Remanufacturing transaction platform input mode selection method based on block chain technology

Abstract

The invention relates to the technical field of information technology, and discloses a block chain technology-based remanufacturing transaction platform input mode selection method, which comprises the steps of collecting and preprocessing product life cycles, user behaviors and market feedback data through edge computing nodes deployed at participants of a remanufacturing industry chain; generating local model updating information; local acquisition and preprocessing of multi-source data are realized by deploying edge computing nodes, distributed collaborative modeling is completed on the premise of guaranteeing data privacy in combination with a federated learning algorithm, a market trend and a platform operation state are analyzed by using a global AI prediction model, and a preliminary mode prediction result is output; and then, by introducing a carbon integral and green value evaluation mechanism, the environmental protection attributes are quantized and are subjected to dynamic weighted fusion with a prediction result, a comprehensive input mode recommendation scheme is generated, automatic execution of mode switching and excitation operation is further realized through an intelligent contract, and continuous iterative optimization is performed on the model based on platform feedback data.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for selecting an input mode for a remanufacturing trading platform based on blockchain technology. Background Technology

[0002] A blockchain-based remanufacturing trading platform input mode selection method is an innovative technical solution that aims to optimize the trading process of remanufactured products and improve the platform's intelligence level and environmental benefits by integrating advanced technologies such as blockchain, edge computing, federated learning, and smart contracts.

[0003] Compared with existing technologies, this invention has significant innovation and practicality in integrating blockchain, artificial intelligence, edge computing, and green manufacturing. However, it still faces several technical challenges and potential problems. Existing remanufacturing trading platforms typically rely on centralized systems for data management and model decision-making, making it difficult to guarantee data transparency, security, and trust mechanisms for multi-party collaboration. Furthermore, they lack effective quantification and incentive mechanisms for green value, resulting in deficiencies in environmental orientation and user participation. In addition, traditional prediction models often employ centralized training methods, posing risks of data privacy leaks and struggling to adapt to dynamic changes in distributed scenarios. While this invention introduces federated learning and edge computing to address these issues, it still faces certain technical bottlenecks in model aggregation efficiency, handling of node heterogeneity, and communication overhead. Simultaneously, although carbon credit calculation and green value assessment mechanisms enhance the platform's sustainable development attributes, they still rely on high-quality underlying data support for the uniformity of assessment standards, data authenticity, and incentive fairness. Incomplete or biased data collection from edge nodes will directly affect the accuracy of global model predictions and recommendations. Summary of the Invention

[0004] In view of the problems existing in the current method for selecting the input mode of a remanufacturing trading platform based on blockchain technology, this invention is proposed.

[0005] Therefore, the purpose of this invention is to provide a method for selecting the input mode of a remanufacturing trading platform based on blockchain technology.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: including, a. Collect and preprocess product lifecycle, user behavior and market feedback data by deploying edge computing nodes in various participants in the remanufacturing industry chain to generate local model update information; b. Upload the local model update information generated by each node to the federated learning coordination node in the blockchain network. The coordination node then aggregates the models using the federated learning algorithm to generate a global AI prediction model. c. Based on the global AI prediction model, analyze market demand trends and platform operation status, and output preliminary prediction results of platform investment mode; d. Calculate carbon credits and generate green value assessment data based on the carbon emissions, resource utilization rate and environmental attributes of remanufactured products, and record them in the blockchain; e. Combine the green value assessment data to dynamically weight and integrate the preliminary prediction results to generate a comprehensive investment model recommendation scheme, which includes the priority and applicable scope of B2B, B2C or C2C models; f. Encapsulate the recommended investment mode scheme into a smart contract and deploy it to a blockchain platform. The smart contract will automatically perform mode switching, carbon credit distribution and incentive operations based on market conditions, user credit and green contribution. g. Based on the platform's operational feedback data, the AI ​​prediction model and the green value assessment model are continuously iterated and optimized to form a closed-loop evolution mechanism, thereby achieving adaptive and sustainable improvement of the investment mode selection method.

