Intelligent environmental protection data tracking method and system based on block chain technology

Abstract

The invention discloses an intelligent environmental protection data tracking method and system based on a block chain technology, and relates to the field of tracking methods, and the method comprises the steps: creating an environmental protection alliance chain; automatic uploading of all environmental protection information is completed; triggering the smart contract to verify all the environmental protection information; and under-chain abnormal node detection is performed on the environmental protection nodes based on the environmental protection information, and abnormal information tracing is completed in an environmental protection alliance chain. According to the method and the device, the abnormal behaviors in the garbage collection process can be effectively and accurately tracked and positioned.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data tracking, and in particular, to a method and system for intelligent environmental protection data tracking based on blockchain technology. Background Art

[0002] In recent years, with the improvement of the public's environmental protection awareness, people have begun to pay more attention to the classification, recycling, and reuse of garbage in order to reduce environmental pollution and promote the sustainable use of resources. Among them, due to the large quantity and complex composition of industrial waste, improper treatment is extremely likely to cause long-term and serious pollution to the environment. Therefore, the treatment process of industrial waste needs to be closely supervised, and abnormal conditions in the treatment process need to be quickly captured and traced.

[0003] With the rapid development of intelligent technology, methods for intelligently tracing abnormal behaviors in the industrial waste treatment process have emerged. Traditional anomaly recognition technologies deploy Internet of Things (IoT) devices at each node in the industrial waste treatment process, collect real-time data of the industrial waste treatment process through the IoT devices in real time, and perform anomaly recognition and anomaly tracing on the real-time data based on machine learning algorithms to achieve all-round monitoring of the industrial waste treatment process. Traditional anomaly recognition technologies can, to a certain extent, capture abnormal behaviors in the industrial waste treatment process and trace their sources. However, the communication between devices may be delayed or interrupted, which may lead to the loss of monitoring data of abnormal conditions, thus affecting the accuracy of anomaly recognition results. At the same time, there is also a risk that the monitoring data of IoT devices may be tampered with. If the tampered monitoring data is used to identify abnormal behaviors in the industrial waste treatment process, it may also lead to deviations in the anomaly recognition results. Summary of the Invention

[0004] The embodiments of the present application provide a method and system for intelligent environmental protection data tracking based on blockchain technology to solve the problem that it is difficult to accurately track and locate abnormal behaviors in the garbage recycling process in the prior art.

[0005] To achieve the above object, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, a method for intelligent environmental protection data tracking based on blockchain technology is provided, and the method includes:

[0007] Create an environmental protection alliance chain for the industrial waste treatment process based on blockchain technology;

[0008] Obtain the environmental protection information of each environmental protection node in the industrial waste treatment process, and complete the automatic uploading of all environmental protection information to the chain according to the consensus mechanism preset in the environmental protection alliance chain;

[0009] Trigger the smart contract based on the preset information compliance logic in the environmental protection alliance chain to verify all environmental protection information;

[0010] If all environmental protection information passes the verification, conduct off-chain abnormal node detection on environmental protection nodes based on the environmental protection information, and complete the traceability of abnormal information in the environmental protection alliance chain according to the abnormal node detection results.

[0011] Optionally, the off-chain abnormal node detection on environmental protection nodes based on the environmental protection information and the completion of the traceability of abnormal information in the environmental protection alliance chain according to the abnormal node detection results include the following steps:

[0012] Mark the first abnormal label for environmental protection nodes based on the environmental protection information and using the threshold method;

[0013] Conduct correlation analysis on each environmental protection node using the correlation algorithm, construct an environmental protection time series diagram according to the correlation analysis results, and mark the second abnormal label for environmental protection nodes based on the environmental protection time series diagram;

[0014] Complete the traceability of abnormal information in the environmental protection alliance chain based on the abnormal environmental protection information corresponding to the environmental protection nodes marked with abnormal labels.

[0015] Optionally, the environmental protection information includes garbage weight information, garbage treatment information, garbage type information, garbage transportation information, and garbage time information.

[0016] Optionally, the steps of conducting correlation analysis on each environmental protection node using the correlation algorithm, constructing an environmental protection time series diagram according to the correlation analysis results, and marking the second abnormal label for each environmental protection node based on the environmental protection time series diagram include the following steps:

[0017] Conduct time series analysis on the environmental protection information to obtain the environmental protection sequences of each environmental protection node;

[0018] Calculate the sequence correlation between all environmental protection sequences using the correlation algorithm;

[0019] Construct an environmental protection time series diagram with the environmental protection sequence as the sequence node and the sequence correlation as the sequence edge;

[0020] Extract the environmental protection main backbone loop in the environmental protection time series diagram based on the recursive algorithm;

[0021] Construct an environmental protection feature matrix of the environmental protection information based on the environmental protection main backbone loop, and judge whether there are abnormal nodes in the environmental protection main backbone loop based on the environmental protection feature matrix;

[0022] If there are abnormal nodes in the environmental protection main backbone loop, identify abnormal nodes for all environmental protection nodes based on the environmental protection main backbone loop, and mark the second abnormal label for environmental protection nodes based on the abnormal node identification results.

[0023] Optionally, the steps for extracting the environmental protection backbone loop from the environmental protection time series diagram based on the recursive algorithm are as follows:

[0024] Use the prim algorithm to divide all sequence edges in the environmental protection time series diagram into backbone tree edges and non-backbone tree edges, and generate the time series backbone tree of the environmental protection time series diagram based on all backbone tree edges;

[0025] Select the top node in the time series backbone tree as the initial node;

[0026] Use the recursive algorithm to perform a depth search on all neighbor nodes of the initial node to obtain multiple path node sets;

[0027] Calculate the path lengths between the initial node and the terminal nodes in all path node sets;

[0028] Mark the terminal nodes in the path node sets with the maximum and the second maximum path lengths as the first terminal node and the second terminal node respectively;

[0029] If there is a connected edge between the first terminal node and the second terminal node, and all connected edges are non-backbone tree edges, then mark all sequence nodes in the path node sets corresponding to the first terminal node and the second terminal node, as well as the initial node, as backbone link nodes;

[0030] Mark all sequence nodes except all backbone link nodes among all sequence nodes as non-backbone link nodes;

[0031] For any non-backbone link node, if there are connected edges between the non-backbone link node and multiple backbone link nodes, and all connected edges are non-backbone tree edges, then mark the non-backbone link node as a backbone link node;

[0032] Extract the environmental protection backbone loop from the environmental protection time series diagram based on all backbone link nodes.

[0033] Optionally, the steps for constructing the environmental protection feature matrix of environmental protection information based on the environmental protection backbone loop and determining whether there are abnormal nodes in the environmental protection backbone loop are as follows:

[0034] According to the first abnormal label, screen abnormal nodes from the backbone link nodes in the environmental protection backbone loop to obtain several abnormal link nodes;

[0035] Analyze the node attributes of all abnormal link nodes, and extract the abnormal node features of all abnormal link nodes according to the node attributes. The abnormal node features include abnormal degree features, abnormal connectivity features, and abnormal clustering features;

[0036] Use the Laplace matrix formula to construct the backbone loop matrix of the environmental protection backbone loop, and extract the backbone loop spectral features of the environmental protection backbone loop according to the backbone loop matrix;

[0037] Analyze the node connection relationship of the environmental protection backbone loop, and extract the backbone geometric features of the environmental protection backbone loop according to the node connection relationship;

[0038] Use the graph partitioning algorithm to decompose the environmental protection backbone loop into multiple loop sub-structures;

[0039] Analyze the structural connection relationship between all loop sub-structures, and extract the connection strength feature and connection path feature between loop sub-structures based on the structural connection relationship;

[0040] Construct an environmental protection feature matrix based on the abnormal node feature, backbone loop spectrum feature, backbone geometric feature, connection strength feature and connection path feature;

[0041] Calculate the matrix similarity between the environmental protection feature matrix and the preset reference environmental protection feature matrix. If the matrix similarity is greater than the preset similarity threshold, it is determined that there are no abnormal nodes in the environmental protection backbone loop;

[0042] If the matrix similarity is less than or equal to the similarity threshold, it is determined that there are abnormal nodes in the environmental protection backbone loop.

[0043] Optionally, the identification of abnormal nodes for all environmental protection nodes based on the environmental protection backbone loop includes the following steps:

[0044] Use the graph partitioning algorithm to decompose the environmental protection backbone loop into multiple loop sub-structures;

[0045] Identify structurally abnormal edges for loop sub-structures according to sequence correlation;

[0046] If the structurally abnormal edge identification result shows that there are abnormal edges in the loop sub-structure, input the abnormal edges into the pre-constructed abnormal undirected graph;

[0047] Determine whether the abnormal undirected graph after all abnormal edges are input is a bipartite graph;

[0048] If the abnormal undirected graph is a bipartite graph, use the Hungarian algorithm to analyze the maximum matching of the abnormal undirected graph, and output the maximum matching result as the abnormal node identification result;

[0049] If the abnormal undirected graph is not a bipartite graph, use the greedy strategy to identify abnormal nodes in the abnormal undirected graph to obtain the abnormal node identification result.

[0050] Optionally, the tracing of abnormal information for the abnormal environmental protection information corresponding to the environmental protection nodes marked with abnormal labels is completed in the environmental protection alliance chain, including the following steps:

[0051] Receive the data interaction request of the target user, and verify the user identity authentication information included in the data interaction request;

[0052] If the user identity authentication information is verified successfully, asymmetric encryption is performed on the garbage feedback information and user credibility in the data interaction request to obtain user-encrypted information;

[0053] Upload the user-encrypted information to the environmental protection alliance chain;

[0054] Analyze the data upload frequency of the data interaction request based on the block timestamp on the environmental protection alliance chain;

[0055] Allocate node abnormality degrees to all environmental protection nodes according to whether the environmental protection nodes are marked with the first abnormal label and / or the second abnormal label;

[0056] When the data upload frequency is greater than the preset frequency threshold or the node abnormality degree of any environmental protection node is greater than the preset abnormality degree threshold, perform abnormal node verification on all environmental protection nodes based on the data interaction request and environmental protection information;

[0057] Complete the traceability of abnormal information in the environmental protection alliance chain based on the abnormal node verification result.

