Data storage management method and platform for extended insurance service

Abstract

The invention discloses a data storage management method and platform for a delay protection service, and relates to the technical field of security data management, and the method comprises the steps: carrying out the wavelet packet transformation and modal decomposition of an equipment operation data set, and obtaining a time-frequency domain feature matrix; performing random mask encryption on the time-frequency domain feature matrix by applying a quantum chaotic encryption algorithm, outputting an encrypted equipment state vector, inputting the encrypted equipment state vector into a federal global health model, and performing distributed feature learning on a privacy learning layer by applying a differential privacy method to obtain an equipment feature latent variable; and the global aggregation layer performs cross-node parameter fusion by applying a secure multi-party computing protocol to form federal consensus parameters. Through the quantum chaotic encryption algorithm and the federal global health model, the anti-attack ability and privacy protection strength of the data in the transmission and storage process are enhanced, and meanwhile, the intelligent level and privacy security guarantee ability of the data storage management method are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of secure data management, and in particular to a data storage management method and platform for extended warranty services. Background Art

[0002] With the rapid development of the Internet of Things (IoT), devices are becoming increasingly intelligent, generating massive amounts of operational data. This data can effectively analyze device health and predict potential failures, providing accurate decision-making for extended warranty services. Existing storage management methods collect and centrally store device operational data with simple encryption, achieving traceability and basic security to a certain extent, fulfilling the fundamental requirements for data retention and access in extended warranty services.

[0003] Existing data storage and management methods face two major challenges: First, they lack efficient and accurate feature extraction and encryption mechanisms when processing data, making device operating data vulnerable to attack or leakage during transmission and storage, affecting the credibility of extended warranty services. Second, data value assessment is overly static, failing to fully consider the fluctuations in data value over time and the impact of external conditions on data value, resulting in irrational resource allocation in extended warranty services. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a data storage management method for extended warranty services to solve the problems of lack of effective encryption mechanism and overly static value assessment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a data storage and management method for extended warranty services, comprising: performing wavelet packet transform and modal decomposition on a device operation data set to obtain a time-frequency domain feature matrix; applying a quantum chaos encryption algorithm to perform random mask encryption on the time-frequency domain feature matrix to output an encrypted device state vector; The encrypted device state vector is input into the federated global health model. The privacy learning layer applies differential privacy to perform distributed feature learning to obtain device feature latent variables. The global aggregation layer applies secure multi-party computing protocols to perform cross-node parameter fusion to form federated consensus parameters. Attention weighting and normalized activation are performed on the device feature latent variables and federated consensus parameters to output a device health assessment report. Extract the risk level weight, access frequency value, and legal compliance weight from the equipment health assessment report, perform dynamic weighted fusion and time-series attenuation correction, and generate a multi-dimensional value assessment vector. Anchor the multi-dimensional value assessment vector on the blockchain to form an extended warranty smart contract. Perform semantic parsing and instruction mapping on the extended warranty smart contract to generate storage policy instructions. Based on the storage policy instructions, migrate and store the encrypted device state vector to form a storage location update record. Perform Merkle tree notarization on the storage location update record to obtain a storage management audit report.

[0007] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: the output of the encrypted device state vector specifically includes the following steps: Perform wavelet packet transform on the equipment operation data set to obtain multi-scale energy spectrum features; perform modal decomposition on the multi-scale energy spectrum features to obtain the time-frequency domain feature matrix; The quantum chaos encryption algorithm is applied to perform phase space mapping on the time-frequency domain characteristic matrix to form a chaotic perturbation basis, which is then encrypted with a random mask to output the encrypted device state vector.

[0008] As a preferred solution of the data storage management method for extended warranty service of the present invention, the specific construction process of the federal global health model is as follows: A privacy learning layer is built through a differentially private neural network, and a global aggregation layer is built through a federated averaging architecture. The privacy learning layer and the global aggregation layer are hierarchically integrated and stacked to build a federated global health model.

[0009] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: the acquisition of device characteristic latent variables specifically includes the following steps: The encrypted device state vector is input into the federated global health model. The privacy learning layer performs gradient quantization on the encrypted device state vector to form an encrypted gradient parameter. Differential privacy is used to perform Gaussian noise perturbation and distributed feature learning on the encrypted gradient parameters to obtain device feature latent variables.

[0010] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: the forming of the federal consensus parameters specifically includes the following steps: The global aggregation layer performs aggregation calculations on the device feature latent variables to form encrypted aggregate features; According to the secure multi-party computing protocol, the encrypted aggregation features are fused across nodes to form the federated consensus parameters.

[0011] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: the output device health assessment report specifically includes the following steps: Apply a multi-head attention mechanism to perform attention weighting on device feature latent variables and federated consensus parameters, and output health status features; The health status features are normalized and activated to generate a health score, which is then structured and encoded to output a device health assessment report.