[0007] As a preferred embodiment of the blockchain-based remanufacturing transaction platform investment mode selection method described in this invention, the edge computing nodes are deployed among multiple participants in the remanufacturing industry chain, including product recyclers, testing and dismantling centers, remanufacturing plants, sales and service platforms, and end-user devices, for collecting local data and performing preprocessing.

[0008] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the product lifecycle data includes product model, manufacturing time, service life, maintenance records, dismantling status, and carbon emission estimates. The data is derived from RFID chips, QR codes, sensors, and historical maintenance systems.

[0009] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the user behavior data includes user purchase frequency, product evaluation, complaints and suggestions, credit rating and recycling willingness, and the data is collected through the sales platform, after-sales service system and user terminal equipment.

[0010] As a preferred embodiment of the input mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the federated learning aggregation adopts the FedAvg algorithm, in which the coordinating node in the blockchain network performs a weighted average of the local model update information from each edge node to generate a global AI prediction model.

[0011] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the green value assessment data is calculated using the following indicators: carbon emissions, material recycling rate, energy consumption intensity, and environmental contribution. The assessment results are recorded in the blockchain for subsequent investment mode recommendations.

[0012] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the dynamic weighted fusion mechanism adopts a linear weighting method to fuse the preliminary prediction results output by the global AI prediction model with green value assessment data to generate a comprehensive investment mode recommendation scheme.

[0013] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the smart contract automatically executes investment mode switching, carbon credit distribution, and penalty mechanisms based on market conditions, user credit rating, and green contribution, ensuring the fairness and incentive of platform operation.

[0014] As a preferred embodiment of the investment mode selection method for a remanufacturing trading platform based on blockchain technology described in this invention, the platform operation feedback data includes model prediction accuracy, user satisfaction, carbon credit utilization rate, and transaction success rate. The data is used for periodic iterative optimization of the AI ​​prediction model and the green value assessment model.

[0015] As a preferred embodiment of the input mode selection method for a remanufacturing transaction platform based on blockchain technology described in this invention, the edge computing node performs cleaning, feature extraction, encryption, and digital signature processing on the collected data locally, generates a data hash value, and uploads it to the blockchain platform to achieve data traceability and privacy protection.

[0016] The beneficial effects of this invention are as follows: By deploying edge computing nodes, local collection and preprocessing of multi-source data are achieved. Combined with federated learning algorithms, distributed collaborative modeling is completed while ensuring data privacy. A global AI prediction model is used to analyze market trends and platform operation status, outputting preliminary model prediction results. Then, by introducing carbon credits and green value assessment mechanisms, environmental protection attributes are quantified and dynamically weighted and integrated with the prediction results to generate a comprehensive investment model recommendation scheme. Furthermore, smart contracts are used to automate the execution of model switching and incentive operations. Based on platform feedback data, the model is continuously iterated and optimized to form a closed-loop evolution mechanism, thereby achieving deep integration and collaborative improvement of the platform in terms of data security, intelligent decision-making, green orientation, and adaptive optimization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the data preprocessing and local model update process of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0021] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0022] Example Reference Figure 1 As an embodiment of the present invention, a method for selecting the input mode of a remanufacturing trading platform based on blockchain technology is provided, the apparatus comprising:

[0023] a. Collect and preprocess product lifecycle, user behavior and market feedback data by deploying edge computing nodes in various participants in the remanufacturing industry chain to generate local model update information; b. Upload the local model update information generated by each node to the federated learning coordination node in the blockchain network. The coordination node then aggregates the models using the federated learning algorithm to generate a global AI prediction model. c. Analyze market demand trends and platform operation status based on a global AI prediction model, and output preliminary prediction results of platform investment mode; d. Calculate carbon credits and generate green value assessment data based on the carbon emissions, resource utilization rate and environmental attributes of remanufactured products, and record them in the blockchain; e. Combine the green value assessment data with the preliminary prediction results for dynamic weighted integration to generate a comprehensive investment model recommendation scheme. The recommendation scheme includes the priority and applicable scope of B2B, B2C or C2C models. f. Encapsulate the recommended investment model into a smart contract and deploy it to the blockchain platform. The smart contract will automatically execute mode switching, carbon credit distribution and incentive operations based on market conditions, user credit and green contribution. g. Based on platform operation feedback data, continuously iterate and optimize the AI ​​prediction model and green value assessment model to form a closed-loop evolution mechanism, and realize the adaptive and sustainable improvement of the input mode selection method.