[0058] Optionally, the abnormal node verification of all environmental protection nodes based on the data interaction request and environmental protection information includes the following steps:

[0059] Assign weights to the corresponding garbage feedback information according to the user credibility, and assign feedback abnormality degrees to each environmental protection node according to the garbage feedback information with weights assigned;

[0060] Fusion the feedback abnormality degree and the node abnormality degree by weighting to obtain the node comprehensive abnormality degree;

[0061] Extract the environmental protection transaction relationships among all environmental protection nodes in the industrial waste treatment process;

[0062] Take the environmental protection nodes as graph nodes, the node comprehensive abnormality degree and environmental protection information as node features, and the environmental protection transaction relationship as edge features to construct an environmental protection transaction graph;

[0063] Connect the corresponding environmental protection nodes in the environmental protection transaction graphs of adjacent time nodes with time edges to obtain an environmental protection transaction spatio-temporal graph;

[0064] Based on the graph neural network model and the spatio-temporal recurrent gate, construct an abnormal node verification model. The abnormal node verification model includes a spatial feature extraction module, a temporal feature extraction module, and an abnormal node verification module. The spatial feature extraction module is constructed based on the enhanced graph convolutional network model, and the temporal feature extraction module is constructed based on the spatio-temporal recurrent gate;

[0065] Use the pre-constructed training set to train the abnormal node verification model;

[0066] Input the environmental protection transaction spatio-temporal graph into the trained abnormal node verification model to extract spatial features and temporal features, obtaining environmental protection transaction spatial features and environmental protection transaction temporal features;

[0067] The abnormal node verification module performs dimension conversion and information integration on the environmental protection transaction spatial features and environmental protection transaction temporal features, and then outputs the abnormal node verification result.

[0068] In a second aspect, the present application provides a smart environmental protection data tracking system based on blockchain technology, which is characterized by including:

[0069] A memory configured to store instructions; and

[0070] A processor configured to call instructions from the memory and, when executing the instructions, be capable of implementing the smart environmental protection data tracking method based on blockchain technology according to any one of the first aspect.

[0071] Through the above technical solutions, by constructing an environmental protection alliance chain for the industrial waste treatment process, it is ensured that environmental protection information cannot be tampered with, enhancing the credibility of environmental protection information, contributing to improving the accuracy of abnormal node identification. Each environmental protection node can share real-time data, promoting information circulation and improving the efficiency of abnormal node identification. At the same time, using smart contracts in the environmental protection alliance chain to verify environmental protection information ensures the rationality and standardization of information, which is also beneficial to the accuracy and efficiency of subsequent abnormal node identification. In addition, introducing the temporal correlation between environmental protection nodes to identify abnormal nodes in the industrial waste treatment process, analyzing the time correlation between environmental protection nodes can help identify potential risks and problems in the industrial waste treatment process, further increasing the accuracy of abnormal node identification. In summary, the present application provides a method for identifying abnormal behaviors in the industrial waste treatment process from multiple dimensions, which can timely capture abnormal behaviors in the industrial waste treatment process, enhance the standardization of the industrial waste treatment process, and thus avoid environmental pollution to a certain extent.

[0072] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flowchart of a smart environmental protection data tracking method based on blockchain technology provided by an embodiment of the present application;

[0074] Figure 2 It is a structural diagram of a temporal backbone tree provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0075] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. It should be understood that the specific implementation manners described herein are only for explaining and illustrating the embodiments of this application, and are not used to limit the embodiments of this application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by this application.

[0076] It should be noted that if there are directional indications (such as up, down, left, right, front, back...) involved in the embodiments of this application, then such directional indications are only used to explain the relative positional relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If this specific posture changes, then the directional indications will also change accordingly.

[0077] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of this application, then such descriptions of "first", "second", etc. are only for descriptive purposes, and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions between various embodiments can be combined with each other, but it must be based on the ability of those of ordinary skill in the art to implement. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0078] Figure 1 A schematic flowchart of a smart environmental protection data tracking method based on blockchain technology according to an embodiment of this application is schematically shown. As Figure 1 shown, an embodiment of this application provides a smart environmental protection data tracking method based on blockchain technology. The method may include the following steps:

[0079] S101. Create an environmental protection consortium chain for the industrial waste treatment process based on blockchain technology;

[0080] The industrial waste treatment process includes multiple links such as waste collection, waste transportation, waste sorting, and waste treatment. The environmental protection nodes include industrial waste manufacturing nodes, industrial waste transfer nodes, industrial waste treatment nodes, and industrial waste recycling nodes. Specifically, the industrial waste manufacturing nodes refer to various types of factories that generate industrial waste, such as chemical plants, paper mills, steel mills, smelters, etc. These factories generate a large amount of industrial waste, such as smelting slag, electroplating waste, construction waste, ceramic waste, and muck. The industrial waste transfer nodes refer to waste recycling enterprises responsible for centrally collecting various types of waste, collecting the waste generated by different factories, and conducting preliminary sorting according to the type of industrial waste. For example, industrial waste can be divided into recyclable waste, non-recyclable waste, hazardous waste, and harmless waste. The industrial waste treatment nodes refer to enterprises that process or harmlessly treat industrial waste. The enterprises corresponding to the industrial waste treatment nodes will classify industrial waste more rigorously, and different types of waste have different treatment methods. For example, non-recyclable industrial waste is landfilled or incinerated for power generation, and special industrial waste containing chemical substances, heavy metals, or radioactive substances is professionally destroyed. The industrial waste recycling nodes refer to re-production enterprises that reprocess and sell recyclable industrial waste. For example, some metal processing enterprises can remelt and purify metal waste and convert it into new metal products. Some recycling enterprises focus on the recycling and reprocessing of waste plastic films. Through processes such as crushing, rinsing, kneading, and pulverizing and drying, waste plastic films are converted into polyethylene high-pressure PE particles, which become the raw materials for products such as fruit baskets and urban drainage pipes, realizing the reuse of waste plastic films.

[0081] Since industrial waste often contains a large amount of harmful substances, such as heavy metals and organic substances, if these substances are not treated and recycled in accordance with the standard process, they will cause great pollution to the environment. Therefore, it is necessary to supervise the key nodes in the process of industrial waste recycling and treatment. Since a large number of enterprises are involved in the process of industrial waste recycling and treatment, the amount of relevant information data of different enterprises for waste treatment is also huge. In the traditional industrial waste recycling supervision system, data and information among different participants (such as the public, waste recyclers, waste treatment enterprises, re-production enterprises, regulatory agencies, etc.) often exist in isolation, lacking a mechanism for sharing and communication. Once some enterprises in the industrial waste treatment process do not recycle waste in accordance with the standard waste treatment method in order to save costs, such as chemical plants secretly dumping hazardous industrial waste, on the one hand, it is difficult to detect abnormal conditions in the industrial waste treatment process in a timely manner, and on the other hand, even if abnormal conditions in the waste treatment process are detected, it is difficult to trace the responsible enterprise.

[0082] Due to the advantages of blockchain technology such as decentralization, immutability, high efficiency, and reliability, it can connect all links in the industrial waste treatment process, from collection, classification, transportation to treatment, achieving seamless docking and real-time update of information. Since the data on the blockchain is immutable, this ensures the authenticity and integrity of the data in the industrial waste treatment process. Due to the traceability of the blockchain, when abnormal conditions occur in the industrial waste treatment process, it is possible to track the abnormal conditions and locate the responsible entity for the abnormal conditions, effectively promoting the accuracy and compliance of industrial waste recycling and treatment, preventing the malicious behavior of individual nodes, promoting the fairness and efficiency of resource allocation, and driving the effect of industrial waste recycling and treatment.

[0083] The basic architecture of the blockchain mainly has six layers, including the data layer, network layer, consensus layer, incentive layer, contract layer, and service layer. Each layer completes a part of the core tasks, and the layers cooperate with each other to achieve a decentralized trust model. The data layer represents the physical form of blockchain technology, is the basic technical structure for designing the blockchain ledger, and describes the composition of the blockchain. The main function of the network layer is to enable communication between nodes in the blockchain network and achieve distributed information recording. The consensus layer is responsible for enabling highly dispersed and mutually distrustful nodes to reach a consensus on a certain aspect efficiently in a decentralized system through information exchange. This is the core idea of the blockchain. The incentive layer refers to the reward scheme for the accounting nodes, which encourages the accounting nodes to actively participate in processes such as data verification and data packaging by providing some reward measures, mobilizing the participation enthusiasm of the whole network nodes. In this application, the system will reward the target users who participate in data interaction, improve the enthusiasm of the masses, and use the power of the masses to supervise the industrial waste treatment process. The contract layer mainly includes different script codes, algorithm mechanisms, and smart contracts, etc., making the ledger flexibly programmable. In this embodiment, a smart contract based on the preset information compliance logic is used to verify whether the environmental protection information meets the information upload specifications, whether the information is complete, and whether the information conforms to the basic logic.