[0012] As a preferred solution of the data storage management method for extended warranty service of the present invention, the forming of the extended warranty smart contract specifically includes the following steps: Perform singular value decomposition on the device health assessment report to extract risk level weights, access frequency values, and legal compliance weights; Apply the entropy weight method to dynamically weight and fuse the risk level weight, access frequency value, and legal compliance weight, and perform time decay correction using the time decay factor to generate a multi-dimensional value assessment vector. The multi-dimensional value assessment vector is hashed and anchored to the blockchain to form an extended warranty smart contract.

[0013] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: the storage location update record specifically includes the following steps: Perform semantic analysis on the warranty extension smart contract to form storage classification parameters, and then perform instruction mapping on the storage classification parameters to generate storage policy instructions; According to the storage policy instructions, the encrypted device state vector is migrated and encrypted stored across chains to form a distributed storage credential, and the distributed storage credential is hashed to form a storage location update record.

[0014] As a preferred solution of the data storage management method for extended warranty service of the present invention, wherein: obtaining the storage management audit report specifically includes the following steps: Perform hash block and random sampling on the storage location update records to obtain a hash node set; Perform Merkle tree notarization on the hash node set and obtain storage management audit report.

[0015] In a second aspect, the present invention provides a data storage and management system for extended warranty services, including a quantum encryption module for performing wavelet packet transform and modal decomposition on a device operation data set to obtain a time-frequency domain feature matrix; applying a quantum chaos encryption algorithm to perform random mask encryption on the time-frequency domain feature matrix to output an encrypted device state vector; The health assessment module is used to input the encrypted device state vector into the federated global health model. The privacy learning layer applies differential privacy to perform distributed feature learning to obtain device feature latent variables. The global aggregation layer applies secure multi-party computing protocols to perform cross-node parameter fusion to form federated consensus parameters. The device feature latent variables and federated consensus parameters are attention-weighted and normalized, and the device health assessment report is output. The contract generation module is used to extract the risk level weight, access frequency prediction value, and legal compliance weight from the equipment health assessment report, and perform dynamic weighted fusion and time-series attenuation correction to generate a multi-dimensional value assessment vector. The multi-dimensional value assessment vector is then anchored on the blockchain to form an extended warranty smart contract. The storage execution module is used to perform semantic parsing and multi-objective optimization on the extended warranty smart contract and generate storage policy instructions. According to the storage policy instructions, the encrypted device state vector is migrated and stored to form a storage location update record. The storage location update record is notarized using Merkle tree to obtain a storage management audit report.

[0016] The present invention achieves the following beneficial effects: Using wavelet packet transform and quantum chaos encryption algorithms for time-frequency domain feature extraction and random mask encryption significantly enhances data anti-attack capabilities and privacy protection during transmission and storage, thereby ensuring the data credibility of extended warranty services. Furthermore, distributed feature learning and cross-node parameter fusion, implemented through a federated global health model, enable accurate assessment of device health status and coordinated optimization of consensus parameters, enhancing the rationality of resource allocation for extended warranty services and further improving the intelligence and privacy protection capabilities of data storage and management methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flowchart of a data storage management method for extended warranty services.

[0019] Figure 2 Schematic diagram of the data storage management system used for extended warranty services.

[0020] Figure 3 Flowchart for quantum chaos encryption.

[0021] Figure 4 Flowchart generated for federated consensus parameters. DETAILED DESCRIPTION

[0022] In order to make the above objectives, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0023] In the following description, a large number of specific details are set forth in order to facilitate a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the concept of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0024] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an embodiment that is independent of or mutually exclusive with other embodiments.

[0025] Reference Figures 1 to 4 For one embodiment of the present application, the embodiment provides a data storage management method for warranty service, comprising the following steps: S1, wavelet packet transform and modal decomposition are performed on the equipment operation data set to obtain a time-frequency domain feature matrix; a quantum chaotic encryption algorithm is applied to perform random mask encryption on the time-frequency domain feature matrix to output an encrypted equipment state vector.

[0026] Specifically, the following steps are included, S1.1, collect the equipment operation data set and perform preprocessing.

[0027] In specific operation, The equipment operation data set includes vibration signals, temperature parameters and operating conditions; the vibration signals are collected by an acceleration sensor, the temperature parameters are collected by a temperature and humidity sensor, and the operating conditions are collected by an operating monitoring software (such as SCADA).

[0028] Next, the equipment operation data set is preprocessed, further, for the vibration signals, baseline correction is performed through zero-mean processing to eliminate baseline drift of the vibration signals, and interference suppression is performed by applying wavelet denoising to improve the purity of the vibration signals, and resolution resampling is performed simultaneously by using linear interpolation to ensure time consistency; For the temperature parameters, missing data is completed by using adjacent point interpolation to restore the continuity of the temperature parameters, and wave fluctuation is smoothed by using sliding average to weaken random disturbance, and dimension unification is performed simultaneously by using range normalization to eliminate the influence of dimension; For the operating conditions, time alignment is performed by using NTP timestamp protocol to ensure time consistency, and format unification is performed simultaneously by using Z-Score standardization to output the preprocessed equipment operation data set.

