A digital asset management method and system based on multi-objective optimization

Abstract

This invention discloses a digital asset management method and system based on multi-objective optimization. The method first constructs an enhanced sharding management network architecture, introducing an organizational affinity analysis method based on interaction frequency and a four-dimensional resource dynamic monitoring mechanism, combined with an autoregressive differential moving average model to achieve load prediction. Second, it establishes a multi-objective optimization framework based on organizational affinity, load balancing, and privacy protection, employing a differential privacy protection mechanism and dynamic privacy budget management, and achieving Pareto optimality through gradient descent. Finally, it designs an intelligent sharding decision algorithm and an enhanced load balancing strategy, constructing a hierarchical traceability architecture based on Merkle trees, and combining zero-knowledge proof and commitment mechanisms to achieve trusted traceability throughout the entire lifecycle. This protects the privacy of inter-organizational relationships and the security of system load information, optimizes system resource allocation and cross-shard collaboration efficiency, and maximizes the overall performance and availability of digital asset sharing services.

Description

Technical Field

[0001] This invention belongs to the field of digital asset sharing and traceability service technology, specifically relating to a digital asset management method and system based on multi-objective optimization. Background Technology

[0002] Digital assets refer to assets owned or controlled by enterprises or individuals, existing in the form of electronic data, and possessing characteristics such as valueability, tradability, and liquidity. With the development of big data and artificial intelligence technologies, the value and importance of digital assets are increasingly prominent, becoming a crucial component of corporate strategic resources. Digital asset sharing refers to the sharing or trading of data resources among different organizations, enterprises, or individuals. It helps enhance data value, promote cooperation, optimize the allocation of social resources, and improve efficiency.

[0003] Traditionally, the management and circulation of digital assets typically employ a centralized architecture, relying on third-party platforms for registration, storage, and sharing. However, under a centralized architecture, data faces risks such as tampering and leakage during transmission and storage, and may encounter issues like transaction fraud and single points of failure, leading to a lack of protection for asset owners' rights and triggering a crisis of trust. Therefore, constructing a decentralized digital asset sharing model to achieve asset ownership confirmation, trading, and traceability based on privacy protection has become a crucial issue that urgently needs to be addressed.

[0004] Blockchain technology, with its distributed and immutable characteristics, provides a decentralized solution for digital asset management. A blockchain-based digital asset management system helps achieve distributed notarization and trusted traceability in the digital asset transaction process, protecting copyright and ownership rights. However, with the increasing demand for traceability, blockchain-based traceability management solutions face a serious scalability crisis, especially when processing large-scale, multi-dimensional traceability data, where such systems exhibit significant performance bottlenecks. Sharding technology is an effective way to improve blockchain scalability, allowing transaction and traceability data to be distributed across different subsets (i.e., shards) for parallel processing, significantly increasing overall system throughput. However, existing sharding technologies often focus only on improving system efficiency and load balancing, neglecting the privacy protection needs unique to digital asset sharing, the diversity of resources, and the complexity and diversity of needs in intra- and inter-organizational collaboration.

[0005] Undoubtedly, digital asset sharing systems have stringent requirements for both privacy protection and resource utilization. On the one hand, the asset traceability chain typically involves multiple organizations, necessitating both the verifiability of traceability information and the protection of each party's commercial privacy. On the other hand, traceability queries are characterized by suddenness and unevenness, requiring the system to efficiently allocate computing resources to cope with varying workloads. Traditional sharding strategies typically employ random allocation, or rely on graph relationships or single load metrics, making it difficult to simultaneously meet these requirements. Furthermore, while considerable research has focused on improving privacy protection in digital asset sharing scenarios, related solutions overly rely on stacking multiple cryptographic methods, resulting in high computational and communication costs. Therefore, it is necessary to explore novel sharding strategies that can effectively balance system load, protect traceability privacy, and ensure efficient cross-domain and cross-organizational collaboration. Summary of the Invention

[0006] The purpose of this invention is to provide a digital asset management method and system based on multi-objective optimization.