[0024] Specifically, by deploying edge computing nodes across all participants in the remanufacturing industry chain, distributed collection and local modeling of product lifecycle, user behavior, and market feedback data are achieved. Combined with blockchain and federated learning technologies, multi-party collaborative training is completed while ensuring data privacy and security, generating a global AI prediction model for analyzing market demand and platform operational status, and outputting preliminary investment pattern prediction results. Furthermore, a green value assessment mechanism is introduced, quantifying environmental indicators such as carbon emissions and resource utilization rates into carbon credits and green assessment data, and dynamically weighting and integrating them with the prediction results to generate a comprehensive recommendation scheme that balances market benefits and environmental orientation. This scheme forms a decentralized and trustworthy execution mechanism through automatic execution mode switching and incentive operations via smart contracts. Simultaneously, the AI ​​model and green assessment model are continuously iterated and optimized based on platform operation feedback data, constructing a closed-loop evolution mechanism that enables the system to have adaptive optimization and long-term sustainable improvement capabilities.

[0025] Edge computing nodes are deployed across multiple participants in the remanufacturing industry chain, including product recyclers, testing and dismantling centers, remanufacturing plants, sales and service platforms, and end-user devices, to collect local data and perform preprocessing.

[0026] Furthermore, edge computing nodes are deployed across multiple key participants in the remanufacturing industry chain, including product recyclers, testing and dismantling centers, remanufacturing plants, sales and service platforms, and end-user devices. This enables end-to-end data collection and local preprocessing from product source to end-user. This deployment not only improves the real-time and comprehensiveness of data acquisition but also reduces data transmission latency and centralized processing pressure through localized processing, ensuring data privacy and system security. During operation, each node collects heterogeneous data based on its role and uniformly transforms it into local model update information that can be used for federated learning, providing high-quality input for subsequent global modeling. In the future, this can be expanded to support more types of participants, such as logistics service providers and regulatory agencies, further enhancing the platform's collaborative capabilities and governance level. It can also improve local processing efficiency by introducing more advanced edge AI algorithms, providing solid support for building an intelligent, green, and trustworthy remanufacturing ecosystem.

[0027] Product lifecycle data includes product model, manufacturing time, service life, maintenance records, dismantling status, and carbon emission estimates. The data comes from RFID chips, QR codes, sensors, and historical maintenance systems.

[0028] The product lifecycle data is collected through multiple sources, including RFID chips, QR codes, sensors, and historical maintenance systems. It covers key information such as product model, manufacturing time, service life, maintenance records, dismantling status, and carbon emission estimates, enabling accurate tracking and digital representation of the product's entire lifecycle status. This data system provides foundational support for subsequent AI modeling, green assessment, and model recommendation during operation, enhancing the platform's ability to judge the feasibility and environmental value of remanufacturing. In the future, it can be further expanded to deep integration with IoT devices to improve the automation level of data collection and combine with blockchain to achieve product traceability and full-process carbon footprint supervision, providing a data foundation for building a green supply chain and carbon trading system.

[0029] User behavior data includes user purchase frequency, product reviews, complaints and suggestions, credit rating, and recycling willingness. The data is collected through sales platforms, after-sales service systems, and user terminal devices.