[0084] Specifically, take the environmental protection nodes in the industrial waste treatment process, such as waste manufacturing and waste recycling enterprises, as the nodes on the chain, determine the permissions of each environmental protection node, select a suitable consensus algorithm (such as PoW, PoS, PBFT, etc.) to maintain network security and data consistency, design the logic and functions of the smart contract according to business requirements, such as for user credibility allocation, environmental protection information verification, and penalty for illegal behaviors, and deploy the smart contract to the blockchain network to ensure its executability and transparency. Upload the environmental protection data (such as waste weight information, waste treatment information, waste type information, etc.) of each environmental protection node to the chain to ensure the immutability and transparency of the data, and set up a data sharing mechanism so that each node can access and use relevant data to promote cooperation.

[0085] S102. Obtain the environmental protection information of each environmental protection node in the industrial waste treatment process, and complete the automatic uploading of all environmental protection information to the chain according to the preset consensus mechanism in the environmental protection alliance chain;

[0086] The environmental protection information includes waste weight information, waste treatment information, waste type information, waste transportation information, and waste time information. The waste weight information refers to the weight of each batch of industrial waste when it arrives at each environmental protection node. The waste treatment information includes waste treatment methods, waste treatment processes, waste treatment frequencies, etc. For example, the waste treatment methods are incineration for power generation and landfill composting. The treatment process of some non-hazardous solid waste includes a series of processes such as compression, crushing, and landfill. The waste type information includes recyclable waste, non-recyclable waste, hazardous waste, etc. Among them, recyclable waste includes waste plastics, waste glass, etc., non-recyclable waste includes chemical containers, construction waste, etc., and hazardous waste includes waste optoelectronic materials, waste paint, waste solvents, etc. The waste transportation information includes waste transportation routes, waste transportation methods, etc. The waste time information refers to the generation time of each batch of industrial waste at the industrial waste manufacturing node and the time when it arrives at all other environmental protection nodes.

[0087] After obtaining the environmental protection information of each environmental protection node in the industrial waste treatment process, through blockchain technology, according to the preset consensus algorithm, the environmental protection information data is automatically recorded on the environmental protection alliance chain in an immutable manner. The preset consensus mechanism is used to ensure that all environmental protection nodes reach an agreement on the environmental protection information and is the basis for realizing the decentralization of the blockchain. There are various algorithms such as Proof of Work (PoW), Proof of Stake (PoS), Practical Byzantine Fault Tolerance (PBFT), etc.

[0088] Taking the Practical Byzantine Fault Tolerance (PBFT) algorithm as an example, the core idea of the PBFT algorithm is to ensure that the system can still reach a consensus decision in the case of failures or malicious behaviors of some nodes through the voting and confirmation mechanisms of multiple replica nodes. When an environmental protection node generates or updates environmental protection information, this information will be packaged into a transaction request and sent to the primary node in the environmental protection consortium chain. After receiving the transaction request, the primary node will package it into a block and generate a pre-prepare message, which contains information such as the block hash value and sequence number. Then, the primary node broadcasts this pre-prepare message to all other replica nodes in the environmental protection consortium chain. The replica nodes will verify the received pre-prepare message for its validity and consistency. If the verification passes, the replica nodes will generate a prepare message and broadcast it to all other nodes in the environmental protection consortium chain, indicating that this node is ready to execute the transaction request. When the replica nodes collect prepare messages sent by more than 2 / 3 of the nodes, they will consider that the transaction request has been verified by the majority of nodes and is ready to be executed. At this time, these nodes will generate a commit message and broadcast it to all nodes in the environmental protection consortium chain. After the primary node and the replica nodes receive a sufficient number of commit messages, they will execute the operations in the transaction request, record the environmental protection information on the blockchain, and then broadcast the execution result to the requester. The requester needs to wait and verify the same execution results from different replica nodes to ensure the final certainty of the transaction. Through the consensus mechanism of the PBFT algorithm, all environmental protection information submitted by environmental protection nodes will be verified and recorded on the blockchain, ensuring the consistency and reliability of the information, and thus ensuring the accuracy of the identification results when identifying abnormal nodes subsequently.

[0089] S103. Trigger the intelligent contract based on the preset information compliance logic in the environmental protection consortium chain to verify all environmental protection information;

[0090] An intelligent contract is an automated protocol defined in the form of computer code, stored on the blockchain, and automatically executed when preset conditions are met. Its core function is to complete transactions or operations according to pre-set rules without manual intervention, allowing for trusted transactions without a third party. These transactions are traceable and irreversible. The operation of intelligent contracts depends on blockchain technology and a decentralized network. The specific workflow includes: writing code using an intelligent contract language and deploying it to the blockchain network, setting the trigger conditions of the intelligent contract, such as triggering the intelligent contract when the environmental protection consortium chain receives a request for environmental protection information from an environmental protection node, verifying whether the transaction request triggers the conditions through the consensus mechanism. If the trigger conditions are met, the contract code will be immediately executed. Finally, the intelligent contract writes the contract execution result into the blockchain to ensure immutability and traceability.

[0091] Specifically, first verify whether the environmental protection information is complete, that is, check whether there is any missing important information in the environmental protection information. For example, if important information such as the garbage weight information or the garbage time information in the environmental protection information is missing, it is determined that the integrity verification of the environmental protection information fails. After completing the information integrity verification, continue to verify the information standardization. The format of the environmental protection information is verified according to the pre-stored information format template. For example, the standard format of the garbage time information is YYYY-MM-DD HH:MM, where YYYY, MM, and DD represent the year, month, and day respectively, and HH and MM represent the hour and minute respectively. If the garbage time information is 2025-2-111, the minute is missing, and the month and date also do not conform to the standard. Therefore, it is determined that the information standardization verification fails. When both the integrity verification and the standardization verification of the environmental protection information pass, a simple logical verification of the environmental protection information is performed. For example, if the garbage time information is later than the current verification time or the garbage weight information is negative, which obviously does not conform to the basic logical rules, it is determined that the logical verification of the environmental protection information fails. The preset information compliance logic is the information format template and the information basic logical rules mentioned above.

[0092] Through the strict verification of the smart contract, the environmental protection information is stored and transmitted on the blockchain in an accurate, complete, and standardized form, which provides a solid data foundation for subsequent detection of abnormal nodes based on this information, and improves the efficiency of abnormal node verification to a certain extent.

[0093] S104. If all the environmental protection information passes the verification, then perform off-chain abnormal node detection on the environmental protection nodes based on the environmental protection information, and complete the traceability of abnormal information in the environmental protection alliance chain according to the abnormal node detection results.

[0094] In this embodiment, if all environmental protection information passes the information verification, the off-chain abnormal node detection of the environmental protection nodes is immediately performed. Specifically, first, the threshold method is used to verify whether the weight difference between the average weight of garbage disposal and the standard weight of garbage disposal of each environmental protection node during the period to be inspected is greater than the preset difference threshold, and a first abnormal label is marked for the environmental protection nodes with the weight difference greater than the preset threshold. Then, the time series analysis of the environmental protection information of each environmental protection node is carried out, and the sequence correlation between the environmental protection information of different environmental protection nodes is calculated according to the time series analysis nodes. An environmental protection time series graph is constructed between all environmental protection nodes based on the sequence correlation. The abnormal nodes are identified for all environmental protection nodes based on the environmental protection time series graph, and a second abnormal label is marked for the environmental protection nodes based on the abnormal node identification result. Since there are interaction behaviors between the environmental protection nodes in the industrial waste treatment process, for example, the industrial waste manufacturing node needs to centrally transport the industrial waste it generates to the industrial waste transfer node, and the industrial waste transfer node needs to preliminarily classify the collected industrial waste and then transport it to other environmental protection nodes for classification and treatment. Therefore, the environmental protection nodes are interconnected and interact with each other. Therefore, the abnormal nodes can be identified according to the correlation between the environmental protection nodes, and a second abnormal label is marked for the environmental protection nodes with abnormalities.

[0095] In addition, in order to further master the garbage disposal status of each environmental protection node, this embodiment introduces a mass supervision module. After the target user completes user registration and login, they can upload industrial waste feedback information, such as the behavior of a certain factory privately landfilling harmful industrial waste. As a broad supervision subject, the masses have a wide and in-depth supervision scope, can discover some problems that are difficult for professional supervision agencies to reach, improve the efficiency of abnormal node identification, and at the same time, by encouraging the masses to actively participate in and understand environmental protection work, can improve the environmental protection awareness of the masses, thus forming a good atmosphere in which the whole society pays attention to and participates in environmental protection work together. Secondly, this bottom-up supervision method helps to promote the scientific process of environmental protection work and improve the quality and efficiency of environmental protection work.

[0096] Starting from the environmental protection nodes marked with abnormal labels, trace all subsequent on-chain environmental protection information of this environmental protection node, and at the same time compare the abnormal environmental protection information with the on-chain environmental protection information of adjacent time nodes to identify continuous abnormalities, count their abnormal frequencies, and confirm whether it is the first abnormality or repeated violation. Based on all of the above, an abnormal report is generated, and finally the abnormal report is encrypted and stored on the chain. At the same time, the environmental protection relevant department will hold the enterprise corresponding to this environmental protection node accountable according to the abnormal report and urge it to make rectifications.

[0097] In one of the embodiments, the off-chain abnormal node detection of the environmental protection nodes is performed based on the environmental protection information, and the abnormal information tracing in the environmental protection alliance chain is completed according to the abnormal node detection result, including the following steps:

[0098] Based on environmental protection information and using the threshold method, mark the first abnormal label for environmental protection nodes;

[0099] Use the correlation algorithm to perform correlation analysis on each environmental protection node, construct an environmental protection time series graph according to the correlation analysis results, and mark the second abnormal label for environmental protection nodes based on the environmental protection time series graph;

[0100] Complete the traceability of abnormal information in the environmental protection alliance chain based on the abnormal environmental protection information corresponding to the environmental protection nodes marked with abnormal labels.

