Cleaning and spraying data management system based on blockchain

Abstract

The present invention relates to the field of blockchain data management technology, and discloses a blockchain-based cleaning and spraying data management system, which includes cleaning parameter acquisition, data sharding encryption, cross-chain feature verification, dynamic decision reconstruction, and execution instruction generation modules. The system obtains the spraying equipment operating status data stream through distributed nodes, generates a data verification feature matrix set through time and space dimension division and feature sharding encryption, and then obtains an optimized verification feature matrix set through consensus mechanism screening and feature correlation verification. It generates a data chain decision feature graph based on hash time lock fusion, and finally generates a device control instruction set based on this. The system uses blockchain technology to achieve data security storage, cross-chain verification and dynamic decision-making, improves the efficiency of cleaning and spraying data management and the accuracy of decision-making, is suitable for intelligent data management and equipment control in industrial scenarios, and solves problems such as weak data security and delayed decision-making in traditional systems.

Description

Technical Field

[0001] The present invention relates to the technical field of blockchain data management, and specifically to a cleaning and spraying data management system based on blockchain. Background Art

[0002] In the industrial cleaning and spraying sectors, efficient management and reliable control of equipment operating data are key to ensuring production quality and improving operational efficiency. Traditional data management systems suffer from weak data security, insufficient correlation analysis of multi-source data, and delayed decision-making responses. These issues make it difficult to meet the demands of modern intelligent manufacturing for real-time data security and intelligent decision-making.

[0003] From a data security perspective, traditional systems typically utilize centralized storage architectures, making data vulnerable to single points of failure, cyberattacks, and human tampering. For example, during the operation of spray equipment, if key parameters such as pressure, temperature, and spray flow rate are illegally tampered with, this can lead to quality issues such as uneven coating thickness and incomplete cleaning, and even cause equipment failure or safety incidents. Furthermore, traditional encryption methods primarily target a single data dimension and lack a layered encryption mechanism for spatiotemporally correlated data, making it difficult to address complex data leakage risks.

[0004] When it comes to data processing and analysis, traditional systems often lack the ability to effectively integrate and deeply mine multi-source, heterogeneous data. During operation, cleaning and spraying equipment generates multi-dimensional data streams encompassing time series, spatial location, and equipment status. Traditional methods typically perform only simple time-series analysis on this data, ignoring the data's correlations across time and space. For example, fluctuations in equipment operating parameters over different time periods may potentially correlate with changes in the workshop's ambient temperature and humidity. Traditional systems are unable to effectively capture these correlations, limiting the accuracy and reliability of decision-making models.

[0005] From a decision-making and control perspective, traditional systems often rely on preset rules or static models for decision-making, lacking the ability to respond to dynamic changes in real time. When sudden changes occur in the equipment's operating status or production environment, static decision-making models struggle to quickly adjust control strategies, potentially leading to decreased production efficiency or wasted resources. For example, during the spray coating process, if the workpiece material undergoes subtle changes, traditional systems are unable to adjust spray parameters based on real-time data, potentially resulting in wasted paint or substandard coating quality.

[0006] Furthermore, traditional systems have significant shortcomings in cross-device and cross-system data exchange. With the development of the Industrial Internet, cleaning and spraying equipment often requires data sharing and collaboration with systems in other production processes. However, the closed architecture of traditional systems hinders data flow, making cross-chain data verification and collaborative decision-making difficult. For example, in scenarios where multiple production lines collaborate, data from each device cannot be effectively interoperated and verified, potentially leading to disconnected production processes and difficulties in quality traceability. Summary of the Invention

[0007] The purpose of the present invention is to provide a blockchain-based cleaning and spraying data management system to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a blockchain-based cleaning and spraying data management system, the system comprising:

[0009] Cleaning parameter acquisition module, used to obtain the operating status data stream of the spraying equipment through distributed nodes;

[0010] A data sharding encryption module, configured to perform spatiotemporal dimension division and feature sharding encryption on the operating status data stream to generate a set of data verification feature matrices;

[0011] A cross-chain feature verification module, used to perform consensus mechanism screening and feature correlation verification on the set of data verification feature matrices to obtain a set of optimized verification feature matrices;

[0012] A dynamic decision reconstruction module, configured to perform feature fusion based on a hash time lock on the set of the optimized verification feature matrices to generate a data chain decision feature graph;

[0013] An execution instruction generation module is used to generate a device control instruction set based on the data chain decision feature diagram.

[0014] Preferably, the data sharding encryption module includes:

[0015] A data stream segmentation unit, configured to dynamically segment the running status data stream according to a preset block size to form a data shard set;

[0016] The feature encryption unit is used to input each shard in the data shard set into a data shard encryptor based on a Merkle tree to generate a set of data verification feature matrices.

[0017] Preferably, the cross-chain feature verification module includes:

[0018] a feature matrix deconstruction unit, configured to perform chain parsing on the data verification feature matrix along a timestamp dimension to obtain a set of feature verification vectors;

[0019] a consensus association calculation unit, configured to calculate a consensus verification value between any two feature vectors in the set of feature verification vectors to generate a feature association consensus matrix;

[0020] a dynamic screening unit, configured to adjust the weights of the feature association consensus matrix according to the hash differences of adjacent blocks in the set of feature verification vectors to generate a consensus constraint association matrix;

[0021] A feature optimization unit is used to perform cross-chain convolution aggregation on the consensus constraint association matrix and the set of feature verification vectors to obtain the optimized verification feature matrix.

[0022] Preferably, the consensus association calculation unit includes:

[0023] A feature mapping subunit, configured to map each feature vector in the set of feature verification vectors to an orthogonal hash space to obtain a set of mapped feature vectors;

[0024] The consensus correlation analysis subunit is used to calculate the hash similarity between any two mapped feature vectors in the set of mapped feature vectors to generate the feature correlation consensus matrix composed of multiple consensus correlation values.

[0025] Preferably, the feature optimization unit is specifically implemented as follows:

[0026] Performing chain convolution processing on the consensus constraint association matrix to generate a consensus constraint feature tensor;

[0027] Inputting the set of feature verification vectors and the consensus constraint feature tensor into a bidirectional timestamp encoding network to generate a set of chained context feature vectors;

[0028] Matrix reorganization is performed on the set of chained context feature vectors to generate the optimized verification feature matrix.