[0029] S1.2, perform wavelet packet transform on the preprocessed equipment operation data set to obtain a multi-scale energy spectrum feature; perform modal decomposition on the multi-scale energy spectrum feature to obtain a time-frequency domain feature matrix.

[0030] In specific operation, The db4 wavelet basis function is used to perform wavelet packet transform on the preprocessed equipment operation data set. Further, the preprocessed equipment operation data set is decomposed into multiple layers (such as 5 layers), and a multi-layer frequency domain spectrum is obtained. The multi-layer frequency domain spectrum is subjected to frequency band separation and numerical normalization to generate sub-band coefficients. The energy of each sub-band coefficient is integrated to form a frequency spectrum energy distribution. The frequency spectrum energy distribution is subjected to frequency band fusion and feature extraction to obtain a multi-scale energy spectrum feature.

[0031] Subsequently, the multi-scale energy spectrum feature is subjected to modal decomposition using Hilbert transform. Further, the multi-scale energy spectrum feature is subjected to frequency separation, and the first L (such as L = 5) instantaneous frequency components are extracted as dominant frequency components. The dominant frequency components are subjected to bandpass filtering to eliminate irrelevant frequency interference, and a time-frequency modal component is generated. The time-frequency modal component is subjected to parameterized sampling to obtain time-frequency characteristic parameters, and the time-frequency characteristic parameters are subjected to matrix reconstruction to generate a time-frequency domain feature matrix. The row elements in the time-frequency domain feature matrix correspond to time parameters, and the column elements represent time-frequency characteristic parameters of different modes.

[0032] S1.3, apply a quantum chaotic encryption algorithm to the time-frequency domain feature matrix to perform phase space mapping to form a chaotic disturbance base, and encrypt the chaotic disturbance base with a random mask to output an encrypted equipment state vector.

[0033] In specific operation, The quantum chaotic encryption algorithm is applied to the time-frequency domain feature matrix to perform phase space mapping. Further, the time-frequency domain feature matrix is input into a trusted execution environment (such as Intel SGX) to ensure zero exposure during the encryption process. A quantum random number generator (QRNG) is used for chaotic initialization. For example, the energy entropy value and the center frequency of the frequency domain feature matrix are extracted. The energy entropy value is set as a chaotic sensitive factor, and the center frequency is set as a phase space boundary parameter to construct a Lorenz chaotic factor. Then, the Lorenz chaotic factor is subjected to phase space mapping. Further, the Lorenz chaotic factor is subjected to trajectory fitting to generate a chaotic state trajectory. The chaotic state trajectory is subjected to discretization processing to obtain a chaotic parameter sequence. The chaotic parameter sequence is subjected to amplitude modulation and phase perturbation to output a chaotic disturbance base. It should be noted that trajectory fitting refers to the process of state iteration and curve fitting of the Lorenz chaotic factor by numerical integration fitting method. Phase perturbation refers to the process of phase shift and nonlinear superposition of the chaotic parameter sequence.

[0034] Next, the chaotic perturbation basis is encrypted with a random mask. Furthermore, the chaotic perturbation basis is masked and superimposed using a random masking factor to obtain a masked chaotic matrix. The masked chaotic matrix is ​​hashed and dimensionally compressed to generate encrypted masked data. Missing values ​​in the encrypted masked data are filled using bilinear interpolation to obtain a standard ciphertext vector. The standard ciphertext vector is associated with a device identifier to generate an encrypted device state vector. It should be noted that the random mask factor is defined based on the dynamic sensitivity parameter of the chaotic perturbation basis, and the exemplary value range is [0.02, 0.05]; hash encryption refers to the process of one-way hash encryption of the masked chaotic matrix; device identification association refers to the process of matching and storing the standard ciphertext vector and the device unique code (collected through FIDO2 authentication software).

[0035] S2. Input the encrypted device state vector into the federated global health model. The privacy learning layer applies differential privacy to perform distributed feature learning to obtain device feature latent variables. The global aggregation layer applies secure multi-party computing protocols to perform cross-node parameter fusion to form federated consensus parameters. Attention weighting and normalized activation are performed on the device feature latent variables and federated consensus parameters to output a device health assessment report.

[0036] The specific steps include: S2.1. Build and train a federated global health model.