[0101] In this embodiment, use the environmental protection alliance chain to query the historical environmental protection information submitted by non-abnormal nodes, that is, the historical environmental protection information is the environmental protection information of all environmental protection nodes in the historical time period, and in this historical time period, all environmental protection nodes meet the garbage treatment specifications and have no abnormal labels. Therefore, the historical environmental protection information in this historical time period is used as the standard environmental protection information. Extract the standard garbage weight information and standard garbage time information in the standard environmental protection information, count the garbage treatment times of each environmental protection node in the historical time period according to the standard garbage time information, and calculate the standard garbage weight of each environmental protection node according to the standard garbage weight information and the garbage treatment times. Calculate the average garbage treatment weight of each environmental protection node in the current time period according to the environmental protection information, calculate the weight difference between the standard garbage weight and the average garbage treatment weight of the same environmental protection node respectively, and mark the first abnormal label for the environmental protection nodes with the weight difference greater than the preset threshold. Use the threshold method to initially screen abnormal nodes for each environmental protection node, which is convenient for further analysis of abnormal nodes of environmental protection nodes. When there is any environmental protection node with a first abnormal label, the relevant department will send a reminder message to the enterprise corresponding to the environmental protection node, reminding the relevant enterprise to recycle industrial garbage according to the specifications, dispelling the fluke mentality of some enterprises, and effectively urging each enterprise to actively carry out garbage recycling operations to a certain extent.

[0102] Then perform time series analysis on the environmental protection information of each environmental protection node, calculate the serial correlation between the environmental protection information of different environmental protection nodes according to the time series analysis nodes, and construct an environmental protection time series graph for all environmental protection nodes based on the serial correlation. Identify abnormal nodes for all environmental protection nodes based on the environmental protection time series graph, and mark the second abnormal label for environmental protection nodes based on the abnormal node identification results. Since there are interaction behaviors between each environmental protection node in the industrial garbage treatment process, for example, the industrial garbage manufacturing node needs to centrally transport the industrial garbage it generates to the industrial garbage transfer node, and the industrial garbage transfer node needs to preliminarily classify the collected industrial garbage and then transport it to other environmental protection nodes for classification and treatment. Therefore, each environmental protection node is interconnected and mutually influential. Therefore, abnormal nodes can be identified according to the correlation between each environmental protection node, and the second abnormal label can be marked for the environmental protection nodes with abnormalities.

[0103] Starting from the environmental protection node marked with an abnormal tag, trace all subsequent environmental protection information uploaded by this environmental protection node. At the same time, compare the abnormal environmental protection information with the environmental protection information uploaded at adjacent time nodes, identify continuous anomalies, and count their anomaly frequencies. At the same time, confirm whether it is the first anomaly or a repeated violation. Based on all of the above, generate an anomaly report, and finally encrypt the anomaly report and store it on the blockchain. At the same time, the environmental protection department will hold the enterprise corresponding to this environmental protection node accountable according to the anomaly report and urge it to make rectifications.

[0104] In one embodiment, the environmental protection information includes garbage weight information, garbage treatment information, garbage type information, garbage transportation information, and garbage time information.

[0105] In this embodiment, the environmental protection information includes garbage weight information, garbage treatment information, garbage type information, garbage transportation information, and garbage time information. The garbage weight information refers to the weight of each batch of industrial garbage when it arrives at each environmental protection node. The garbage treatment information includes garbage treatment methods, garbage treatment processes, garbage treatment frequencies, etc. For example, the garbage treatment method is incineration for power generation or landfill composting. The treatment process of some non-hazardous solid garbage includes a series of processes such as compression, crushing, and landfill. The garbage type information includes recyclable garbage, non-recyclable garbage, hazardous garbage, etc. Among them, recyclable garbage includes waste plastics, waste glass, etc., non-recyclable garbage includes chemical containers, construction waste, etc., and hazardous garbage includes waste optoelectronic materials, waste paint, waste solvents, etc. The garbage transportation information includes garbage transportation routes, garbage transportation methods, etc. The garbage time information refers to the generation time of each batch of industrial garbage at the industrial garbage manufacturing node and the time when it arrives at all other environmental protection nodes.

[0106] In one embodiment, a correlation algorithm is used to perform correlation analysis on each environmental protection node. According to the correlation analysis results, an environmental protection time series graph is constructed. The steps for marking the second abnormal node for each environmental protection node based on the environmental protection time series graph are as follows:

[0107] Perform time series analysis on the environmental protection information to obtain the environmental protection sequences of each environmental protection node;

[0108] Use the correlation algorithm to calculate the sequence correlation between all environmental protection sequences;

[0109] Construct an environmental protection time series graph with the environmental protection sequence as the sequence node and the sequence correlation as the sequence edge;

[0110] Extract the environmental protection backbone loop in the environmental protection time series graph based on the recursive algorithm;

[0111] Construct an environmental protection feature matrix of the environmental protection information based on the environmental protection backbone loop, and judge whether there are abnormal nodes in the environmental protection backbone loop based on the environmental protection feature matrix;

[0112] If there are abnormal nodes in the environmental protection backbone loop, all environmental protection nodes are identified for abnormal nodes based on the environmental protection backbone loop, and a second abnormal label is marked for the environmental protection nodes based on the abnormal node identification result.

[0113] In this embodiment, according to the garbage time information in the environmental protection information, the environmental protection information of each environmental protection node is converted into a series of continuous data sequences, that is, the environmental protection sequence. For example, the environmental protection sequence can be expressed as G = {g1, g2, g3,..., g T}, where g T represents the sequence point of the environmental protection sequence at time node T, and the environmental protection information corresponding to time node T is included in the sequence point. The sequence correlation between each environmental protection sequence is calculated using the correlation algorithm. Commonly used correlation algorithms include the Pearson correlation coefficient, the Spearman rank correlation coefficient, etc. Taking the Pearson correlation coefficient as an example, the Pearson correlation coefficient is suitable for measuring the linear correlation degree between two sequences, and its value range is between -1 and 1. The closer the value is to 1 or -1, the stronger the correlation; the closer it is to 0, the weaker the correlation. The sequence correlation between all environmental protection sequences can be calculated using the Pearson correlation coefficient calculation formula. The environmental protection sequence is used as the sequence node, and the sequence correlation is used as the sequence edge to construct the environmental protection time series graph.

[0114] Then, the largest loop structure in the environmental protection time series graph, that is, the environmental protection backbone loop, is extracted. Specifically, first use the prim algorithm to extract the time series backbone tree from the environmental protection time series graph. The edges in the time series backbone tree are used as the backbone tree edges, and the edges not located in the time series backbone tree are used as non-backbone tree edges. Based on the time series backbone tree, the top node of the time series backbone tree is used as the initial node, and the recursive algorithm is used to perform a depth search on the neighbor nodes of the initial node to obtain the path node set corresponding to each neighbor node. Refer to Figure 2 , node 13 is the initial node, and nodes 20, 21, 19, 23, 16 are the neighbor nodes of the initial node. Calculate the path length between the initial node and the terminal node in each path node set, and sort the path node sets from long to short according to the path length. Each time, take out the two longest path node sets in order and mark them as the first terminal node and the second terminal node respectively. Analyze whether there is a connected edge between the first terminal node and the second terminal node, that is, whether the paths formed by the two path node sets can be connected in the environmental protection time series graph. Refer to Figure 2, there is a connected edge between node 13 and node 30. Then, it is judged whether the connected edge is a non-trunk tree edge. If there is a connected edge between the first terminal node and the second terminal node, and all the connected edges are non-trunk tree edges, then all the sequence nodes and the initial node in the path node sets corresponding to the first terminal node and the second terminal node are marked as trunk link nodes. If there is no connected edge between the first terminal node and the second terminal node, or the connected edge is not a non-trunk tree edge, then the first terminal node in the path node set with the maximum path length is removed, the second terminal node is used as the first terminal node, and the terminal node in the path node set with the third largest path length is used as the second terminal node, and the above judgment steps are continued until two path node sets that meet the judgment conditions are found.

[0115] Mark all the sequence nodes except all the trunk link nodes among all the sequence nodes as non-trunk link nodes, and analyze whether there is a connected edge between the non-trunk link nodes and multiple trunk link nodes (which can be two trunk link nodes), and whether the connected edge is a non-trunk tree edge. If there is a connected edge and all the connected edges are non-trunk tree edges, then the non-trunk link node is re-marked as a trunk link node. Traverse all the non-trunk link nodes to obtain the final set of trunk link nodes, and construct the environmental protection main loop in the environmental protection time series diagram according to all the trunk link nodes in the set of trunk link nodes.

[0116] The environmental protection main loop can display the mutual association of environmental protection nodes in the time series. By analyzing the environmental protection main loop, it can be judged whether the change trends of different sequence nodes are consistent, and whether there is a certain periodicity or regularity between different sequence nodes. In the process of similarity analysis, if the association pattern of a certain sequence node is significantly different from that of other sequence nodes, or its position in the environmental protection main loop does not match the expectation, then this sequence node may be regarded as an abnormal node. In addition, in the face of a large amount of environmental protection information, directly performing similarity analysis may be very time-consuming and complex. Constructing the environmental protection main loop can be used as a preprocessing step to help us narrow the analysis scope and focus on key nodes and association patterns, thereby improving the efficiency of abnormal node detection.

