Immersive visual product processing process management system

Abstract

The invention discloses an immersive visual product processing process management system, belongs to the field of product processing data management systems, and constructs a set of production transparency system integrating immersive interaction, real-time data encryption transmission, multi-chain collaborative evidence storage and intelligent contract automatic verification. The whole process closed loop of live broadcast interaction-production execution-data evidence storage-anti-counterfeiting verification is realized, a credible trust infrastructure is provided for customized production in a live broadcast e-commerce scene, and novel consumption trust ecological construction of "what you see is what you get and what you get is what you get is what you get is what you get is what you get is what you get is what you get.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of product processing data management systems, and more particularly to an immersive visual product processing progress management system. BACKGROUND

[0002] Under the dual driving of global consumption upgrading and industrial digital transformation, live streaming e-commerce has become the core link connecting the production end and the consumption end. Consumers have higher requirements for product production process transparency, precise execution of customization needs, and trust mechanisms for genuine product traceability. Traditional manufacturing industries face the following core pain points:

[0003] Insufficient production process transparency. Although existing production management systems use flat data dashboards or video live streaming modes for display, consumers cannot perceive the entire product production process in real time from an immersive perspective. In the context of customized production, personalized needs such as process parameter adjustment and appearance customization are easily subject to information decay or execution deviation when transmitted through traditional order systems, resulting in a mismatch between the delivery result and customer expectations.

[0004] Data storage and anti-counterfeiting technology lags behind. Traditional production data storage relies on centralized databases, which are vulnerable to data tampering and lack real-time verification methods. Although blockchain technology has been introduced in the anti-counterfeiting field, it is mostly limited to single-product traceability labels and is not deeply integrated with dynamic production data such as equipment operating parameters and process rhythms. Counterfeiters can evade inspection by forging static labels or tampering with partial production data. When customized disputes arise, there is a lack of a credible evidence chain covering the entire link from "demand instructions-production execution-product delivery," making it difficult to define responsibilities.

[0005] To address the above problems, it is necessary to build a production transparency system that integrates immersive interaction, real-time data encryption transmission, multi-chain collaborative storage, and intelligent contract automated verification, achieving a full-process closed loop of "live interaction-production execution-data storage-anti-counterfeiting verification" and providing a credible trust infrastructure for customized production in the live streaming e-commerce scenario, promoting the construction of a new consumer trust ecosystem based on "what you see is what you get, what you get is what you have witnessed." Therefore, we propose an immersive visual product processing progress management system to solve the above problems. SUMMARY

[0006] 1. Technical problems to be solved

[0007] In view of the problems in the prior art, the purpose of the present application is to provide an immersive visual product processing process management system, which builds a production transparency system integrating immersive interaction, real-time data encryption transmission, multi-chain collaborative evidence storage and intelligent contract automatic verification, realizes a whole-process closed loop of "live interaction-production execution-data evidence storage-fake-proof verification", provides a trusted trust infrastructure for customized production in a live e-commerce scenario, and promotes the construction of a new consumption trust ecological system of "what you see is what you get, what you get is what you prove".

[0008] 2. Technical solution

[0009] To solve the above problems, the present application adopts the following technical solution.

[0010] An immersive visual product processing process management system, comprising:

[0011] A user interaction layer is provided with an e-commerce platform VR live room, the VR live room includes a three-dimensional scene rendering module, a customization instruction receiving module, a data verification display module, and a real-time verification function module for generating verification data and synchronously writing to a blockchain evidence storage layer, the verification data generated by the real-time verification function module is synchronously written to the blockchain evidence storage layer through a secure channel of the instruction processing layer;

[0012] An instruction processing layer includes an instruction encryption transmission unit and a two-way authentication feedback unit, the instruction encryption transmission unit generates a 256-bit encryption key based on the ECC elliptic curve algorithm, constructs a transmission layer security channel using the TLS-1.3 protocol, and realizes data encryption transmission by matching the ChaCha20-Poly1305 encryption suite, and the encrypted data is issued to a PLC device cluster through an industrial Internet of Things gateway;

[0013] A production data acquisition layer includes an industrial Internet of Things gateway and an edge computing node, the industrial Internet of Things gateway establishes a communication connection with the PLC device cluster through the OPCUA protocol, and the edge computing node is deployed with a lightweight convolutional neural network model, which is configured to identify production key nodes in a video stream in real time and generate a spatiotemporal feature vector;

[0014] The customization instruction receiving module is used to parse user instructions into structured data and transmit them to the instruction processing layer, and the instruction encryption transmission unit issues the structured data to the PLC device cluster after ECC encryption;

[0015] The blockchain storage layer adopts a three-layer distributed ledger architecture of a core production chain, a supply chain collaboration chain and a user verification chain. The core production chain adopts a hardware security module to manage the key life cycle, realizes the secure generation, storage and regular rotation of the key, ensures the physical level security protection of the encryption process, and synchronously writes the verification data generated by the real-time verification function module into the core production chain for storage. The spatiotemporal feature vector generated by the edge computing node is transmitted to the blockchain storage layer through the security channel of the instruction processing layer. After the spatiotemporal feature vector hash value is verified by the smart contract of the blockchain storage layer, the timing data visualization unit of the user trust enhancement layer automatically updates the chart based on the verification result. The core production chain stores the production process parameters. The supply chain collaboration chain synchronously transmits the data hash value to the core production chain after verifying the raw material traceability data by the smart contract. The user verification chain records the digital certificate digest and the digital twin ID.