[0030] Ideally, user behavior data is collected through sales platforms, after-sales service systems, and user terminal devices, covering dimensions such as purchase frequency, product reviews, complaints and suggestions, credit rating, and recycling willingness. This comprehensively reflects users' preferences and willingness to participate in the use of remanufactured products. This type of data provides key support for the market adaptability analysis of platform investment models, user credit assessment, and the design of green incentive mechanisms, enhancing the personalization and feasibility of recommendation schemes. In the future, it can be further expanded to combine user profiles and behavior prediction models to improve user participation and platform stickiness. Furthermore, by linking with the carbon credit system, it can achieve quantitative incentives for users' green consumption behavior and promote the formation of a user-driven sustainable remanufacturing ecosystem.

[0031] Federated learning aggregation uses the FedAvg algorithm, in which the coordinating nodes in the blockchain network perform a weighted average of the local model update information from each edge node to generate a global AI prediction model.

[0032] It should be noted that by setting up a coordinating node in the blockchain network and weighting the local model update information from each edge computing node to generate a global AI prediction model, this mechanism achieves multi-party collaborative modeling while ensuring data privacy. It effectively solves the problems of data silos, high transmission costs, and privacy leaks in traditional centralized training. In the specific implementation process, each edge node independently trains its local model based on locally collected product lifecycle data, user behavior data, and market feedback data, and only uploads model parameter updates rather than the original data. The coordinating node performs weighted aggregation based on weight factors such as the amount of data and model contribution of each node to generate a global model with greater generalization ability. Compared with the existing technology that relies on a central server for centralized data processing, this invention has significantly improved model training efficiency, data security, and modeling accuracy. In the future, differential privacy, homomorphic encryption, and other technologies can be further introduced to enhance the security of the federated learning process, and combined with a dynamic weight adjustment mechanism to improve the model convergence speed and stability, providing solid technical support for building an intelligent, trustworthy, and sustainable remanufacturing trading platform.

[0033] Green value assessment data is calculated using the following indicators: carbon emissions, material recycling rate, energy consumption intensity, and environmental contribution. The assessment results are recorded in the blockchain for subsequent investment model recommendations.

[0034] The system constructs a green value assessment system using key indicators such as carbon emissions, material recycling rate, energy consumption intensity, and environmental contribution. This system quantifies the environmental benefits of remanufactured products and records the assessment results on the blockchain to ensure data immutability and traceability. This provides a green-oriented basis for recommending investment models on the platform. This mechanism breaks through the limitations of traditional remanufacturing trading platforms that only focus on economic benefits and ignore environmental attributes. It achieves the organic integration of environmental value and market decision-making. In the specific implementation process, each participant collects data such as carbon footprint and resource utilization throughout the product lifecycle through edge computing nodes. After uploading the data to the blockchain, the smart contract automatically triggers the green assessment process, generating standardized green value assessment data. This data serves as an important input to the dynamic weighted fusion model, directly affecting the recommendation priority and applicable scope of different investment models such as B2B, B2C, and C2C.

[0035] The dynamic weighted fusion mechanism uses a linear weighting method to fuse the preliminary prediction results output by the global AI prediction model with the green value assessment data to generate a comprehensive investment model recommendation scheme.

[0036] In summary, a dynamic fusion mechanism is constructed using a linear weighting method. This mechanism weights and fuses the market trend predictions output by the global AI prediction model with green value assessment data to generate a comprehensive investment model recommendation scheme that balances economic benefits and environmental orientation. During implementation, this mechanism achieves flexible matching between AI prediction results and green assessment indicators by setting dynamic weighting factors (such as market demand fluctuation coefficients, policy guidance indices, and green priority thresholds). This ensures that the recommendation results respond to market changes while reflecting the value orientation of green manufacturing. Compared to traditional remanufacturing platforms that rely solely on single economic indicators or static rules for model selection, this scheme has significant advantages in model fusion strategies, multi-objective decision-making mechanisms, and green incentive responses. Blockchain recording of the fusion process and weight adjustment logs further guarantees the transparency and auditability of the recommendation mechanism. In the future, nonlinear fusion models, multi-objective optimization algorithms, or reinforcement learning mechanisms can be introduced to enhance the adaptive capabilities and decision-making intelligence of the recommendation system, enabling the platform to possess greater flexibility and sustainable development capabilities under complex market environments and policy orientations.