[0117] In one embodiment, extracting the environmental protection main loop in the environmental protection time series diagram based on the recursive algorithm includes the following steps:

[0118] Use the prim algorithm to divide all the sequence edges in the environmental protection time series diagram into trunk tree edges and non-trunk tree edges, and generate the time series trunk tree of the environmental protection time series diagram based on all the trunk tree edges;

[0119] Select the top node in the time series trunk tree as the initial node;

[0120] Use the recursive algorithm to perform a depth search on all the neighbor nodes of the initial node to obtain multiple path node sets;

[0121] Calculate the path length between the initial node and the terminal nodes in the set of all path nodes;

[0122] Mark the terminal nodes in the path node sets with the maximum path length and the second maximum path length as the first terminal node and the second terminal node respectively;

[0123] If there is a connected edge between the first terminal node and the second terminal node, and all the connected edges are non-trunk tree edges, then mark all the sequence nodes in the path node sets corresponding to the first terminal node and the second terminal node, as well as the initial node, as trunk link nodes;

[0124] Mark all the sequence nodes except all the trunk link nodes among all the sequence nodes as non-trunk link nodes;

[0125] For any non-trunk link node, if there are connected edges between the non-trunk link node and multiple trunk link nodes, and all the connected edges are non-trunk tree edges, then mark the non-trunk link node as a trunk link node;

[0126] Extract the environmental protection main loop in the environmental protection time series diagram based on all the trunk link nodes.

[0127] In this embodiment, the Prim algorithm is a greedy algorithm for searching the minimum spanning tree in a weighted connected graph. The minimum spanning tree refers to finding a subgraph that contains all vertices in an undirected graph, and the sum of the weights of all edges in this subgraph is the smallest, and it does not contain any loops. The specific steps for generating the time series main tree (minimum spanning tree) of the environmental protection time series diagram are as follows: Randomly select any sequence node in the environmental protection time series diagram as the starting node, count the sequence correlation between the starting node and its corresponding neighbor nodes, mark the sequence edge with the minimum sequence correlation as the trunk tree edge, then use the neighbor node connected to the trunk tree edge as the starting node, continue to count the sequence correlation between the starting node and its corresponding neighbor nodes, and continue to obtain the trunk tree edge until all the sequence nodes are traversed to obtain the final set of trunk tree edges, and construct the time series main tree of the environmental protection time series diagram according to all the trunk tree edges in the set of trunk tree edges. At the same time, mark the sequence edges except the trunk tree edges among all the sequence edges as non-trunk trees.

[0128] Select the top node in the time-series backbone tree as the initial node. The top node refers to the node located at the top of the time-series backbone tree without a parent node. Use a recursive algorithm to perform a depth search on all neighbors of the initial node. Common recursive algorithms include depth-first search (DFS), binary search algorithm, and regular matching algorithms. Taking depth-first search (DFS) as an example, first construct a set of path nodes for storing the depth search path. Starting from the initial node, start a depth search on its neighbor nodes. For each newly visited sequence node, add it to the set of path nodes and continue the in-depth search until no further in-depth search is possible, that is, all sequence nodes have been visited, to obtain the final set of path nodes.

[0129] Calculate the path lengths between the initial node and the terminal nodes in each set of path nodes, and sort the sets of path nodes in descending order according to the path lengths. The terminal node is the sequence node that is finally added to the set of path nodes. Each time, take out the two longest sets of path nodes in order and mark them as the first terminal node and the second terminal node respectively. Analyze whether there is a connecting edge between the first terminal node and the second terminal node, that is, whether the paths formed by the two sets of path nodes can be connected in the environmental protection time-series diagram. Then judge whether the connecting edge is a non-backbone tree edge. If there is a connecting edge between the first terminal node and the second terminal node and all connecting edges are non-backbone tree edges, then mark all the sequence nodes in the sets of path nodes corresponding to the first terminal node and the second terminal node and the initial node as backbone link nodes. If there is no connecting edge between the first terminal node and the second terminal node, or the connecting edge is not a non-backbone tree edge, then remove the first terminal node in the set of path nodes with the largest path length, use the second terminal node as the first terminal node, and use the terminal node in the set of path nodes ranked third in terms of path length as the second terminal node, and continue to execute the above judgment steps until two sets of path nodes that meet the judgment conditions are found.

[0130] Mark all the sequence nodes except all the backbone link nodes among all the sequence nodes as non-backbone link nodes. Analyze whether there is a connecting edge between the non-backbone link nodes and multiple backbone link nodes (which can be two backbone link nodes) and whether the connecting edge is a non-backbone tree edge. If there is a connecting edge and all connecting edges are non-backbone tree edges, then re-mark the non-backbone link node as a backbone link node. Traverse all non-backbone link nodes to obtain the final set of backbone link nodes, and construct the environmental protection backbone loop in the environmental protection time-series diagram according to all the backbone link nodes in the set of backbone link nodes.

[0131] The environmental protection backbone loop can display the mutual associations of environmental protection nodes in a time series. By analyzing the environmental protection backbone loop, it is possible to determine whether the change trends of nodes in different sequences are consistent, and whether there is a certain periodicity or regularity between nodes in different sequences. During the similarity analysis process, if the association pattern of a certain sequence node is significantly different from that of other sequence nodes, or its position in the environmental protection backbone loop does not match the expectation, then this sequence node may be regarded as an abnormal node. Additionally, in the face of a large amount of environmental protection information, directly performing similarity analysis may be very time-consuming and complex. Constructing the environmental protection backbone loop can be used as a preprocessing step to help us narrow the analysis scope and focus on key nodes and association patterns, thereby improving the efficiency of abnormal node detection.

[0132] In one embodiment, constructing an environmental protection feature matrix of environmental protection information based on the environmental protection backbone loop, and determining whether there are abnormal nodes in the environmental protection backbone loop based on the environmental protection feature matrix includes the following steps:

[0133] Screen abnormal nodes from the backbone nodes in the environmental protection backbone loop according to the first abnormal label to obtain a number of abnormal link nodes;

[0134] Analyze the node attributes of all abnormal link nodes, and extract the abnormal node features of all abnormal link nodes according to the node attributes. The abnormal node features include abnormal degree features, abnormal connectivity features, and abnormal clustering features;

[0135] Use the Laplace matrix formula to construct the backbone loop matrix of the environmental protection backbone loop, and extract the backbone loop spectral features of the environmental protection backbone loop according to the backbone loop matrix;

[0136] Analyze the node connection relationship of the environmental protection backbone loop, and extract the backbone geometric features of the environmental protection backbone loop according to the node connection relationship;

[0137] Use the graph partitioning algorithm to decompose the environmental protection backbone loop into multiple loop sub-structures;

[0138] Analyze the structural connection relationship between all loop sub-structures, and extract the connection strength features and connection path features between the loop sub-structures based on the structural connection relationship;

[0139] Construct an environmental protection feature matrix based on the abnormal node features, backbone loop spectral features, backbone geometric features, connection strength features, and connection path features;

[0140] Calculate the matrix similarity between the environmental protection feature matrix and a preset reference environmental protection feature matrix. If the matrix similarity is greater than the preset similarity threshold, it is determined that there are no abnormal nodes in the environmental protection backbone loop;

[0141] If the matrix similarity is less than or equal to the similarity threshold, it is determined that there are abnormal nodes in the environmental protection backbone loop.

[0142] In this embodiment, the backbone nodes with the first abnormal tag in the environmental protection backbone loop are marked as abnormal nodes. Calculate the node degree and node clustering coefficient of the abnormal nodes. The node degree refers to the number of edges connected to the node, reflecting the tightness of the connections around the node. The node clustering coefficient is the ratio of the actual number of edges between the neighbor nodes of the node to the maximum possible number of edges, reflecting the tightness of the connections between its neighbor nodes. The node degree and node clustering coefficient are used as the abnormal degree feature and abnormal clustering feature respectively. The abnormal connectivity feature refers to the connectivity between abnormal nodes. For any two abnormal nodes, if there is a path such that the two abnormal nodes are mutually reachable, then the two abnormal nodes are said to be connected. The abnormal connectivity feature can reflect the degree of mutual influence between abnormal nodes in the environmental protection backbone loop. The more abnormal nodes with connectivity, the higher the degree of mutual influence between abnormal nodes, and thus the greater the impact on the entire garbage disposal process. Subsequently, the punishment intensity for the enterprises corresponding to the abnormal nodes can refer to their influence degree.

[0143] The Laplacian matrix is a real symmetric matrix. Its diagonal elements represent the degrees of the backbone nodes (i.e., the number of edges connected to the backbone nodes), and the non-diagonal elements represent the connection relationships between the backbone nodes (if there is a connection edge, it is -1, otherwise it is 0). Based on this, the backbone loop matrix of the environmental protection backbone loop is constructed, and then the eigenvalue decomposition method is used to calculate the eigenvalues and eigenvectors of the backbone loop matrix. The distribution of eigenvalues can reflect the overall structural characteristics of the environmental protection backbone loop, and the eigenvectors can reflect the correlation degree and mutual influence between each backbone node in the environmental protection backbone loop. The eigenvalues and eigenvectors are integrated as the backbone loop spectrum features of the environmental protection backbone loop. The node connection relationship refers to the connection relationship between the backbone nodes. Calculate the longest path and the shortest path between the backbone nodes with connection relationships, and use the longest path and the shortest path as the backbone geometric features of the environmental protection backbone loop. The longest path can reflect the overall scale or scalability of the environmental protection backbone loop, and the shortest path reflects the connectivity and distance between the backbone nodes.