[0037] In specific operations, In the TensorFlow framework, the differential privacy neural network is called and initialized using the noise_multiplier parameter. For example, the privacy budget is set to 0.5, the privacy relaxation factor is set to 1e-5, and the regularization coefficient is set to 0.3. A normalized activation layer is added after the differential privacy neural network to normalize features to eliminate dimensionality differences, and gradient normalization is performed using the GELU activation function to complete the construction of the privacy learning layer. The federated averaging architecture is called and initialized using the FedAvg parameter. For example, the learning rate is set to 0.01, the maximum communication rounds to 100, and the client sampling rate to 0.8. A secure multi-party computing protocol is embedded in the federated averaging architecture to protect parameter aggregation privacy, and a multi-head attention mechanism is used for dynamic weight allocation to complete the construction of the global aggregation layer. Residual connections are used to fuse cross-layer features between the privacy learning layer and the global aggregation layer to obtain an integrated feature representation. KL divergence is used to align the integrated feature representation to obtain regularized features. Softmax activation is performed on the regularized features to generate attention allocation weights. Based on the attention allocation weights, the privacy learning layer and the global aggregation layer are hierarchically integrated and stacked to complete the construction of the federated global health model. Next, the federated global health model is trained. The historical encrypted device state vectors are further divided into a sample set, a training set, and a validation set. On the sample set, a data enhancer is used to expand and enhance features to form enhanced samples. On the training set, the AdamW optimizer is used to fine-tune the parameters of the enhanced samples, while gradient clipping is simultaneously applied to constrain gradients to obtain updated federated global health model parameters. On the validation set, the cross-entropy loss function is applied to quantify the loss of the updated federated global health model parameters to obtain the validation loss. When the validation loss exceeds the convergence threshold for several consecutive rounds (e.g., five times), training terminates and the trained federated global health model is output. It should be noted that the convergence threshold is defined based on the volatility of historical validation loss, and an exemplary value range is [0.001, 0.005].

[0038] S2.2. The privacy learning layer applies differential privacy method to perform distributed feature learning to obtain device feature latent variables.

[0039] In specific operations, The encrypted device state vector is input into the federated global health model through a secure data channel (such as a TLS 1.3 encrypted channel). The privacy learning layer applies a multi-layer differential privacy neural network to quantize the gradient of the encrypted device state vector. Furthermore, the first layer performs feature dimensionality reduction mapping on the encrypted device state vector to obtain encrypted low-dimensional features, and performs L2 gradient clipping on the encrypted low-dimensional features to generate normalized gradients. The second layer performs sliding window cumulative integration on the normalized gradient to obtain aggregated gradient parameters. The aggregated gradient parameters are then subjected to exponential weighted smoothing to output stable gradient parameters. The third layer performs homomorphic encryption and gradient sparsification on the stable gradient parameters to form encrypted gradient parameters. It should be noted that L2 gradient clipping refers to the process of gradient scaling and norm constraint on the state vector of the encryption device; homomorphic encryption refers to the process of encrypting and protecting stable gradient parameters and securely aggregating them using the Paillier encryption algorithm.

[0040] Use differential privacy to perturb the encrypted gradient parameters with Gaussian noise. Furthermore, use a dynamic noise multiplier to inject Gaussian noise and scale the noise to obtain the noise-perturbed gradient. Perform homomorphic decryption and reconstruction on the noise-perturbed gradient to output a global gradient update representation. It should be noted that the dynamic noise multiplier is defined based on the privacy budget consumption rate of the historical encrypted gradient parameter, with an exemplary value range of [0.5, 1.2]; homomorphic decryption reconstruction refers to the process of private key decryption and gradient normalization of the noise perturbation gradient.

[0041] Then, distributed feature learning is performed on the global gradient update representation. Furthermore, feature projection transformation is performed on the global gradient update representation to obtain a low-dimensional gradient embedding representation. The low-dimensional gradient embedding representation is distributed aligned and sparsely encoded to obtain a regularized feature representation. The regularized feature representation is nonlinearly transformed using the Swish activation function to output the device feature latent variable. It should be noted that sparse coding refers to the process of embedding low-dimensional gradients for feature selection and coefficient sparsification.

[0042] S2.3. The global aggregation layer applies a secure multi-party computing protocol to perform cross-node parameter fusion to form federated consensus parameters.

[0043] In specific operations, The global aggregation layer performs aggregation calculations on the device feature latent variables to form encrypted aggregate features. Furthermore, the encrypted feature vectors in the device feature latent variables are extracted and normalized to obtain standard encryption parameters. Local weighted aggregation is performed on the standard encryption parameters to obtain local aggregated ciphertexts. The local aggregated ciphertexts are sharded with distributed keys to obtain encrypted sharded data. The Lagrangian basis function is used to perform aggregation calculations on the encrypted sharded data to obtain encrypted aggregated vectors. The specific mathematical format is as follows: ; in, represents the encrypted aggregate vector, Indicates the encrypted shard data index, Indicates the total number of encrypted shard data. Indicates the The weight of the encrypted shard data, Indicates the The parameter update amount of each encrypted shard data; It should be noted that the weight of the encrypted shard data is defined based on the stability contribution rate of the historical encrypted aggregation vector, and the exemplary value range is [0.1, 0.9]. The parameter update amount is obtained by performing a feature space transformation operation on the encrypted shard data.

[0044] The encrypted aggregate vector is Gaussian filtered and dimensionally compressed to form an encrypted aggregate feature. Based on the secure multi-party computing protocol, the encrypted aggregate feature is subjected to cross-node parameter fusion. Furthermore, the encrypted aggregate feature is threshold decrypted to obtain collaborative decryption nodes. Dynamic weighted averaging of the collaborative decryption nodes is performed to obtain node fusion decryption parameters. Cross-validation consensus is performed on the node fusion decryption parameters to form federated consensus parameters. It should be noted that threshold decryption refers to the process of sharding encrypted aggregate features into t nodes using Shamir secret sharing and performing collaborative decryption on the nodes; cross-validation consensus refers to the gradient distribution verification of node fusion decryption parameters through zero-knowledge proof to improve the global convergence of node fusion decryption parameters.

