, obtained from the blockchain historical records or the pesticide application database, in hours or days. The time point of pesticide application is usually a known or recorded data.

[0203] : Detection time deviation period difference, , the time difference between the detection time and the application time, is used to describe the periodic fluctuation characteristics.

[0204] : Periodic fluctuation influence factor, set according to experimental or empirical values, with a value range of 0<η≤1, describing the additional impact of the periodic fluctuation characteristics of the detection time on the time weight. For example, periodic degradation peaks or residual fluctuations may occur after pesticide application.

[0205] : Periodic fluctuation period of pesticide residues, obtained through experiments or model fitting, in hours or days. For example, some pesticides may show 24-hour periodic changes under the influence of the temperature difference between day and night.

[0206] Time deviation influence items:

[0207]

[0208] The core part is , represents the absolute deviation between the detection time and the administration time.

[0209] The smaller the deviation, the closer the detection time is to the administration time and the higher the weight.

[0210] pass (Pesticide degradation characteristic time constant) adjustment to ensure that the formula conforms to the degradation law of pesticides.

[0211] The role of exponential function:

[0212] use Describe the nonlinear effects of time deviation.

[0213] When the detection time is close to the administration time, the exponential term tends to 0 and the weight tends to the maximum value. .

[0214] When the detection time is far away from the administration time, the exponential term tends to infinity and the weight tends to zero.

[0215] Periodic fluctuation term:

[0216]

[0217] Describe the periodic residual fluctuations using The function describes the periodic changes of pesticide residues (such as the influence of temperature difference between day and night).

[0218] Controls the length of the cycle. For example, a 24-hour cycle represents a day-night cycle.

[0219] Controls the fluctuation amplitude. The larger the value, the more significant the impact of periodic fluctuations on the weight.

[0220] The periodic fluctuation term introduces the periodic fluctuation characteristics, so that the time weight not only depends on the time deviation, but also reflects the impact of periodic changes.

[0221] When the application time is approaching, , the exponential term tends to 0, the weight .

[0222] When away from the application time, that is, when , the exponential term tends to infinity, the weight .

[0223] when When it is a multiple of the period (such as ), The term reaches its maximum value and the weight is periodically enhanced.

[0224] when When the deviation is a multiple of the period, The term decreases and the weight is weakened periodically.

[0225] when When increases, the impact of time deviation on the weight becomes more sensitive, and the weight changes quickly.

[0226] when When it decreases, the influence of time deviation on weight is weakened and the weight changes smoothly.

[0227] This formula dynamically describes the impact of time factors on pesticide residues: the weights are dynamically adjusted through time deviation (the distance between detection time and application time) and pesticide degradation characteristics (time constant) to accurately reflect the impact of time on pesticide residues. Taking into account periodic characteristics: pesticide residues are affected by periodic fluctuations in the environment (such as the temperature difference between day and night), and the formula introduces this characteristic through the cosine function to make the weight more realistic. Wide applicability: The formula parameters are flexible and can be adjusted according to the degradation characteristics and environmental conditions of different pesticides, and are applicable to a variety of pesticides and application scenarios.

[0228] Step S13344, obtain the regional pesticide application frequency, and calculate the weight of the regional relationship edge based on the regional pesticide application frequency.

[0229] Specifically, step S1333 adds relationship edges between nodes of the pesticide residue toxicity knowledge graph to establish semantic connections between nodes. A detection relationship edge is established between the pesticide node and the test sample node, indicating that the pesticide residue is detected in the test sample. An environmental impact relationship edge is established between the test sample node and the environment node, indicating the impact of the environmental conditions of the test sample on the pesticide residue level. A time relationship edge is established between the time node and the test sample node, indicating the time point when the detection behavior occurs. A regional relationship edge is established between the regional node and the test sample node, indicating the geographical area from which the test sample is collected.

[0230] By adding multi-dimensional relationship edges, the pesticide residue toxicity knowledge graph can fully characterize the semantic information of pesticide residue detection data and reveal the intrinsic connection between pesticide residues and samples, environment, time, and region. For example, through the detection relationship edge, you can easily query the residue status of a certain pesticide in different agricultural products; through the environmental impact relationship edge, you can analyze the impact of factors such as temperature and humidity on the behavior of pesticide residues; through the time relationship edge, you can track the dynamic trend of pesticide residue levels; through the regional relationship edge, you can achieve a visual presentation of regional pesticide residue conditions. Rich semantic associations help to deeply understand the laws of pesticide residue behavior and provide more comprehensive and accurate decision-making support for pesticide residue supervision.

[0231] In step S1334, weights are assigned to the relationship edges between the nodes of the pesticide residue toxicity knowledge graph to quantitatively characterize the strength of the node association. The ratio of the second pesticide residue concentration to the pesticide residue threshold is used as the weight of the detection relationship edge to intuitively reflect the severity of the excessive pesticide residue in the sample. For example, if the concentration of aldicarb in a sample is 0.8 mg / kg, and the residue threshold of aldicarb is 0.5 mg / kg, the weight of the detection relationship edge is 0.8 / 0.5=1.6, indicating that the aldicarb residue in the sample exceeds the standard by 60%. The larger the weight, the more serious the excessive pesticide residue and the higher the food safety risk.

[0232] Obtain the frequency of regional pesticide application, and calculate the weight of the regional relationship edge based on the frequency of regional pesticide application. The application frequency (number of applications per mu per year) can be normalized to the range of 0-1 as the weight. The higher the application frequency, the greater the weight, indicating that the intensity of pesticide use in the region is high and the risk of pesticide residues is high. The environmental sensitivity of the region can also be considered, and the weights of ecologically fragile areas (such as wetlands and water source protection areas) can be added to highlight the severity of their pesticide residue pollution. Regional weights can reflect the differences in pesticide residue risks in different regions and achieve focused supervision of key areas and high-risk areas.

[0233] Step S2000, compare the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk levels of n2 pesticide types; combine the first region information and the first time information of the first pesticide residue detection data to obtain the second pesticide residue detection data of the same region and the same period in the blockchain; perform cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the region; based on the risk benchmark, dynamically calibrate the original risk level to generate a calibrated risk level;

[0234] Furthermore, step S2000 includes:

[0235] Step S2100, comparing the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk levels of n2 pesticide types; the original risk levels include four levels: safe, slightly exceeded, moderately exceeded, and severely exceeded;

[0236] Furthermore, if Figure 8 As shown, step S2100 includes:

[0237] Step S2110, comparing the first pesticide residue concentration with the pesticide residue threshold of the corresponding pesticide node in the pesticide residue toxicity knowledge graph, and calculating a first excess concentration ratio; the first excess concentration ratio is equal to the first pesticide residue concentration divided by the pesticide residue threshold;

[0238] Step S2120, calculating the first risk index according to the toxicity index and hazard index of the corresponding pesticide node in the pesticide residue toxicity knowledge graph and the first excessive concentration ratio ;

[0239] Step S2130: The first risk index The original risk levels of the n2 pesticide types are compared with the preset risk level thresholds and divided into original risk levels.