[0007] In a first aspect, the present invention provides a digital asset management method based on multi-objective optimization, the method comprising:

[0008] Initialize a differential privacy budget and a multi-objective weight vector, and assign different organizations to different shards; perform affinity grouping based on the frequency of interactions between organizations to obtain multiple affinity groups; obtain the weighted load of different shards through the resource state groups and corresponding resource weight vectors of each shard; perform differential privacy protection on the weighted load based on the differential privacy budget to obtain the privacy-protected weighted load; construct affinity policies and enhanced load balancing policies respectively; if the affinity group is not empty, obtain the optimal shard through the affinity policy; the affinity policy obtains the corresponding primary shard based on the affinity group, and introduces a multi-objective weight vector to obtain a dynamic threshold; if the final shard load of the primary shard is less than or equal to the dynamic threshold, the primary shard is taken as the optimal shard;

[0009] If the affinity group is empty or the final shard load of the primary shard is greater than the dynamic threshold, an enhanced load balancing strategy is adopted to obtain the optimal shard. The enhanced load balancing strategy obtains a stationary load sequence through an autoregressive summation moving average model, and obtains the final shard load based on the stationary load sequence and the privacy-preserving weighted load using a dynamic weighting mechanism. A load balancing score is obtained based on the final shard load, and a multi-objective weighted score is calculated based on the load balancing score and the multi-objective weight vector to select the optimal shard.

[0010] An optimization model is constructed based on the multi-objective weight vector corresponding to the multi-objective, and the multi-objective weight vector is updated according to the optimization model corresponding to the optimal partition; a dynamic monitoring mechanism is constructed, and the above process of obtaining the optimal partition is re-executed when the system index exceeds the preset threshold.

[0011] Construct a hierarchical traceability architecture based on hash trees, and combine zero-knowledge proof and commitment mechanisms to achieve trusted traceability throughout the entire lifecycle.

[0012] Preferably, the multi-objective weight vector includes the weights of the data locality objective, the load balancing objective, and the privacy protection objective.

[0013] Preferably, the dynamic threshold in the affinity strategy is obtained using the following method:

[0014]

[0015] in, This is the minimum value of all current weighted loads; The benchmark adjustment coefficient; This indicates the benefits of data locality; This represents the load difference between the current shard and the primary shard. Indicates the level of privacy protection; These are the weights for locality objectives, load balancing objectives, and privacy protection objectives, respectively.

[0016] Preferably, the multi-objective weighted score is the sum of the products of the data locality score, the load balancing score, and the privacy protection score, respectively, and their corresponding multi-objective weights.

[0017] Preferably, the specific process of affinity grouping is as follows: calculate the affinity score between different organizations based on the frequency of interaction between organizations, construct an affinity topology map based on a threshold, and use a community detection algorithm to assign organizations belonging to the same community to the same affinity group.

[0018] As a preferred approach, after obtaining the weighted load, an adaptive weight adjustment algorithm is used to identify bottleneck resources, increase the weight corresponding to the bottleneck resources, and proportionally reduce the weights of other resources so that the sum of the weights of all resources after adjustment is 1; the weighted load is then updated based on the adjusted resource weights.

[0019] Preferably, the weighting coefficients in the dynamic weighting mechanism The expression for dynamically adjusting based on prediction accuracy is:

[0020]

[0021] in, These are the initial weighting coefficients; This is the root mean square error; and This is a hyperparameter.

[0022] Preferably, the optimization model includes a function to minimize the cross-shard transaction ratio, a function to minimize the load imbalance, and a function to minimize the privacy leakage risk; the resource status group includes CPU utilization, memory usage, storage capacity, and network bandwidth.

[0023] Preferably, the privacy-protected weighted load ;in, Weighted load; Lap (·) represents the Laplace distribution; Δ L Indicates the maximum range of weighted load variation; The privacy budget is used to protect the privacy of the load; the final sharded load in the affinity strategy is a weighted load after privacy protection.