[0045] S2.4. Perform attention weighting and normalized activation on the device feature latent variables and federated consensus parameters, and output the device health assessment report.

[0046] In specific operations, The multi-head attention mechanism is applied to the device feature latent variables and the federated consensus parameters for attention weighting and normalized activation to generate a health score. Furthermore, the device feature latent variables are linearly projected to generate a query vector, and the federated consensus parameters are simultaneously subjected to bilinear mapping and feature decoupling to obtain key vectors and value vectors. The query vector, key vector, and value vector are weighted by the scaled dot product formula to obtain weighted health features, and residual connections are performed on the weighted health features to output health status features. The health status features are normalized to eliminate dimensional differences, and the GELU activation function is applied for nonlinear transformation to generate a preliminary health score. The specific mathematical formula is as follows: ; in, For the initial health score, is the query vector; is the key vector, represents the transpose operation, represents a value vector, represents the scaling factor; It should be noted that the scaling factor is defined based on the key vector dimension, and the exemplary value range is [64, 256].

[0047] The health score is compressed to the [0, 1] interval using the Sigmoid function to form a standard health score. A sliding window statistical method is then used to perform trend analysis on the standard health score. Furthermore, a sliding difference is performed on the standard health score to obtain its mean and standard deviation. The mean and standard deviation are then subjected to an exponential moving average to generate a trend feature vector. Based on the trend feature vector, abnormal pattern recognition is performed on the standard health score. For example, when the trend feature vector exceeds the abnormal threshold, the corresponding standard health score is defined as an abnormal health state. This is then integrated to obtain a set of abnormal fluctuation patterns. It should be noted that the abnormal threshold is defined based on the dynamic fluctuation range of the historical trend feature vector, and an exemplary value range is [0.15, 0.85].

[0048] The Protobuf serialization protocol is used to structure the abnormal fluctuation pattern set and the standard health score. Furthermore, the abnormal fluctuation pattern and the standard health score are feature-concatenated and binary-encoded to obtain a compressed health data stream. The compressed health data stream is then Base64-encoded to obtain a transmittable health data packet. The transmittable health data packet is then structured and encapsulated in JSON format to generate a device health assessment report. It should be noted that the Protobuf serialization protocol is called through the Java language code library; Base64 transcoding refers to the process of mapping compressed health data streams to ASCII characters.

[0049] S3. Extract the risk level weight, access frequency value, and legal compliance weight from the equipment health assessment report, and perform dynamic weighted fusion and time-series attenuation correction to generate a multi-dimensional value assessment vector; anchor the multi-dimensional value assessment vector on the blockchain to form an extended warranty smart contract.

[0050] The specific steps include: S3.1. Perform singular value decomposition on the equipment health assessment report to extract risk level weights, access frequency values, and legal compliance weights.

[0051] In specific operations, The device health assessment report is dimensionally compressed and missing values ​​are filled to output a standardized health assessment matrix. The standardized health assessment matrix is ​​then decomposed into left singular values ​​and right singular values ​​by eigenvalue decomposition. The first k left singular values ​​are extracted and integrated to obtain a set of left singular vectors. The first left singular vector in the left singular vector set is weightedly fused with the standard health score to obtain a risk feature vector. The risk feature vector is then Softmax normalized to obtain the risk level weight. The second left singular vector is linearly combined with the access timestamp to obtain the access timing feature. The access timing feature is then Min-Max scaled to generate the access frequency value. The third left singular vector is inner-product projected with the compliance label to obtain the compliance association feature. The compliance association feature is then Sigmoid activated to output the legal compliance weight. It should be noted that access timestamps are collected from device operation logs; compliance tags are extracted from device historical compliance records (collected via Splunk Enterprise Edition software).

[0052] S3.2. Apply the entropy weight method to dynamically weight and fuse the risk level weight, access frequency value and legal compliance weight, and perform time series attenuation correction through the time attenuation factor to generate a multi-dimensional value assessment vector.

[0053] In specific operations, Perform dynamic weighted fusion on the risk level weight, access frequency value and legal compliance weight to obtain a multi-indicator weighted matrix; perform entropy weight calculation on the multi-indicator weighted matrix to obtain the information entropy value of each indicator. The specific mathematical formula is as follows: ; in, Indicates the indicator index, No. The information entropy value of an indicator, Indicates the total number of devices, Indicates the device index, Indicates the The device is in Entropy weight on each indicator; The information entropy value is corrected for time series decay through the time decay factor. For example, the information entropy value within the last 7 days is assigned a high decay weight (such as 0.7), and the information entropy value over the past 30 days is assigned a low decay weight (such as 0.3). According to the assigned decay weights, the information entropy value is weightedly aggregated to obtain the time series weighted entropy value. It should be noted that the time decay factor is defined based on the dynamic fluctuation range of the historical information entropy value, and an exemplary value range is [0.1, 0.9].