[0240] Four risk level thresholds are preset: safety boundary value , slightly exceeding the boundary value , moderately exceeded the limit value and severely exceeded boundary values ;

[0241] when When , the original risk level is safe;

[0242] when When the original risk level is slightly exceeded;

[0243] when When , the original risk level is moderately exceeded;

[0244] when When the original risk level is severely exceeded.

[0245] Specifically, the core of step S2100 is to calculate the first risk index as a quantitative basis for dividing risk levels. The first risk index comprehensively considers three key factors:

[0246] First excess concentration ratio: reflects the relative degree of excess of pesticide residue concentration. The first excess concentration ratio is equal to the first pesticide residue concentration divided by the pesticide residue threshold, which directly represents the excess multiple. For example, if the residue concentration of a certain pesticide in vegetables is 0.2 mg / kg, and the corresponding MRL is 0.1 mg / kg, then the first excess concentration ratio is 2, indicating that the pesticide residue exceeds the standard by 2 times.

[0247] Toxicity index: reflects the toxicological characteristics of the pesticide itself. The toxicity index is obtained by querying the authoritative toxicology database. Common indicators include acute reference dose (ARfD), acceptable daily intake (ADI), etc. For example, for highly toxic pesticides, the ARfD value is low, indicating that it is more harmful to the human body and the risk is higher when the pesticide residue exceeds the standard.

[0248] Hazard index: reflects the potential harmful consequences of excessive pesticide residues. Hazard indexes are obtained by analyzing historical poisoning cases caused by pesticide residues. For example, if excessive pesticide residues have caused a large number of consumers to be poisoned and sent to the hospital, the hazard index should be set to a higher value to highlight the risk.

[0249] When calculating the first risk index, different weights are given to the three factors to weigh the impact of the degree of excess, toxicity and hazard on the risk. For example, assuming that the weight of the first excess concentration ratio is 0.5, the weight of the toxicity index is 0.3, the weight of the hazard index is 0.2, and the three indicators of a certain pesticide residue are 2, 0.8, and 0.6 respectively, then the first risk index =2×0.5+0.8×0.3+0.6×0.2=1.36.

[0250] Finally, the first risk index is compared with the preset risk level threshold to determine the original risk level of the pesticide residue. The risk level threshold is a critical value obtained based on expert experience and historical data statistics and can be adjusted dynamically. For example, it can be set:

[0251] when When ≤1, the original risk level is safe;

[0252] When 1< When ≤2, the original risk level is slightly exceeded;

[0253] When 2< When ≤3, the original risk level is moderately exceeded;

[0254] when When >3, the original risk level is severely exceeded.

[0255] Taking the above example, the first risk index is 1.36, which is between 1 and 2, so the original risk level of the pesticide residue is classified as slightly exceeding the standard.

[0256] Step S2100 achieves the standardization of pesticide residue risk level classification by quantitatively calculating the first risk index and comparing it with a unified threshold. Compared with relying solely on detection values ​​or manual experience judgment, this method is more objective and consistent, and reduces judgment bias. Regulatory authorities can quickly screen out high-risk agricultural products based on risk levels and conduct targeted supervision and sampling. Agricultural product producers and operators can also optimize pesticide use plans and control pesticide residue risks accordingly. Consumers can use risk level labels to independently choose safe agricultural products.

[0257] In summary, step S2100 constructs a rapid risk assessment model for pesticide residues by comparing the correlation between pesticide residue data and toxicity knowledge, providing scientific and quantitative technical support for the quality and safety management of agricultural products. This method makes full use of the massive data generated by the Internet of Things pesticide residue detection device and combines it with pesticide toxicology knowledge to form an objective, dynamic and comprehensive risk analysis capability, which has good promotion and application value.

[0258] Step S2200, combining the first region information and the first time information of the first pesticide residue detection data, obtaining the second pesticide residue detection data of the same region and time period in the blockchain;

[0259] Specifically, the purpose of step S2200 is to retrieve historical test data related to the current test data from the blockchain to provide a reference for subsequent risk assessment. The first area information includes the geographical location coordinates of the test sample collection, the name of the area, the area type and environmental characteristics, etc. The first time information is the specific time point when the test operation is performed, usually accurate to seconds. Using this information, other test data in the same area and similar time can be queried in the blockchain to form a targeted data set.

[0260] For example, if the current test sample is collected from a farmers' market in Shanghai and the test time is 10:30 on December 4, 2024, you can search the blockchain for all test records of the farmers' market and its surrounding areas in early December 2024, and extract key information such as pesticide residue concentration, test sample type, environmental factors, etc. as the second pesticide residue test data.

[0261] By obtaining historical data from the same region and time period, we can have a more comprehensive understanding of the overall situation of pesticide residues in the region and determine whether the current test results are in line with the overall regional trend. This regional and time period data combination can reduce the randomness of individual cases and improve the reliability of risk assessment.

[0262] In summary, step S2200 provides regionalized and time-series data support for pesticide residue risk assessment by making full use of historical data stored in blockchain, thereby improving the pertinence and accuracy of the assessment results. This step is an important data basis for risk assessment and lays the foundation for subsequent cluster analysis and threshold determination.

[0263] Step S2300, performing cluster analysis on the second pesticide residue detection data to determine the risk benchmarks of n2 pesticide types in the area;

[0264] Furthermore, if Fig. 9 As shown, step S2300 includes:

[0265] Step S2310, classifying the second pesticide residue detection data according to the second pesticide type to generate n2 second pesticide residue data sets;

[0266] Step S2320, clustering each second pesticide residue data set, dividing the residue concentration into three risk clusters of high, medium, and low;

[0267] Step S2330, calculating the means of the three risk clusters of high, medium and low, and generating a second pesticide residue concentration threshold; the second pesticide residue concentration threshold includes a high risk concentration threshold, a medium risk concentration threshold and a low risk concentration threshold;

[0268] Further, step S2330 includes:

[0269] Step S2331, extracting the mean residual concentration of the high-risk cluster as the high-risk concentration threshold;

[0270] Step S2332, extracting the mean value of the residual concentration of the medium risk cluster as the medium risk concentration threshold;

[0271] Step S2333, extracting the mean residual concentration of the low-risk cluster as the low-risk concentration threshold.

[0272] Step S2340: Using the second pesticide residue concentration threshold as the risk benchmark for n2 pesticide types in the area.

[0273] Specifically, step S2300 uses a clustering algorithm to perform risk classification on the second pesticide residue detection data, automatically discovers data clusters with different residue concentration levels, and uses this to determine the risk baseline value of each type of pesticide in the area. Cluster analysis is an unsupervised learning method that can divide a data set into multiple subsets that are similar internally and significantly different externally, and is suitable for automatic classification of risk levels.

[0274] In this step, the test data are first grouped according to the type of pesticide to obtain n2 pesticide residue concentration data sets. Then, each data set is clustered independently and divided into three risk clusters: high, medium, and low. The K-means algorithm is used here to achieve clustering by minimizing the sum of the squares of the distances between the samples in the cluster and the cluster center.