[0016] The anti-counterfeiting verification layer includes a physical feature extraction module and a smart contract execution module. The physical feature extraction module is configured with a quantum dot spectrum analyzer and a nanoscale microscope. The spectrum data and nanoscale pattern features are collected by the quantum dot spectrum analyzer and the nanoscale microscope and synchronously stored in the blockchain storage layer. The smart contract execution module is developed based on the Solidity language and is configured with a dynamic parameter verification engine and a multi-level exception handling mechanism. The multi-level exception handling mechanism includes a fault-tolerant verification rule library for storing exception judgment models and processing rules. The spectrum data collected by the physical feature extraction module is uploaded to the smart contract execution module after Base64 encoding through the gRPC protocol. After receiving the spectrum data, the smart contract execution module automatically triggers the hash comparison with the standard template stored in the blockchain.

[0017] The user trust enhancement layer obtains the raw material traceability data, equipment operation parameters and cross-chain storage information from the blockchain storage layer, including a geographic information visualization unit, a timing data visualization unit and a blockchain browser component, for realizing the raw material source display, equipment operation parameter analysis and cross-chain data query.

[0018] Further, the three-dimensional scene rendering module constructs a virtual factory environment based on the Unity3D engine. The customized instruction receiving module analyzes the barrage input by using the natural language processing technology. The data verification display module supports the WebGL technology to realize the interactive display of the production process digital twin view.

[0019] The instruction encryption transmission unit generates a 256-bit encryption key based on the ECC elliptic curve algorithm. The bidirectional authentication feedback unit realizes the identity verification of the PLC device and the blockchain node by using the asymmetric encryption technology, forming a closed-loop verification link of the instruction transmission.

[0020] The core production chain of the blockchain storage layer is built based on the Hyperledger Fabric consortium chain framework, the PBFT consensus algorithm is configured to realize high-efficiency transaction processing capability, and the zero-knowledge proof technology is adopted to protect the privacy of the production process parameters stored by the blockchain storage layer. The supply chain collaboration chain of the blockchain storage layer is built based on the Corda blockchain platform, the verification of the synchronized raw material traceability data is automatically performed by the smart contract, the user verification chain is built based on the Ethereum side chain technology, the Merkle Patricia tree structure is adopted to store the digital certificate digest, and the non-forgery certificate of the ERC-721 standard is supported.

[0021] Further, the real-time verification function module of the VR live room includes:

[0022] A random challenge code generation unit generates a 128-bit random number as a challenge code based on the SHA-3 algorithm;

[0023] An industrial vision feedback unit captures the challenge code image displayed on the LED screen in real time through a high-speed camera deployed in the production site;

[0024] A timestamp anchoring unit binds the challenge code image hash value with a timestamp accurate to the nanosecond level and then writes it into the blockchain;

[0025] The random challenge code generation unit integrates the WebSocket push function to push the 128-bit random number to the LED screen in the production site in real time, the industrial vision feedback unit transmits the captured challenge code image to the timestamp anchoring unit through the gigabit Ethernet, the timestamp anchoring unit performs SHA-256 hash operation on the challenge code image to generate anchoring data containing the nanosecond-level timestamp, and the anchoring data is written into the core production chain through the PBFT consensus mechanism of Hyperledger Fabric.

[0026] Further, the digital twin ID generation mechanism corresponding to the user verification chain includes:

[0027] A multi-dimensional feature extraction unit extracts the first 8 bits of the device MAC address, the production timestamp, and the 6-bit random number generated by the random number generator;

[0028] A hash fusion unit fuses the multi-dimensional features to generate a 160-bit unique identifier by using the Keccak-256 algorithm;

[0029] A three-way writing unit synchronously writes the unique identifier into the PLC device register, the physical anti-fake label storage area, and the blockchain smart contract storage mapping;

[0030] The 160-bit identifier generated by the hash fusion unit is written into the PLC device register through the communication protocol, written into the EEPROM storage area of the physical anti-counterfeiting label through the NFC interface, and synchronized to the user verification chain of the blockchain storage layer through the setID function of the Solidity smart contract.

[0031] Further, the intelligent video hash aggregation technology of the edge computing node includes:

[0032] A dynamic time slice division algorithm automatically adjusts the time slice length to millisecond precision by real-time acquisition of process beat signals fed back by the PLC device through the OPCUA protocol;

[0033] A space-time feature extraction network extracts space-time feature vectors of video frames using a 3D convolutional neural network;

[0034] A Merkle tree construction module constructs the feature vectors within the time slice into a Merkle tree structure to generate a root hash value;

[0035] The dynamic time slice division algorithm adjusts the time slice length according to the process beat signals fed back by the PLC device, the space-time feature extraction network pulls video streams through the RTSP protocol, the generated feature vectors are transmitted to the Merkle tree construction module through the Kafka message queue, and after the root hash value is distributed stored by IPFS, the hash address is written into the block header of the core production chain.

[0036] Further, the dynamic parameter verification engine of the smart contract execution module specifically includes:

[0037] A parameter classifier divides process parameters into a basic parameter set and a customized parameter set based on a machine learning algorithm;

[0038] A dynamic rule generator generates parameter constraint rules in real time according to customer instructions;

[0039] A dual verification executor simultaneously performs on-chain rule verification and off-chain physical feature comparison.