[0054] Then, the multi-index weighted matrix, time series weighted entropy value and time decay factor are normalized and linearly combined to generate a preliminary evaluation vector; the feature dimension integration is performed on the preliminary evaluation vector to form a multidimensional value evaluation vector.

[0055] S3.3. Hash the multi-dimensional value assessment vector and anchor it to the blockchain to form a warranty extension smart contract.

[0056] In specific operations, The multidimensional value assessment vector is hashed and encrypted using the SHA-256 algorithm. Furthermore, the multidimensional value assessment vector is serialized and encoded, for example, the numerical field of the multidimensional value assessment vector is encoded as 0, and the classification label field of the multidimensional value assessment vector is encoded as 1, to obtain a binary data stream; a hash digest is performed on the binary data stream to generate a hash value of a fixed length (e.g., 256 bits); the fixed-length hash value is converted into hexadecimal to obtain a hash encrypted string. The hash encrypted string serves as a unique fingerprint of the multidimensional value assessment vector and can uniquely identify data integrity; It should be noted that hash digest refers to the process of performing a one-way hash transformation on a binary data stream.

[0057] The hashed string is then anchored to the blockchain through a smart contract (such as an Ethereum contract written in Solidity). Furthermore, the smart contract records the key metadata of the hashed string (such as the evaluation vector version and generation time), and simultaneously triggers the PBFT consensus mechanism to write the key metadata to the blockchain's distributed ledger, outputting the blockchain transaction hash record. It should be noted that the PBFT consensus mechanism is a practical Byzantine fault-tolerant algorithm that deterministically stores key metadata through a three-phase protocol (pre-prepare, prepare, and commit).

[0058] To enhance traceability, blockchain transaction hash records and multi-dimensional value assessment vectors are synchronously stored in an offline database (such as MongoDB) to establish an on-chain and off-chain bidirectional index. Based on the on-chain and off-chain bidirectional index, blockchain transaction hash records are digitally signed and bound to prevent end-to-end malicious tampering, generating secure blockchain transaction records. The secure blockchain transaction records are then encapsulated with contract logic to generate an extended warranty smart contract. It should be noted that contract logic encapsulation refers to the process of mapping business rules to secure blockchain transaction records.

[0059] S4. Perform semantic parsing and instruction mapping on the extended warranty smart contract to generate storage policy instructions. Based on the storage policy instructions, migrate and store the encrypted device state vector to form a storage location update record. Perform Merkle tree notarization on the storage location update record to obtain a storage management audit report.

[0060] The specific steps include: S4.1. Perform semantic analysis on the warranty extension smart contract to form storage classification parameters, perform instruction mapping on the storage classification parameters, and generate storage policy instructions.

[0061] In specific operations, Perform semantic parsing on the extended warranty smart contract. Furthermore, extract and count the state variable data of the extended warranty smart contract and calculate the weekly average access frequency of the state variable data. Identify hot and cold data based on the weekly average access frequency. For example, identify state variable data with a low weekly average access frequency (e.g., <10 times) as cold data, and state variable data with a high weekly average access frequency (e.g., ≥100 times) as hot data. Output the hot and cold data identifiers. Synchronously perform time series sampling on the extended warranty smart contract to obtain the contract access timestamp; perform sliding window integration on the contract access timestamp to generate the contract access interval; apply the density threshold to perform density interval division on the contract access interval. For example, when the contract access interval is greater than the density threshold, it is divided into a high-frequency density interval; when the contract access interval is less than the density threshold, it is divided into a low-frequency density interval. The low-frequency density interval and the high-frequency density interval are classified and weighted to obtain the activity distribution; integrate the hot and cold data identifiers with the activity distribution to obtain the storage grading parameters.

[0062] It should be noted that the density threshold is defined based on the dynamic volatility of historical contract access intervals, and an exemplary value range is [100ms, 5s].

[0063] The storage tier parameters are then mapped to instructions. Furthermore, the storage tier parameters are predefined to operation types. For example, hot data is defined as a memory operation type, cold data as a disk operation type, low-frequency density intervals are defined as batch processing operations, and high-frequency density intervals are defined as real-time processing operations. The predefined operation types are mapped to command templates to generate an instruction rule set. The instruction rule set is formatted and encapsulated to output storage policy instructions. It should be noted that command template mapping refers to the process of selecting a template and binding parameters for a predefined operation type.

[0064] S4.2. According to the storage policy instructions, the encrypted device state vector is migrated and encrypted and stored across chains to form a distributed storage credential. The distributed storage credential is hashed and recorded to form a storage location update record.