[0037] The smart contract automatically executes input mode switching, carbon credit distribution, and penalty mechanisms based on market conditions, user credit rating, and green contribution, ensuring the fairness and incentive of the platform's operation.

[0038] Ideally, by leveraging smart contract technology, market conditions, user credit ratings, and green contributions are used as core decision-making factors to automatically execute key operations such as switching investment modes, distributing carbon credits, and penalizing violations. This constructs a decentralized and trustworthy platform operation mechanism. During implementation, based on the immutability of blockchain, this mechanism ensures that all decision-making processes and execution results are open and transparent, effectively improving the platform's governance efficiency and operational fairness. Smart contracts dynamically adjust the activation priority of B2B, B2C, or C2C modes based on market trends output by a global AI prediction model. Simultaneously, they control users' trading permissions and carbon credit acquisition limits based on their credit ratings, further differentiating incentives based on green contributions to form a positive feedback mechanism. This enhances users' enthusiasm for participating in green remanufacturing. Compared to traditional platforms that rely on manual intervention or centralized systems for rule enforcement, this solution demonstrates significant advantages in automated control, rule transparency, and precise incentives. In the future, it can be expanded to a multi-level contract linkage mechanism, introduce a dynamic reward and punishment model and a cross-chain governance protocol, further enhance the platform's adaptability and ecosystem synergy in complex business scenarios, and provide solid technical support for building an intelligent, green and trustworthy remanufacturing transaction system.

[0039] The platform's operational feedback data includes model prediction accuracy, user satisfaction, carbon credit utilization rate, and transaction success rate. This data is used to periodically iterate and optimize the AI ​​prediction model and the green value assessment model.

[0040] Furthermore, by collecting platform operation feedback data, including key indicators such as model prediction accuracy, user satisfaction, carbon credit utilization rate, and transaction success rate, a closed-loop feedback optimization mechanism is constructed. This mechanism is used to periodically iterate and optimize the AI ​​prediction model and the green value assessment model, thereby achieving adaptive evolution and performance improvement of the system. In the specific implementation process, each edge computing node continuously collects user behavior, market response, and transaction execution data, and uploads them to the model optimization module through the blockchain network. The smart contract triggers the model retraining and parameter tuning process according to a preset cycle (such as weekly or monthly), and updates the global model in combination with the federated learning mechanism to ensure its continuous adaptability to market dynamics and user preferences. Compared with the existing remanufacturing trading platform's problems of lagging model updates and lack of feedback mechanisms, this solution shows significant advantages in model self-learning ability, feedback data-driven optimization, and dynamic correction of green assessment. By introducing a data-driven optimization loop, the platform's intelligence level and operational efficiency have been improved, as well as the timeliness and accuracy of green value assessment. This further promotes the remanufacturing transaction model towards efficiency, intelligence, and sustainability. In the future, it can be further expanded to include reinforcement learning mechanisms, multi-objective optimization algorithms, and anomaly detection models to enhance the system's robustness and decision-making quality in complex environments.

[0041] Edge computing nodes clean, extract features, encrypt and digitally sign the collected data locally, generate data hash values ​​and upload them to the blockchain platform to achieve data traceability and privacy protection.

[0042] In this system, edge computing nodes locally clean, extract features, encrypt, and digitally sign the collected raw data, generating a unique data hash value which is then uploaded to the blockchain platform. This ensures data traceability and integrity verification while protecting data privacy. This mechanism overcomes the shortcomings of traditional centralized data processing models, such as easy data leakage, tampering, and difficulty in traceability. Local preprocessing effectively reduces data redundancy and transmission load, while encryption and signature technologies ensure the authenticity and reliability of data sources and the immutability of content. In practice, each node performs structured processing on product lifecycle data, user behavior data, and market feedback information according to its role. After extracting key features, only model updates or hash digests are uploaded, avoiding centralized exposure of raw sensitive information and significantly improving system security and compliance. Compared to the high risks of centralized data storage and insufficient user privacy protection commonly found in existing technologies, this solution introduces lightweight processing and secure encapsulation mechanisms at the edge, achieving pre-emptive and decentralized data governance.