[0024] Secondly, this invention provides a digital asset management system based on multi-objective optimization, which is used to execute the aforementioned digital asset management method. The digital asset management system includes a digital asset application layer, an enhanced sharding management layer, an optimal sharding selection layer, and a digital asset traceability layer. The digital asset application layer includes digital asset owners, users, and smart contracts, responsible for asset registration, access, and transaction management. The enhanced sharding management layer includes a multi-dimensional resource monitoring module, a differential privacy protection layer, a predictive load analyzer, an organizational affinity manager, and an asset traceability collaboration engine. The optimal sharding selection layer implements intelligent sharding decisions based on a multi-objective optimization strategy. The digital asset traceability layer integrates zero-knowledge proofs, sharding information exchange, and cross-sharding traceability protocols. The smart contracts are responsible for data access permission verification, usage condition agreements, and payment allocation management, and automatically execute related operations and record them on the blockchain. Related information includes asset metadata, access permission rules, privacy protection parameters, sharding allocation strategies, and traceability record update rules.

[0025] The beneficial effects of this invention are:

[0026] 1. This invention establishes a three-objective optimization framework of data locality, load balancing, and privacy protection, achieving synergistic optimization of privacy protection and system performance. Compared with traditional privacy protection schemes, this invention deeply integrates privacy protection technology with sharding optimization algorithms, achieving a significant improvement in system performance while ensuring user privacy. At the same time, this invention effectively reduces the load standard deviation through adaptive weight adjustment and dynamic threshold calculation.

[0027] 2. This invention utilizes an intelligent sharding strategy to distribute digital assets and related operations across multiple shards for parallel processing, significantly improving system throughput and resolving the scalability bottleneck of traditional digital asset sharing systems. Simultaneously, this invention establishes a complete digital asset lifecycle traceability system, addressing the lack of a comprehensive traceability mechanism in existing digital asset management systems, making it difficult to track the creation, transfer, and usage history of assets. Furthermore, based on the immutability of blockchain, this invention establishes a complete traceability chain from asset creation to destruction, supporting fine-grained operational auditing and compliance tracking, providing technical support for the legal recognition and intellectual property protection of digital assets. Attached Figure Description

[0028] Figure 1 This is the overall flowchart of the present invention.

[0029] Figure 2 This is a diagram of the architecture of the digital asset sharing service system based on sharded blockchain in this invention.

[0030] Figure 3 This is a sequence diagram of the digital asset sharing service in this invention. Detailed Implementation

[0031] The present invention will be further described below with reference to the accompanying drawings.

[0032] Example 1

[0033] like Figure 1 , 2 As shown in Figure 3, a digital asset management method based on multi-objective optimization employs a digital asset management system comprising a multi-dimensional resource monitoring module, a differential privacy protection module, a predictive load analyzer, an organizational affinity manager, and a digital asset traceability collaboration engine. The multi-dimensional resource monitoring module is responsible for real-time collection of resource status data for each shard across four dimensions: CPU utilization, memory usage, storage capacity, and network bandwidth. The differential privacy protection layer integrates a Laplace noise mechanism and privacy budget management functionality. The predictive load analyzer uses a time-series model for load prediction. The organizational affinity manager calculates inter-organizational affinity scores based on interaction frequency.

[0034] like Figure 2 As shown, this digital asset management method includes the following steps:

[0035] Step S1: Initialization

[0036] Construct sharding registry contracts and organization registration contracts respectively. The sharding registry contract shards digital asset information, with each shard configured including a shard ID, node list, load threshold, and privacy level. The organization registration contract assigns different organizations to different shards. Establish a cross-shard communication protocol, configure message routing mechanisms and atomic commit protocols. Initialize a multi-objective weight vector λ = [λ1, λ2, λ3], where λ1 is the weight of the data locality objective; λ2 is the weight of the load balancing objective; and λ3 is the weight of the privacy protection objective. Set initial balancing parameters. β 0 is 0.5. Set the differential privacy budget. ε and δ Configure the Laplace noise generator. Deploy the asset tracking contract, including modules for asset creation, transfer, and historical tracking.