[0275] The basic steps of the K-means algorithm are as follows:

[0276] 1. Randomly select 3 samples as the initial cluster centers.

[0277] 2. Calculate the Euclidean distance of each sample to the three cluster centers and divide the samples into the clusters with the closest distance.

[0278] 3. Recalculate the coordinates of the center point of each cluster (i.e., the mean residual concentration of all samples in the cluster).

[0279] 4. Repeat steps 2 and 3 above until the cluster center no longer changes or the maximum number of iterations is reached.

[0280] Through clustering, the natural distribution pattern of pesticide residue concentration can be discovered and the risk level can be automatically quantified. This data-driven classification method can objectively reflect the distribution characteristics of pesticide residues in the area and determine a realistic risk benchmark.

[0281] After completing the risk cluster division, step S2330 further quantifies the criteria for determining each risk level. The means of the high, medium, and low risk clusters correspond to the typical residual concentration values ​​of the high, medium, and low risk levels, respectively, and can be used as quantitative thresholds for dividing risk levels.

[0282] The second pesticide residue concentration thresholds include:

[0283] High-risk concentration threshold: The mean residue concentration of the high-risk cluster. Residue concentrations exceeding this threshold are considered high risk and may cause significant harm to human health and require focused supervision.

[0284] Medium risk concentration threshold: The mean residue concentration of the medium risk cluster. Residue concentrations between the high risk concentration threshold and the low risk concentration threshold are considered medium risk and require close attention and appropriate control measures.

[0285] Low-risk concentration threshold: The mean residual concentration of the low-risk cluster. Residual concentrations below this threshold are considered low risk, with little impact on human health, and management can be moderately relaxed.

[0286] Compared with the traditional unified threshold, this adaptive threshold determination method can fully consider the distribution differences of different pesticide residues in different regions. For example, in areas where pesticides are used less, the residue concentration is generally low, and the low-risk threshold may be far below the national standard; while in areas where pesticides are used frequently, even if they meet the national standard, the residue level may be higher than other areas, and a stricter threshold needs to be set. The data-driven threshold generation method can automatically adapt to such differences and formulate more refined and locally adapted risk management strategies.

[0287] In general, step S2300 applies cluster analysis method to automatically discover risk clusters of pesticide residues in a data-driven manner and generate objective and quantitative risk benchmarks. By calculating the concentration mean of different risk clusters, the residue thresholds adapted to local conditions can be obtained, providing a reference for subsequent risk judgment and early warning. This adaptive risk grading method can significantly improve the scientificity and effectiveness of pesticide residue supervision.

[0288] The risk benchmark is a risk classification standard for the residual concentration level of a specific type of pesticide in a certain area. It serves as a reference point for dynamic risk assessment and is used to determine the risk level of the current test data. The high, medium and low concentration thresholds of each pesticide are used as the risk benchmark for the type of pesticide in the area for dynamic calibration and risk assessment, which helps to carry out precise management and supervision by region and pesticide type.

[0289] Step S2400, dynamically calibrating the original risk level in step S2100 based on the risk benchmark to generate a calibrated risk level;

[0290] Further, step S2400 includes:

[0291] Step S2410: Obtain a second pesticide residue concentration threshold value of the same type as the first pesticide, and generate a first risk level adjustment coefficient based on a comparison between the first pesticide residue concentration and the second pesticide residue concentration threshold value. ;

[0292] Further, step S2410 includes:

[0293] Step S2411, when the first pesticide residue concentration Less than low risk concentration threshold When the first risk level adjustment coefficient The value is 1;

[0294] Step S2412, when the first pesticide residue concentration Greater than high risk concentration threshold When the first risk level adjustment coefficient The value is -1;

[0295] Step S2413, when the first pesticide residue concentration Greater than or equal to the low risk concentration threshold And less than or equal to the high risk concentration threshold When the first risk level adjustment coefficient The value is 0.

[0296] Step S2420, obtaining the first area type in the first area information, and obtaining the risk level distribution probability of the area type from the pesticide residue toxicity knowledge graph ,in Safe, Mild, Moderate, Severe , generate the second risk level adjustment coefficient ;

[0297] Further, step S2420 includes:

[0298] Step S2421, if the original risk level is safe, then ;in, It indicates the probability that the pesticide residue test result is at a safe level under this area type; Indicates the probability that the pesticide residue test result is slightly above the standard level; It indicates the probability that the pesticide residue test result is moderately exceeded the standard level; Indicates the probability that the pesticide residue test result is a severe excess level;

[0299] Step S2422: If the original risk level is slightly exceeded, ;

[0300] Step S2423: If the original risk level is moderately exceeded, ;

[0301] Step S2424: If the original risk level is severely exceeded, .

[0302] Step S2430: Combine the first risk level adjustment coefficient and the second risk level adjustment factor , the original risk level is adjusted to generate a calibrated risk level.

[0303] The adjusting of the original risk level to generate a calibrated risk level includes:

[0304] Assume the original risk level is , the risk level after calibration is , the risk level adjustment step is ;

[0305] Calculate the total adjustment factor ;

[0306] when hour, ;

[0307] when hour, ;

[0308] when hour, .

[0309] Specifically, step S2400 dynamically calibrates the original risk level in step S2100 based on the risk benchmark to generate a calibrated risk level. This step makes full use of the historical test data accumulated on the blockchain, combines the pesticide residue concentration and regional characteristics of the current test sample, and adaptively adjusts the original risk level to improve the accuracy and practicality of the risk assessment results.

[0310] Specifically, step S2410 obtains the pesticide residue concentration threshold value of the same type of pesticide as the sample to be evaluated, takes it as a reference benchmark, and generates the first risk level adjustment coefficient based on the comparison of the actual detected pesticide residue concentration with the threshold value. The pesticide residue concentration threshold here refers to the maximum residue standard allowed for pesticides in agricultural products. For example, for paraquat pesticides in vegetables, the maximum residue limit (MRL) is 0.05 mg / kg. Assuming that the paraquat residue concentration detected in a batch of spinach is 0.03 mg / kg, which is lower than the MRL, the first risk level adjustment coefficient is 1, indicating a tendency to lower the risk level. On the contrary, if the detected concentration is 0.08 mg / kg, which is higher than the MRL, the first risk level adjustment coefficient is -1, indicating a tendency to increase the risk level. By introducing the pesticide residue threshold value as a comparison benchmark, the risk level of the residue concentration can be objectively reflected, providing a quantitative basis for risk level adjustment.