[0065] In specific operations, Migrate the encrypted device state vector from the source chain (such as Ethereum) to the target chain (such as Polkadot) through a cross-chain protocol (such as the IBC protocol). Furthermore, lock the encrypted device state vector on the source chain, mark the integrity of the encrypted device state vector, and output the marked encrypted data packet. Then, migrate the marked encrypted data packet to the target chain through a cross-chain relay, and synchronize the inter-chain state to ensure the integrity of the data transmission. It should be noted that inter-chain state synchronization refers to the process of confirming the data integrity of the target chain for the marked encrypted data packet.

[0066] During the cross-chain migration process, the encrypted data packet is encrypted and stored through the storage policy instruction. Furthermore, the storage sharding parameters of the storage policy instruction are extracted. According to the storage sharding parameters, the encrypted data packet is sharded to obtain an encrypted data shard set. The encrypted data shard set is registered for storage location to obtain shard storage location information. The shard storage location information is stored on the blockchain to generate a distributed storage certificate. It should be noted that the storage location registration refers to the process of binding the storage node and recording the location information of the encrypted data fragment set.

[0067] Then the distributed storage credential is hashed and recorded, and further, the distributed storage credential is hashed and mapped to obtain a location credential hash value; the location credential hash value is timestamped to form a time location marker credential, and the time location marker credential is archived on the chain to obtain a storage location update record.

[0068] S4.3, Hash block and random sampling are performed on the storage location update record to obtain a hash node set, and a Merkle tree storage is performed on the hash node set to obtain a storage management audit report.

[0069] In specific operation, The storage location update record is hashed and blocked according to a fixed size (such as 1MB), and a block hash set is obtained; the block hash set is randomly sampled to generate a candidate sampling pool, and a verification sample is extracted from the candidate sampling pool according to a certain proportion (such as 10%), and a hash node set is output.

[0070] Then the hash node set is executed to perform Merkle tree storage, and further, the hash node set is sorted in descending order to obtain an ordered hash node, and the ordered hash node is used as a leaf node of the Merkle tree, and the leaf nodes of the Merkle tree are recursively stacked to build a complete Merkle tree structure; the complete Merkle tree structure is written into the blockchain through the smart contract (such as Ethereum Solidity contract) to generate a Merkle tree storage record; The storage location update record and the Merkle tree storage record are mapped to obtain an associated field set, and the associated field set is audited and signed to form an audit intermediate record; the audit intermediate record is executed to fill in the report field, and a storage management audit report is output.

[0071] The embodiment also provides a data storage management system for extended warranty service, comprising: a quantum encryption module for performing wavelet packet transform and modal decomposition on a device running data set to obtain a time-frequency domain feature matrix; a quantum chaotic encryption algorithm is used to perform random mask encryption on the time-frequency domain feature matrix to output an encrypted device state vector; A health assessment module is configured to input the encrypted device state vector into a federal global health model, and a privacy learning layer applies differential privacy law for distributed feature learning to obtain a device feature latent variable; a global aggregation layer applies a secure multi-party computation protocol for cross-node parameter fusion to form a federal consensus parameter; the device feature latent variable and the federal consensus parameter are subjected to attention weighting and normalized activation to output a device health assessment report. The contract generation module is used to extract the risk level weight, access frequency prediction value, and legal compliance weight from the equipment health assessment report, and perform dynamic weighted fusion and time-series attenuation correction to generate a multi-dimensional value assessment vector. The multi-dimensional value assessment vector is then anchored on the blockchain to form an extended warranty smart contract. The storage execution module is used to perform semantic parsing and multi-objective optimization on the extended warranty smart contract and generate storage policy instructions. According to the storage policy instructions, the encrypted device state vector is migrated and stored to form a storage location update record. The storage location update record is notarized using Merkle tree to obtain a storage management audit report.

[0072] This embodiment also provides a computer device suitable for the data storage management method for extended warranty service, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the data storage management method for extended warranty service proposed in the above embodiment.

[0073] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0074] This embodiment further provides a storage medium having a computer program stored thereon. When executed by a processor, the program implements the data storage management method for extended warranty service as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0075] In summary, this invention significantly enhances data security and privacy protection during transmission and storage by using wavelet packet transform and quantum chaos encryption algorithms for time-frequency domain feature extraction and random mask encryption, thereby ensuring the data credibility of extended warranty services. Furthermore, distributed feature learning and cross-node parameter fusion using a federated global health model enable accurate assessment of device health status and coordinated optimization of consensus parameters, enhancing the rationality of resource allocation for extended warranty services and further improving the intelligence and privacy protection capabilities of data storage and management methods.

[0076] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A data storage management method for extended warranty service, characterized by: include, Perform wavelet packet transform and modal decomposition on the equipment operation data set to obtain the time-frequency domain feature matrix; Apply the quantum chaos encryption algorithm to perform random mask encryption on the time-frequency domain feature matrix and output the encrypted device state vector; The encrypted device state vector is input into the federated global health model. The privacy learning layer applies differential privacy to perform distributed feature learning to obtain device feature latent variables. The global aggregation layer applies a secure multi-party computing protocol to fuse cross-node parameters and form federated consensus parameters. Perform attention weighting and normalized activation on device feature latent variables and federated consensus parameters, and output a device health assessment report; Extract the risk level weight, access frequency value, and legal compliance weight from the device health assessment report, perform dynamic weighted fusion and time-series attenuation correction, and generate a multi-dimensional value assessment vector. Anchor the multi-dimensional value assessment vector to the blockchain to form an extended warranty smart contract; Perform semantic parsing and instruction mapping on the extended warranty smart contract to generate storage policy instructions. Based on the storage policy instructions, migrate and store the encrypted device state vector to form a storage location update record. Perform Merkle tree notarization on the storage location update record to obtain a storage management audit report.

