A full-process traceability system for aquatic products based on blockchain technology and its implementation method

Abstract

The present invention discloses a full-process traceability system for aquatic products based on blockchain technology and its implementation method, which belongs to the field of quality control technology. The system includes a data acquisition layer composed of an environmental monitoring unit, a feeding monitoring unit and an RFID identification unit for collecting aquaculture environment data; an edge computing layer including an edge server and an industrial control computer, which pre-process the collected data; a blockchain layer constructed using the Hyperledger Fabric platform to achieve data tamper-proof storage and automatically verify data authenticity through smart contracts; an application layer including aquaculture management subsystem, a processing management subsystem, a logistics management subsystem and a sales management subsystem, and a multi-level authorization mechanism to manage data access rights. The present invention adopts the above-mentioned system and its implementation method to solve problems such as data falsification, high tracing costs and information opacity, and realize full-chain credible traceability of aquatic product quality and safety.

Description

Technical Field

[0001] The present invention relates to the field of quality control technology, and in particular to a full-process traceability system for aquatic products based on blockchain technology and an implementation method thereof. Background Art

[0002] To strengthen the control of aquatic product quality and safety at the source and achieve information-based, digital management across the entire supply chain, it is imperative to establish a traceability system for aquatic product quality and safety that covers every link, including breeding, processing, warehousing, logistics, and sales. Traditional traceability methods rely primarily on centralized databases to store product information, which presents challenges such as data silos, susceptibility to tampering, and high traceability costs. The recent rise of blockchain technology offers a new approach to addressing these challenges.

[0003] Blockchain is a decentralized distributed ledger technology that uses cryptographic principles to connect and store data in blocks across multiple nodes. This allows for trusted data records and consensus without the need for centralized management. Blockchain's decentralized, tamper-resistant, traceable, and highly transparent characteristics are highly compatible with the needs of food safety traceability. Applying blockchain technology to aquatic product quality and safety traceability is expected to address industry pain points such as information silos, reduce counterfeiting, and increase transparency, providing strong support for comprehensive quality control from pond to table.

[0004] Currently, both domestic and international research has begun exploring the application of blockchain technology in food traceability, with some success. However, in the aquatic product sector, mature, full-chain blockchain traceability solutions are still lacking, especially when combined with smart devices and covering multiple scenarios. Therefore, developing a blockchain-based intelligent traceability system applicable to all links in the aquatic product supply chain is crucial for ensuring product quality and safety and promoting the healthy development of the industry. Summary of the Invention

[0005] The purpose of the present invention is to provide a full-process traceability system for aquatic products based on blockchain technology and its implementation method to solve the problems existing in the background technology.

[0006] To achieve the above objectives, the present invention provides a full-process traceability system for aquatic products based on blockchain technology, comprising:

[0007] The data collection layer consists of an environmental monitoring unit, a feeding monitoring unit, and an RFID identification unit, which are used to automatically collect aquaculture environment data;

[0008] The edge computing layer includes edge servers and industrial control computers, which are used to pre-process the collected data;

[0009] The blockchain layer is built on the Hyperledger Fabric platform to achieve tamper-proof data storage and automatically verify data authenticity through smart contracts;

[0010] The application layer includes breeding management subsystem, processing management subsystem, logistics management subsystem and sales management subsystem. The application layer adopts a multi-level authorization mechanism to manage data access rights.

[0011] Preferably, the unit contents of the environmental monitoring unit, the feeding monitoring unit and the RFID identification unit are as follows:

[0012] The environmental monitoring unit includes a dissolved oxygen sensor, a pH sensor, and a temperature sensor. The dissolved oxygen sensor, pH sensor, and temperature sensor are all connected to the industrial control computer via the RS485 bus. Each sensor uses a stainless steel housing with a waterproof rating of IP67. Each sensor probe is vertically installed 30-50 cm below the water surface of the aquaculture pond.

[0013] The feeding monitoring unit includes a feeding amount sensor and a feeding time recorder. The feeding amount sensor is a weighing sensor with a range of 0-100kg, installed at the bottom of the hopper of the feeding equipment, connected in series with the feeding time recorder and connected to the industrial control computer via the RS485 bus;

[0014] The RFID identification unit includes an RFID reader and an RFID electronic tag. The RFID reader uses a high-frequency reader with an operating frequency of 13.56MHz and a reading distance of 0-10cm. It is connected to the industrial control computer through the TCP / IP protocol. The RFID electronic tag uses a waterproof package and has a unique identification code for recording aquatic product breeding batch information.

[0015] Preferably, the specific structure of the feeding monitoring unit is:

[0016] The feed rate sensor adopts a parallel beam structure, the main body is made of 304 stainless steel, the sensitivity is 2mV / V±0.1%, the accuracy level is C3, and it has a built-in temperature compensation circuit. The operating temperature range is -20℃ to +60℃. It is fixed to the four corners of the hopper bottom with four M12 bolts, and the inclination angle is no more than 2°.