[0029] Preferably, the dynamic decision reconstruction module includes:

[0030] A feature matrix compression unit, configured to perform maximum pooling processing of a timestamp dimension on each matrix in the set of optimized verification feature matrices to generate a set of optimized verification feature vectors;

[0031] A hash entropy value calculation unit, configured to calculate a hash entropy value of each feature vector in the set of optimized verification feature vectors to generate a dynamic hash entropy set;

[0032] A reference feature determination unit, configured to select an optimized verification feature vector corresponding to a maximum hash entropy value in the dynamic hash entropy set as an initial decision reference vector;

[0033] a dynamic weight calculation unit, configured to calculate the dynamic decision weight of each feature vector in the set of optimized verification feature vectors based on the hash distance between each feature vector and the initial decision reference vector and the hash entropy value of each feature vector to generate a dynamic weight set;

[0034] A chain reconstruction unit is used to perform weighted aggregation on the set of optimized verification feature vectors using the dynamic weight set to generate the data chain decision feature graph.

[0035] Preferably, the hash entropy value calculation unit is specifically implemented as follows:

[0036] Calculating the mean vector and variance vector of the optimized verification feature vector;

[0037] Performing element-by-element difference calculation on the optimized verification feature vector and the mean vector, and performing a square operation on the difference result to generate a feature hash difference vector;

[0038] Calculating the overall median of the feature hash difference vector; performing a ratio operation on the median and the square value of the variance vector, and inputting the result into a normalization function to obtain the hash entropy value.

[0039] Preferably, the dynamic weight calculation unit is specifically implemented as follows:

[0040] Multiplying the hash entropy value of the optimized verification feature vector and the hash entropy value of the initial decision reference vector by a first decision coefficient to obtain a first dynamic decision factor;

[0041] Multiplying the absolute value of the Euclidean distance between the optimized verification feature vector and the initial decision reference vector by a second decision coefficient to obtain a second dynamic decision factor;

[0042] The first dynamic decision factor and the second dynamic decision factor are weighted and summed to obtain the dynamic decision weight.

[0043] Preferably, the execution instruction generation module is specifically implemented as follows:

[0044] The data chain decision feature graph is input into the instruction generator based on the smart contract to generate the equipment control instruction set, which is used to control the operating parameters of the cleaning spray equipment.

[0045] Preferably, the time-space dimension division includes:

[0046] Dividing the running status data stream into continuous data segments according to timestamps, and adding a spatial grid code to each data segment to form a spatiotemporally associated slice set;

[0047] The feature shard encryption performs cross-signature verification on the shard set through the asymmetric keys of the distributed nodes.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] In terms of data security protection, the system implements spatial and temporal dimension division and feature shard encryption of the operating status data stream through the data shard encryption module. The data stream is dynamically divided into a set of data shards according to the preset block size, and each shard is encrypted using the Merkle tree structure. Combined with the asymmetric key cross-signature verification mechanism of distributed nodes, a multi-level data security protection system is formed. This encryption method not only achieves fine-grained protection of data, but also uses the distributed storage characteristics of the blockchain to disperse the data shards across different nodes, significantly reducing the risk of single-point leakage and ensuring the integrity and immutability of data during collection, transmission, and storage. For example, when a data shard of a node is attacked, other nodes can still verify the authenticity of the data through the consensus mechanism to ensure the stable operation of the system.

[0050] In the data processing and verification phase, the cross-chain feature verification module achieves in-depth mining and optimization of multi-dimensional data features through chain parsing, consensus association calculation, dynamic screening, and cross-chain convolution aggregation of the data verification feature matrix. The feature verification vector set is parsed along the timestamp dimension, the hash similarity between any two feature vectors is calculated to generate a feature association consensus matrix, and the weights are adjusted according to the hash differences between adjacent blocks, effectively capturing the correlation and dynamic changes of data in the spatiotemporal dimensions. The generation of chain context feature vectors through a bidirectional timestamp coding network further enhances the contextual semantic understanding of data features, enabling the system to extract more representative optimized verification feature matrices from complex data streams, providing more accurate data support for subsequent decision-making. For example, when analyzing fluctuations in the pressure parameters of spray equipment, the system can discover its potential correlation with parameters such as paint viscosity and ambient temperature through the feature association consensus matrix, thereby more comprehensively evaluating the equipment's operating status.

[0051] The dynamic decision reconstruction module implements a dynamic decision-making mechanism based on real-time data through operations such as maximum pooling, hash entropy calculation, dynamic weight allocation, and weighted aggregation. The eigenvector corresponding to the maximum hash entropy value is selected as the initial decision benchmark, and the dynamic weight is calculated by combining the hash distance between each eigenvector and the benchmark vector and its own hash entropy value. This enables the system to automatically adjust the decision weight according to the uncertainty and correlation of the data, and generate a data chain decision feature map that is more in line with the actual operating status. This dynamic decision-making mechanism significantly improves the system's adaptability to complex working conditions and can quickly respond to changes in the equipment's operating status or production environment. For example, when the workpiece material changes and the spraying parameters need to be adjusted, the system can re-aggregate the eigenvectors based on the dynamic weights calculated in real time, quickly generate a new decision feature map, and guide the equipment to adjust the operating parameters to avoid quality problems or resource waste caused by static decisions.

[0052] The execution instruction generation module uses smart contracts to convert decision-making feature maps into equipment control instruction sets, achieving automated integration between decision-making and control. The programmability and automatic execution of smart contracts ensure the accuracy and timeliness of instruction generation. Furthermore, the blockchain consensus mechanism ensures the immutability and traceability of instructions, ensuring that control instructions can be securely and reliably transmitted to the cleaning and spraying equipment, enabling precise control of operating parameters. For example, during the cleaning process, the system can generate instructions to adjust parameters such as cleaning fluid flow and pressure based on real-time data, ensuring the stability and consistency of the cleaning effect.