[0037] In this embodiment, digital assets can be any one of digital certificates, intellectual property documents, or medical and health records.

[0038] Digital certificates include, but are not limited to, academic certificates, professional qualification certificates, and identity authentication certificates; intellectual property documents include, but are not limited to, patent application documents, trademark registration materials, and copyrighted works; and medical and health records include, but are not limited to, electronic medical records, examination reports, and medication records.

[0039] Step S2: Affinity Grouping

[0040] Based on the frequency of interactions between different organizations, obtain organizational information. With other organizations Affinity score Its expression is:

[0041]

[0042] in, Indicates organization o i and organization o j The number of interactions between them; and This represents the total number of interactions within the organization.

[0043] Based on affinity score A ij Constructing the affinity topology graph G A = (V, E), if the affinity scores of the two organizations are... A ij Greater than or equal to the preset threshold θ A Then add a border between the two organizations. Affinity topology graphs are obtained using community detection algorithms.G A The community structure within the community is used to assign organizations from the same community to the same affinity group G.

[0044] Step S3: Obtain the weighted load of the shards

[0045] Build each shard separately s i Corresponding initial resource state group The resource status group includes CPU utilization, memory utilization, storage capacity, and network bandwidth. The initial resource status group is then analyzed. The resource states in the data are normalized and represented as follows:

[0046]

[0047] in, This indicates the partitioning after normalization. s i The corresponding resource status group Resource status; This indicates the fragmentation before normalization. s i The corresponding resource status group Resource status; and These are the resource states in the group before normalization. The maximum and minimum values ​​of each resource state.

[0048] Fragments are obtained based on the normalized resource status. s i Weighted load L i Its expression is:

[0049]

[0050] in, For fragmentation s i The corresponding resource status group; For fragmentation s i The corresponding resource weight vector; and They are shards s i CPU utilization and corresponding resource weights; and They are shards s i Memory usage and corresponding resource weights; and They are shardss i Storage capacity and corresponding resource weights. and They are shards s i The network bandwidth and corresponding resource weights; .

[0051] The bottleneck resources in the resource status group are obtained by using an adaptive resource weight adjustment algorithm. It is represented as:

[0052]

[0053] in, This represents the global weight coefficient corresponding to each usage rate; This represents the value of the input variable that maximizes the objective function.

[0054] Increase the weight of the bottleneck resource and decrease the weight of other resources proportionally, so that the sum of the weights of all resources after adjustment is 1. Represented as:

[0055]

[0056] in, This is the weight adjustment amount.

[0057] Based on the adjusted resource weights Update weighted load L i .

[0058] Step S4: Construct a differential privacy protection mechanism

[0059] Differential privacy protection mechanisms include load privacy protection and affinity privacy protection; to counter load analysis attacks, differential privacy protection is applied to load information, as shown below:

[0060]

[0061] in, Weighted load with privacy protection; Lap (·) represents the Laplace distribution; Δ L Indicates the maximum range of weighted load variation; To cover the privacy budget for privacy protection.

[0062] To counter organizational affinity attacks, differential privacy protection is applied to affinity scores, as shown below:

[0063]

[0064] in, Affinity score after privacy protection; ΔA Indicates the maximum range of variation in affinity score; Privacy budget for affinity privacy protection.

[0065] Using dynamic privacy budget management, the budget consumption model is constructed as follows:

[0066]

[0067] in, This indicates the total privacy budget for the next iteration; This represents the total privacy budget for the current iteration; Indicates the privacy budget consumption coefficient; This indicates the total privacy budget consumed in the current iteration; This represents the privacy budget recovery factor.

[0068] In some embodiments, advanced combinatorial analysis is used to calculate the total privacy loss. Its expression is:

[0069]

[0070] in, This represents the current iteration number; Confidence level; Based on the parameters of privacy loss.

[0071] Obtain the privacy parameters for the current iteration based on the current privacy loss. Its expression is:

[0072]

[0073] in, Basic privacy parameters; For adjustment coefficients; For the front t The cumulative privacy loss across iterations; This is the upper limit of the total budget.

