Watermark embedding method and device for standard data of power grid and electronic equipment

Abstract

The embodiment of the invention provides a watermark embedding method and device for standard data of a power grid and electronic equipment. The method comprises the following steps: acquiring initial data of power equipment of a power grid; wherein the initial data represents related power data generated in the operation process of power equipment of the power grid; carrying out topological feature extraction on the initial data to obtain a topological feature vector; according to a preset watermark library and the topological feature vector, determining watermark distribution information; wherein the topological feature vectors represent topological structure features of the initial data; the preset watermark library comprises watermark information; the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data; and according to the watermark distribution information, performing embedding processing on the initial data to obtain a watermark-embedded data set. The method is used for achieving the watermark embedding effect suitable for the standard data of the power grid.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a method and device for embedding a watermark into standard data of a power grid, and an electronic device. BACKGROUND

[0002] With the development of digitalization of the power grid, standard data of power equipment has become a key digital asset. In order to effectively trace the data source, confirm the data ownership and verify the integrity when necessary, a watermark needs to be embedded into such data.

[0003] In the prior art, a fixed watermark information is usually used and implanted into a fixed position in the data. This method has poor robustness and is easily affected by data processing operations and thus invalid. At the same time, it also destroys the structure and statistical characteristics of the data itself, thereby interfering with subsequent data analysis and application.

[0004] Therefore, there is an urgent need for a scheme for embedding a watermark into standard data of a power grid. SUMMARY

[0005] Embodiments of the present application provide a method and device for embedding a watermark into standard data of a power grid, and an electronic device, to achieve the effect of embedding a watermark into standard data of a power grid.

[0006] In a first aspect, the embodiments of the present application provide a method for embedding a watermark into standard data of a power grid, comprising:

[0007] obtaining initial data of power equipment of a power grid; wherein the initial data represents related power data generated during operation of the power equipment of the power grid;

[0008] extracting topological features from the initial data to obtain a topological feature vector; determining watermark distribution information according to a preset watermark library and the topological feature vector; wherein the topological feature vector represents topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents embedding positions and intensity distributions of the watermark information in the initial data;

[0009] performing embedding processing on the initial data according to the watermark distribution information to obtain a data set with embedded watermarks.

[0010] In a possible implementation, extracting topological features from the initial data to obtain a topological feature vector comprises:

[0011] performing data standardization processing on the initial data to obtain standardized data;

[0012] performing similarity processing on the standardized data according to a preset similarity function to obtain an adjacency matrix, wherein the adjacency matrix represents a similarity relationship between power equipment nodes in the standardized data;

[0013] performing processing on the adjacency matrix to obtain the topological feature vector.

[0014] In a possible implementation, the processing on the adjacency matrix to obtain the topological feature vector comprises:

[0015] performing feature extraction processing on the adjacency matrix according to a topological invariant algorithm to obtain a topological invariant, wherein the topological invariant represents a topological feature of the adjacency matrix;

[0016] performing data coding processing on the topological invariant to obtain the topological feature vector.

[0017] In a possible implementation, the determination of the watermark distribution information according to the preset watermark library and the topological feature vector comprises:

[0018] determining watermark information from the preset watermark library and performing block processing on the watermark information to obtain watermark block data;

[0019] performing perturbation processing on the topological feature vector according to a preset perturbation function to obtain a change degree of the topological feature vector, and determining a topological sensitive vector based on the change degree, wherein the topological sensitive vector represents a data sensitivity degree of the topological feature vector;

[0020] determining the watermark distribution information according to the watermark block data, the topological sensitive vector, and the topological feature vector.

[0021] In a possible implementation, the determination of the watermark distribution information according to the watermark block data, the topological sensitive vector, and the topological feature vector comprises:

[0022] performing comparison processing on a parameter in the topological sensitive vector and a preset threshold to determine a watermark intensity vector, wherein the watermark intensity vector represents watermark intensity information at different positions in the topological sensitive vector;

[0023] determining a watermark embedding position according to a topological complexity in the topological feature vector;

[0024] determining the watermark distribution information according to the watermark block data, the watermark intensity information, and the watermark embedding position.

[0025] In a possible implementation, the embedding processing on the initial data according to the watermark distribution information to obtain a data set with embedded watermark comprises:

[0026] The initial data is processed sequentially in the time domain and then in the wavelet transform domain to obtain frequency domain data.

[0027] Based on the watermark distribution information, the frequency domain data is embedded to obtain a dataset with embedded watermarks.

[0028] In one possible implementation, acquiring initial data from the power grid's electrical equipment includes:

[0029] Acquire standard data of power equipment in the power grid; wherein, the standard data represents the relevant power data generated during the operation of the power equipment in the power grid initially collected;

[0030] The standard data is preprocessed to obtain processed data; the data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation.

[0031] Based on the encryption algorithm, the processed data is subjected to data fingerprint encryption to obtain the initial data.

[0032] Secondly, embodiments of this application provide a watermark embedding device for standard power grid data, comprising:

[0033] An acquisition module is used to acquire initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid;

[0034] The determination module is used to extract topological features from the initial data to obtain a topological feature vector; and to determine watermark distribution information based on a preset watermark library and the topological feature vector; wherein, the topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data.