2. The data storage management method for extended warranty service according to claim 1, wherein: The output encryption device state vector specifically includes the following steps: Perform wavelet packet transform on the equipment operation data set to obtain multi-scale energy spectrum features; perform modal decomposition on the multi-scale energy spectrum features to obtain the time-frequency domain feature matrix; The quantum chaos encryption algorithm is applied to perform phase space mapping on the time-frequency domain characteristic matrix to form a chaotic perturbation basis, which is then encrypted with a random mask to output the encrypted device state vector.

3. The data storage management method for extended warranty service according to claim 2, wherein: The specific construction process of the federal global health model is as follows: A privacy learning layer is built through a differentially private neural network, and a global aggregation layer is built through a federated averaging architecture. The privacy learning layer and the global aggregation layer are hierarchically integrated and stacked to build a federated global health model.

4. The data storage management method for extended warranty service according to claim 1, wherein: The method of obtaining the device characteristic latent variables specifically includes the following steps: The encrypted device state vector is input into the federated global health model. The privacy learning layer performs gradient quantization on the encrypted device state vector to form an encrypted gradient parameter. Differential privacy is used to perform Gaussian noise perturbation and distributed feature learning on the encrypted gradient parameters to obtain device feature latent variables.

5. The data storage management method for extended warranty service according to claim 4, wherein: The formation of the federal consensus parameters specifically includes the following steps: The global aggregation layer performs aggregation calculations on the device feature latent variables to form encrypted aggregate features; According to the secure multi-party computing protocol, the encrypted aggregation features are fused across nodes to form the federated consensus parameters.

6. The data storage management method for extended warranty service according to claim 1, wherein: The output device health assessment report specifically includes the following steps: Apply a multi-head attention mechanism to perform attention weighting on device feature latent variables and federated consensus parameters, and output health status features; The health status features are normalized and activated to generate a health score, which is then structured and encoded to output a device health assessment report.

7. The data storage management method for extended warranty service according to claim 1, wherein: The formation of the warranty extension smart contract specifically includes the following steps: Perform singular value decomposition on the device health assessment report to extract risk level weights, access frequency values, and legal compliance weights; Apply the entropy weight method to dynamically weight and fuse the risk level weight, access frequency value, and legal compliance weight, and perform time decay correction using the time decay factor to generate a multi-dimensional value assessment vector. The multi-dimensional value assessment vector is hashed and anchored to the blockchain to form an extended warranty smart contract.

8. The data storage management method for extended warranty service according to claim 7, wherein: The storage location update record specifically includes the following steps: Perform semantic analysis on the warranty extension smart contract to form storage classification parameters, and then perform instruction mapping on the storage classification parameters to generate storage policy instructions; According to the storage policy instructions, the encrypted device state vector is migrated and encrypted stored across chains to form a distributed storage credential, and the distributed storage credential is hashed to form a storage location update record.

9. The data storage management method for extended warranty service according to claim 1, wherein: The acquisition of the storage management audit report specifically includes the following steps: Perform hash block and random sampling on the storage location update records to obtain a hash node set; Perform Merkle tree notarization on the hash node set and obtain storage management audit report.

10. A data storage management system for extended warranty service, based on the data storage management method for extended warranty service according to any one of claims 1 to 9, characterized in that: include, Quantum encryption module, used to perform wavelet packet transform and modal decomposition on the device operation data set to obtain the time-frequency domain feature matrix; Apply the quantum chaos encryption algorithm to perform random mask encryption on the time-frequency domain feature matrix and output the encrypted device state vector; The health assessment module is used to input the encrypted device state vector into the federated global health model. The privacy learning layer applies differential privacy to perform distributed feature learning to obtain device feature latent variables. The global aggregation layer applies a secure multi-party computing protocol to fuse cross-node parameters and form federated consensus parameters. Perform attention weighting and normalized activation on device feature latent variables and federated consensus parameters, and output a device health assessment report; The contract generation module is used to extract the risk level weight, access frequency prediction value, and legal compliance weight from the device health assessment report, and perform dynamic weighted fusion and time-series attenuation correction to generate a multi-dimensional value assessment vector; Anchor the multi-dimensional value assessment vector to the blockchain to form an extended warranty smart contract; The storage execution module is used to perform semantic parsing and multi-objective optimization on the extended warranty smart contract and generate storage policy instructions. According to the storage policy instructions, the encrypted device state vector is migrated and stored to form a storage location update record. The storage location update record is notarized using Merkle tree to obtain a storage management audit report.

Images