[0311] Step S2420 queries the historical risk distribution data of the same type as the sample area to be evaluated from the pesticide residue toxicity knowledge graph, and generates a second risk level adjustment coefficient based on Bayesian inference. Area type refers to the category of place to which the sample collection area belongs, such as farmland, orchard, vegetable greenhouse, etc. The pesticide residue risk level of an area depends not only on the result of a single test, but also on the historical risk probability distribution of the area. The risk level distribution probability of an area type refers to the probability obtained by statistically analyzing the risk level distribution of pesticide residues in a specific area type (such as farmland, orchard, market, etc.) based on historical data and knowledge graphs. For example, if the knowledge graph shows that in the historical test data of a certain farmland area, 80% of the samples have a safe risk level, 15% are slightly exceeded, 5% are moderately exceeded, and there are no severely exceeded samples, it can be inferred that the overall risk of the area is low. When the original risk of a new test sample is judged to be moderately exceeded, the second adjustment coefficient =5%-0%=5%>0, indicating a tendency to lower the risk level. This prior probability inference based on the knowledge graph can fully tap into regional background information and make the risk assessment results more comprehensive and objective.

[0312] Specifically, if the original risk level is safe, the second risk level adjustment coefficient It is equal to the probability of the safety level under the area type minus the sum of the probabilities of the three levels of slight excess, moderate excess and severe excess. The meaning of this calculation formula is that if the probability of the safety level under the area type is greater and the probability of the other three excess levels is smaller, the second risk level adjustment coefficient The larger the risk, the lower the overall risk of pesticide residues in this area type, and the possibility of appropriately raising the safety level. On the contrary, if the probability of exceeding the standard level in this area type is high, then The smaller it is, the higher the risk of pesticide residues in this area type, and the possibility of lowering the safety level may be required.

[0313] If the original risk level is slightly exceeded, the second risk level adjustment coefficient It is equal to the probability of slight exceeding the standard in this area type minus the sum of the probabilities of moderate exceeding the standard and severe exceeding the standard. It means that if the probability of slight exceeding the standard in this area type is greater, and the probability of moderate exceeding the standard and severe exceeding the standard is smaller, then The greater the risk, the more likely it is that the pesticide residue risk in this area type is mainly slightly exceeded, and the possibility of slightly exceeded level can be appropriately increased. On the contrary, if the probability of moderate and severe exceeded level in this area type is high, then The smaller it is, the higher the risk of pesticide residues in this area type, and the possibility of lowering the level of slight exceeding the standard needs to be lowered.

[0314] If the original risk level is moderately exceeded, the second risk level adjustment coefficient It is equal to the probability of moderate exceeding the standard level minus the probability of severe exceeding the standard level under the area type; its meaning is that if the probability of moderate exceeding the standard under the area type is greater and the probability of severe exceeding the standard is smaller, then The larger the value, the higher the pesticide residue risk in this area type is, and the probability of moderate excess can be appropriately adjusted. The smaller it is, the more serious the risk of pesticide residues in this area type is, and the possibility of lowering the moderate exceedance level needs to be achieved.

[0315] If the original risk level is severely exceeded, the second risk level adjustment coefficient is equal to the probability of severely exceeded level under the area type; its meaning is that if the probability of severely exceeded level under the area type is greater, The larger it is, the more serious the pesticide residue risk is for this area type, and there is a possibility that the severe exceeding standard level needs to be maintained or increased.

[0316] Step S2430 comprehensively considers the first risk level adjustment coefficient and the second risk level adjustment coefficient, adjusts the original risk level, and obtains the final calibrated risk level. The core idea of ​​the adjustment is: when the sum of the two adjustment coefficients is greater than or equal to 1, the risk level is adjusted down. When the sum of the two adjustment coefficients is less than or equal to -1, the risk level is increased. When the sum of the adjustment coefficients is between -1 and 1, the original risk level remains unchanged. Usually 1 is taken, that is, each adjustment can only increase or decrease one level at most. This adjustment method makes appropriate corrections to the original risk level based on the reference residual concentration and regional characteristics, taking into account both the degree of excess of the test sample itself and the overall risk level of the region, thus avoiding the risk assessment results being too absolute.

[0317] The introduction of a dynamic calibration mechanism will help improve the scientificity and rationality of pesticide residue risk assessment. On the one hand, by comparing the pesticide residue concentration thresholds, the impact of the quantified residue level on the risk level can be clearly defined, objectively reflecting the sample's own risk of exceeding the standard; on the other hand, by using the knowledge graph to perform prior statistics on the regional risk probability distribution, the regional background information can be fully absorbed to balance the one-sidedness of individual test results. The risk level after dynamic calibration can be closer to the actual risk status of the test sample, providing a reliable reference for the decision-making of the regulatory authorities. For example, if the original risk level of a batch of spinach is slightly exceeded, but it is adjusted to a safe level after calibration, the regulatory authorities can consider releasing it for sale to avoid unnecessary losses; on the contrary, if the calibration raises the risk level from moderate to severe, measures such as removal from shelves and recalls should be taken immediately to prevent the risk from expanding.

[0318] In summary, the dynamic calibration mechanism introduced in step S2400 uses the risk benchmark as a reference to flexibly adjust the original risk level, which improves the objectivity of risk assessment while also enhancing the practical value of the assessment results. By quantitatively analyzing the differences in pesticide residue concentrations and regional risk distribution characteristics, the actual risk level of the test samples is dynamically reflected, providing a scientific basis for precise policy implementation. This dynamic calibration method based on knowledge graphs and online learning is helpful for upgrading and optimizing pesticide residue detection and food safety governance systems.

[0319] Step S2500: associate the first pesticide residue detection data, the original risk level, and the calibrated risk level, upload them to the blockchain for storage, and continuously update the pesticide residue toxicity knowledge graph.

[0320] Specifically, the purpose of step S2500 is to incorporate the latest pesticide residue test results and risk assessment results into the blockchain and knowledge graph to achieve data accumulation and iterative upgrade of knowledge. Associating the first pesticide residue test data, the original risk level and the calibrated risk level means organizing these three types of information into a complete data structure or record to clearly express the logical relationship between them. By uploading the associated data to the blockchain for storage, the authenticity, integrity and traceability of the pesticide residue test data can be ensured by leveraging the decentralized and tamper-proof characteristics of the blockchain. This is crucial for pesticide residue supervision and food safety management. Through smart contracts, data uploading can be automatically triggered and the legitimacy of the uploading data can be verified. Continuously updating the pesticide residue toxicity knowledge graph means expanding and updating the nodes, edges, attributes, etc. in the knowledge graph based on the latest pesticide residue test data uploaded to the chain. For example, new test sample nodes, regional nodes and timestamp nodes can be added, the risk level distribution probability of pesticide nodes can be updated, and the association strength between different nodes can be adjusted. Through continuous graph updates, the knowledge graph can always reflect the latest status and patterns of pesticide residues, providing dynamic knowledge support for risk assessment.

[0321] Step S2500 can achieve closed-loop management of the entire process of pesticide residue data and improve data value density through blockchain storage and knowledge graph update. On the one hand, blockchain storage makes the process of data uploading, verification, and storage transparent and reliable, providing a reliable data infrastructure for regulatory authorities, production companies, consumers, etc. On the other hand, knowledge graph updates allow the internal connections and laws of data to be continuously mined and strengthened, which helps to scientifically assess pesticide residue risks and accurately formulate control measures.