[0040] Further, the smart contract execution module calls the standard spectrum template of the blockchain storage layer to perform similarity comparison of spectral data and nano-pattern features. The standard spectrum template is acquired by a quantum dot spectrum analyzer during the first piece inspection stage of the production data acquisition layer, written into the core production chain storage through SHA-256 hash operation, and used as the benchmark data for anti-counterfeiting verification. The similarity comparison of the nano-pattern features is performed by generating micro-nano structures with a resolution of 100 nm on the surface of the label through electron beam lithography technology.

[0041] Further, the core production chain serves as the core layer of the three-layer distributed ledger architecture, and is configured with a main chain and a backup chain as internal redundant nodes. Atomic-level data synchronization is achieved through state channel technology to ensure data redundancy and consistency. The supply chain collaboration chain adopts a distributed consensus algorithm, and the user verification chain achieves data synchronization based on side chain technology. In the three-layer distributed ledger architecture of the blockchain storage layer, the node configuration mechanism of each chain includes:

[0042] The core production chain adopts an identity-based access control model, and the nodes are deployed in a physically isolated dedicated network.

[0043] After the supply chain collaboration chain verifies the raw material traceability data through the smart contract, it broadcasts the data hash value to the core production chain every 10 minutes through a notarization node. After the core production chain verifies the hash consistency through the Fabric cross-chain protocol, data mapping and storage are completed.

[0044] The user verification chain adopts Plasma side chain technology. The Plasma side chain receives digital certificate update events from the main chain through the state channel, synchronizes to the MerklePatricia tree storage node in RLP encoding format, and realizes more than 10,000 query responses per second through the state channel.

[0045] Further, the fault-tolerant verification rule library of the smart contract execution module contains a three-level abnormality judgment model, including a first-level abnormality processing sub-rule, a second-level abnormality processing sub-rule, and a third-level abnormality processing sub-rule.

[0046] The first-level abnormality processing sub-rule is used to predict missing data through a Kalman filter for single-process data interruption. The Kalman filter subscribes to the historical data topic of the PLC device through OPCUA, predicts the missing data, and returns it to the cache area of the edge computing node through the CoAP protocol.

[0047] The second-level abnormality processing sub-rule is used to reconstruct the production process using a Petri net model for cross-process logical contradiction.

[0048] The third-level abnormality processing sub-rule is used to trigger a Bayesian network for abnormality traceability analysis for multi-link data missing. The Bayesian network calls the multi-chain data of the blockchain storage layer through the gRPC service, locates the abnormal source through the evidence propagation algorithm, sends warning information to the user interaction layer through the message bus, and triggers the process parameter adjustment instruction to the PLC device cluster through the instruction processing layer.

[0049] Further, the user trust enhancement layer includes:

[0050] The geographic information visualization unit realizes the three-dimensional geographic information display of the raw material source based on CesiumJS, and supports user interactive tracking of the raw material transportation path.

[0051] The time series data visualization unit constructs a time series analysis chart of device operation parameters using D3.js.

[0052] The blockchain browser component supports cross-chain data query of Ethereum and consortium chain, and provides block verification, transaction tracking and smart contract calling functions.

[0053] 3. Advantages

[0054] Compared with the prior art, the advantages of the present application are:

[0055] (1) The present scheme, through the three-dimensional scene rendering module of the VR live room and the WebGL interactive digital twin view, consumers can break through the perspective limitation of traditional flat billboards and video live, and can real-time view the production line equipment running state, process parameter adjustment, material flow, etc. Key nodes in an immersive perspective, making the production transparency from traditional "partial segment display" to "full process real-time perceptible" effect, improving the consumer's sense of participation in customized production;

[0056] (2) The present scheme, the custom instruction receiving module uses natural language processing technology to analyze the barrage input, converts user personalized needs such as process parameter adjustment and appearance customization into structured data, transmits it to the PLC device cluster through the 256-bit encryption key generated by the ECC elliptic curve algorithm and the security channel constructed by the TLS-1.3 protocol, avoids information decay caused by manual transcription in traditional order system, and at the same time, the two-way authentication feedback unit realizes the identity authentication of the device and the blockchain node through asymmetric encryption, forms a closed loop link of "instruction issuing-execution feedback", and improves the accuracy of customized product delivery meeting expectations;

[0057] (3) The present scheme, the core production chain is based on HyperledgerFabric consortium chain, adopts PBFT consensus algorithm and hardware security module to manage key life cycle, ensures the physical level security protection of core data such as production process parameters and device operation data, and through the real-time verification function module, writes the dynamic data of the production site into the blockchain, forms the "time-space-operation" bound tamper-proof evidence, and resists the risk of counterfeiters tampering with static data;

[0058] (4) In this scheme, the supply chain collaboration chain synchronizes the raw material traceability data hash value to the core production chain every 10 minutes, the user verification chain stores the digital twin ID and certificate digest, forming a full-link notarization system covering "demand instruction-raw material procurement-production execution-finished product delivery". When a dispute occurs, cross-chain data query can be performed through the blockchain browser component, and the spatial-temporal feature vector hash value generated by the edge computing node can be automatically verified by the smart contract, improving the convenience of responsibility definition. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 The figure is a schematic diagram of the system architecture of the present application;

[0060] Figure 2 The figure is a schematic diagram of the user interaction layer and principle of the present application;

[0061] Figure 3 The figure is a schematic diagram of the instruction processing layer and production data acquisition layer and principle of the present application;

[0062] Figure 4 The figure is a schematic diagram of the blockchain notarization layer and anti-counterfeiting verification layer and principle of the present application;