[0017] The feeding time recorder uses an industrial-grade single-chip microcomputer controller with a main frequency of 100MHz, a real-time clock chip, and a time accuracy of ±1s / day. It is equipped with a 3.5-inch LCD display, an IP65 protection grade, and uses a 24V DC power supply. The built-in UPS backup power supply can continuously work for 4 hours.

[0018] The feeding amount sensor and the feeding time recorder are connected by a 4-core shielded cable with an outer diameter of 8 mm and a length of no more than 5 meters. The terminal blocks are crimp terminals, and the signal line uses a twisted pair structure to suppress common-mode interference. The RS485 bus uses a T-type connector to achieve daisy-chain connection, a 120Ω terminal resistor is installed at the end of the bus, and the communication rate is set to 9600 bps.

[0019] Preferably, the edge server and industrial control computer contents are as follows:

[0020] The industrial control computer uses the Linux real-time operating system and collects sensor data through a data acquisition program. The data acquisition program includes a data verification module, a data filtering module, and a data compression module. The data verification module uses a CRC checksum algorithm to ensure data integrity, the data filtering module uses a median filter algorithm to remove outliers, and the data compression module uses a lossless compression algorithm to reduce data transmission volume.

[0021] The edge server is deployed in a Docker containerized environment and runs an edge computing framework, which includes a data parsing engine, a feature extraction engine, and a data fusion engine. The data parsing engine parses the collected raw data into a standard format, the feature extraction engine uses a sliding time window method to extract data features, and the data fusion engine uses a Kalman filter algorithm to fuse multi-source data.

[0022] The edge server interacts with the blockchain layer through message queues, and uses the publish / subscribe model to achieve asynchronous data transmission. The message queue is implemented using RabbitMQ, and SSL encryption is used to ensure data transmission security.

[0023] Preferably, the specific implementation architecture of the industrial control computer is:

[0024] The Linux real-time operating system uses the RTLinux kernel and is configured with the real-time priority scheduling policy SCHED_FIFO, with an interrupt response time of less than 100 microseconds. The system adopts a dual-partition structure, in which the system partition is mounted in read-only mode, and the data partition uses a journaling file system to ensure data integrity.

[0025] The data acquisition program is designed based on a multi-threaded architecture, including device driver threads, data acquisition threads, and data processing threads:

[0026] The device driver thread is responsible for managing the communication interfaces of RS485 devices and TCP / IP devices, and adopts the epoll mechanism to achieve high concurrent IO processing, supporting concurrent access of up to 1,000 devices.

[0027] The data acquisition thread adopts polling mode to perform regular sampling on each sensor. The sampling period can be configured in the range of 100ms-1000ms, and a double buffer mechanism is used to avoid data loss.

[0028] The data processing thread receives collected data through a pipeline mechanism and uses zero-copy technology to improve data processing efficiency. The processed data interacts with the application through shared memory.

[0029] Industrial control computers also include a watchdog program that monitors the system's operating status and automatically restarts related processes when a system anomaly is detected. It also records system logs through the syslog mechanism and supports unified management of remote log servers.

[0030] Preferably, the Hyperledger Fabric platform adopts a consortium chain architecture, including an ordering node cluster, a peer node cluster, and a channel management module; the ordering node cluster uses the Raft consensus algorithm to maintain blockchain ledger consistency, the peer node cluster is responsible for maintaining the chain code and state database, and the channel management module is responsible for managing data isolation between different organizations;

[0031] The specific implementation architecture of the Hyperledger Fabric platform is as follows:

[0032] The sorting node cluster consists of a master node and several slave nodes, which communicate with each other through the Raft consensus protocol:

[0033] The master node is responsible for receiving transaction requests and sorting them by timestamp. It uses a batch processing mechanism to package transactions into blocks, with a block size threshold set at 2MB or 500 transactions. Slave nodes maintain synchronization with the master node through a heartbeat mechanism with a heartbeat interval of 100ms. When a master node failure is detected, an election mechanism is automatically triggered. Nodes communicate using the gRPC protocol, using TLS1.3 encrypted channels, and support up to 5 nodes running concurrently.

[0034] The peer node cluster adopts a layered architecture design:

[0035] The consensus layer is responsible for verifying transactions and maintaining ledgers, using LevelDB to store transaction data and supporting data storage of up to 10TB. The chaincode layer is responsible for executing smart contracts, using Docker containers to isolate the runtime environment and supporting smart contracts written in Golang, Java, and Node.js. The state database layer is implemented using CouchDB, supporting JSON format data storage and rich query capabilities, with data synchronization latency of no more than 100ms.

[0036] The channel management module adopts policy-based access control:

[0037] Channel configuration uses MSP (Member Service Provider) identity management and supports X.509 certificate authentication. Channel policies are defined using a signature policy language that supports combinational logic such as AND, OR, and NOutOf. Data isolation is implemented based on a multi-ledger mechanism, with each channel maintaining independent ledger data.