[0144] The graph partitioning algorithms include methods such as random hash point splitting, greedy algorithm, and spectral graph partitioning. Taking spectral graph partitioning as an example, spectral graph partitioning is based on the eigenvalues and eigenvectors of the graph for partitioning. It divides the vertices of the graph into two or more parts by calculating the eigenvectors of the Laplacian matrix of the graph. The spectral graph partitioning methods include ratio cut, minimum cut, maximizing graph partitioning, etc. Therefore, according to the above eigenvalues and eigenvectors, the backbone nodes with close connections in the environmental protection backbone loop can be divided into the same loop substructure, making the connections between loop substructures as sparse as possible, while the connections within the loop substructure are as dense as possible. The structural connection relationship refers to the connection relationship between loop substructures, and the connection strength feature refers to the tightness of the connection between loop substructures, which can be measured by calculating the number of common backbone links and edges between them. The connection path feature can be measured by the number of paths existing between two loop substructures.

[0145] Finally, after normalizing the abnormal node features, backbone loop spectral features, backbone geometric features, connection strength features, and connection path features, they are combined into a multi-dimensional feature vector, and the multi-dimensional feature vector is used as the row (or column) of the matrix to construct the environmental protection feature matrix. The matrix similarity between the environmental protection feature matrix and the reference environmental protection feature matrix is calculated using a similarity calculation formula. Commonly used similarity calculation formulas include Euclidean distance, Manhattan distance, cosine similarity, etc. If the matrix similarity is greater than the preset similarity threshold, it indicates that there are no abnormal nodes in the environmental protection backbone loop. If the matrix similarity is less than or equal to the similarity threshold, it indicates that there are abnormal nodes in the environmental protection backbone loop.

[0146] The reference environmental protection feature matrix is constructed through the same processing steps as the environmental protection feature matrix using the historical environmental protection information mentioned above. Among them, for the extraction steps of the abnormal degree feature, abnormal connectivity feature, and abnormal clustering feature, the abnormal environmental protection nodes with the first abnormal label in the environmental protection time series graph are mapped to the reference environmental protection time series graph constructed based on the historical environmental protection information, and the features of the corresponding nodes of the abnormal environmental protection nodes in the reference environmental protection time series graph are extracted to obtain the reference degree feature, reference connectivity feature, and reference clustering feature. This method can make the calculated matrix similarity more accurate and reliable, thus making the final abnormal node detection result more accurate.

[0147] In addition, if there are abnormal nodes in the environmental protection backbone loop, an abnormal node matrix can also be constructed according to the abnormal degree feature, abnormal connectivity feature, and abnormal clustering feature. At the same time, a reference node matrix is constructed according to the reference degree feature, reference connectivity feature, and reference clustering feature. The node position of the abnormal node is initially judged according to the similarity between the abnormal node matrix and the reference node matrix, which is used for result verification with the abnormal node recognition result of the subsequent abnormal node recognition step, and the abnormal node recognition result is initially verified, thereby further improving the reliability and accuracy of the abnormal node recognition result.

[0148] In one of the embodiments, the identification of abnormal nodes for all environmental protection nodes based on the environmental protection backbone ring includes the following steps:

[0149] Use the graph partitioning algorithm to decompose the environmental protection backbone ring into multiple ring substructures;

[0150] Identify structurally abnormal edges for the ring substructures according to sequence correlation;

[0151] If the result of structurally abnormal edge identification shows that there are abnormal edges in the ring substructure, input the abnormal edges into the pre-constructed abnormal undirected graph;

[0152] Judge whether the abnormal undirected graph after all abnormal edges are input is a bipartite graph;

[0153] If the abnormal undirected graph is a bipartite graph, use the Hungarian algorithm to analyze the maximum matching of the abnormal undirected graph, and output the maximum matching result as the abnormal node identification result;

[0154] If the abnormal undirected graph is not a bipartite graph, use the greedy strategy to identify abnormal nodes for the abnormal undirected graph to obtain the abnormal node identification result.

[0155] In this embodiment, methods such as random hash point segmentation, greedy algorithm, and spectral graph partitioning can be used to decompose the environmental protection backbone ring into multiple ring substructures. Analyze the sequence correlation between any two child nodes in the ring substructure. If the sequence correlation is less than the preset correlation threshold, the connection edge between the above two child nodes is input into the pre-constructed abnormal undirected graph. In other words, abnormal edge recording is performed in the pre-constructed abnormal undirected graph because the correlation between normal child nodes should be stable, and abnormality will lead to the break of the association, that is, the sequence correlation is low. Traverse all ring substructures to obtain an abnormal undirected graph containing all suspected abnormal edges. Then analyze whether the abnormal undirected graph is a bipartite graph. A bipartite graph is a special model in graph theory. If all vertices of the abnormal undirected graph can be divided into two non-overlapping subsets, and the two vertices associated with each edge in the graph belong to these two different vertex sets respectively, then the abnormal undirected graph is called a bipartite graph, otherwise it is not a bipartite graph.

[0156] If the abnormal undirected graph is a bipartite graph, the Hungarian algorithm is used to find an augmenting path to gradually increase the number of edges in the matching until the maximum matching is reached. In the maximum matching, the unmatched nodes (i.e., the nodes not in any matching edges) can be regarded as abnormal nodes because they do not form stable pairing relationships with other nodes and may represent abnormalities or isolated points in the abnormal undirected graph. An augmenting path is a path from the left part to the right part such that the sum of the weights of all edges on the path is the largest. Once an augmenting path is found, it is added to the current matching until no more augmenting paths can be found, and the maximum matching result is obtained. If the abnormal undirected graph is not a bipartite graph, the greedy strategy is used to iteratively select the node with the highest degree in the abnormal undirected graph as the abnormal node and remove the edges connected to it until there are no abnormal edges, and the final abnormal node recognition result is obtained. The core idea of the greedy algorithm is to make the optimal choice at each step, hoping to achieve the global optimal solution through a series of locally optimal choices.

[0157] In addition, abnormal substructure recognition is performed between loop substructures through the same abnormal recognition steps as above to obtain the abnormal substructure recognition result. The abnormal node recognition result is preliminarily verified according to the abnormal substructure recognition result and the node positions of the abnormally judged nodes above. If the abnormal node recognition result matches the abnormal substructure recognition result, that is, the abnormal nodes exceeding the preset node quantity threshold are located in the abnormal substructure, it indicates that its preliminary verification result passes. If there are abnormal nodes exceeding the preset abnormal node quantity threshold in the abnormal node recognition result and the node positions are the same as those of the preliminarily judged abnormal nodes, it can also indicate that its preliminary verification result passes.

[0158] By detecting and verifying abnormal nodes in the industrial waste treatment process through different methods, enterprises with abnormal behaviors in the industrial waste treatment process can be accurately identified, which is convenient for environmental protection-related departments to conduct key supervision and severe warnings on these enterprises with abnormal behaviors, urge the enterprises to make rectifications, and thus standardize the industrial waste recycling and treatment process. This measure can reduce the exploitation and consumption of natural resources, achieve the effective utilization and conservation of resources, and promote sustainable development. At the same time, since there are also harmful substances in industrial waste, standardizing the industrial waste recycling and treatment process can also reduce the pollution and damage of harmful substances to the environment.

[0159] In one embodiment, the abnormal information traceability based on the abnormal environmental protection information corresponding to the environmental protection nodes marked with abnormal labels is completed in the environmental protection alliance chain, including the following steps:

[0160] Receive the data interaction request of the target user and verify the user identity authentication information contained in the data interaction request;

[0161] If the user identity authentication information is verified successfully, asymmetric encryption is performed on the spam feedback information and user credibility in the data interaction request to obtain user-encrypted information;

[0162] The user-encrypted information is uploaded to the environmental protection alliance chain;

[0163] Based on the block timestamps on the environmental protection alliance chain, the data upload frequency of the data interaction request is analyzed;

[0164] Node abnormality degrees are assigned to all environmental protection nodes according to whether the environmental protection nodes are marked with the first abnormal label and / or the second abnormal label;

[0165] When the data upload frequency is greater than the preset frequency threshold or the node abnormality degree of any environmental protection node is greater than the preset abnormality degree threshold, abnormal node verification is performed on all environmental protection nodes based on the data interaction request and environmental protection information;

[0166] Based on the abnormal node verification results, abnormal information tracing is completed in the environmental protection alliance chain.

[0167] In this embodiment, after receiving the data interaction request of the target user, identity and permission verification are performed on the user identity authentication information in the data interaction request. For example, it is detected whether the target user has completed registration or whether its permissions have changed. If the user identity authentication information is verified successfully, asymmetric encryption is performed on the spam feedback information and user credibility in the data interaction request. Asymmetric encryption is an encryption method that uses two different keys (public key and private key) to encrypt and decrypt data. In asymmetric encryption, the public key can be made public to anyone and is used to encrypt data or verify digital signatures; while the private key must be strictly confidential and is only held by the key owner and is used to decrypt data or generate digital signatures. Using asymmetric encryption can ensure the data security of the spam feedback information and user credibility. Logical verification is performed on the user-encrypted information. If the verification passes, the user-encrypted information will be automatically uploaded to the chain.

[0168] Since each block on the environmental protection alliance chain contains a timestamp, this timestamp records the specific time when the block was added to the chain. By analyzing these timestamps, the time point when the data interaction request was recorded and uploaded to the chain can be traced. Further, the difference between adjacent block timestamps can be statistically analyzed, and this difference represents the time interval of data upload, that is, the data upload frequency. For example, if the timestamps of two adjacent blocks on a certain environmental protection alliance chain are T1 and T2 respectively, then the time interval between these two blocks is T2 - T1. If the time interval T2 - T1 is less than the preset time interval threshold, it means that the data upload frequency is high; conversely, if the time interval T2 - T1 is greater than or equal to the preset time interval threshold, it means that the data upload frequency is low.