[0063] Figure 5 The figure is a schematic diagram of the user trust enhancement layer and principle of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0065] Embodiment 1:

[0066] Please refer to Figures 1-5 An immersive visual product processing process management system based on a layered architecture of a user interaction layer, an instruction processing layer, a production data acquisition layer, a blockchain notarization layer, an anti-counterfeiting verification layer and a user trust enhancement layer, realizes full-process management and trusted verification of product processing process through collaborative interaction of each layer module and unit, and the specific principle is as follows:

[0067] I. User instruction analysis and secure transmission

[0068] 1. User interaction layer instruction processing User inputs instructions such as bullet screen and customization parameters through e-commerce platform VR live room, and customization instruction receiving module uses natural language processing technology to analyze unstructured instructions into structured data and trigger real-time verification function module to generate verification data.

[0069] Real-time verification process:

[0070] The random challenge code generation unit generates a 128-bit random challenge code based on the SHA-3 algorithm and pushes it to the production site LED screen display through WebSocket;

[0071] The industrial vision feedback unit captures the LED screen challenge code image through a high-speed camera and transmits it to the timestamp anchoring unit through a gigabit Ethernet;

[0072] The timestamp anchoring unit performs SHA-256 hash operation on the image, binds the hash value with the nanosecond-level timestamp, and writes it into the core production chain of the block chain storage layer through the PBFT consensus mechanism of HyperledgerFabric.

[0073] 2. The structured instruction encrypted transmission structure of the instruction processing layer is processed by the instruction encryption transmission unit:

[0074] Based on the ECC elliptic curve algorithm, a 256-bit encryption key is generated, and a secure channel is constructed using the TLS-1.3 protocol and the ChaCha20-Poly1305 encryption suite. The instruction data is encrypted and then sent to the PLC device cluster through the industrial Internet of Things gateway. At the same time, the two-way authentication feedback unit verifies the identity of the PLC device and the block chain node through asymmetric encryption technology, forming an instruction transmission closed loop verification link.

[0075] II. Production data acquisition and feature extraction

[0076] 1. Industrial Internet of Things data access Industrial Internet of Things gateway establishes communication with PLC device cluster through OPCUA protocol, real-time acquisition of device running parameters (such as process beat signal, sensor data), and synchronization to edge computing node.

[0077] 2. Edge computing intelligent analysis The edge computing node deploys a lightweight convolutional neural network model, pulls the production site video stream through the RTSP protocol, and extracts the spatiotemporal feature vector of the video frame using a 3D convolutional neural network.

[0078] Dynamic time slice division algorithm automatically adjusts the time slice length according to the process beat signal (millisecond level precision) feedback by PLC, and transmits the feature vector in the time slice to the Merkle tree construction module through Kafka message queue, generates root hash value, and stores it through IPFS distributed storage, and the hash address is written into the block header of the core production chain.

[0079] III. Block chain storage and cross-chain collaboration

[0080] 1. Three-layer chain data storage architecture

[0081] Core production chain: Store production process parameters, real-time verification data, spatio-temporal feature vector root hash generated by edge computing, configure main chain and backup chain to realize atomic-level data synchronization through state channel.

[0082] Supply chain collaboration chain: Verify raw material traceability data through smart contract, broadcast data hash value to core production chain every 10 minutes through notary node, complete mapping storage after verification by Fabric cross-chain protocol.

[0083] User verification chain: Record digital certificate digest and digital twin ID, digital twin ID generation process is as follows:

[0084] Multi-dimensional feature extraction unit extracts device MAC first 8 bits, production timestamp, and 6-bit random number;

[0085] Hash fusion unit generates 160-bit unique identifier through Keccak-256 algorithm, synchronously writes into PLC register, physical anti-fake label EEPROM (NFC interface), and blockchain smart contract storage mapping.

[0086] 2. Smart contract triggers data linkage Blockchain storage layer verifies spatio-temporal feature vector hash value through smart contract, automatically triggers user trust enhancement layer time series data visualization unit to update chart, supply chain collaboration chain verifies raw material traceability data, and periodically synchronizes hash value to core production chain.

[0087] Four, Anti-fake verification and exception handling

[0088] 1. Physical feature acquisition and on-chain comparison Physical feature extraction module acquires product physical features through quantum dot spectrum analyzer (acquires spectrum data) and nanoscale microscope (acquires 100nm resolution micro-nano pattern features), data is encoded through Base64 and transmitted to smart contract execution module through gRPC protocol.

[0089] Smart contract verification process:

[0090] Dynamic parameter verification engine divides process parameter types based on machine learning, generates constraint rules in real time according to customer instructions, and simultaneously performs on-chain rule verification (compares with SHA-256 hash standard template stored in core production chain first piece inspection stage) and off-chain physical feature comparison;

[0091] Multi-level exception handling mechanism:

[0092] First-level exception (single-process data interruption):

[0093] Kalman filter subscribes to historical data through OPCUA, predicts missing data, and returns to edge computing node cache through CoAP protocol;

[0094] Second-level exception (cross-process logical contradiction):

[0095] Petri net model reconstructs production process;

[0096] Three-level anomaly (multi-link data missing) :

[0097] Bayesian network calls multi-chain data through gRPC, locates the source of anomaly, sends warning through message bus, and triggers instruction processing layer to adjust PLC process parameters.