[0053] Furthermore, through cross-chain feature verification and the collaborative work of distributed nodes, the system enables cross-device and cross-system data interoperability and verification, breaking the closed architecture of traditional systems and providing strong support for multi-device collaborative operations and quality traceability in the Industrial Internet environment. Through the blockchain's chain structure and timestamp mechanism, the system can fully record the data generation, processing, and decision-making process, providing a reliable basis for transparent management of the production process and rapid tracing of quality issues. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a working principle diagram of the blockchain-based cleaning and spraying data management system described in the present invention;

[0055] Figure 2 This is the design diagram of the data sharding encryption module;

[0056] Figure 3 This is the design diagram of the cross-chain feature verification module;

[0057] Figure 4 Design diagram of the refactoring module for dynamic decision making. DETAILED DESCRIPTION

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0059] See also Figures 1-4 The present invention relates to a blockchain-based cleaning and spraying data management system, which includes:

[0060] The present invention provides a blockchain-based cleaning and spraying data management system. The system includes a cleaning parameter acquisition module, a data sharding encryption module, a cross-chain feature verification module, a dynamic decision reconstruction module, and an execution instruction generation module. These modules work together to implement blockchain management of cleaning and spraying data. The specific implementation method is as follows:

[0061] The cleaning parameter acquisition module, deployed via distributed nodes within the spray equipment network, captures real-time data streams on the spray equipment's operating status, including but not limited to equipment operating temperature, pressure, spray flow rate, operation timestamps, and equipment geographic coordinates. The distributed nodes utilize a blockchain-based P2P network architecture to ensure decentralized and real-time data collection. Each node performs data collection and preliminary verification to prevent data loss due to single points of failure.

[0062] Data Sharding and Encryption Module: This module performs spatiotemporal segmentation and feature-based sharding encryption on the collected operational status data stream. Spatiotemporal segmentation involves slicing the data stream into continuous data segments based on timestamps. Each segment is then assigned a spatial grid code (e.g., a grid index based on the device's GPS coordinates) to form a collection of spatiotemporally correlated shards. Feature-based sharding encryption uses asymmetric keys from distributed nodes to cross-signature and verify the shard collection. A Merkle tree structure is then used to generate a set of data verification feature matrices, enabling encrypted storage and rapid verification of data shards.

[0063] Cross-chain feature verification module: This module performs consensus screening and feature relevance verification on a collection of data verification feature matrices. First, the feature matrices submitted by each node are screened for validity through a consensus mechanism (such as the improved PBFT algorithm), eliminating matrices corresponding to abnormal data. Then, based on hash similarity calculations and cross-chain convolution aggregation of feature vectors, the relevance between features is verified, generating a collection of optimized verification feature matrices to ensure the consistency and reliability of data features.

[0064] Dynamic Decision Reconstruction Module: This module performs feature fusion based on hash time locks on the set of optimized and verified feature matrices. It compresses the matrix dimensions through maximum pooling of the timestamp dimension to generate a set of feature vectors. The module then determines the initial decision reference vector based on hash entropy calculations. The weights of each feature vector are dynamically calculated by combining hash distance and entropy. Finally, a weighted aggregation is performed to generate a data chain decision feature graph, enabling dynamic reconstruction of data features and decision association.

[0065] Execution instruction generation module: The data chain decision feature graph is input into the instruction generator based on the smart contract, and the equipment control instruction set is generated through the preset control logic (such as PID control algorithm parameters, equipment start and stop rules, etc.). The instruction set is directly used to control the operating parameters of the cleaning and spraying equipment, such as spray pressure adjustment, cleaning liquid flow control, operation path planning, etc., to realize data-driven equipment automation control.

[0066] The present invention will be further described below in conjunction with Examples 1 to 5:

[0067] Example 1:

[0068] This embodiment involves the specific implementation of a data sharding encryption module and a cross-chain feature verification module. The data sharding encryption module includes a data stream segmentation unit and a feature encryption unit. The data stream segmentation unit dynamically segments the operational data stream. Specifically, it divides the continuous data stream into multiple data shards according to a preset block size (this size can be adaptively adjusted based on network bandwidth, data flow, and other actual conditions, for example, initially set to 512KB). Each data shard must carry information such as a timestamp, a spatial grid code, and a shard sequence number. The timestamp identifies the time of data collection. The spatial grid code can be generated using a geographic information system (GIS) gridding algorithm based on the device's GPS coordinates (for example, dividing the device's area into grid cells of a specific resolution, such as 50m×50m, with each grid cell corresponding to a unique code). The shard sequence number indicates the order of the data shards in the original data stream. This information ensures the spatiotemporal correlation of the data shards, facilitating subsequent data processing and verification.

[0069] The feature encryption unit feeds a set of data shards into a Merkle-tree-based data shard encryptor for processing. Specifically, a hash algorithm (such as the SHA-256 hash algorithm) is used to generate a corresponding hash value for each data shard. These hash values ​​are then combined layer by layer according to the Merkle tree construction rules to ultimately construct a Merkle tree. During this process, the hash value of each data shard forms its hash path within the Merkle tree. This hash path information is integrated to form a set of data verification feature matrices. The root hash value of the Merkle tree is stored along with the data shards in each blockchain node. When data integrity verification is required, a simple comparison of the root hash value quickly determines whether the data has been tampered with. This approach enables encrypted storage and rapid verification of data shards, ensuring the security and reliability of data during storage and transmission.

[0070] The cross-chain feature verification module includes a feature matrix deconstruction unit, a consensus association calculation unit, a dynamic screening unit, and a feature optimization unit. The feature matrix deconstruction unit needs to perform chain parsing on the data verification feature matrix along the timestamp dimension. Specifically, each data verification feature matrix is ​​decomposed into a series of feature verification vectors in chronological order. These vectors are arranged in chronological order to form a set of feature verification vectors. Each vector corresponds to the device operation characteristics under a time slice. For example, within a 100ms time window, the device's operating temperature, pressure, flow rate and other characteristic data will be integrated into a feature verification vector. In this way, the matrix data is converted into a vector sequence that is easy to analyze, laying the foundation for subsequent feature correlation verification.

[0071] The consensus association calculation unit's processing flow begins by mapping each eigenvector in the set of feature verification vectors to an orthogonal hash space. Mapping to an orthogonal hash space can be achieved through specific mathematical transformations, such as projecting the eigenvectors using a set of linearly independent hash functions to convert them into frequency-domain eigenvectors or other forms of eigenvectors. This reduces the linear correlation between the eigenvectors, making the features easier to distinguish and analyze in the hash space. After the mapping is completed, the hash similarity between any two mapped eigenvectors in the set of mapped eigenvectors needs to be calculated. This hash similarity can be calculated using a variety of algorithms, such as the cosine similarity algorithm and the Hamming distance algorithm. The calculated hash similarity values ​​generate a feature association consensus matrix, whose rows and columns correspond to the indices of the feature verification vectors. The element values ​​in the matrix reflect the strength of the association between the two eigenvectors. A larger value indicates a stronger association between the features, and a smaller value indicates a weaker association. This matrix provides an intuitive understanding of the interrelationships between the features.