[0038] Smart contracts include data verification contracts, storage contracts, and access control contracts. The data verification contract uses a digital signature algorithm and timestamp verification mechanism to ensure the credibility of the data source. The storage contract uses a Merkle tree structure and IPFS distributed storage to prevent data tampering. The access control contract implements fine-grained permission management based on the ABAC (Attribute-Based Access Control) model.

[0039] It also includes a blockchain network management module, which is responsible for node certificate management, blockchain network monitoring and performance optimization. Certificate management adopts the PKI architecture, network monitoring is implemented using Prometheus, and performance optimization is achieved by dynamically adjusting block size and transaction concurrency.

[0040] Preferably, the contents of the breeding management subsystem, processing management subsystem, logistics management subsystem and sales management subsystem are as follows:

[0041] The aquaculture management subsystem adopts a B / S architecture and includes an aquaculture environment monitoring module, a feed delivery management module, and a growth cycle management module. The aquaculture environment monitoring module uses WebSocket to push environmental data in real time, the feed delivery management module uses a deep learning algorithm to achieve intelligent feeding amount prediction, and the growth cycle management module uses GIS technology to achieve visual management of aquaculture ponds.

[0042] The processing management subsystem and logistics management subsystem adopt a microservice architecture and are implemented using the Spring Cloud framework. The processing management subsystem includes a processing process tracking module, a quality inspection module, and an inventory management module; the logistics management subsystem includes a transportation monitoring module, a temperature monitoring module, and a distribution management module. Data exchange between these modules is carried out through RESTful APIs.

[0043] The sales management subsystem integrates mobile payment interfaces and traceability code query functions, uses the OAuth2.0 protocol for user authentication, and verifies product traceability information through blockchain smart contracts. The multi-level authorization mechanism is implemented based on the RBAC (role-based access control) model, including four permission levels: system administrator, enterprise administrator, operator, and consumer. Different permission levels correspond to different data access scopes and operation permissions.

[0044] Preferably, the contents of the breeding environment monitoring module, the feed delivery management module and the growth cycle management module are as follows:

[0045] The aquaculture environment monitoring module adopts a layered architecture design:

[0046] The data collection layer pushes real-time data via a WebSocket server, uses a heartbeat detection mechanism to maintain connection status, and supports up to 1,000 concurrent connections. The data display layer uses the ECharts chart library to visualize data, supporting the linkage display of multiple indicators such as temperature, pH value, and dissolved oxygen, with a refresh rate configurable from 1 to 60 seconds. The alarm management layer uses a rule engine to implement threshold alarms and supports multiple channels such as SMS, email, and WeChat to push alarm information.

[0047] The feed delivery management module is designed based on a deep learning framework:

[0048] The data preprocessing layer uses a sliding time window to segment data, with a window size of 24 hours and a step size of 1 hour. The model training layer uses an LSTM neural network structure, consisting of three layers of LSTM units, 128 neurons per layer, and a dropout rate of 0.5. The prediction and analysis layer uses the Adam optimizer with a learning rate of 0.001 to predict the optimal feeding amount based on historical feeding data and environmental parameters.

[0049] The growth cycle management module is implemented based on GIS technology:

[0050] The map engine layer uses the OpenLayers framework to support pond boundary drawing and spatial analysis; the data management layer uses PostGIS extension to implement spatial data storage, supporting latitude and longitude and plane coordinate systems; the business function layer integrates the visualization of attribute information such as pond number, breeding species, and stocking density.

[0051] Preferably, the specific implementation architecture of the sales management subsystem is:

[0052] The mobile payment interface adopts a unified payment gateway design:

[0053] The payment gateway layer integrates the SDKs of mainstream payment channels such as Alipay and WeChat Pay, and adopts the adapter model to achieve interface unification. The transaction processing layer adopts the distributed transaction framework TCC model to ensure the atomicity of the payment process. The reconciliation and settlement layer adopts a scheduled task method, performing transaction reconciliation at 2:00 am every day and supporting automatic error verification.

[0054] The traceability code query function is based on QR code technology:

[0055] The QR code generation layer uses the QR Code encoding standard, with a Level Q error correction level and supports 2048 bytes of information storage. The data encoding layer uses the Base64 encoding format, which contains information such as product ID, production batch, and timestamp. The parsing and verification layer verifies the authenticity and validity of the code by calling the blockchain interface through the REST API.

[0056] OAuth2.0 authentication adopts a layered design:

[0057] The authentication service layer supports password mode and authorization code mode. The access token is valid for 2 hours, and the refresh token is valid for 30 days. The resource service layer uses JWT format tokens and is signed using the RSA-256 algorithm. The user management layer is connected to the blockchain smart contract to realize the on-chain storage of user identity information.