[0169] Junk feedback information refers to the abnormal information about garbage disposal reported by the public. When the public discovers that an enterprise is suspected of illegal dumping of industrial waste or illegal discharge of industrial wastewater, etc., they can report it to the environmental protection supervision platform and provide relevant clues. The user credibility is assigned according to the accuracy of the junk feedback information of the target user. When the target user successfully uploads junk feedback information to the environmental protection supervision platform, the platform will give integral rewards to the target user. If the junk feedback information uploaded by the target user is subsequently confirmed as correct by the environmental protection department, the platform will give multiple integral rewards to the target user and increase the user credibility of the target user. The user credibility can be used to measure the accuracy of the junk feedback information uploaded by the target user.

[0170] The node abnormality degree is assigned to all environmental protection nodes according to whether the environmental protection node is marked with the first abnormal label and the second abnormal label. If the environmental protection node is marked with both the first abnormal label and the second abnormal label, the node abnormality degree of this environmental protection node is the highest. If the environmental protection node is only marked with the second abnormal label, the node abnormality degree of this environmental protection node is the second highest. If the environmental protection node is only marked with the first abnormal label, the node abnormality degree of this environmental protection node is the lowest. When the data uploading frequency is greater than the preset frequency threshold or the node abnormality degree of any environmental protection node is greater than the preset abnormality degree threshold, the abnormal node verification is performed on all environmental protection nodes based on the data interaction request and environmental protection information. Subsequently, starting from the verified abnormal node, all subsequent uploaded environmental protection information of this abnormal node will be traced. At the same time, the abnormal environmental protection information will be compared with the environmental protection information uploaded at the adjacent time node to identify continuous abnormalities, count their abnormal frequencies, and confirm whether it is the first abnormality or repeated violation. Based on all the above, an abnormal report is generated, and finally the abnormal report is encrypted and stored on the chain. At the same time, the environmental protection department will pursue the enterprise corresponding to this environmental protection node according to the abnormal report and urge it to rectify.

[0171] Due to the large population base and wide distribution of the public, some problems that are difficult for professional supervision agencies to reach can be discovered, which can help the environmental protection department quickly locate and investigate abnormal behaviors in the industrial waste treatment process and accurately locate the abnormal behaviors and trace the responsible enterprises. By combining the junk feedback information uploaded by the public with objective environmental protection information, the supervision of the industrial waste treatment process is realized from different directions. While improving the effectiveness of environmental protection work and the work efficiency of relevant departments, by encouraging the public to actively report abnormal garbage disposal behaviors, more people can be involved in environmental protection work, forming a good atmosphere of pooling wisdom and joint supervision. This not only helps to improve the effectiveness of environmental protection work, but also enhances the public's environmental awareness and sense of responsibility, helps the public integrate environmental awareness into daily life, and promotes the public to form an environmental awareness of actively classifying and recycling domestic waste in daily life.

[0172] In one embodiment, the abnormal node verification of all environmental protection nodes based on the data interaction request and environmental protection information includes the following steps:

[0173] Assign weights to the corresponding garbage feedback information according to the user credibility, and assign feedback abnormality degrees to each environmental protection node according to the garbage feedback information with weights assigned;

[0174] Fusion the feedback abnormality degree and the node abnormality degree with weights to obtain the comprehensive node abnormality degree;

[0175] Extract the environmental protection transaction relationships among all environmental protection nodes in the industrial waste treatment process;

[0176] Take the environmental protection nodes as graph nodes, the comprehensive node abnormality degree and environmental protection information as node features, and the environmental protection transaction relationship as edge features to construct an environmental protection transaction graph;

[0177] Connect the corresponding environmental protection nodes in the environmental protection transaction graphs of adjacent time nodes with time edges to obtain an environmental protection transaction spatio-temporal graph;

[0178] Based on the graph neural network model and the spatio-temporal recurrent gate, construct an abnormal node verification model. The abnormal node verification model includes a spatial feature extraction module, a temporal feature extraction module, and an abnormal node verification module. The spatial feature extraction module is constructed based on the enhanced graph convolutional network model, and the temporal feature extraction module is constructed based on the spatio-temporal recurrent gate;

[0179] Use the pre-constructed training set to train the abnormal node verification model;

[0180] Input the environmental protection transaction spatio-temporal graph into the trained abnormal node verification model to extract spatial features and temporal features, and obtain the environmental protection transaction spatial features and environmental protection transaction temporal features;

[0181] Output the abnormal node verification result through the abnormal node verification module after dimension conversion and information integration of the environmental protection transaction spatial features and environmental protection transaction temporal features.

[0182] In this embodiment, first, weights are assigned to the garbage feedback information uploaded by the target user according to the user credibility of the target user. The higher the user credibility, the higher the assigned weight. According to the number of times each environmental protection node is feedback by all target users and the weights of the corresponding garbage feedback information, a feedback anomaly degree is assigned to each environmental protection node. After assigning weight coefficients to the feedback anomaly degree and the node anomaly degree, data fusion is performed to obtain the comprehensive node anomaly degree of each environmental protection node. The environmental protection transaction relationship refers to whether there is a transaction relationship between environmental protection nodes. For example, within a certain period of time, a certain factory transports the generated industrial garbage to a certain garbage recycling enterprise and entrusts the garbage recycling enterprise to carry out garbage recycling and treatment work. Then, there is an environmental protection transaction relationship between the factory and the garbage recycling enterprise. In another period of time, if the factory entrusts other garbage recycling enterprises to carry out garbage recycling and treatment work, then there is no environmental protection transaction relationship between the factory and the garbage recycling enterprise. Taking the environmental protection node as a graph node, the comprehensive node anomaly degree and environmental protection information as node features, and the environmental protection transaction relationship as edge features, an environmental protection transaction graph is constructed. Since the environmental protection transaction relationship between different environmental protection nodes is dynamically changing, the environmental protection transaction graph also changes dynamically, including node addition, node deletion, edge addition, and edge deletion. For example, if a certain factory goes bankrupt due to poor management, new graph nodes will appear. If the entrustment relationship between two enterprises is terminated, the corresponding edge needs to be deleted.

[0183] The abnormal node verification model is constructed based on the graph neural network model and the spatio-temporal recurrent gate. The graph neural network model is constructed based on the enhanced graph convolutional network model and the spatio-temporal recurrent gate, and includes a spatial feature extraction module, a temporal feature extraction module, and an abnormal node verification module. The spatial feature extraction module is constructed based on the enhanced graph convolutional network model and is used to extract the topological spatial features of the environmental protection transaction graph. Among them, the definition formula of the enhanced graph convolution is as follows:

[0184]

[0185] Among them, represents the learnable coefficient matrix of the k-order neighbors of the (l - 1)-th layer of graph convolution, W (l-1) is the coefficient matrix of the (l - 1)-th layer, is the k-order adjacency matrix, σ is the activation function, and K is the total order.

[0186] The enhanced graph convolutional network model is a model that combines reinforcement learning and graph convolutional neural networks. Using the enhanced graph convolutional network model can increase the receptive field range of the convolutional operation, thereby better aggregating spatial features.

[0187] The temporal feature extraction module is constructed based on the spatio-temporal recurrent gate to ensure the balance of feature information in the spatio-temporal distribution. The temporal feature extraction module includes a time stability gate θ t and a spatial stability gate Responsible for capturing the dependency information of the dynamic environmental protection transaction graph in time and space respectively, and its expression is as follows:

[0188] θ t =σ(TX i +Vψ t-1 )

[0189]

[0190] where X i is the feature vector at time t, T and V are parameter matrices, ψ t-1 is the hidden state at t - 1, diag is the diagonalization operation, σ is the activation function, and g t * is the graph convolution of the single - layer structure.

[0191] Use the pre - constructed training set to perform model iterative training on the abnormal node verification model, continuously adjust the model parameters until the preset maximum number of iterations is reached. Input the environmental protection transaction spatio - temporal graph into the trained abnormal node verification model to extract spatial features and temporal features. Input the obtained environmental protection transaction spatial features and environmental protection transaction temporal features into the abnormal node verification module, and obtain the final abnormal node verification result after passing through a fully - connected layer and an activation function.

[0192] The environmental protection relevant department can punish the enterprises with abnormal behaviors according to the final abnormal node verification result, urge the relevant enterprises to rectify the industrial waste treatment process, and will conduct key supervision on the relevant enterprises with abnormal behaviors during the subsequent supervision process. This measure is conducive to reducing the exploitation and consumption of natural resources, realizing the effective utilization and conservation of resources, and promoting sustainable development. At the same time, since there are also harmful substances in industrial waste, standardizing the industrial waste recycling and treatment process can also reduce the pollution and damage of harmful substances to the environment.

[0193] This application also discloses a smart environmental protection data tracking system based on blockchain technology, which is characterized by including:

[0194] A memory configured to store instructions; and

[0195] A processor configured to call instructions from the memory and be able to implement the smart environmental protection data tracking method based on blockchain technology as described above when executing the instructions.

[0196] Among them, the processor may adopt a central processing unit (CPU). Of course, according to the actual usage, other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. may also be adopted. The general-purpose processor may adopt a microprocessor or any conventional processor, etc. The present application does not make any restrictions on this.

[0197] Among them, the memory may be an internal storage unit of the computer device, for example, the hard disk or memory of the computer device, or may also be an external storage device of the computer device, for example, a plug-in hard disk, a smart media card (SMC), a secure digital card (SD) or a flash card (FC), etc. equipped on the computer device. Moreover, the memory may also be a combination of the internal storage unit and the external storage device of the computer device. The memory is used to store computer programs and other programs and data required by the computer device. The memory may also be used to temporarily store the data that has been output or will be output. The present application does not make any restrictions on this.

[0198] The embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to make a machine execute the above-mentioned method for tracking intelligent environmental protection data based on blockchain technology.