[0072] The dynamic screening unit adjusts the weights of the feature association consensus matrix based on the hash differences between adjacent blocks in the set of feature verification vectors. The hash differences between adjacent blocks can be calculated by computing the forward difference, that is, the rate of change of the hash values ​​of two adjacent blocks. Features with significant hash differences indicate significant time series variability and are therefore likely key features, so their weights in the feature association consensus matrix should be increased. Features with smaller hash differences have their weights reduced accordingly. This weight adjustment generates a consensus constraint association matrix that highlights the impact of key features, allowing subsequent feature optimization processes to focus more on features that are crucial for decision-making.

[0073] The implementation of the feature optimization unit is relatively complex. First, a chain convolution operation is performed on the consensus constraint correlation matrix. This chain convolution can be performed using a three-dimensional convolution approach (the convolution kernel size can be customized, such as 3×3×3). This convolution operation extracts feature correlation patterns in both spatial and temporal dimensions, generating a consensus constraint feature tensor. The set of feature verification vectors and the consensus constraint feature tensor are then input into a bidirectional timestamp encoding network, which can be a network structure capable of processing time series data, such as a Bi-LSTM network. The network's input layer receives the feature vectors and timestamp encoding information. The hidden layer processes the time series using bidirectional recurrent neurons, capturing feature dependencies between past and future time steps and thus fully mining the contextual information in the data. The output layer generates a set of chained contextual feature vectors containing contextual information. Finally, a matrix reorganization operation is performed on the set of chained contextual feature vectors, which arranges the multidimensional vectors into a two-dimensional matrix in chronological order. This method converts the chained contextual feature vector set into an optimized verification feature matrix. The rows of the matrix correspond to time series, and the columns correspond to feature dimensions. This processing achieves spatiotemporal correlation optimization and cross-chain aggregation of features, so that the optimized feature matrix can more accurately reflect the changing patterns of equipment operating status, providing a more reliable basis for subsequent dynamic decision-making reconstruction.

[0074] During the entire processing of the data sharding encryption module and the cross-chain feature verification module, each unit works closely together. The data stream segmentation unit provides a structured data sharding set for the feature encryption unit, which implements data encryption and verification through the Merkle tree structure. Each unit in the cross-chain feature verification module then parses, maps, calculates similarity, adjusts weights, and optimizes features of the data verification feature matrix in turn, gradually screening and optimizing features to ensure that the final optimized verification feature matrix set can accurately and reliably reflect the feature correlation and spatiotemporal characteristics of the equipment operating status data, laying a solid foundation for the subsequent processing of the entire blockchain-based cleaning and spraying data management system. Through such a design, the system can efficiently process and verify data in a distributed environment, ensuring the integrity, security, and reliability of the data, thereby achieving effective management and utilization of the operating data of cleaning and spraying equipment.

[0075] Example 2:

[0076] This embodiment involves a further refinement of the consensus association calculation unit and the feature optimization unit. The consensus association calculation unit includes a feature mapping subunit and a consensus correlation analysis subunit. The core function of the feature mapping subunit is to map each feature vector in the set of feature verification vectors to an orthogonal hash space to reduce the linear correlation between features, so as to facilitate subsequent similarity calculation and feature analysis. In specific implementation, a set of predefined orthogonal hash function groups can be used. These hash functions have the characteristic of being linearly independent and can project the original feature vector from the original feature space into a multi-dimensional orthogonal hash space. For example, for the original feature verification vector containing multi-dimensional features such as temperature, pressure, and flow, it can be converted into a frequency domain feature vector or other forms of orthogonal feature vectors through the action of the orthogonal hash function group, so that in the new hash space, the correlation between different feature dimensions is significantly reduced, thereby more clearly showing the independent characteristics and mutual relationships of each feature.

[0077] The consensus correlation analysis subunit calculates the hash similarity between any two mapped feature vectors based on the mapped feature vectors, thereby generating a feature correlation consensus matrix. Hash similarity can be calculated using various algorithms, such as the Hamming distance algorithm or the cosine similarity algorithm. Taking the cosine similarity algorithm as an example, its basic principle is to measure the similarity between two vectors by calculating the cosine of the angle between them. A cosine value closer to 1 indicates a more similar vector, i.e., a stronger correlation between the features; conversely, a cosine value closer to -1 or 0 indicates a weaker or irrelevant correlation. In practice, the mapped feature vectors are first normalized to eliminate the influence of dimensionality differences. Then, the cosine similarity values ​​are calculated pairwise between any two vectors. These values ​​are arranged in the order of the feature vector indices to form a two-dimensional matrix, the feature correlation consensus matrix. The rows and columns of this matrix correspond to different feature verification vectors, and each element in the matrix is ​​a value between 0 and 1, intuitively reflecting the strength of the correlation between any two feature vectors and providing a quantitative basis for subsequent feature screening and optimization.

[0078] The feature optimization unit's processing flow includes chain convolution of the consensus constraint association matrix, feature extraction using a bidirectional timestamp coding network, and matrix reorganization. First, the consensus constraint association matrix undergoes chain convolution. This process employs a three-dimensional convolution approach, using a convolution kernel of a specific size (e.g., 3×3×3) to perform sliding convolutions on the matrix in both spatial and temporal dimensions to extract feature association patterns. Chain convolution captures the local associations between adjacent feature vectors in the matrix, as well as the dynamic changes in features across different time slices, generating a consensus constraint feature tensor. This tensor not only incorporates the feature association information in the original matrix but also enhances the spatial and temporal correlations between features through convolution operations, providing richer context for subsequent feature fusion.

[0079] Next, the set of feature verification vectors and the consensus-constrained feature tensor are input into a bidirectional timestamp encoding network. This network typically employs a recurrent neural network (RNN) variant, such as a Bi-LSTM network. Its core advantage lies in its ability to simultaneously capture feature dependencies between past and future time steps in time series data. The network's input layer receives two inputs: the mapped feature verification vectors and the corresponding timestamp encoding information. The timestamp encoding can use one-hot encoding or other suitable encoding methods to identify the temporal order of the feature vectors. The hidden layer consists of bidirectional recurrent neurons. The forward neurons process the time series from front to back, capturing feature information from past time steps; the backward neurons process the time series from back to front, capturing feature information from future time steps. The outputs of these two vectors are fused in the hidden layer to generate an intermediate feature representation containing contextual information. Based on the outputs of the hidden layers, the output layer generates a set of chained contextual feature vectors. Each vector integrates the feature dependencies before and after the time point, more comprehensively reflecting the time series characteristics of the device's operating status.