[0058] A method for implementing a full-process traceability system for aquatic products based on blockchain technology, comprising the following steps:

[0059] S1. Automatically collect multi-source heterogeneous data such as water quality parameters, feeding records, and breeding batches during the breeding process through equipment such as environmental monitoring units, feeding monitoring units, and RFID identification units, realizing automation of data collection, with an automation rate of over 95%;

[0060] S2. Industrial control computers serve as edge gateways for data collection, performing pre-processing such as verification, filtering, and compression on the raw collected data. Edge servers further perform data cleaning, feature extraction, and multi-source fusion, improving data quality while minimizing data transmission volume and reducing the storage and computing burden of the blockchain.

[0061] S3. Upload the integrated and optimized structured data to the blockchain network in a unified data format through the API interface provided by the smart contract for evidence storage, making the data traceable and tamper-proof throughout its life cycle;

[0062] S4. Each business application subsystem (including breeding, processing, warehousing, logistics, etc.) uses blockchain data and combines cutting-edge technologies such as big data analysis and artificial intelligence to conduct real-time monitoring of production and operation activities, intelligent early warning, and process optimization, thereby improving the level of refined quality management and supply chain collaboration efficiency;

[0063] S5. Managers can use the Web management backend, and consumers can use mobile apps and other terminals to enter the product traceability code to query the quality and safety information of each link such as product production, processing, warehousing, and transportation in real time, thereby achieving full visualization and control of product quality.

[0064] Therefore, the present invention adopts the above-mentioned blockchain-based aquatic product full-process traceability system and implementation method, which has the following beneficial effects:

[0065] (1) Introducing blockchain technology into the field of aquatic product quality and safety traceability, and utilizing the blockchain's decentralization, tamper-proofing, traceability, and high transparency, it solves the industry's pain points such as data falsification, high traceability costs, and information opacity in traditional traceability systems from the source, and realizes information sharing and business collaboration in all links of the aquatic product supply chain;

[0066] (2) By combining automatic identification technologies such as RFID with IoT sensing devices, the automated collection of key quality data during the aquaculture process has been achieved, with an automation rate of over 95%. By giving aquatic products an electronic "identity card", a digital archive covering the entire life cycle has been constructed, greatly improving traceability efficiency and accuracy.

[0067] (3) Introducing the concept of edge computing to perform hierarchical data processing. By deploying edge gateways and edge servers at the source of data generation, data can be processed locally and key features can be extracted. This not only reduces the pressure of data transmission, but also reduces the storage and computing burden of the blockchain, making the blockchain system lighter and greatly improving the robustness and scalability of the system.

[0068] (4) The application layer integrates cutting-edge technologies such as big data and artificial intelligence, and realizes the refinement and intelligence of production management through in-depth analysis of on-chain data and data-driven intelligent decision-making. In particular, in the breeding process, data modeling is used to optimize feeding strategies and warn of disease risks, making the breeding process more scientific and standardized, significantly improving breeding efficiency and yield rate;

[0069] (5) The introduction of blockchain and smart contracts provides a new solution for identity authentication and permission management for all supply chain participants. Based on smart contracts and fine-grained authorization mechanisms, it not only effectively protects business secrets but also promotes data sharing and collaboration across organizations, embodying the perfect fusion of centralized control and decentralized incentives.

[0070] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 Schematic diagram of the system architecture of the present invention;

[0072] Figure 2 This is a schematic diagram of the structure of the RFID identification unit of the present invention;

[0073] Figure 3 This is a flow chart of edge computing layer data processing in the present invention;

[0074] Figure 4 This is a schematic diagram of the blockchain layer architecture of the present invention;

[0075] Figure 5 The figure is a flow chart of the implementation method of the present invention. DETAILED DESCRIPTION

[0076] The following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but rather merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort shall fall within the scope of protection of the present invention.

[0077] See also Figure 1 , a full-process traceability system for aquatic products based on blockchain technology, including a data collection layer, edge computing layer, blockchain layer and application layer designed from bottom to top.

[0078] The data acquisition layer at the bottom layer is responsible for automating and digitizing data. This layer includes an environmental monitoring unit, a feeding monitoring unit, and an RFID identification unit.

[0079] In the breeding process, the environmental monitoring unit collects water temperature, pH value, dissolved oxygen and other breeding environment parameters in real time through various sensors placed in the breeding pond. Figure 1 As shown in Figure 1, the unit primarily consists of a temperature sensor, a pH sensor, and a dissolved oxygen sensor. Each sensor is connected to an industrial control computer via an RS485 bus. The sensor housing is made of IP67-rated stainless steel for strong corrosion resistance. The probe is suspended vertically 30 to 50 cm below the water surface to prevent measurement deviations caused by sludge deposition.

[0080] At the same time, the feeding monitoring unit is used to record the feeding data during the breeding process. Figure 1 As shown, the unit includes a weighing sensor with a range of 0 to 100 kg, which is installed at the hopper outlet of the feeding equipment to collect the weight of the feed each time in real time; it also includes a feeding time recorder, which is connected in series with the weighing sensor and transmits the collected data to the industrial control computer through the RS485 bus, which records the time and amount of each feeding.