[0074] Step S5: Optimal partitioning decision

[0075] Step S5-1: Constructing an affinity strategy

[0076] If the affinity group G of the organization is not empty, then obtain the main slice s_aff of that affinity group and calculate the dynamic threshold, which is expressed as:

[0077]

[0078] in, This is the minimum value of all current loads; The benchmark adjustment coefficient; This indicates the benefits of data locality; This represents the load difference between the current shard and the primary shard. Indicates the level of privacy protection.

[0079] If the final shard load (weighted load after privacy protection) Laff of the primary shard s_aff is less than or equal to the dynamic threshold threshold, then the candidate shard saff is selected as the optimal shard.

[0080] Step S5-2: Construct an enhanced load balancing strategy

[0081] If the organization's affinity group G is empty or the weighted load value Laff of the primary s_aff is greater than the dynamic threshold, an enhanced load balancing strategy is executed. Multidimensional resource weight adjustments are performed on all available shards. An ARIMA (Autoregressive Differential Moving Average) model is used for sharding. s i Weighted load time series Process to obtain a stable load sequence ARIMA p , d , q The model is represented as:

[0082]

[0083] in, B For the shift operator; d It is the difference order; For weighted load time series; This is the white noise error term; Indicates the autoregressive component. ; These are the autoregressive coefficients; The order of the autoregressive component; Indicates the difference part. ; This represents the moving average portion. ; The moving average coefficient; q This represents the order of the moving average component.

[0084] The final fragment load is calculated using a dynamic weighting mechanism. Its expression is:

[0085]

[0086] in, α These are the weighting coefficients; For fragmentation s i Stable load.

[0087] Weighting coefficient α The expression for dynamically adjusting based on prediction accuracy is:

[0088]

[0089] in, These are the initial weighting coefficients; This is the root mean square error; and This is a hyperparameter used to control the adjustment range.

[0090] Based on the final fragment load Obtain the load balancing score, and based on the load balancing score, obtain the multi-objective weighted score (composite_score) corresponding to the primary shard. The expression for this score is:

[0091] composite_score =λ1×locality_score + λ2×load_score + λ3×privacy_score

[0092] Among them, locality_score represents the local score of the data; load_score represents the load balancing score; and privacy_score represents the privacy protection score.

[0093] Select the fragment with the lowest score as the optimal fragment.

[0094] Step S6: Parameter Update

[0095] An optimization model containing three objective functions is constructed, as shown in Equation (16).

[0096]

[0097] in, This function represents the function that minimizes the proportion of cross-shard transactions. This represents the function that minimizes load imbalance. This represents a function that minimizes the risk of privacy breaches.

[0098] Evaluate the current optimization model in each iteration. F (t) = [ f 1 (t) , f 2 (t) , f 3 (t)First, check if the weights are Pareto-dominated. Then, use gradient descent for adaptive weight updates to achieve the Pareto optimal solution. The weight update process is as follows:

[0099] ▽λ = Compute Weight Gradient(F (t) , λ (t) )

[0100] λ (t+1) = Project To Simplex(λ (t) -α▽λ)

[0101] Where ▽λ is the gradient vector of the current multi-objective weights; α is the learning rate parameter, used to control the step size of weight updates; λ (t+1) This is the multi-objective weight vector for the next iteration; Compute Weight Gradients indicates that the weight gradients are calculated based on the current objective function value; Project To Simplex indicates that the updated weights are projected onto the simplex constraint space; Convert Weights To Parameter indicates that the multi-objective weights are converted into affinity-load balancing parameters.

[0102] Step S7: Dynamic sharding adjustment and load rebalancing.

[0103] A dynamic monitoring mechanism is established. When monitoring indicators exceed preset thresholds (load imbalance exceeding 20%, cross-shard transaction ratio exceeding 30%, privacy budget consumption exceeding 80%), the process of obtaining the optimal shard is re-executed. An incremental sharding adjustment strategy is adopted to minimize data migration costs and ensure service continuity.