[0043] It should be noted that by deploying edge computing nodes across all participants in the remanufacturing industry chain, distributed collection and local preprocessing of product lifecycle, user behavior, and market feedback data are achieved. Combined with mechanisms such as data cleaning, feature extraction, encryption, and digital signatures, data hashes are generated and uploaded to the blockchain to ensure data privacy, integrity, and traceability. Based on this, federated learning coordination nodes are used to weighted aggregate the local model update information uploaded by each edge node, generating a global AI prediction model for analyzing market demand trends and platform operational status, outputting preliminary investment pattern prediction results. Simultaneously, a green value assessment system is constructed based on carbon emissions, material recycling rate, energy consumption intensity, and environmental contribution. Carbon credits are calculated based on indicators such as [list of indicators], and the assessment results are stored on the blockchain. Through a dynamic weighted fusion mechanism, AI prediction results are organically combined with green assessment data to generate a comprehensive recommendation scheme that balances market benefits and environmental orientation. This scheme is then encapsulated as a smart contract and deployed to the blockchain platform. The smart contract automatically executes mode switching, carbon credit distribution, and incentive operations based on market conditions, user credit ratings, and green contributions. During platform operation, feedback data is continuously collected, including model prediction accuracy, user satisfaction, carbon credit utilization rate, and transaction success rate. This data is used to periodically iterate and optimize the AI ​​model and green assessment model, building a closed-loop evolution mechanism that enables the system to have continuous adaptive optimization and long-term sustainable improvement capabilities.

[0044] In summary, by leveraging edge computing nodes to achieve end-to-end data collection and local preprocessing, the real-time nature, security, and privacy of data are effectively guaranteed. Combining blockchain and federated learning technologies enables multi-party collaborative modeling, overcoming the limitations of data silos and centralized processing in traditional platforms, significantly improving model prediction accuracy and system scalability. By introducing a green value assessment mechanism, environmental indicators such as carbon emissions, material recycling rates, and energy consumption intensity are quantified into carbon credits and recorded on the blockchain, introducing green guidance factors into platform decision-making and promoting the low-carbon and sustainable development of remanufacturing transactions. A dynamic weighted fusion mechanism organically combines AI prediction results with green assessment data to generate comprehensive recommendation schemes that balance market benefits and environmental value, enhancing the scientific rigor and flexibility of platform decision-making. Through automatic smart contract execution mode switching, carbon credit distribution, and incentive mechanisms, a decentralized and trustworthy operating system is constructed, enhancing the platform's fairness and execution efficiency. Simultaneously, based on platform operation feedback data, the AI ​​model and green assessment model are continuously iterated and optimized, forming a closed-loop evolution mechanism that enables the system to have self-learning and self-optimization capabilities, further improving the platform's intelligence level and long-term sustainable development capabilities.

[0045] It is important to note that the constructions and arrangements of this application shown in several different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who consult this disclosure will readily understand that many modifications are possible (e.g., changes in the size, dimensions, structure, shape, and proportions of various elements, as well as parameter values ​​(e.g., temperature, pressure, etc.), mounting arrangements, use of materials, color, orientation, etc.) without substantially departing from the novel teachings and advantages of the subject matter described in this application). For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of elements may be inverted or otherwise altered, and the nature or number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of the invention. The order or sequence of any process or method steps may be changed or rearranged according to alternative embodiments. In the claims, any "device plus function" clause is intended to cover the structure described herein that performs the function, and not only structurally equivalent but also equivalent in structure. Other substitutions, modifications, alterations, and omissions may be made in the design, operation, and arrangement of the exemplary embodiments without departing from the scope of the invention. Therefore, the present invention is not limited to the specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0046] Furthermore, in order to provide a concise description of exemplary embodiments, not all features of actual embodiments (i.e., those features that are not relevant to the currently considered best mode for carrying out the invention, or those features that are not relevant to implementing the invention) may be omitted.