[0081] For the identification of aquatic products, this embodiment uses radio frequency identification (RFID) technology. Figure 2 As shown in the figure, the RFID identification unit consists of an RFID reader and an RFID electronic tag. The RFID reader uses a high-frequency reader with an operating frequency of 13.56MHz, with a reading distance of up to 10cm, and is connected to the industrial control computer via the TCP / IP protocol.

[0082] Each aquatic product is equipped with an RFID tag. This waterproof tag contains a unique identification code that records information such as the aquaculture batch and stocking date. During the aquaculture process, when aquatic products pass under the RFID reader, the reader automatically reads the tag information and transmits it to the industrial computer, enabling automatic identification of the aquatic product and digital management of aquaculture records.

[0083] The data collection layer submits the collected multi-source heterogeneous data to the upper edge computing layer for preprocessing. Figure 1 As shown in the figure, the edge computing layer is mainly composed of industrial control computers and edge servers deployed in the farm area.

[0084] The industrial control computer serves as the gateway for data collection. It uses the Linux real-time operating system and performs periodic sampling of each sensor node through polling. The sampling period can be configured between 100 and 1000 milliseconds as needed. Considering the large volume of raw collected data, the computer's built-in data collection program incorporates modules such as data verification, data filtering, and data compression to improve data quality and transmission efficiency.

[0085] like Figure 3 As shown in the figure, the data verification module adopts the cyclic redundancy check (CRC) algorithm to verify the collected data frames, eliminate the data with transmission errors, and ensure the integrity of the data; the data filtering module adopts the median filtering algorithm to remove dirty data and wild points caused by equipment failure or external interference, thereby improving the accuracy of the data; the data compression module adopts a lossless compression algorithm (such as run-length encoding) to compress the data volume as much as possible without affecting the data accuracy, thereby reducing the bandwidth occupied by data transmission.

[0086] On the other hand, the edge computing layer also deploys several edge servers. Figure 1 As shown, the edge server is deployed using the popular Docker containerization technology and runs a self-developed edge computing framework. This framework further aggregates and integrates multi-source data, mines key features, and enhances data value, providing a unified data entry point for blockchain-based on-chain evidence storage.

[0087] like Figure 3 As shown in Figure 2, the core of the framework includes a data parsing engine, a feature extraction engine, and a data fusion engine. The data parsing engine is responsible for semantically parsing and formatting heterogeneous data collected by industrial control computers (such as environmental monitoring data, feeding monitoring data, and RFID identification data), and outputting data in a unified JSON format.

[0088] On this basis, the feature extraction engine employs a variety of data mining algorithms to extract key features that are crucial for predicting aquatic product quality and safety risks. For example, a sliding time window method is used to determine whether there has been a risk of water quality deterioration within the past 24 hours. Another example is correlation analysis, which determines the strength of the correlation between feeding amount and water quality, providing a reference for intelligent feeding decisions.

[0089] The data fusion engine uses multi-sensor fusion algorithms to fuse data from multiple sources to create more comprehensive and reliable environmental portraits. For example, it uses a Kalman filter algorithm to fuse measurements of parameters such as temperature, pH, and dissolved oxygen to produce a comprehensive water quality score, providing a reference for decision-making in optimizing the aquaculture process.

[0090] Furthermore, to support real-time data processing and reliable delivery, the edge server and blockchain layer are loosely coupled via a message queue middleware. This middleware uses the currently popular RabbitMQ, employing a publish / subscribe model and persistence mechanisms to ensure data delivery even in unexpected situations such as network outages. Furthermore, SSL encryption is used throughout the transmission channel to prevent data eavesdropping and tampering.

[0091] After the data passes through the layers of screening and optimization at the edge computing layer, it enters the blockchain layer for on-chain storage. Figure 4 As shown in the figure, the blockchain layer is established in the form of a consortium chain and is built using the most mature Hyperledger Fabric blockchain platform.

[0092] From an architectural perspective, a blockchain network primarily consists of a cluster of ordering nodes, a cluster of peer nodes, and a channel management system. Peer nodes are responsible for running smart contracts (i.e., chaincodes) and maintaining ledger data. Multiple peer nodes form a cluster, representing an organization (such as a farm, processing plant, or distributor) joining the blockchain network.

[0093] Ordering nodes are key hubs in the network, responsible for receiving transaction requests, sorting them into blocks, and distributing them to peer nodes for verification, thereby achieving global consensus on the network. Ordering nodes can also form clusters, using the Raft protocol to automatically elect a master node, with other nodes acting as slaves. Master and slave nodes synchronize in real time through a heartbeat mechanism. If the master node fails, a slave node can immediately take over, thus achieving high availability of the ordering service.