[0098] Five, user trust enhancement and data visualization User trust enhancement layer obtains multi-chain data from blockchain storage layer, and realizes visualization and query through the following modules:

[0099] Geographic information visualization unit (CesiumJS) :

[0100] Three-dimensional display of raw material source, supporting interactive traceability of transportation path;

[0101] Time series data visualization unit (D3.js) :

[0102] Constructing device operation parameter time series analysis chart;

[0103] Blockchain browser component:

[0104] Supporting cross-chain query of Ethereum and consortium chain, providing block verification, transaction tracking and smart contract calling functions.

[0105] Architectural logical relationship explains that each layer realizes data flow through secure channels (TLS-1.3, gRPC, etc.) and cross-chain protocols (Fabric cross-chain, Plasma state channel), forming a closed loop link of "user instruction input → encrypted transmission → production data collection → blockchain storage → anti-fake verification → trusted data visualization", ensuring the transparency, security and traceability of product processing.

[0106] Example 2:

[0107] In view of the above example 1, further description is made, please refer to Figures 1-5 Taking the intelligent electronic device production scene as the background, the deployment of an immersive visualization product processing management system by a consumer electronics manufacturer is described, which realizes user remote monitoring, real-time instruction interaction, production data storage, supply chain traceability and terminal product anti-fake verification of TWS earphone production line. Taking the adjustment of earphone cavity injection parameters by users through e-commerce platform VR live room and the verification of product authenticity as an example, the following shows the system full-link technical implementation principle:

[0108] I. User interaction layer:

[0109] VR live room instruction interaction and real-time verification

[0110] 1. VR virtual factory construction

[0111] The three-dimensional scene rendering module constructs a 1:1 scale virtual injection molding workshop based on the Unity3D engine, synchronizes the physical workshop equipment layout, pipeline state, and process parameters (such as injection molding machine temperature, pressure dynamic numerical value) in real time.

[0112] Users can enter the live room through the PC end WebGL interface or VR headsets, and see the digital twin model of the injection molding machine executing the earphone cavity injection molding process. The interface right side floating data verification display module displays the blockchain stored production key node hash value and timestamp in real time.

[0113] 2. User instruction analysis and transmission

[0114] Users input instructions in the live room:

[0115] "Adjust the injection molding temperature to 220°C and maintain for 5 seconds", the custom instruction receiving module is parsed into structured data by natural language processing (NLP) technology:

[0116] {

[0117] "device": "injection molding machine A-01",

[0118] "action": "set",

[0119] "parameter": "temperature",

[0120] "value": 220,

[0121] "duration": 5,

[0122] "timestamp": "2025-05-19T16:00:00Z"

[0123] }

[0124] Structured data is transmitted to the instruction processing layer, and the instruction encryption transmission unit generates a 256-bit temporary encryption key based on the ECC elliptic curve algorithm, encrypts it through the TLS-1.3 protocol + ChaCha20-Poly1305 suite, and sends it to the PLC device cluster (injection molding machine controller) through the industrial Internet of Things gateway (supports OPCUA protocol) to execute parameter adjustment.

[0125] 3. Real-time verification function execution

[0126] The random challenge code generation unit generates a 128-bit random number (such as `0x3b7f9a4d2c1e85...`) based on the SHA-3 algorithm and pushes it to the production site LED screen in real time through WebSocket, displaying the content as "Challenge Code: 3B7F9A4D".

[0127] The industrial vision feedback unit deploys a high-speed camera (resolution 1920x1080, frame rate 500fps) to capture LED screen images and transmits them to the timestamp anchoring unit through gigabit Ethernet.

[0128] The timestamp anchoring unit performs SHA-256 hash operation on the image (generating hash value `0x5d2c8f...`) and binds the nanosecond-level timestamp (`2025-05-19T16:00:02.123456789Z`), writes it to the core production chain through the PBFT consensus mechanism of HyperledgerFabric (consensus completed within 2 seconds), and users can click the "Verify" button in the live room to view the on-chain record through the blockchain browser.

[0129] II. Production data collection and edge computing

[0130] 1. Real-time collection of equipment data

[0131] The industrial Internet of Things gateway establishes communication with the injection molding machine PLC through OPCUA protocol and collects process parameters in real time:

[0132] Injection temperature 220°C, pressure 110MPa, injection time 5 seconds, and equipment running state (such as servo motor speed, mold opening and closing times).

[0133] The edge computing node deployed on the workshop edge server pulls the production site video stream (resolution 1080p, frame rate 30fps) through RTSP protocol, and its lightweight 3D convolutional neural network model identifies key nodes such as injection filling, pressure maintaining and cooling in real time, generating a spatiotemporal feature vector (dimension 128) containing spatial features (mold filling form) and temporal features (duration of each stage).

[0134] 2. Intelligent video hash aggregation

[0135] The dynamic time slicing algorithm divides the video stream into 15 1-second time slices according to the process beat signal (injection cycle 15 seconds) fed back by the PLC, and generates 1 feature vector for each time slice.

[0136] The Merkle tree construction module constructs 15 feature vectors into a Merkle tree, generates a root hash value (`0xa2f5...`), and after distributed storage through IPFS, writes the hash address to the block header (block height 123456) of the core production chain.

[0137] Three, blockchain storage layer:

[0138] Three-layer architecture collaborative storage

[0139] 1. Core production chain (Hyperledger Fabric consortium chain)

[0140] Store injection process parameters, Merkle root hash generated by edge computing, challenge code anchored data for real-time verification. Use hardware security module (HSM) to manage keys, automatically rotate encryption keys every 7 days to ensure physical level security.