[0080] Finally, a matrix reorganization operation is performed on the set of chain context feature vectors. The specific method of matrix reorganization is to arrange the multi-dimensional chain context feature vectors in chronological order and convert them into a two-dimensional optimized verification feature matrix. For example, if each chain context feature vector is an n-dimensional vector and there are m time step vectors in total, then the reorganized matrix is ​​m rows and n columns, where each row corresponds to a feature vector of a time step and each column corresponds to a feature dimension. Through this matrix reorganization operation, the scattered feature vectors are integrated into a structured matrix form, which facilitates the subsequent dynamic decision reconstruction module to perform feature fusion and decision analysis. The optimized verification feature matrix not only retains the spatiotemporal information of the original features, but also enhances the correlation and context dependency between features through chain convolution processing and feature extraction of the bidirectional timestamp coding network, so that each element in the matrix can more accurately reflect the comprehensive information of the device operation status at a specific time and feature dimension.

[0081] In the collaborative process between the consensus association calculation unit and the feature optimization unit, the feature mapping subunit provides more efficient feature representation through orthogonal hash space conversion for feature similarity calculation. The consensus correlation analysis subunit provides a clear basis for feature screening by quantifying the correlation strength between features. The feature optimization unit achieves spatiotemporal correlation optimization and cross-chain aggregation of features through deep feature extraction using chained convolution processing and a bidirectional timestamp coding network. This layer-by-layer processing approach, from spatial feature conversion to similarity calculation and then to deep feature fusion, forms a complete feature verification and optimization process. This ensures that the cross-chain feature verification module can output a high-quality set of optimized and verified feature matrices, providing reliable feature data support for dynamic decision-making and equipment control within the entire blockchain-based cleaning and spraying data management system. Through this design, the system can effectively mine and utilize deep feature correlations in equipment operation data in complex distributed environments, improving data processing accuracy and scientific decision-making, thereby achieving intelligent and precise management of cleaning and spraying equipment.

[0082] Example 3:

[0083] This embodiment relates to the specific implementation of a dynamic decision reconstruction module, which includes a feature matrix compression unit, a hash entropy value calculation unit, a benchmark feature determination unit, a dynamic weight calculation unit, and a chain reconstruction unit. The feature matrix compression unit needs to perform maximum pooling processing on the timestamp dimension for each matrix in the set of optimized verification feature matrices. Specifically, for each optimized verification feature matrix, its rows correspond to time series and its columns correspond to feature dimensions. The maximum pooling processing is to take the maximum value of all time point data of each feature dimension (column) in the timestamp dimension (i.e., row direction), thereby compressing the two-dimensional matrix into a one-dimensional optimized verification feature vector to form a set of optimized verification feature vectors. This processing method can retain the key peak information of each feature dimension in the time series, eliminate redundant time details, realize the compression of matrix dimensions and the extraction of key features, and facilitate subsequent entropy value calculation and weight distribution.

[0084] The hash entropy value calculation unit is used to calculate the hash entropy value of the optimized verification feature vector. The processing flow is as follows: First, the mean vector and variance vector of the optimized verification feature vector are calculated. The mean vector Each element of is the average value of the corresponding feature dimension, reflecting the average level of the feature; the variance vector Each element of is the variance of the corresponding feature dimension, reflecting the degree of fluctuation of the feature. Next, the optimized verification feature vector With the mean vector Perform element-by-element difference calculation to obtain the difference vector , and then square the difference result to generate the feature hash difference vector , each element of the vector represents the square of the deviation of the corresponding eigenvalue from the mean. The larger the value, the farther the eigenvalue deviates from the mean. Then, the overall median of the feature hash difference vector is calculated As a robust statistic, the median can reduce the impact of extreme values ​​on the results and characterize the central tendency of the feature distribution. Finally, the median and the square of the variance vector are combined. Perform ratio operation, get the intermediate result and input normalization function (such as Sigmoid function) to map the result to interval, get the hash entropy value The calculation formula of hash entropy value can be expressed as:

[0085] ;

[0086] in, Represents a normalization function, which is used to convert the ratio result into a standardized entropy value. A larger entropy value indicates a higher uncertainty of the feature, and a higher weight may need to be given in the decision-making process.

[0087] The function of the benchmark feature determination unit is to select the optimized verification feature vector corresponding to the maximum hash entropy value from the dynamic hash entropy set as the initial decision benchmark vector This vector corresponds to the characteristic state with the highest uncertainty in the data. Using it as a reference point for decision-making can ensure that the decision-making process fully considers the most variable and uncertain features in the data, avoiding decision-making bias caused by ignoring key changing features. For example, if the hash entropy value of a certain characteristic vector is the largest, it means that the corresponding equipment operating state fluctuates greatly or is more dispersed in the time series, which may reflect abnormal operating conditions or critical operating stages in the equipment operation. Setting it as a reference vector will help highlight the impact of such features in subsequent weight calculations.

[0088] The dynamic weight calculation unit needs to calculate the dynamic decision weight of each feature vector based on the hash distance between the optimized verification feature vector and the initial decision reference vector and the hash entropy value of each feature vector. The specific steps are as follows: First, the hash entropy value of the optimized verification feature vector is calculated. Hash entropy value of the initial decision reference vector Multiply by the first decision coefficient (For example, the initial setting is 0.6, which can be adjusted according to actual needs), and the first dynamic decision factor is obtained , this factor reflects the impact of the difference in entropy between the eigenvector and the reference vector on the weight. ,but If it is positive, the weight is increased; otherwise, the weight is decreased. Secondly, the absolute value of the Euclidean distance between the optimized verification feature vector and the initial decision reference vector is calculated. , multiplied by the second decision coefficient (For example, the initial setting is 0.4, which can be adjusted dynamically), and the second dynamic decision factor is obtained , which reflects the influence of the distance between the feature vector and the reference vector in the feature space on the weight. The larger the distance, the greater the difference in features, and the different contributions to the decision. Finally, the first dynamic decision factor and the second dynamic decision factor are weighted and summed to obtain the dynamic decision weight. , the calculation formula is:

[0089] ;

[0090] in, and is the weight coefficient, which is used to adjust the influence of entropy value difference and spatial distance on weight. The sum of the two is usually 1 (e.g. ) to ensure a balanced weight calculation. In this way, the dynamic decision weight of each feature vector takes into account both its uncertainty (entropy) and its difference from the baseline vector (spatial distance), making the weight allocation more scientific and reasonable.