[0094] The channel management system is used to achieve data isolation. Because all parties involved in aquatic product traceability strongly prioritize confidentiality for commercial data, it's impossible for all members to see all data. Therefore, the platform sets up different channels based on permissions. Each channel corresponds to a ledger, and only members within that channel can see the data within the ledger. This allows for data sharing while protecting the privacy of all parties.

[0095] The core functions of a blockchain are implemented by a series of smart contracts deployed on peer-to-peer nodes. For example, the data verification contract uses asymmetric encryption to sign data, combined with a timestamp mechanism to form a data summary, which is then uploaded to the blockchain network for storage, ensuring non-repudiation and tamper-proofing once the data is on-chain. The storage contract stores the data as an object in the InterPlanetary File System (IPFS) and stores the data hash as a pointer on the blockchain, enabling on-chain and off-chain collaboration and addressing the limited storage capacity of the blockchain. The access control contract, based on a fine-grained attribute-based access control (ABAC) model, intelligently matches access policies based on the requester's identity and data attributes, automating permissions management. These basic contracts can be flexibly assembled to form complex business logic.

[0096] On this basis, the blockchain management system provides a series of supporting services, including member management, network monitoring, and performance optimization. Member management is implemented based on PKI, with the CA server issuing digital certificates to each node and managing the entire certificate lifecycle, including application, renewal, and revocation, thus achieving trusted identity authentication. Network monitoring is implemented based on DevOps tools such as Prometheus, allowing administrators to view indicators such as the operating status and resource usage of each node in real time. Performance optimization can be achieved by adjusting parameters such as block size and transaction concurrency to find the optimal balance between blockchain security and performance.

[0097] After the blockchain layer stores the collected and integrated data in a trustworthy manner, it can be called by the application layer. Figure 1 As shown, the application layer includes different business subsystems for various aspects of production management, including breeding management system, processing management system, logistics management system and sales management system.

[0098] The aquaculture management system utilizes a browser / server (B / S) architecture to provide information-based aquaculture services to farmers. For example, the environmental monitoring module displays pond water quality parameters in real time. When a parameter exceeds the standard, the system automatically issues an alert and links to a knowledge base, providing best practices for pond management. The feeding management module uses algorithms such as reinforcement learning, combining aquaculture environment and fish monitoring data to learn and optimize feeding strategies, enabling intelligent feeding. The growth cycle management module uses WebGIS to visualize pond distribution, visually displaying the growth cycle of each batch of aquatic products, from seedlings to adult fish, facilitating coordinated scheduling.

[0099] Subsystems such as processing management and logistics management utilize a microservices architecture developed based on the Spring Cloud framework. This architecture facilitates the expansion of new functionality, with loosely coupled services enabling independent development, testing, deployment, and scalability. For example, the processing management system includes microservices for process monitoring, quality and safety testing, and warehouse inventory management, which are integrated with the workshop's MES, quality inspection system, and WMS. The logistics management system includes microservices for transport scheduling, temperature control monitoring, and delivery tracking, which are integrated with the TMS, cold chain system, and courier company. Microservices exchange data using RESTful APIs in JSON format.

[0100] The terminal's sales management system primarily serves consumers, providing product quality and safety information inquiries. Consumers simply scan the QR code on the product packaging with their mobile phone to access the entire product chain, from cultivation to sales, and verify its authenticity. The system utilizes the OAuth 2.0 protocol for user authentication and authorization, accommodating multiple login methods (such as mobile phone number, WeChat ID, and facial recognition) to enhance the user experience. Furthermore, the system incorporates a mobile payment SDK, supporting convenient payment methods such as scan-to-pay and one-click payment, completing a closed-loop online sales process.

[0101] like Figure 5 , also provides a method for implementing a full-process traceability system for aquatic products based on blockchain technology, including the following steps:

[0102] Step 1: Data collection: Through the equipment of environmental monitoring unit, feeding monitoring unit, and RFID identification unit, multi-source heterogeneous data such as water quality parameters, feeding records, and breeding batches in the breeding process are automatically collected to realize the automation of data collection, and the collection automation rate reaches more than 95%.

[0103] Step 2: Edge computing: The industrial control computer acts as the edge gateway for data collection, performing pre-processing such as verification, filtering, and compression on the original collected data. The edge server further performs data cleaning, feature extraction, and multi-source fusion on the data, thereby improving data quality while minimizing data transmission volume and reducing the storage and computing burden of the blockchain.

[0104] Step 3: Upload the optimized structured data to the blockchain network for storage in a unified data format through the API interface provided by the smart contract, so as to achieve traceability and non-tamperability of the data throughout its life cycle.

[0105] Step 4: Intelligent analysis: Each business application subsystem (including breeding, processing, warehousing, logistics, etc.) calls blockchain data and combines cutting-edge technologies such as big data analysis and artificial intelligence to conduct real-time monitoring of production and operation activities, intelligent early warning, and process optimization, thereby improving the level of refined quality management and supply chain collaboration efficiency.