[0322] Step S3000, set up a smart contract on the blockchain and automatically trigger the corresponding execution strategy according to the calibrated risk level.

[0323] Furthermore, step S3000 includes:

[0324] Step S3100: When the calibrated risk level is safe, the smart contract automatically sends a safety notification to the producer and the regulator;

[0325] Step S3200: When the risk level after calibration is slightly exceeded or moderately exceeded, the smart contract automatically sends a rectification notice to the producer and sets a rectification deadline; if the rectification is not completed within the rectification deadline, an early warning is automatically sent to the regulator;

[0326] Step S3300: When the risk level after calibration is severely exceeded, the smart contract automatically sends a recall notice to the manufacturer, sends an early warning to the regulator, and publishes a severe excess notice on the blockchain.

[0327] Step S3400, generate a test report, the content of which includes test results, risk level, traceability information, etc., and is securely accessed through the blockchain interface.

[0328] Specifically, smart contracts are an automatically executed program deployed on the blockchain that can automatically execute corresponding operations according to preset conditions without human intervention. When the detected pesticide residue risk level changes, smart contracts can automatically trigger corresponding execution strategies, such as sending notifications to relevant parties and issuing announcements, greatly improving the efficiency and reliability of risk management.

[0329] Step S3100 automatically sends safety notifications to producers and regulators through smart contracts based on the safety level. Producers refer to producers or suppliers of agricultural products, and regulators refer to government departments responsible for the supervision and management of agricultural product quality and safety. When the test results show that the risk level of pesticide residues is safe, it is very important to notify producers and regulators in a timely manner. On the one hand, producers need to know the safety status of their agricultural products in order to strengthen quality control; on the other hand, regulators need to understand the safety status of agricultural products in their jurisdiction in order to carry out targeted work. Traditional notifications may have lags and omission risks. Automatically sending notifications through smart contracts can achieve timely, accurate, and non-omission information transmission, helping to ensure the quality of agricultural products. For example, if the vegetables grown by a vegetable farmer are tested for pesticide residue safety, the smart contract of the blockchain system will automatically send a notification to the vegetable farmer and the local Food and Drug Administration to inform the safety status of the batch of vegetables. The vegetable farmer can sell them with peace of mind, and the regulatory bureau can also optimize the sampling plan accordingly.

[0330] Step S3200: For the situation of slight or moderate excess, a rectification notice is automatically sent to the producer through the smart contract, and a rectification deadline is set; if the rectification is not completed within the rectification deadline, an early warning is automatically sent to the regulator. The purpose of this measure is to urge the producer to take rectification measures as soon as possible to control the risk of pesticide residues. Although slight and moderate excesses have not reached the high-risk level of severe excesses, they have exceeded the pesticide residue limit specified in the national standard and need to be given sufficient attention. Automatically sending rectification notices and setting rectification deadlines through blockchain smart contracts can urge producers to attach great importance to the problem and complete the rectification within the specified time. If the producer fails to complete the rectification within the rectification deadline, the smart contract will automatically send an early warning to the regulator, reminding the regulator to strengthen supervision and urge rectification. This automatic early warning mechanism can achieve a benign interaction between regulators and producers and form a pattern of co-governance of agricultural product quality and safety. For example, if the strawberries on a farm are found to have a slightly excessive level of a certain pesticide residue, the smart contract will automatically send a rectification notice to the farm, requiring it to complete the rectification within one week; if the farm fails to complete the rectification after one week, the smart contract will automatically send an early warning to the local Agriculture and Rural Affairs Bureau, which can then call in the farm manager for a talk and urge him to make the rectification as soon as possible.

[0331] Step S3300: For high-risk situations with severe excess levels, three measures are automatically taken through smart contracts: sending a recall notice to the producer, sending an early warning to the regulator, and publishing a severe excess notice on the blockchain. Since agricultural products with severe excesses have seriously threatened the health of consumers, they must be blocked from entering the market. The smart contract automatically sends a recall notice to the producer, requiring it to immediately initiate the recall procedure and stop the sale and circulation of problematic agricultural products. At the same time, it automatically sends an early warning to the regulator, prompting the regulator to pay close attention and take control measures. The purpose of publishing a severe excess notice on the blockchain is to inform all relevant parties of the risk information and prevent problematic agricultural products from entering the market. Blockchain is difficult to tamper with and traceable. Publishing an announcement on the blockchain can ensure the authenticity and credibility of the announcement information. For example, if the imported apples in a supermarket are found to have seriously exceeded the standard for aldicarb residues, reaching the level of severe excess, the smart contract will immediately trigger three measures: send a recall notice to the apple supplier, requiring it to immediately recall the problematic apples; send an early warning to the local market supervision and administration bureau, prompting regulatory intervention; and publish a severe excess announcement on the blockchain to remind other relevant parties to stop selling the batch of problematic apples.

[0332] Step S3400 provides for generating a test report, the contents of which include test results, risk level, traceability information, etc., and are securely accessed through the blockchain interface. The test report is the final result of the pesticide residue detection activity, and needs to fully and accurately reflect the entire process and results of the test. Including test results, risk level, traceability information, etc. in the report can provide data support for subsequent risk disposal, quality management, regulatory decisions, etc. The test report is securely accessed through the blockchain interface, utilizing the security characteristics of the blockchain. Blockchain has the characteristics of being difficult to tamper with, highly available, and traceable. Placing the test report in the blockchain system and through strict authority management can ensure the authenticity and security of the report and prevent data from being maliciously tampered with. For example, a third-party testing agency conducted a pesticide residue test on a batch of eggs and formed a test report; the report content includes sample information, test items and results, risk level (safety), test time and location, sample traceability information, etc.; the report data is written into the blockchain, and the public can query the report hash value through the blockchain browser to verify the authenticity of the report; the authorized quality supervision department can obtain the complete report content through a dedicated on-chain interface for quality supervision.

[0333] In summary, step S3000 uses blockchain smart contracts to automatically trigger corresponding measures according to the risk level of pesticide residues, which is an efficient, reliable and transparent pesticide residue management solution. It can take targeted response measures in a timely manner when there is a risk in the quality of agricultural products, control risks to the maximum extent, ensure the quality and safety of agricultural products, and safeguard the rights and interests of consumers. Through the joint participation of producers, regulators, the public and other parties, a pattern of co-governance of agricultural product quality and safety is formed, which is conducive to the green development of agriculture and rural revitalization. This solution makes full use of the unique advantages of blockchain technology and provides new ideas for solving the problem of pesticide residues. With the development of digital agriculture, emerging information technologies such as blockchain will play an increasingly important role in the governance of agricultural product quality and safety.