[0091] The chain reconstruction unit uses the dynamic weight set to perform weighted aggregation on the set of optimized verification feature vectors to generate a data chain decision feature graph. The corresponding dynamic decision weight Multiply to get the weighted eigenvector , then sum up all weighted feature vectors to get the data chain decision feature map The feature map is presented in the form of a matrix, where rows correspond to different decision factors or feature combinations, and columns correspond to feature dimensions. The value of each element in the matrix reflects the comprehensive decision-making impact of the corresponding feature after weighted aggregation. For example, if a feature dimension has a high hash entropy value and distance from the reference vector in the feature vectors of multiple time steps, then its corresponding element value in the feature map will increase significantly due to weighted aggregation, indicating that the feature has a high contribution to the decision and needs to be paid special attention to in subsequent equipment control.

[0092] Throughout the dynamic decision reconstruction module's workflow, the feature matrix compression unit rapidly extracts key features through maximum pooling. The hash entropy calculation unit addresses feature uncertainty through statistics and normalization. The baseline feature determination unit uses high-entropy features as decision anchors. The dynamic weight calculation unit scientifically assigns weights based on entropy differences and spatial distances. Finally, the chain reconstruction unit generates a comprehensive decision feature map through weighted aggregation. These units are interconnected, from feature compression and entropy calculation to weight assignment and feature aggregation, forming a complete dynamic decision chain. This ensures that the system can dynamically adjust its decision basis based on the real-time characteristics of equipment operating data, generating a decision feature map that better reflects actual operating conditions. This provides accurate input data for the instruction generation module, enabling precise control of the operating parameters of the cleaning and spraying equipment. This design not only fully leverages the distributed storage and verification advantages of blockchain data, but also, through a dynamic weighting mechanism and feature fusion algorithm, enhances the system's adaptability to complex operating conditions and improves decision-making flexibility, providing solid technical support for the intelligent management of cleaning and spraying equipment data.

[0093] Example 4:

[0094] This embodiment involves a further implementation of a dynamic weight calculation unit and an execution instruction generation module. The dynamic weight calculation unit introduces an adaptive adjustment mechanism when calculating the first dynamic decision factor to enhance the weight difference of high-entropy features. Specifically, when the hash entropy value of the optimized verification feature vector is greater than the entropy value of the initial decision reference vector, the system automatically increases the first decision coefficient from the initial value (such as 0.6) to 0.8, and vice versa. For example, assuming that the equipment operating pressure data corresponding to a certain feature vector fluctuates greatly in the time series, its hash entropy value is significantly higher than the entropy value of the reference vector. At this time, the first decision coefficient is increased, which increases the first dynamic decision factor of the feature vector, thereby occupying a higher proportion in the final weight and highlighting its impact on the decision. The calculation of the second dynamic decision factor uses the standardized Euclidean distance. The feature vector is first Z-score standardized to eliminate the influence of different feature dimensions (such as Pa for pressure and m³ / h for flow). Taking equipment operating temperature (unit: °C) and spray flow rate (unit: L / min) as examples, both are converted to dimensionless values ​​after normalization. The absolute value of the distance from the reference vector is then calculated and multiplied by the dynamically adjusted second decision coefficient (initial value 0.4, which can fluctuate within ±20% based on the overall data fluctuation range). For example, if the fluctuation range of temperature characteristics suddenly increases within a certain period of time, the system will automatically increase the second decision coefficient to 0.48 to enhance the role of distance in the weight calculation, ensure that the weight distribution can adapt to changes in data distribution, and improve the robustness of the calculation.

[0095] The execution instruction generation module inputs the data-chain decision feature graph into a smart contract-based instruction generator. The smart contract predefines the device control logic. For example, for a common cleaning and spraying equipment combination, when the spray pressure characteristic value in the data-chain decision feature graph exceeds a preset threshold (e.g., 8 MPa), the smart contract triggers a pressure adjustment instruction. The instruction generator calculates specific parameters by combining the characteristic value with a preset control function (e.g., a linear mapping function). For example, if the current characteristic value is 9 MPa and the preset function is "command parameter = characteristic value - threshold," the generated pressure adjustment instruction parameter is 1 MPa, which controls the device to reduce the pressure by 1 MPa. The instruction generator utilizes a modular design and supports control instruction templates for different types of equipment, automatically matching them based on the device type identifier field in the feature graph. For example, if the feature graph contains the identifier "Equipment Type = High-Pressure Washer," the instruction generator automatically calls the control template for the high-pressure washer and generates instructions for adjusting the pressure valve opening and starting and stopping the cleaning fluid pump. If the identifier is "Equipment Type = Electrostatic Spray Gun," the control template for the spray gun is called and generates instructions for adjusting the spray flow rate and adjusting the electrostatic voltage.

[0096] The generated device control instruction set consists of an instruction code, parameter values, an execution timestamp, and a checksum. For example, for a pressure adjustment instruction for a high-pressure cleaner, the instruction code "CMD_001" represents pressure adjustment, the parameter value "-1MPa" indicates a 1MPa pressure reduction, and the execution timestamp "2025-05-30T14:30:00Z" specifies the instruction execution time. The checksum is generated by hashing the instruction code, parameter values, and timestamp using the SHA-256 algorithm to ensure the integrity of the instruction during transmission. The instruction is broadcast to the device control node via the blockchain. Upon receiving the instruction, the node first uses the checksum to verify that the instruction has not been tampered with. If verification is successful, it parses the instruction parameters and executes the corresponding operation. For example, after parsing the pressure adjustment instruction, the control node of the high-pressure cleaner sends an electrical signal to the pressure valve actuator to adjust the valve opening to reduce pressure. The sensor monitors the pressure changes in real time and feeds the execution results back to the cleaning parameter acquisition module, forming a closed-loop control system.