[0106] Step 5: Information query: Managers can use the web management background, and consumers can use mobile apps and other terminals to enter the product traceability code to query the quality and safety information of each link such as product production, processing, warehousing, and transportation in real time, thereby achieving full visualization and control of product quality.

[0107] Therefore, the present invention adopts the above-mentioned blockchain-based aquatic product full-process traceability system and its implementation method, which can not only be applied to the aquatic product field, but can also be extended to other agricultural products such as livestock, poultry, vegetables and fruits. It is of great significance to improve the level of food safety management, ensure the public's dietary health, and promote the modernization of agriculture.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solutions of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A full-process traceability system for aquatic products based on blockchain technology, characterized by: include: The data collection layer consists of an environmental monitoring unit, a feeding monitoring unit, and an RFID identification unit, which are used to automatically collect aquaculture environment data; The edge computing layer includes edge servers and industrial control computers, which are used to pre-process the collected data; The blockchain layer is built on the Hyperledger Fabric platform to achieve tamper-proof data storage and automatically verify data authenticity through smart contracts; The Hyperledger Fabric platform adopts a consortium chain architecture, including an ordering node cluster, a peer node cluster, and a channel management module. The ordering node cluster uses the Raft consensus algorithm to maintain blockchain ledger consistency, the peer node cluster is responsible for maintaining the chain code and state database, and the channel management module is responsible for managing data isolation between different organizations. The specific implementation architecture of the Hyperledger Fabric platform is as follows: The sorting node cluster consists of a master node and several slave nodes, which communicate with each other through the Raft consensus protocol: The master node is responsible for receiving transaction requests and sorting them by timestamp, using a batch processing mechanism to package transactions into blocks; the slave nodes maintain synchronization with the master node through a heartbeat mechanism; The nodes communicate using the gRPC protocol; The peer node cluster adopts a layered architecture: The consensus layer is responsible for verifying transactions and maintaining ledgers, using LevelDB to store transaction data. The chaincode layer is responsible for executing smart contracts, using Docker containers to isolate the operating environment. The state database layer is implemented using CouchDB. The channel management module adopts policy-based access control: Channel configuration uses MSP to manage identity; channel strategy is defined using signature strategy language; data isolation is implemented based on a multi-ledger mechanism, and each channel maintains independent ledger data; Smart contracts include data verification contracts, storage contracts, and access control contracts: the data verification contract uses a digital signature algorithm and timestamp verification mechanism, the storage contract uses a Merkle tree structure and IPFS distributed storage, and the access control contract implements fine-grained permission management based on the ABAC model; It also includes a blockchain network management module, which is responsible for node certificate management, blockchain network monitoring and performance optimization. Certificate management adopts the PKI architecture, network monitoring uses Prometheus, and performance optimization is achieved by dynamically adjusting block size and transaction concurrency. The application layer includes the breeding management subsystem, processing management subsystem, logistics management subsystem, and sales management subsystem. The application layer uses a multi-level authorization mechanism to manage data access rights; The breeding management subsystem adopts B / S architecture, including breeding environment monitoring module, feed delivery management module and growth cycle management module. The contents are as follows: The aquaculture environment monitoring module adopts a layered architecture: The data collection layer pushes real-time data through the WebSocket server and uses a heartbeat detection mechanism to maintain the connection status; the data display layer uses the ECharts chart library to achieve data visualization; the alarm management layer uses a rule engine to implement threshold alarms; The feed delivery management module is based on a deep learning framework: The data preprocessing layer uses a sliding time window to segment data, with a window size of 24 hours and a step size of 1 hour. The model training layer uses an LSTM neural network structure, consisting of three layers of LSTM units, 128 neurons per layer, and a dropout rate of 0.

5. The prediction and analysis layer uses the Adam optimizer with a learning rate of 0.001 to predict the optimal feeding amount based on historical feeding data and environmental parameters. The growth cycle management module is implemented based on GIS: The map engine layer uses the OpenLayers framework; the data management layer uses PostGIS extension to implement spatial data storage; the business function layer integrates the visualization of pond number, breeding species, and stocking density attribute information.

2. The blockchain-based aquatic product traceability system according to claim 1 is characterized in that: The contents of the environmental monitoring unit, feeding monitoring unit and RFID identification unit are as follows: The environmental monitoring unit includes a dissolved oxygen sensor, a pH sensor and a temperature sensor, all of which are connected to the industrial control computer via the RS485 bus. Each sensor probe is vertically installed 30-50 cm below the water surface of the aquaculture pond; The feeding monitoring unit includes a feeding amount sensor and a feeding time recorder. The feeding amount sensor is a weighing sensor with a range of 0-100kg, installed at the bottom of the hopper of the feeding equipment, connected in series with the feeding time recorder and connected to the industrial control computer via the RS485 bus; The RFID identification unit includes an RFID reader and an RFID electronic tag. The RFID reader uses a high-frequency reader with an operating frequency of 13.56MHz and a reading distance of 0-10cm. It is connected to the industrial control computer through the TCP / IP protocol.