[0047] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for selecting an investment model for a remanufacturing trading platform based on blockchain technology, characterized in that: include, a. Collect and preprocess product lifecycle, user behavior and market feedback data by deploying edge computing nodes in various participants in the remanufacturing industry chain to generate local model update information; b. Upload the local model update information generated by each node to the federated learning coordination node in the blockchain network. The coordination node then aggregates the models using the federated learning algorithm to generate a global AI prediction model. c. Based on the global AI prediction model, analyze market demand trends and platform operation status, and output preliminary prediction results of platform investment mode; d. Calculate carbon credits and generate green value assessment data based on the carbon emissions, resource utilization rate and environmental attributes of remanufactured products, and record them in the blockchain; e. Combine the green value assessment data to dynamically weight and integrate the preliminary prediction results to generate a comprehensive investment model recommendation scheme, which includes the priority and applicable scope of B2B, B2C or C2C models; f. Encapsulate the recommended investment mode scheme into a smart contract and deploy it to a blockchain platform. The smart contract will automatically perform mode switching, carbon credit distribution and incentive operations based on market conditions, user credit and green contribution. g. Based on the platform's operational feedback data, the AI ​​prediction model and the green value assessment model are continuously iterated and optimized to form a closed-loop evolution mechanism, thereby achieving adaptive and sustainable improvement of the investment mode selection method.

2. The method for selecting an investment model for a remanufacturing trading platform based on blockchain technology according to claim 1, characterized in that: The edge computing nodes are deployed across multiple participants in the remanufacturing industry chain, including product recyclers, testing and dismantling centers, remanufacturing plants, sales and service platforms, and end-user devices, to collect local data and perform preprocessing.

3. The method for selecting an investment model for a remanufacturing trading platform based on blockchain technology according to claim 2, characterized in that: The product lifecycle data includes product model, manufacturing time, service life, maintenance records, dismantling status, and carbon emission estimates. The data is derived from RFID chips, QR codes, sensors, and historical maintenance systems.

4. The method for selecting an investment model for a remanufacturing trading platform based on blockchain technology according to claim 3, characterized in that: The user behavior data includes user purchase frequency, product reviews, complaints and suggestions, credit rating, and recycling intention. The data is collected through the sales platform, after-sales service system, and user terminal devices.

5. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology according to claim 4, characterized in that: The federated learning aggregation uses the FedAvg algorithm, in which the coordinating nodes in the blockchain network perform a weighted average of the local model update information from each edge node to generate a global AI prediction model.

6. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology according to claim 5, characterized in that: The green value assessment data is calculated using the following indicators: carbon emissions, material recycling rate, energy consumption intensity, and environmental contribution. The assessment results are recorded in the blockchain for subsequent investment model recommendations.

7. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology according to claim 6, characterized in that: The dynamic weighted fusion mechanism uses a linear weighting method to fuse the preliminary prediction results output by the global AI prediction model with the green value assessment data to generate a comprehensive investment model recommendation scheme.

8. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology according to claim 7, characterized in that: The smart contract automatically executes input mode switching, carbon credit distribution, and penalty mechanisms based on market conditions, user credit rating, and green contribution, ensuring the fairness and incentive of the platform's operation.

9. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology as described in claim 8, characterized in that: The platform's operational feedback data includes model prediction accuracy, user satisfaction, carbon credit utilization rate, and transaction success rate. This data is used for periodic iterative optimization of the AI ​​prediction model and the green value assessment model.

10. The method for selecting an investment mode for a remanufacturing trading platform based on blockchain technology according to claim 9, characterized in that: The edge computing node performs local cleaning, feature extraction, encryption, and digital signature processing on the collected data, generates a data hash value, and uploads it to the blockchain platform to achieve data traceability and privacy protection.

Images