[0097] During the dynamic weight calculation process, the combination of an adaptive adjustment mechanism and standardized processing can effectively cope with feature changes under different working conditions. For example, when the equipment switches from conventional cleaning mode to fine spray mode, the volatility of the flow characteristics may increase significantly, and its hash entropy value will increase accordingly and exceed the reference vector. At this time, the first decision coefficient is automatically increased, which increases the weight of the flow characteristics. The system pays more attention to changes in flow data when making decisions, thereby generating control instructions that better meet the needs of fine spraying. The application of standardized Euclidean distance avoids weight deviations caused by dimensional differences, ensuring that different features such as temperature, pressure, and flow have a fair competitive basis in weight calculation.

[0098] The smart contract mechanism of the execution instruction generation module ensures the transparency and tamper-proof nature of the control logic. For example, the smart contract pre-sets the logic of "automatically starting the cooling system when the temperature exceeds 60°C." This logic is stored in code on each node of the blockchain and cannot be modified by any single node without authorization. When the temperature characteristic value in the characteristic graph reaches 65°C, the instruction generator automatically triggers the cooling system start instruction, and the instruction is synchronized to all nodes through the blockchain network to ensure the consistency and reliability of equipment control. In addition, the modular instruction template design improves the compatibility of the system and can quickly adapt to new types of cleaning and spraying equipment. Simply add the corresponding control template to the system to achieve integrated management of the new equipment.

[0099] The dynamic weight calculation unit achieves flexible adjustment of feature weights through adaptive mechanisms and standardized methods. The execution instruction generation module ensures the accuracy and compatibility of instructions through smart contracts and modular design. This design not only improves the system's responsiveness to complex working conditions but also ensures the credibility and traceability of control instructions through the distributed nature of blockchain, providing a complete technical solution for the intelligent and automated management of cleaning and spraying equipment. From the weight calculation of feature vectors to the generation and execution of instructions, each link is based on actual data characteristics and equipment control requirements, forming a complete closed loop from data acquisition and processing to decision-making and execution, ensuring the efficient and reliable operation of the system.

[0100] Example 5:

[0101] This embodiment involves the underlying implementation of spatiotemporal dimension division and feature slicing encryption. Spatiotemporal dimension division is achieved through a dual mechanism of timestamp slicing and spatial grid coding. Taking a spraying production line as an example, the equipment is equipped with a GPS positioning module and a high-precision clock to collect operating status data streams (such as pressure, flow, temperature, operating coordinates, etc.) in real time. Timestamp slicing can be set to 100ms intervals (adjustable according to the data update frequency), dividing the continuous data stream into multiple time windows, each window corresponding to a data segment. For example, at 10:00:00 on May 30, 2025, the system generates a data segment with a timestamp of "20250530100000", which contains the equipment operation data within 100ms before and after that moment.

[0102] Spatial grid coding uses a geographic information system (GIS) gridding algorithm to divide the equipment's operating area into 50m x 50m grid cells (resolution can be adjusted based on the operating range). For example, if a device's current GPS coordinates are (longitude 118.78°, latitude 31.95°), they are mapped to a grid cell numbered "A05" through the gridding algorithm. Each data segment is assigned a corresponding grid code, forming a spatiotemporally correlated slice identifier, such as "20250530100000-001_A05," where "20250530100000" is the timestamp, "001" is the slice sequence number (indicating the first data slice within the time window), and "A05" is the spatial grid code. This identification method provides each data slice with clear spatiotemporal coordinates, facilitating subsequent retrieval and verification of data by region or time range.

[0103] Feature shard encryption uses asymmetric keys from distributed nodes to perform cross-signature verification on a set of shards. Assume that a blockchain network consists of three distributed nodes: Node A, Node B, and Node C. Each node possesses a unique public-private key pair (public key Puk, private key Prk). When a device collects the data shard "20250530100000-001_A05," the source node (e.g., Node A) first digitally signs the shard using its private key Prk_A, generating signature S1. It then transmits the shard and signature S1 to its neighboring node, Node B. After receiving the data shard, Node B verifies the validity of signature S1 using Node A's public key Puk_A. It generates a digest of the shard content using a hash algorithm, then decrypts signature S1 using Puk_A, comparing the two to ensure consistency. If verification is successful, Node B re-signs the shard using its private key Prk_B, generating signature S2, and transmits the shard, S1, and S2 to Node C. Node C repeats the above verification process, verifying the signatures of node A and node B in turn. After the verification is passed, it adds its own signature S3, and finally forms a shard set containing multi-node cross-signatures.

[0104] The cross-signed shards are organized into a Merkle tree structure. Each shard's hash value contains the original data and all signature information. For example, if the shard content is D and the signatures are S1, S2, and S3, the shard hash value H(D, S1, S2, S3) is generated using the SHA-256 algorithm. The bottom-level leaf nodes are the hash values ​​of each shard. Adjacent leaf nodes are combined to generate the hash value of the parent node above, until a root hash value is generated. The root hash value is stored synchronously with the shard set in the blocks of each node in the blockchain. To verify data integrity, simply compare the root hash values ​​stored on each node. To trace the data's origin, the signature chain (S1 → S2 → S3) of each shard can be used to determine the sequence of nodes involved in the verification, ensuring that the data cannot be tampered with and its origin can be traced.

[0105] For example, consider the cross-regional operation of cleaning and spraying equipment. Suppose the equipment moves from grid A05 to grid B07. The corresponding data stream is divided into multiple shards in chronological order, such as "20250530100000-001_A05," "20250530100000-002_B07," and so on. Each shard undergoes cross-signature verification by adjacent nodes during transmission, ensuring that data generated by the equipment operating in different regions is reliably recorded. For example, when the equipment enters grid B07, the newly generated shard is cross-signed by nodes B, C, and D, forming a new signature chain, demonstrating the distributed node verification coverage in the spatial dimension.

[0106] In a data retrieval scenario, for example, to query equipment operating data for grid A05 between 10:00 AM and 10:01 AM on May 30, 2025, the corresponding shard set can be quickly located using the spatiotemporal identifier. The system then filters out all eligible shards based on the timestamps "20250530100000" to "20250530100100" and the grid code "A05." The integrity of each shard is then verified using the Merkle tree hash path to ensure the accuracy of the retrieved data.

[0107] The combination of a cross-signature mechanism and a Merkle tree structure not only enables encrypted storage of data shards but also enhances system security through distributed node consensus verification. For example, if a malicious node attempts to tamper with the contents of a shard, it must simultaneously tamper with the signatures of all participating signing nodes, a feasibility that is virtually impossible under the blockchain consensus mechanism. Furthermore, the spatial and temporal division of data provides a clear temporal index, facilitating data aggregation and analysis based on business needs (such as regional operation statistics and time period performance analysis), thereby improving the system's data management efficiency.