3. The aquatic product traceability system based on blockchain technology according to claim 2 is characterized in that: The specific structure of the feeding monitoring unit is: The feeding amount sensor adopts a parallel beam structure with a built-in temperature compensation circuit. It is fixed to the four corners of the hopper bottom by bolts, with an inclination angle of no more than 2°. The feeding time recorder adopts industrial-grade single-chip microcomputer controller, has a real-time clock chip, and is equipped with a 3.5-inch LCD display; The feeding amount sensor and the feeding time recorder are connected using a 4-core shielded cable.

4. The aquatic product traceability system based on blockchain technology according to claim 1 is characterized in that: The contents of edge servers and industrial control computers are as follows: The industrial control computer uses the Linux real-time operating system and collects sensor data through a data acquisition program. The data acquisition program includes a data verification module, a data filtering module, and a data compression module. The data verification module uses a CRC checksum algorithm to ensure data integrity, the data filtering module uses a median filter algorithm to remove outliers, and the data compression module uses a lossless compression algorithm to reduce data transmission volume. The edge server is deployed in Docker containers and runs the edge computing framework, which includes a data parsing engine, a feature extraction engine, and a data fusion engine. The data parsing engine parses the collected raw data into a standard format, the feature extraction engine uses a sliding time window method to extract data features, and the data fusion engine uses a Kalman filter algorithm to fuse multi-source data. The edge server interacts with the blockchain layer through message queues, and uses publish or subscribe mode to achieve asynchronous data transmission. The message queue is implemented using RabbitMQ, and SSL encryption is used to ensure data transmission security.

5. The blockchain-based aquatic product traceability system according to claim 4 is characterized by: The specific implementation architecture of industrial control computers is: The data acquisition program is based on a multi-threaded architecture, including device driver threads, data acquisition threads, and data processing threads: The device driver thread is responsible for managing the communication interface of RS485 devices and TCP / IP devices, and adopts the epoll mechanism to achieve high concurrent IO processing; The data acquisition thread adopts polling mode to perform regular sampling on each sensor. The sampling period can be configured in the range of 100ms-1000ms, and a double buffer mechanism is used to avoid data loss. The data processing thread receives collected data through the pipeline mechanism, and the processed data interacts with the application through shared memory; Industrial control computers also include a watchdog program that monitors the system's operating status and automatically restarts related processes when a system anomaly is detected. It also records system logs through the syslog mechanism.

6. The aquatic product full traceability system based on blockchain technology according to claim 1 is characterized in that: The contents of the breeding management subsystem, processing management subsystem, logistics management subsystem and sales management subsystem are as follows: The processing management subsystem and logistics management subsystem adopt a microservice architecture and are implemented using the Spring Cloud framework. The processing management subsystem includes a processing process tracking module, a quality inspection module, and an inventory management module; the logistics management subsystem includes a transportation monitoring module, a temperature monitoring module, and a distribution management module. Data exchange between these modules is carried out through RESTful APIs. The sales management subsystem integrates mobile payment interface and traceability code query function; the multi-level authorization mechanism is implemented based on the RBAC model, including four authority levels: system administrator, enterprise administrator, operator and consumer.

7. The aquatic product traceability system based on blockchain technology according to claim 6 is characterized in that: The specific implementation architecture of the sales management subsystem is: The mobile payment interface uses a unified payment gateway, including the payment gateway layer, transaction processing layer, and reconciliation and settlement layer; The traceability code query function is implemented based on the QR code method, including the QR code generation layer, data encoding layer and parsing verification layer; OAuth2.0 authentication adopts a layered architecture, including authentication service layer, resource service layer and user management layer.

8. A method for implementing a full-process traceability system for aquatic products based on blockchain technology, applying a full-process traceability system for aquatic products based on blockchain technology as described in any one of claims 1 to 7, characterized in that: The following steps are involved: S1. Data collection: Through the equipment of environmental monitoring unit, feeding monitoring unit and RFID identification unit, water quality parameters, feeding records and breeding batches in the breeding process are automatically collected, realizing the automation of data collection, and the collection automation rate reaches more than 95%; S2, edge computing, industrial control computers serve as edge gateways for data collection, performing verification, filtering, and compression preprocessing operations on the original collected data; The edge server cleans the data, extracts features, and performs multi-source fusion processing; S3: Uploading the optimized structured data to the blockchain network for evidence storage in a unified data format through the API provided by the smart contract; S4, intelligent analysis: each business application subsystem calls blockchain data and combines big data analysis and artificial intelligence to conduct real-time monitoring of production and operation activities, intelligent early warning, and process optimization; S5. Information query: managers can use the web management background, and consumers can use mobile terminals to input product traceability codes to query the quality and safety information of each link of product production, processing, warehousing and transportation in real time.

Images