[0141] Configure the main chain and backup chain redundant nodes, realize atomic level data synchronization through state channel technology, and guarantee data consistency (synchronization delay <100ms).

[0142] 2. Supply chain collaboration chain (Corda blockchain)

[0143] Earphone shell raw material supplier uploads ABS plastic particle traceability data through Corda platform:

[0144] Place of origin (Taiwan Taizhou, China), melt index test report (23g / 10min), batch number (202505

[0145] 01), smart contract automatically verifies data integrity, and broadcasts data hash value (`0x4e9b...`) to the core production chain every 10 minutes through notarization nodes.

[0146] The core production chain verifies the hash consistency through the Fabric cross-chain protocol, and completes the mapping storage of raw material data and production batch.

[0147] 3. User verification chain (Ethereum sidechain)

[0148] Digital twin ID generation:

[0149] Multi-dimensional feature extraction unit extracts the first 8 bits of injection molding machine MAC address (`00:0C:29`), production timestamp (`2025-05-19T16:00:00`), 6-bit random number (`789456`), and generates a 160-bit ID (`0xf3a7...`) through Keccak-256 algorithm.

[0150] ID is synchronized and written to PLC device register (address DB100.DBW0), NFC anti-counterfeit label EEPROM storage area (storage space 1KB), and stored in user verification chain through Solidity smart contract, supporting ERC-721 standard certificate query.

[0151] Four, anti-counterfeiting verification layer:

[0152] Spectral analysis and smart contract verification

[0153] 1. Standard template generation (first piece inspection phase)

[0154] The quantum dot spectrometer collects the spectral data of the first piece of earphone shell (wavelength range 400-800 nm, resolution 0.5 nm), and after SHA-256 hash operation (hash value `0xb2d5...`) writes into the core production chain as a standard spectral template.

[0155] The nanoscale microscope collects the 100 nm resolution micro-nano structure (such as "TWS-001" dot matrix pattern) generated by electron beam lithography on the surface of the earphone label, and stores it as binary data (size 512 KB) and uploads it.

[0156] 2. Terminal product anti-counterfeiting verification (consumer side)

[0157] The user uses the mobile phone NFC function to read the earphone label, triggering the physical feature extraction module:

[0158] The quantum dot spectrometer (integrated into the brand APP's external device) collects the shell spectral data on site, Base64 encoded (string length 1024 bytes) and uploaded to the smart contract execution module through the gRPC protocol.

[0159] The nanoscale microscope (mobile phone camera with AI vision algorithm) takes pictures of the label micro-nano structure, and performs pixel-level comparison with the on-chain standard template (similarity calculation error ≤0.1%).

[0160] The dynamic parameter verification engine divides the spectral peak (such as reflectivity at 550 nm) into basic parameters based on machine learning, and the micro-nano structure dot array spacing into custom parameters, and generates real-time constraints (such as peak deviation ≤±1.5 nm, spacing error ≤±5 nm).

[0161] Dual verification executor synchronously executes:

[0162] On-chain verification: spectral data hash and standard template hash comparison (match returns `true`);

[0163] Off-chain comparison: micro-nano structure similarity calculation (result 97.3%, higher than threshold 95%).

[0164] After verification, the user APP displays "genuine product", and associates with the display of raw material origin (Taiwan, China), production time (May 19, 2025 16:00), equipment ID (injection molding machine A-01) and other cross-chain data.

[0165] 3. Abnormal processing (assuming spectral comparison fails)

[0166] Triggering a three-level exception handling sub-rule:

[0167] The Bayesian network calls the core production chain (process parameters), supply chain collaboration chain (raw material batch), and user verification chain (device ID) data, and locates the abnormal source as "the melt index of raw material batch 202505-01 exceeds the standard" through the evidence propagation algorithm.

[0168] The system sends warning information to the factory control room through the message bus, and automatically triggers the process parameter adjustment instruction (reduce the injection pressure to 100 MPa, and extend the cooling time to 8 seconds) to the PLC device cluster, blocking the abnormal production.

[0169] Five, user trust enhancement layer:

[0170] Data visualization and cross-chain query

[0171] 1. Geographic information visualization

[0172] The geographic information visualization unit loads the three-dimensional map of the raw material supplier (Taiwan Taichung plant) based on CesiumJS. Users can view the raw material transportation path (GPS track uploaded by the logistics party through the supply chain collaboration chain) by clicking on the marker. The path nodes display the transportation time (May 15, 2025 10:00-16:00) and container temperature and humidity data (average temperature 25℃, humidity 45%).

[0173] 2. Time series data visualization

[0174] The time series data visualization unit uses D3.js to draw the temperature fluctuation curve of the injection molding machine. The horizontal axis represents time (precision minute level), and the vertical axis represents temperature value. Abnormal points (such as sudden temperature drop) are marked in red. Clicking on the marker can view the corresponding blockchain timestamp and operation record.

[0175] 3. Blockchain browser interaction

[0176] The user inputs the digital twin ID (`0xf3a7...`) and queries across chains through the blockchain browser component:

[0177] Core production chain:

[0178] View injection molding process parameter block (height 123456), Merkle root hash, and challenge code verification record;

[0179] User verification chain:

[0180] View digital certificate digest and NFT format anti-counterfeiting certificate (compliant with ERC-721 standard, tamper-proof);

[0181] Supply chain collaboration chain:

[0182] Trace raw material procurement transactions (supplier signature, quality inspection report hash).