[0104] Step S8: Digital asset sharing services and full lifecycle traceability.

[0105] Step S8-1: Attribute-based access control and privacy protection transactions.

[0106] Users submit asset access requests, including identity credentials, asset ID, and permission level. The system verifies access permissions using a zero-knowledge proof mechanism without disclosing specific permission information. Access requests are routed to the corresponding shard node based on the sharding decision. All access operations are recorded on the blockchain, applying differential privacy to protect user behavior patterns while ensuring audit traceability. A commitment mechanism, Crel(o), is employed. i , o j ) = H(o i / / o j / / r) Protect the privacy of inter-organizational relationships; among which, HFor cryptographic hash functions; r The salt value is randomized; the zero-knowledge proof mechanism uses the zk-SNARKs protocol to verify complex access control policies without disclosing specific permission information.

[0107] Step S8-2: Full lifecycle traceability system based on Merkle tree (hash tree).

[0108] A unique traceability identifier is generated for each digital asset, establishing a complete record chain from creation to destruction. Key operational events are recorded, including asset creation, modification, access, sharing, transfer, and deletion. Each event includes a timestamp, operator identity hash, and operation content hash. A Merkle tree structure is used to organize traceability data, ensuring data integrity and tamper-proof capabilities. The traceability data adopts a hierarchical storage architecture: frequently accessed data is stored on the main blockchain, while historical data is distributed through IPFS (InterPlanetary File System), with integrity verification maintained through hash pointers.

[0109] In summary, the goal of this invention is to ensure secure and efficient asset management and trusted traceability in complex, multi-organizational collaborative digital asset sharing environments. It not only maximizes the protection of inter-organizational relationship privacy and load information security, reducing the risk of privacy breaches, but also intelligently optimizes system resource allocation and load balancing, significantly improving cross-shard operation efficiency and maximizing the overall performance and availability of digital asset sharing services. Through the deep integration of organization affinity awareness, multi-dimensional resource monitoring, differential privacy protection, and predictive load management, this invention achieves synergistic optimization of privacy protection and system performance, providing a complete technical solution for secure large-scale digital asset sharing.

[0110] Example 2

[0111] This embodiment provides an electronic device, specifically, the electronic device includes a memory and a processor. The memory stores executable code, and when the processor executes the executable code, it implements the digital asset management method in Embodiment 1.

[0112] The memory may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk drive. Communication between this system network element and at least one other network element is achieved through at least one communication interface (which can be wired or wireless), such as the Internet, wide area network, local area network, or metropolitan area network.

[0113] The bus can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc.

[0114] The memory is used to store programs. After receiving an execution instruction, the processor executes the program. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor or implemented by the processor.

[0115] The processor may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above methods can be completed by integrated logic circuits in the processor's hardware or by software instructions. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0116] Example 3

[0117] The computer program product of the readable storage medium provided in this embodiment includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the digital asset management method in Embodiment 1. For specific implementation, please refer to the foregoing method embodiments, which will not be repeated here.