[0334] Example 2

[0335] This embodiment provides a blockchain-based management system for all-in-one fluorescent quantitative test strips based on embodiment 1. Fig.10 As shown, including:

[0336] Data acquisition module: used to obtain the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip;

[0337] Data storage module: used to upload the first pesticide residue detection data to the blockchain for storage;

[0338] Knowledge graph construction module: used to construct the knowledge graph of pesticide residue toxicity;

[0339] Risk level classification module: used to compare the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk level of n2 pesticide types;

[0340] Calibration module: used to combine the first area information and the first time information of the first pesticide residue detection data to obtain the second pesticide residue detection data of the same area and time period in the blockchain; perform cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the area; based on the risk benchmark, dynamically calibrate the original risk level to generate a calibrated risk level;

[0341] Execution module: used to set up smart contracts on the blockchain and automatically trigger corresponding execution strategies based on the calibrated risk level.

[0342] In the data acquisition module, the first pesticide residue detection data includes a first detection sample type, a first pesticide type, a first pesticide residue concentration, a first environmental factor, a first area information and a first time information; the first area information includes a first geographic coordinate, a first area name, a first area type and a first environmental feature; the first time information includes a first timestamp; the first pesticide type includes n2 pesticide types;

[0343] The step of obtaining the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip comprises:

[0344] Step S1110, collecting a sample to be tested, using an all-in-one fluorescent quantitative test strip to test it, and obtaining a first pesticide residue concentration;

[0345] Step S1120, recording the type information of the sample to be tested as the first test sample type, and taking the type of pesticide residue to be tested as the first pesticide type;

[0346] Step S1130, recording the geographical location information of the sample to be detected as the first regional information; recording the detection time as the first time information;

[0347] Step S1140, collecting environmental parameters during detection as the first environmental factor.

[0348] In the data storage module, uploading the first pesticide residue detection data to the blockchain for storage includes:

[0349] Step S1210, encoding the first pesticide residue detection data according to a predetermined data format, generating transaction data on the blockchain, and digitally signing the transaction data;

[0350] Step S1220, other blockchain nodes verify the digital signature of the transaction data;

[0351] Step S1230, through the consensus algorithm, the blockchain network reaches a consensus on the transaction data and packages it into a new block;

[0352] In step S1240, the new block is linked to the existing blockchain.

[0353] In the knowledge graph construction module, the construction of the pesticide residue toxicity knowledge graph includes:

[0354] Step S1310, obtaining the pesticide residue historical transaction data on the blockchain, extracting the first key indicator of pesticide residue from the pesticide residue historical transaction data, wherein the first key indicator of pesticide residue includes a second pesticide type, a second test sample type, a second pesticide residue concentration, a second environmental factor, second regional information, and second time information; the second regional information includes a second geographic coordinate, a second regional name, a second regional type, and a second environmental feature; the second time information includes a second timestamp; and the second pesticide type includes n2 pesticide types;

[0355] Step S1320, obtaining a second key indicator of pesticide residue, wherein the second key indicator of pesticide residue includes a pesticide residue threshold, a toxicity index, a hazard index, and a risk level label;

[0356] Step S1330, constructing a pesticide residue toxicity knowledge graph based on the first key indicator of pesticide residue and the second key indicator of pesticide residue.

[0357] The step S1330 includes:

[0358] Step S1331, creating nodes of a pesticide residue toxicity knowledge graph according to the first key indicator of pesticide residue, wherein the nodes of the pesticide residue toxicity knowledge graph include pesticide nodes, test sample nodes, environment nodes, time nodes and region nodes;

[0359] Step S1332, expanding the attributes of the nodes of the pesticide residue toxicity knowledge graph according to the first key indicator of pesticide residue and the second key indicator of pesticide residue;

[0360] Step S1333, adding relationship edges between nodes of the pesticide residue toxicity knowledge graph;

[0361] Step S1334, assigning weights to the relationship edges between the nodes of the pesticide residue toxicity knowledge graph.

[0362] The step S1331 includes:

[0363] Step S13311, creating a pesticide node according to the second pesticide type;

[0364] Step S13312, creating a detection sample node according to the second detection sample type;

[0365] Step S13313, creating an environment node according to the second environment factor;

[0366] Step S13314, creating a time node according to the second time information;

[0367] Step S13315: Create a regional node according to the second regional information.

[0368] The step S1332 includes:

[0369] Step S13321, expanding the attributes of the pesticide node according to the pesticide residue threshold, toxicity index, hazard index and risk level label;

[0370] Step S13322, expanding the attributes of the detection sample node according to the second pesticide residue concentration;

[0371] Step S13323, expanding the attributes of the time node according to the second timestamp;

[0372] Step S13324, expanding the attributes of the area node according to the second geographic coordinates, the second area name, the second area type and the second environmental characteristics.

[0373] The step S1333 includes:

[0374] Step S13331, establishing a detection relationship edge between the pesticide node and the detection sample node;

[0375] Step S13332, establishing an environmental impact relationship edge between the detection sample node and the environment node;

[0376] Step S13333, establishing a time relationship edge between the time node and the detection sample node;

[0377] Step S13334, establishing a regional relationship edge between the regional node and the detection sample node.

[0378] The step S1334 includes:

[0379] Step S13341, using the ratio of the second pesticide residue concentration to the pesticide residue threshold as the weight of the detection relationship edge;

[0380] Step S13342, calculating the weight of the environmental impact relationship edge according to the second environmental factor and the second pesticide residue concentration;

[0381] Step S13343, obtaining the pesticide application time, and calculating the weight of the time relationship edge according to the second timestamp of the time node and the pesticide application time;

[0382] Step S13344, obtain the regional pesticide application frequency, and calculate the weight of the regional relationship edge based on the regional pesticide application frequency.

[0383] In the risk level classification module, the original risk level includes four levels: safe, slightly exceeded, moderately exceeded, and severely exceeded;

[0384] The original risk levels for determining n2 pesticide types include:

[0385] Step S2110, comparing the first pesticide residue concentration with the pesticide residue threshold of the corresponding pesticide node in the pesticide residue toxicity knowledge graph, and calculating a first excess concentration ratio; the first excess concentration ratio is equal to the first pesticide residue concentration divided by the pesticide residue threshold;

[0386] Step S2120, calculating the first risk index according to the toxicity index and hazard index of the corresponding pesticide node in the pesticide residue toxicity knowledge graph and the first excessive concentration ratio ;

[0387] Step S2130: The first risk index The original risk levels of the n2 pesticide types are compared with the preset risk level thresholds and divided into original risk levels.

[0388] Four risk level thresholds are preset: safety boundary value , slightly exceeding the boundary value , moderately exceeded the limit value and severely exceeded boundary values ;

[0389] when When , the original risk level is safe;

[0390] when When the original risk level is slightly exceeded;

[0391] when When , the original risk level is moderately exceeded;

[0392] when When the original risk level is severely exceeded.

[0393] In the calibration module, the risk benchmarks for determining n2 pesticide types in the area include:

[0394] Step S2310, classifying the second pesticide residue detection data according to the second pesticide type to generate n2 second pesticide residue data sets;

[0395] Step S2320, clustering each second pesticide residue data set, dividing the residue concentration into three risk clusters of high, medium, and low;

[0396] Step S2330, calculating the means of the three risk clusters of high, medium and low, and generating a second pesticide residue concentration threshold; the second pesticide residue concentration threshold includes a high risk concentration threshold, a medium risk concentration threshold and a low risk concentration threshold;

[0397] Step S2340: Using the second pesticide residue concentration threshold as the risk benchmark for n2 pesticide types in the area.