[0183] This embodiment fully demonstrates the whole process closed loop of the system from user interaction → instruction encryption transmission → production data acquisition → blockchain storage → anti-fake verification → trust display, embodies the landing feasibility of the system in the field of industrial internet and consumer electronics, and significantly improves the production transparency and user trust.

[0184] The above merely provides the preferred embodiments of the present application; however, the protection scope of the present application is not limited thereto. Any person skilled in the art, according to the technical solution of the present application and the improved concept thereof, can make equivalent replacements or changes within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. An immersive visual product process management system, characterized by, The application relates to a user interaction layer, an instruction processing layer, a production data acquisition layer and a blockchain storage layer. The instruction processing layer comprises an instruction encryption transmission unit and a bidirectional authentication feedback unit, the instruction encryption transmission unit generates a 256-bit encryption key based on an ECC elliptic curve algorithm, adopts a TLS-1.3 protocol to construct a transmission layer security channel, and realizes data encryption transmission by matching a ChaCha20-Poly1305 encryption suite; the encrypted data is issued to a PLC device cluster through an industrial internet of things gateway. The production data acquisition layer comprises an industrial internet of things gateway and an edge computing node, the industrial internet of things gateway establishes a communication connection with the PLC device cluster through an OPCUA protocol, and the edge computing node is provided with a lightweight convolutional neural network model which is configured to identify production key nodes in a video stream in real time and generate a space-time feature vector. The customized instruction receiving module is used for analyzing user instructions into structured data and transmitting the structured data to the instruction processing layer, the instruction encryption transmission unit issues the structured data to the PLC device cluster after ECC encryption. The blockchain storage layer adopts a three-layer distributed ledger architecture of a core production chain, a supply chain cooperation chain and a user verification chain, the core production chain adopts a hardware security module to manage a key life cycle, realizes safe generation, storage and regular replacement of the key, ensures physical-level safety protection of the encryption process, the verification data generated by the real-time verification function module is synchronously written into the core production chain for storage, the space-time feature vector generated by the edge computing node is transmitted to the blockchain storage layer through the safety channel of the instruction processing layer, the blockchain storage layer verifies the space-time feature vector hash value through an intelligent contract, and automatically triggers the time sequence data visualization unit of the user trust enhancement layer to update a chart based on the verification result, wherein the core production chain stores production process parameters, the supply chain cooperation chain verifies raw material traceability data through an intelligent contract, and regularly synchronizes data hash values to the core production chain, and the user verification chain records a digital certificate digest and a digital twin ID. ​ The anti-counterfeiting verification layer includes a physical feature extraction module and an intelligent contract execution module. The physical feature extraction module is configured with a quantum dot spectrum analyzer and a nanoscale microscope. Spectrum data and nanoscale pattern features are collected by the quantum dot spectrum analyzer and the nanoscale microscope and are synchronously stored in the blockchain storage layer. The intelligent contract execution module is developed based on the Solidity language and is configured with a dynamic parameter verification engine and a multi-level exception handling mechanism. The multi-level exception handling mechanism includes a fault-tolerant verification rule library for storing exception judgment models and processing rules. The spectrum data collected by the physical feature extraction module is uploaded to the intelligent contract execution module after being encoded by Base64 and through the gRPC protocol. After receiving the spectrum data, the intelligent contract execution module automatically triggers a hash comparison with the standard template stored in the blockchain; The user trust enhancement layer obtains raw material traceability data, equipment operating parameters, and cross-chain storage information from the blockchain storage layer, including a geographic information visualization unit, a time series data visualization unit, and a blockchain browser component, for displaying raw material sources, analyzing equipment operating parameters, and querying cross-chain data.

2. The immersive visual product process management system of claim 1, wherein: The three-dimensional scene rendering module builds a virtual factory environment based on the Unity3D engine. The customized instruction receiving module uses natural language processing technology to analyze the barrage input. The data verification display module supports WebGL technology to realize interactive display of the production process digital twin view. The instruction encryption transmission unit generates a 256-bit encryption key based on the ECC elliptic curve algorithm. The two-way authentication feedback unit implements identity verification between the PLC device and the blockchain node through asymmetric encryption technology, forming a closed-loop verification link for instruction transmission. The core production chain of the blockchain storage layer is built based on the Hyperledger Fabric consortium chain framework and implements efficient transaction processing capability by configuring the PBFT consensus algorithm. The production process parameters stored by the blockchain storage layer are protected by zero-knowledge proof technology. The supply chain collaboration chain of the blockchain storage layer is built based on the Corda blockchain platform and automatically executes the verification of the synchronized raw material traceability data through the smart contract. The user verification chain is built based on the Ethereum sidechain technology and stores digital certificate digests using the Merkle Patricia tree structure, supporting non-forgery certificates based on the ERC-721 standard.

3. The immersive visual product process management system of claim 1, wherein: The real-time verification function module of the VR live room includes: A random challenge code generation unit generates a 128-bit random number as a challenge code based on the SHA-3 algorithm. An industrial vision feedback unit captures challenge code images displayed on the LED screen in real time through high-speed cameras deployed on the production site. A timestamp anchoring unit binds the challenge code image hash value with a timestamp accurate to the nanosecond level and writes it into the blockchain. The random challenge code generation unit integrates WebSocket push function, and pushes 128-bit random number to a production site LED screen in real time. The industrial visual feedback unit transmits a captured challenge code image to the timestamp anchoring unit through a gigabit Ethernet. The timestamp anchoring unit performs SHA-256 hash operation on the challenge code image to generate anchoring data containing a nanosecond timestamp, and writes the anchoring data into a core production chain through a PBFT consensus mechanism of Hyperledger Fabric.