[0199] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0200] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0201] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction means that implements the function specified in one or more of the processes and / or blocks Figure 1 in one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0202] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing steps for implementing the function specified in one or more of the processes and / or blocks Figure 1 in one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0203] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0204] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM) and / or non-volatile memory such as read only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

[0205] Computer-readable media includes both permanent and non-permanent, removable and non-removable media implemented by any method or technology for storing information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0206] It should also be noted that the term "including", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, commodity or device comprising the element.

[0207] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A smart environmental protection data tracking method based on blockchain technology, characterized in that: The method comprises the following steps: Create an environmental protection alliance chain for industrial waste treatment processes based on blockchain technology; Obtain environmental protection information of each environmental protection node in the industrial waste treatment process, and complete the automatic chaining of all environmental protection information according to the consensus mechanism preset in the environmental protection alliance chain; Trigger the smart contract based on preset information compliance logic in the environmental protection alliance chain to verify all environmental protection information; If all environmental information verification passes, off-chain abnormal node detection will be performed on the environmental protection node based on the environmental protection information, and abnormal information tracing will be completed in the environmental protection alliance chain according to the abnormal node detection results.

2. The method according to claim 1, characterized in that: The off-chain abnormal node detection of environmental protection nodes based on environmental protection information and the abnormal information tracing in the environmental protection alliance chain according to the abnormal node detection results include the following steps: Based on the environmental protection information and using a threshold method, a first abnormal label is marked for the environmental protection node; Use the correlation algorithm to perform correlation analysis on each environmental protection node, build an environmental protection time series diagram according to the correlation analysis results, and mark the second abnormal label for the environmental protection node based on the environmental protection time series diagram; Abnormal information tracing is completed in the environmental protection alliance chain based on abnormal environmental protection information corresponding to the environmental protection nodes marked with abnormal labels.

3. The method according to claim 2, characterized in that Environmental protection information includes garbage weight information, garbage disposal information, garbage type information, garbage transportation information and garbage time information.

4. The method according to claim 2, characterized in that: The method of using a correlation algorithm to perform correlation analysis on each environmental protection node, constructing an environmental protection time sequence diagram according to the correlation analysis result, and marking each environmental protection node as a second abnormal node based on the environmental protection time sequence diagram includes the following steps: Perform time series analysis on environmental protection information to obtain the environmental protection sequence of each environmental protection node; The correlation algorithm is used to calculate the serial correlation between all environmental protection sequences; An environmental protection time series graph is constructed with environmental protection sequences as sequence nodes and sequence correlations as sequence edges; Extract the environmental protection trunk ring in the environmental protection timing diagram based on the recursive algorithm; An environmental protection feature matrix of environmental protection information is constructed based on the environmental protection trunk ring, and based on the environmental protection feature matrix, it is determined whether there are abnormal nodes in the environmental protection trunk ring; If there are abnormal nodes in the environmental protection trunk ring, abnormal node identification is performed on all environmental protection nodes based on the environmental protection trunk ring, and a second abnormal label is marked for the environmental protection node based on the abnormal node identification result.

5. The method according to claim 4, characterized in that The method of extracting the environmental protection trunk ring in the environmental protection timing diagram based on the recursive algorithm comprises the following steps: The prim algorithm is used to divide all sequence edges in the environmental timing graph into trunk tree edges and non-trunk tree edges, and the timing trunk tree of the environmental timing graph is generated based on all trunk tree edges; Select the top node in the time series trunk tree as the initial node; Use a recursive algorithm to perform a deep search on all neighbor nodes of the initial node to obtain multiple path node sets; Calculate the path length from the initial node to the terminal node in the set of all path nodes; Mark the terminal nodes in the path node set with the largest path length and the path node set with the second largest path length as the first terminal node and the second terminal node respectively; If there is a connected edge between the first terminal node and the second terminal node, and the connected edges are all non-trunk tree edges, then all sequence nodes and the initial node in the path node set corresponding to the first terminal node and the second terminal node are marked as trunk ring nodes; Mark all sequence nodes except all backbone nodes in all sequence nodes as non-backbone nodes; For any non-backbone ring node, if there are connected edges between the non-backbone ring node and multiple backbone ring nodes, and the connected edges are all non-backbone tree edges, then the non-backbone ring node is marked as a backbone ring node; The environmental protection trunk ring in the environmental protection timing diagram is extracted based on all the trunk ring nodes.

6. The method according to claim 4, characterized in that The steps of constructing an environmental protection feature matrix of environmental protection information based on the environmental protection trunk ring and judging whether there is an abnormal node in the environmental protection trunk ring based on the environmental protection feature matrix include the following steps: According to the first abnormal label, the trunk ring nodes in the environmental protection trunk ring are screened for abnormal nodes to obtain several abnormal ring nodes; Analyze the node attributes of all abnormal ring nodes, and extract the abnormal node features of all abnormal ring nodes according to the node attributes. The abnormal node features include abnormal degree features, abnormal connectivity features and abnormal clustering features; The trunk ring matrix of the environmental protection trunk ring is constructed by using the Laplace matrix formula, and the trunk ring spectrum characteristics of the environmental protection trunk ring are extracted according to the trunk ring matrix; Analyze the node connection relationship of the environmental protection trunk ring, and extract the trunk geometric features of the environmental protection trunk ring according to the node connection relationship; The environmental protection backbone ring is decomposed into multiple ring substructures using the graph partitioning algorithm; Analyze the structural connection relationship between all the ring substructures, and extract the connection strength characteristics and connection path characteristics between the ring substructures based on the structural connection relationship; An environmental protection feature matrix is ​​constructed based on abnormal node features, trunk ring spectrum features, trunk geometry features, connection strength features, and connection path features; Calculate the matrix similarity between the environmental protection feature matrix and the preset reference environmental protection feature matrix. If the matrix similarity is greater than the preset similarity threshold, it is determined that there is no abnormal node in the environmental protection trunk ring. If the matrix similarity is less than or equal to the similarity threshold, it is determined that there are abnormal nodes in the environmental protection backbone ring.

7. The method according to claim 4, characterized in that The abnormal node identification of all environmental protection nodes based on the environmental protection trunk ring includes the following steps: The environmental protection backbone ring is decomposed into multiple ring substructures using the graph partitioning algorithm; Identify structurally abnormal edges of loop substructures based on sequence correlation; If the structural abnormal edge identification result shows that the ring substructure has abnormal edges, the abnormal edges are input into the pre-constructed abnormal undirected graph; Determine whether the abnormal undirected graph after completing all abnormal edge inputs is a bipartite graph; If the abnormal undirected graph is a bipartite graph, the Hungarian algorithm is used to analyze the maximum matching of the abnormal undirected graph, and the maximum matching result is output as the abnormal node identification result; If the abnormal undirected graph is not a bipartite graph, the greedy strategy is used to identify abnormal nodes on the abnormal undirected graph to obtain the abnormal node identification result.

8. The method according to claim 2, characterized in that: The abnormal information tracing based on the abnormal environmental information corresponding to the environmental node marked with the abnormal label in the environmental alliance chain includes the following steps: Receive a data interaction request from a target user and verify the user identity authentication information contained in the data interaction request; If the user identity authentication information is verified, the spam feedback information and user credibility in the data interaction request are asymmetrically encrypted to obtain the user encrypted information; Upload user encrypted information to the environmental protection alliance chain; Analyze the frequency of data on-chain for data interaction requests based on the block timestamp on the environmental protection alliance chain; Allocate node abnormality degrees to all environmental protection nodes according to whether the environmental protection nodes are marked with the first abnormal label and / or the second abnormal label; When the frequency of data uplink is greater than the preset frequency threshold or the node abnormality of any environmental protection node is greater than the preset abnormality threshold, all environmental protection nodes are checked for abnormal nodes based on the data interaction request and environmental protection information; Based on the abnormal node verification results, abnormal information tracing is completed in the environmental protection alliance chain.

9. The method according to claim 8, characterized in that The abnormal node verification of all environmental protection nodes based on the data interaction request and environmental protection information includes the following steps: Assign weights to the corresponding spam feedback information according to the user's credibility, and assign feedback abnormality to each environmental protection node according to the spam feedback information that has completed the weight assignment; The feedback abnormality and node abnormality are weighted and fused to obtain the node comprehensive abnormality. Extract the environmental protection transaction relationship between all environmental protection nodes in the industrial waste treatment process; The environmental protection nodes are used as graph nodes, the node comprehensive abnormality and environmental protection information are used as node features, and the environmental protection transaction relationship is used as the edge feature to construct an environmental protection transaction graph; Connect the corresponding environmental protection nodes in the environmental protection transaction graph of adjacent time nodes with time edges to obtain an environmental protection transaction space-time graph; An abnormal node verification model is constructed based on the graph neural network model and the space-time recurrent gate. The abnormal node verification model includes a spatial feature extraction module, a temporal feature extraction module and an abnormal node verification module. The spatial feature extraction module is constructed based on the enhanced graph convolutional network model, and the temporal feature extraction module is constructed based on the space-time recurrent gate. Use the pre-built training set to train the abnormal node verification model; The environmental protection transaction spatiotemporal graph is input into the trained abnormal node verification model to extract spatial and temporal features, and the environmental protection transaction spatial features and environmental protection transaction temporal features are obtained; The abnormal node verification module performs dimension conversion and information integration on the environmental protection transaction space characteristics and environmental protection transaction time characteristics, and then outputs the abnormal node verification results.

10. A smart environmental protection data tracking system based on blockchain technology, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call instructions from a memory and to implement a smart environmental data tracking method based on blockchain technology according to any one of claims 1 to 9 when executing the instructions.

Images