[0398] The step S2330 includes:

[0399] Step S2331, extracting the mean residual concentration of the high-risk cluster as the high-risk concentration threshold;

[0400] Step S2332, extracting the mean value of the residual concentration of the medium risk cluster as the medium risk concentration threshold;

[0401] Step S2333, extracting the mean residual concentration of the low-risk cluster as the low-risk concentration threshold.

[0402] Step S2340: Using the second pesticide residue concentration threshold as the risk benchmark for n2 pesticide types in the area.

[0403] In the calibration module, the risk level generated after calibration includes:

[0404] Step S2410: Obtain a second pesticide residue concentration threshold value of the same type as the first pesticide, and generate a first risk level adjustment coefficient based on a comparison between the first pesticide residue concentration and the second pesticide residue concentration threshold value. ;

[0405] Step S2420, obtaining the first area type in the first area information, and obtaining the risk level distribution probability of the area type from the pesticide residue toxicity knowledge graph ,in Safe, Mild, Moderate, Severe , generate the second risk level adjustment coefficient ;

[0406] Step S2430: Combine the first risk level adjustment coefficient and the second risk level adjustment factor , the original risk level is adjusted to generate a calibrated risk level.

[0407] The step S2410 includes:

[0408] Step S2411, when the first pesticide residue concentration Less than low risk concentration threshold When the first risk level adjustment coefficient The value is 1;

[0409] Step S2412, when the first pesticide residue concentration Greater than high risk concentration threshold When the first risk level adjustment coefficient The value is -1;

[0410] Step S2413, when the first pesticide residue concentration Greater than or equal to the low risk concentration threshold And less than or equal to the high risk concentration threshold When the first risk level adjustment coefficient The value is 0.

[0411] The step S2420 includes:

[0412] Step S2421, if the original risk level is safe, then ;in, It indicates the probability that the pesticide residue test result is at a safe level under this area type; Indicates the probability that the pesticide residue test result is slightly above the standard level; It indicates the probability that the pesticide residue test result is moderately exceeded the standard level; Indicates the probability that the pesticide residue test result is a severe excess level;

[0413] Step S2422: If the original risk level is slightly exceeded, ;

[0414] Step S2423: If the original risk level is moderately exceeded, ;

[0415] Step S2424: If the original risk level is severely exceeded, .

[0416] The adjusting of the original risk level to generate a calibrated risk level includes:

[0417] Assume the original risk level is , the risk level after calibration is , the risk level adjustment step is ;

[0418] Calculate the total adjustment factor ;

[0419] when hour, ;

[0420] when hour, ;

[0421] when hour, .

[0422] Example 3

[0423] This embodiment discloses an electronic device, which may include one or more processors and one or more memories. The memories store computer-readable codes, which, when executed by the one or more processors, may execute the blockchain-based all-in-one fluorescent quantitative test strip management method as described above.

[0424] The method or system according to the embodiment of the present application can also be implemented with the help of the architecture of an electronic device. The electronic device may include a bus, one or more CPUs, a read-only memory (ROM), a random access memory (RAM), a communication port connected to a network, an input / output component, a hard disk, etc. A storage device in an electronic device, such as a ROM or a hard disk, can store the management method of the all-in-one fluorescent quantitative test strip based on blockchain provided in this application. The management method of the all-in-one fluorescent quantitative test strip based on blockchain may, for example, include: obtaining the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip, uploading the first pesticide residue detection data to the blockchain for storage, and constructing a pesticide residue toxicity knowledge graph; comparing the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk levels of n2 pesticide types; combining the first area information and the first time information of the first pesticide residue detection data to obtain the second pesticide residue detection data of the same area and time period in the blockchain; performing cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the area; based on the risk benchmark, dynamically calibrating the original risk level to generate a calibrated risk level; setting up a smart contract on the blockchain to automatically trigger the corresponding execution strategy according to the calibrated risk level.

[0425] Furthermore, the electronic device may also include a user interface. Of course, the architecture disclosed in the present invention is only exemplary, and when implementing different devices, one or more components in the electronic device disclosed in the present invention may be omitted according to actual needs.

[0426] Example 4

[0427] This embodiment discloses a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor, the management method of the all-in-one fluorescent quantitative test strip based on blockchain of the embodiment of the present application can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache). Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0428] In addition, according to the implementation of the present application, the process described in the above reference flow chart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, the non-transitory machine-readable storage medium stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: obtaining the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip, uploading the first pesticide residue detection data to the blockchain for storage, and constructing a pesticide residue toxicity knowledge map; comparing the first pesticide residue detection data with the pesticide residue toxicity knowledge map to determine the original risk level of n2 pesticide types; combining the first area information and the first time information of the first pesticide residue detection data, obtaining the second pesticide residue detection data of the same area and the same period in the blockchain; performing cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the area; based on the risk benchmark, dynamically calibrating the original risk level to generate a calibrated risk level; setting up a smart contract on the blockchain to automatically trigger the corresponding execution strategy according to the calibrated risk level. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are performed.

[0429] The methods, systems, and devices of the present application may be implemented in many ways. For example, the methods, systems, and devices of the present application may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present application are not limited to the order specifically described above unless otherwise specified. In addition, in some embodiments, the present application may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the method according to the present application. Therefore, the present application also covers recording media storing programs for executing the method according to the present application.

[0430] In addition, the parts of the above-mentioned technical solutions provided in the embodiments of the present application that are consistent with the implementation principles of the corresponding technical solutions in the prior art are not described in detail to avoid excessive redundancy.

[0431] The specific implementation modes as described above further describe the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation mode of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, 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 blockchain-based management method for all-in-one fluorescent quantitative test strips, characterized in that: The method comprises: Obtain the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip, upload the first pesticide residue detection data to the blockchain for storage, and construct a pesticide residue toxicity knowledge graph; The first pesticide residue detection data is compared with the pesticide residue toxicity knowledge graph to determine the original risk levels of n2 pesticide types; the first region information and the first time information of the first pesticide residue detection data are combined to obtain the second pesticide residue detection data of the same region and time period in the blockchain; cluster analysis is performed on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the region; based on the risk benchmark, the original risk level is dynamically calibrated to generate a calibrated risk level; Set up smart contracts on the blockchain to automatically trigger corresponding execution strategies based on the calibrated risk level.

2. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 1, characterized in that: The first pesticide residue detection data includes a first detection sample type, a first pesticide type, a first pesticide residue concentration, a first environmental factor, a first area information and a first time information; the first area information includes a first geographic coordinate, a first area name, a first area type and a first environmental feature; The first time information includes a first timestamp; The first pesticide category includes n2 pesticide categories; The step of obtaining the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip comprises: Collect samples to be tested, use an all-in-one fluorescent quantitative test strip to test, and obtain a first pesticide residue concentration; record the type information of the samples to be tested as the first test sample type, and use the type of pesticide residue tested as the first pesticide type; record the geographical location information of the samples to be tested as the first regional information; record the test time as the first time information; and collect environmental parameters during the test as the first environmental factors.

3. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 2, characterized in that: The construction of the pesticide residue toxicity knowledge graph includes: Obtain the historical transaction data of pesticide residues on the blockchain, and extract the first key indicator of pesticide residues from the historical transaction data of pesticide residues, wherein the first key indicator of pesticide residues includes a second pesticide type, a second test sample type, a second pesticide residue concentration, a second environmental factor, a second regional information, and a second time information; the second regional information includes a second geographic coordinate, a second regional name, a second regional type, and a second environmental feature; the second time information includes a second timestamp; and the second pesticide type includes n2 pesticide types; Obtaining a second key indicator of pesticide residue, wherein the second key indicator of pesticide residue includes a pesticide residue threshold, a toxicity index, a hazard index, and a risk level label; Based on the first key indicator of pesticide residue and the second key indicator of pesticide residue, a knowledge graph of pesticide residue toxicity is constructed.

4. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 3, characterized in that: The construction of the pesticide residue toxicity knowledge graph based on the first key indicator of pesticide residue and the second key indicator of pesticide residue includes: According to the first key indicator of pesticide residue, create nodes of the pesticide residue toxicity knowledge graph, wherein the nodes of the pesticide residue toxicity knowledge graph include pesticide nodes, test sample nodes, environment nodes, time nodes and region nodes; According to the first key indicator of pesticide residue and the second key indicator of pesticide residue, the attributes of the nodes of the pesticide residue toxicity knowledge graph are expanded; Add relationship edges between nodes in the pesticide residue toxicity knowledge graph; Assign weights to the relationship edges between nodes in the pesticide residue toxicity knowledge graph.

5. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 4, characterized in that: The nodes for creating the pesticide residue toxicity knowledge graph include: Create a pesticide node according to the second pesticide type, create a detection sample node according to the second detection sample type, create an environment node according to the second environmental factor, create a time node according to the second time information, and create a region node according to the second region information; The attributes of the nodes of the expanded pesticide residue toxicity knowledge graph include: The attributes of the pesticide node are expanded according to the pesticide residue threshold, toxicity index, hazard indicator and risk level label; the attributes of the detection sample node are expanded according to the second pesticide residue concentration; the attributes of the time node are expanded according to the second timestamp; the attributes of the region node are expanded according to the second geographic coordinates, the second region name, the second region type and the second environmental characteristics.

6. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 5, characterized in that: The adding of relationship edges between nodes of the pesticide residue toxicity knowledge graph includes: Establish a detection relationship edge between the pesticide node and the detection sample node; establish an environmental impact relationship edge between the detection sample node and the environment node; establish a time relationship edge between the time node and the detection sample node; establish a region relationship edge between the region node and the detection sample node; The step of assigning weights to the relationship edges between the pesticide residue toxicity knowledge graph nodes includes: The ratio of the second pesticide residue concentration to the pesticide residue threshold is used as the weight of the detection relationship edge; According to the second environmental factor and the second pesticide residue concentration, the weight of the environmental impact relationship edge is calculated; Obtain the pesticide application time, and calculate the weight of the time relationship edge according to the second timestamp of the time node and the pesticide application time; Obtain the frequency of regional pesticide application and calculate the weight of the regional relationship edge based on the frequency of regional pesticide application.

7. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 6, characterized in that: The original risk levels for determining n2 pesticide types include: Compare the first pesticide residue concentration with the pesticide residue threshold of the corresponding pesticide node in the pesticide residue toxicity knowledge graph to calculate a first excess concentration ratio; the first excess concentration ratio is equal to the first pesticide residue concentration divided by the pesticide residue threshold; According to the toxicity index and hazard index of the corresponding pesticide node in the pesticide residue toxicity knowledge graph, combined with the first excessive concentration ratio, the first risk index is calculated ; The first risk index The original risk levels of the n2 pesticide types are compared with the preset risk level thresholds and divided into original risk levels.

8. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 7, characterized in that: The risk criteria for determining n2 pesticide types in the area include: Classifying the second pesticide residue detection data according to the second pesticide type to generate n2 second pesticide residue data sets; Each second pesticide residue dataset was clustered and the residue concentrations were divided into three risk clusters: high, medium, and low; Calculate the means of the three risk clusters of high, medium and low to generate a second pesticide residue concentration threshold; the second pesticide residue concentration threshold includes a high risk concentration threshold, a medium risk concentration threshold and a low risk concentration threshold; The second pesticide residue concentration threshold is used as the risk benchmark for n2 pesticide types in the area.

9. The method for managing an all-in-one fluorescent quantitative test strip based on blockchain according to claim 8, characterized in that: The dynamically calibrating the original risk level to generate the calibrated risk level includes: Obtain a second pesticide residue concentration threshold value of the same type as the first pesticide, and generate a first risk level adjustment coefficient based on a comparison between the first pesticide residue concentration and the second pesticide residue concentration threshold value. ; Obtain the first area type in the first area information, and obtain the risk level distribution probability of the area type from the pesticide residue toxicity knowledge graph ,in Safe, Mild, Moderate, Severe , generate the second risk level adjustment coefficient ; Combined with the first risk level adjustment coefficient and the second risk level adjustment factor , adjust the original risk level to generate a calibrated risk level; Generating a first risk level adjustment coefficient include: When the first pesticide residue concentration Less than low risk concentration threshold When the first risk level adjustment coefficient The value is 1; when the first pesticide residue concentration Greater than high risk concentration threshold When the first risk level adjustment coefficient The value is -1; when the first pesticide residue concentration Greater than or equal to the low risk concentration threshold And less than or equal to the high risk concentration threshold When the first risk level adjustment coefficient The value is 0.

10. A blockchain-based all-in-one fluorescent quantitative test strip management system, which is used to implement the blockchain-based all-in-one fluorescent quantitative test strip management method according to any one of claims 1 to 9, characterized in that: The system comprises: Data acquisition module: used to obtain the first pesticide residue detection data of the all-in-one fluorescent quantitative test strip; Data storage module: used to upload the first pesticide residue detection data to the blockchain for storage; Knowledge graph construction module: used to construct the knowledge graph of pesticide residue toxicity; Risk level classification module: used to compare the first pesticide residue detection data with the pesticide residue toxicity knowledge graph to determine the original risk level of n2 pesticide types; Calibration module: used to combine the first area information and the first time information of the first pesticide residue detection data to obtain the second pesticide residue detection data of the same area and time period in the blockchain; perform cluster analysis on the second pesticide residue detection data to determine the risk benchmark of n2 pesticide types in the area; based on the risk benchmark, dynamically calibrate the original risk level to generate a calibrated risk level; Execution module: used to set up smart contracts on the blockchain and automatically trigger corresponding execution strategies based on the calibrated risk level.

Images