[0108] The processing flow of the entire embodiment is closely centered around the spatiotemporal attributes of the data and the distributed verification requirements. It assigns clear spatiotemporal coordinates to the data through timestamp slicing and spatial grid encoding, and uses asymmetric key cross-signatures and Merkle tree structures to achieve data encryption and integrity verification. This design not only meets the data segmentation requirements of cleaning and spraying equipment in a dynamic operating environment, but also ensures the credibility and traceability of the data through the distributed characteristics of the blockchain, providing a reliable technical foundation for cross-regional and cross-time data management and application. From the generation of data shards, signature verification to the construction of the Merkle tree, each link is closely connected to form a complete spatiotemporal data encryption and verification system, ensuring that the system can efficiently manage cleaning and spraying data in complex physical spaces and time series.

[0109] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0110] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A cleaning and spraying data management system based on blockchain, characterized in that: include: Cleaning parameter acquisition module, used to obtain the operating status data stream of the spraying equipment through distributed nodes; A data sharding encryption module, configured to perform spatiotemporal dimension division and feature sharding encryption on the operating status data stream to generate a set of data verification feature matrices; A cross-chain feature verification module, used to perform consensus mechanism screening and feature correlation verification on the set of data verification feature matrices to obtain a set of optimized verification feature matrices; A dynamic decision reconstruction module is used to perform feature fusion based on hash time lock on the set of optimized verification feature matrices to generate a data chain decision feature graph, wherein the data chain decision feature graph is presented in the form of a matrix, and each element value in the matrix reflects the comprehensive decision impact of the corresponding feature after weighted aggregation; An execution instruction generation module, configured to generate a device control instruction set according to the data chain decision feature graph; The cross-chain feature verification module includes: a feature matrix deconstruction unit, configured to perform chain parsing on the data verification feature matrix along a timestamp dimension to obtain a set of feature verification vectors; a consensus association calculation unit, configured to calculate a consensus verification value between any two feature vectors in the set of feature verification vectors to generate a feature association consensus matrix; a dynamic screening unit, configured to adjust the weights of the feature association consensus matrix according to the hash differences of adjacent blocks in the set of feature verification vectors to generate a consensus constraint association matrix; A feature optimization unit, configured to perform cross-chain convolution aggregation on the consensus constraint association matrix and the set of feature verification vectors to obtain the optimized verification feature matrix; The feature optimization unit is specifically implemented as follows: Performing chain convolution processing on the consensus constraint association matrix to generate a consensus constraint feature tensor; Inputting the set of feature verification vectors and the consensus constraint feature tensor into a bidirectional timestamp encoding network to generate a set of chained context feature vectors; Matrix reorganization is performed on the set of chained context feature vectors to generate the optimized verification feature matrix.

2. The blockchain-based cleaning and spraying data management system according to claim 1 is characterized in that: The data fragmentation encryption module includes: A data stream segmentation unit, configured to dynamically segment the running status data stream according to a preset block size to form a data shard set; The feature encryption unit is used to input each shard in the data shard set into a data shard encryptor based on a Merkle tree to generate a set of data verification feature matrices.

3. The blockchain-based cleaning and spraying data management system according to claim 1 is characterized in that: The consensus association calculation unit includes: A feature mapping subunit, configured to map each feature vector in the set of feature verification vectors to an orthogonal hash space to obtain a set of mapped feature vectors; The consensus correlation analysis subunit is used to calculate the hash similarity between any two mapped feature vectors in the set of mapped feature vectors to generate the feature correlation consensus matrix composed of multiple consensus correlation values.

4. The blockchain-based cleaning and spraying data management system according to claim 1 is characterized in that: The dynamic decision reconstruction module includes: A feature matrix compression unit, configured to perform maximum pooling processing of a timestamp dimension on each matrix in the set of optimized verification feature matrices to generate a set of optimized verification feature vectors; A hash entropy value calculation unit, configured to calculate a hash entropy value of each feature vector in the set of optimized verification feature vectors to generate a dynamic hash entropy set; A reference feature determination unit, configured to select an optimized verification feature vector corresponding to a maximum hash entropy value in the dynamic hash entropy set as an initial decision reference vector; a dynamic weight calculation unit, configured to calculate the dynamic decision weight of each feature vector in the set of optimized verification feature vectors based on the hash distance between each feature vector and the initial decision reference vector and the hash entropy value of each feature vector to generate a dynamic weight set; A chain reconstruction unit is used to perform weighted aggregation on the set of optimized verification feature vectors using the dynamic weight set to generate the data chain decision feature graph.

5. The blockchain-based cleaning and spraying data management system according to claim 4 is characterized in that: The hash entropy value calculation unit is specifically implemented as follows: Calculating the mean vector and variance vector of the optimized verification feature vector; Performing element-by-element difference calculation on the optimized verification feature vector and the mean vector, and performing a square operation on the difference result to generate a feature hash difference vector; Calculating the overall median of the feature hash difference vector; performing a ratio operation on the median and the square value of the variance vector, and inputting the result into a normalization function to obtain the hash entropy value.

6. The blockchain-based cleaning and spraying data management system according to claim 5 is characterized in that: The dynamic weight calculation unit is specifically implemented as follows: Multiplying the hash entropy value of the optimized verification feature vector and the hash entropy value of the initial decision reference vector by a first decision coefficient to obtain a first dynamic decision factor; Multiplying the absolute value of the Euclidean distance between the optimized verification feature vector and the initial decision reference vector by a second decision coefficient to obtain a second dynamic decision factor; The first dynamic decision factor and the second dynamic decision factor are weighted and summed to obtain the dynamic decision weight.

7. The blockchain-based cleaning and spraying data management system according to claim 6 is characterized in that: The execution instruction generation module is specifically implemented as follows: The data chain decision feature graph is input into the instruction generator based on the smart contract to generate the equipment control instruction set, which is used to control the operating parameters of the cleaning spray equipment.

8. The blockchain-based cleaning and spraying data management system according to claim 1 is characterized in that: The time and space dimension division includes: Dividing the running status data stream into continuous data segments according to timestamps, and adding a spatial grid code to each data segment to form a spatiotemporally associated slice set; The feature shard encryption performs cross-signature verification on the shard set through the asymmetric keys of the distributed nodes.

Images