[0118] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0119] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A digital asset management method based on multi-objective optimization, characterized in that: The method includes: Initialize a differential privacy budget and a multi-objective weight vector, and assign different organizations to different shards; perform affinity grouping based on the frequency of interactions between organizations to obtain multiple affinity groups; obtain the weighted load of different shards through the resource state groups and corresponding resource weight vectors of each shard; perform differential privacy protection on the weighted load based on the differential privacy budget to obtain the privacy-protected weighted load; construct affinity policies and enhanced load balancing policies respectively; if the affinity group is not empty, obtain the optimal shard through the affinity policy; the affinity policy obtains the corresponding primary shard based on the affinity group, and introduces a multi-objective weight vector to obtain a dynamic threshold; if the final shard load of the primary shard is less than or equal to the dynamic threshold, the primary shard is taken as the optimal shard; The multi-objective weight vector includes the weights for the data locality objective, the load balancing objective, and the privacy protection objective. The specific process of assigning different organizations to different shards is as follows: construct shard registry contracts and organization registration contracts respectively; perform data sharding on digital asset information through the shard registry contracts, with each shard configured including shard ID, node list, load threshold and privacy level; and assign different organizations to different shards through the organization registration contracts. The specific process of affinity grouping is as follows: calculate the affinity score between different organizations based on the frequency of interaction between organizations, construct an affinity topology map based on the threshold, and use a community detection algorithm to assign organizations belonging to the same community to the same affinity group; If the affinity group is empty or the final shard load of the primary shard is greater than the dynamic threshold, an enhanced load balancing strategy is adopted to obtain the optimal shard. The enhanced load balancing strategy obtains a stationary load sequence through an autoregressive summation moving average model, and obtains the final shard load based on the stationary load sequence and the privacy-preserving weighted load using a dynamic weighting mechanism. A load balancing score is obtained based on the final shard load, and a multi-objective weighted score is calculated based on the load balancing score and the multi-objective weight vector to select the optimal shard. An optimization model is constructed based on the multi-objective function corresponding to the multi-objective weight vector, and the multi-objective weight vector is updated according to the optimization model corresponding to the optimal partition; a dynamic monitoring mechanism is constructed, and when the system index exceeds the preset threshold, the above process of obtaining the optimal partition is re-executed. Construct a hierarchical traceability architecture based on hash trees, and combine zero-knowledge proof and commitment mechanisms to achieve trusted traceability throughout the entire lifecycle.

2. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: The dynamic threshold in the affinity strategy The method to obtain it is as follows: in, This is the minimum value of all current weighted loads; The benchmark adjustment coefficient; This indicates the benefits of data locality; This represents the load difference between the current shard and the primary shard. Indicates the level of privacy protection; These are the weights for locality objectives, load balancing objectives, and privacy protection objectives, respectively.

3. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: The multi-objective weighted score is the sum of the products of the data locality score, load balancing score, and privacy protection score with their respective multi-objective weights.

4. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: After obtaining the weighted load, an adaptive weight adjustment algorithm is used to identify bottleneck resources, increase the weight corresponding to the bottleneck resources, and reduce the weight of other resources proportionally, so that the sum of the weights of all resources after adjustment is 1; the weighted load is updated based on the adjusted resource weights.

5. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: The weighting coefficients in the dynamic weighting mechanism The expression for dynamically adjusting based on prediction accuracy is: in, These are the initial weighting coefficients; This is the root mean square error; and This is a hyperparameter.

6. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: The optimization model includes functions for minimizing the cross-shard transaction ratio, minimizing the load imbalance, and minimizing the privacy leakage risk; the resource status group includes CPU utilization, memory usage, storage capacity, and network bandwidth.

7. The digital asset management method based on multi-objective optimization according to claim 1, characterized in that: The privacy-protected weighted load ;in, For weighted loads; Lap(·) represents the Laplace distribution; ΔL represents the maximum range of weighted load variation; A privacy budget is allocated for load privacy protection; the final sharded load in the affinity strategy is a privacy-protected weighted load.

8. A digital asset management system based on multi-objective optimization, characterized in that: For performing the digital asset management method based on multi-objective optimization as described in claim 1; The digital asset management system includes a digital asset application layer, an enhanced sharding management layer, an optimal sharding selection layer, and a digital asset traceability layer. The digital asset application layer includes digital asset owners, users, and smart contracts, and is responsible for asset registration, access, and transaction management. The enhanced sharding management layer includes a multi-dimensional resource monitoring module, a differential privacy protection layer, a predictive load analyzer, an organization affinity manager, and an asset traceability collaboration engine. The optimal sharding selection layer achieves intelligent sharding decision-making based on a multi-objective optimization strategy; the digital asset traceability layer integrates zero-knowledge proofs, sharding information exchange, and cross-sharding traceability protocols; the smart contracts are responsible for data access permission verification, usage condition agreement, and payment allocation management, and automatically execute relevant operations and record them on the blockchain; the relevant information includes asset metadata, access permission rules, privacy protection parameters, sharding allocation strategies, and traceability record update rules.

Images