4. The immersive visual product process management system of claim 1, wherein: The digital twin ID generation mechanism corresponding to the user verification chain comprises: a multi-dimensional feature extraction unit for extracting 8-bit MAC address of a device, a production timestamp and 6-bit random number generated by a random number generator; a hash fusion unit for fusing the multi-dimensional features to generate a 160-bit unique identifier by using a Keccak-256 algorithm; a three-way writing unit for synchronously writing the unique identifier into a PLC device register, a physical anti-fake label storage area and a blockchain smart contract storage mapping; the 160-bit identifier generated by the hash fusion unit is written into the PLC device register through a communication protocol, written into an EEPROM storage area of the physical anti-fake label through an NFC interface, and synchronously written into the user verification chain of the blockchain storage layer through a setID function of a Solidity smart contract.

5. The immersive visual product process management system of claim 1, wherein: The intelligent video hash aggregation technology of the edge computing node comprises: a dynamic time slice division algorithm for acquiring a process beat signal fed back by a PLC device in real time through an OPCUA protocol, and automatically adjusting a time slice length to millisecond-level precision; a space-time feature extraction network for extracting a space-time feature vector of a video frame by using a 3D convolutional neural network; a Merkle tree construction module for constructing the feature vector in a time slice into a Merkle tree structure to generate a root hash value; the dynamic time slice division algorithm adjusts the time slice length according to the process beat signal fed back by the PLC device, the space-time feature extraction network pulls a video stream through an RTSP protocol, the generated feature vector is transmitted to the Merkle tree construction module through a Kafka message queue, and the hash address is written into a block header of the core production chain after the root hash value is stored in an IPFS distributed storage.

6. The immersive visual product process management system of claim 1, wherein: The dynamic parameter verification engine of the smart contract execution module specifically comprises: a parameter classifier for dividing process parameters into a basic parameter set and a customized parameter set based on a machine learning algorithm; a dynamic rule generator for generating parameter constraint rules in real time according to customer instructions; a dual verification executor for simultaneously performing on-chain rule verification and off-chain physical feature comparison.

7. The immersive visual product process management system of claim 1, wherein: The intelligent contract execution module calls the standard spectrum template of the blockchain storage layer for similarity comparison of spectral data and nano-pattern features. The standard spectrum template is written into the core production chain storage by SHA-256 hash operation after being collected by a quantum dot spectrum analyzer during product first-piece inspection in the production data collection layer, and serves as benchmark data for anti-counterfeiting verification. The similarity comparison of the nano-pattern features is performed by electron beam lithography technology to generate micro-nano structures with a resolution of 100 nm on the surface of the label.

8. The immersive visual product process management system of claim 1, wherein: The core production chain serves as the core layer of the three-layer distributed ledger architecture, and is configured with a main chain and a backup chain as internal redundant nodes. Atomic-level data synchronization is achieved through state channel technology to ensure data redundancy and consistency. The supply chain collaboration chain adopts a distributed consensus algorithm, and the user verification chain achieves data synchronization based on side chain technology. In the three-layer distributed ledger architecture of the blockchain storage layer, the node configuration mechanisms of each chain include: The core production chain adopts an identity-based access control model, and the nodes are deployed in a physically isolated dedicated network. After the supply chain collaboration chain verifies the raw material traceability data through the intelligent contract, it broadcasts the data hash value to the core production chain every 10 minutes. After the core production chain verifies the hash consistency through the Fabric cross-chain protocol, data mapping storage is completed. The user verification chain adopts Plasma side chain technology. The Plasma side chain receives digital certificate update events from the main chain through the state channel, synchronizes to the MerklePatricia tree storage node in RLP encoding format, and achieves more than 10,000 query responses per second through the state channel.

9. The immersive visual product process management system of claim 1, wherein: The fault-tolerant verification rule library of the intelligent contract execution module includes a three-level abnormality judgment model, including a first-level abnormality processing sub-rule, a second-level abnormality processing sub-rule, and a third-level abnormality processing sub-rule. The first-level abnormality processing sub-rule predicts missing data through a Kalman filter for single-process data interruption. The Kalman filter subscribes to the historical data topic of the PLC device through OPCUA, predicts the missing data, and returns it to the cache area of the edge computing node through the CoAP protocol. The second-level abnormality processing sub-rule reconstructs the production process using a Petri net model for cross-process logical contradiction. The third-level abnormality processing sub-rule triggers a Bayesian network for abnormality source analysis for multi-link data missing. The Bayesian network calls the multi-chain data of the blockchain storage layer through the gRPC service, locates the abnormal source through the evidence propagation algorithm, sends warning information to the user interaction layer through the message bus, and triggers the process parameter adjustment instruction to the PLC device cluster through the instruction processing layer.

10. The immersive visual product process management system of claim 1, wherein: The user trust enhancement layer includes: A geographic information visualization unit that realizes three-dimensional geographic information display of raw material sources based on CesiumJS, supporting user interactive raw material transportation path tracing; A time series data visualization unit that constructs a time series analysis chart of device operating parameters using D3.js. Blockchain browser component, supporting cross-chain data query of Ethereum and consortium chain, providing block verification, transaction tracking and smart contract calling functions.

Images