[0035] The processing module is used to perform embedding processing on the initial data according to the watermark distribution information to obtain a dataset with embedded watermarks.

[0036] In one possible implementation, the determining module includes:

[0037] A standardization processing module is used to perform data standardization processing on the initial data to obtain standardized data;

[0038] A similarity processing module is used to perform similarity processing on the standardized data according to a preset similarity function to obtain an adjacency matrix; wherein, the adjacency matrix represents the similarity relationship between power equipment nodes in the standardized data;

[0039] The feature processing module is used to process the adjacency matrix to obtain the topological feature vector.

[0040] In one possible implementation, the feature processing module includes:

[0041] According to the topological invariant algorithm, the adjacency matrix is ​​subjected to feature extraction processing to obtain topological invariants; wherein, the topological invariants represent the topological features of the adjacency matrix;

[0042] The topological invariants are digitized and encoded to obtain the topological feature vector.

[0043] In one possible implementation, the determining module includes:

[0044] The block processing module is used to determine watermark information from the preset watermark library and to perform block processing on the watermark information to obtain watermark block data.

[0045] The perturbation processing module is used to perturb the topological feature vector according to a preset perturbation function to obtain the degree of change of the topological feature vector; and to determine the topological sensitivity vector based on the degree of change; wherein the topological sensitivity vector characterizes the data sensitivity of the topological feature vector.

[0046] The confirmation submodule is used to determine the watermark distribution information based on the watermark block data, the topology sensitivity vector, and the topology feature vector.

[0047] In one possible implementation, the verification submodule includes:

[0048] The watermark intensity vector is determined by comparing the parameters in the topology-sensitive vector with a preset threshold; wherein the watermark intensity vector represents the watermark intensity information at different positions in the topology-sensitive vector.

[0049] The watermark embedding position is determined based on the topological complexity in the topological feature vector.

[0050] The watermark distribution information is determined based on the watermark block data, the watermark intensity information, and the watermark embedding position.

[0051] In one possible implementation, the processing module includes:

[0052] The initial data is processed sequentially in the time domain and then in the wavelet transform domain to obtain frequency domain data.

[0053] Based on the watermark distribution information, the frequency domain data is embedded to obtain a dataset with embedded watermarks.

[0054] In one possible implementation, the acquisition module includes:

[0055] Acquire standard data of power equipment in the power grid; wherein, the standard data represents the relevant power data generated during the operation of the power equipment in the power grid initially collected;

[0056] The standard data is preprocessed to obtain processed data; the data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation.

[0057] Based on the encryption algorithm, the processed data is subjected to data fingerprint encryption to obtain the initial data.

[0058] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0059] The memory stores computer-executed instructions;

[0060] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0061] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0062] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0063] This application provides a method, apparatus, and electronic device for watermarking standard data of a power grid. It acquires initial data from the power equipment of the power grid, which represents the electrical data generated during the operation of the equipment. Next, topological features are extracted from the initial data to obtain a topological feature vector characterizing the topological structure of the initial data. Then, based on a preset watermark library and the topological feature vector, watermark distribution information is determined. The preset watermark library contains watermark information, while the watermark distribution information clarifies the embedding position and intensity distribution of the watermark information in the initial data, providing a basis for accurate watermark embedding. Finally, based on the watermark distribution information, the initial data is embedded to obtain a dataset with embedded watermarks. This effectively solves the technical problems of unreasonable watermark embedding positions and intensities in existing power grid data watermark embedding methods, which are susceptible to attacks and tampering, leading to difficulties in ensuring data security and integrity. It improves the accuracy and security of watermark embedding in standard power grid data, enhances the data's resistance to attacks during transmission and storage, and ensures the security and reliability of power grid data. Attached Figure Description

[0064] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0065] Figure 1 A flowchart illustrating a watermark embedding method for standard power grid data provided in this application embodiment. Figure 1 ;

[0066] Figure 2 A flowchart illustrating a watermark embedding method for standard power grid data provided in this application embodiment. Figure 2 ;

[0067] Figure 3 A schematic diagram of the structure of a watermark embedding device for standard power grid data provided in this application embodiment. Figure 1 ;

[0068] Figure 4 A schematic diagram of the structure of a watermark embedding device for standard power grid data provided in this application embodiment. Figure 2 ;

[0069] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0070] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0071] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0072] As the digitalization of the power grid deepens, the massive amounts of standard data generated and accumulated by power equipment during operation are becoming increasingly valuable, forming a crucial digital asset for the power industry. This data not only covers basic information, operating status, and performance parameters of power equipment, but also provides indispensable decision-making support for all aspects of power system planning, dispatching, and operation and maintenance. It plays a vital role in ensuring the safe and stable operation of the power grid and improving the reliability and efficiency of power supply.

[0073] In practical applications, in order to effectively trace the source of data, clarify the responsible parties at each stage of data generation, transmission, storage and use; confirm the ownership of data to prevent data from being illegally seized or misused; and verify the integrity of data when necessary to ensure that data has not been tampered with or damaged during the flow of data, thereby guaranteeing the authenticity and reliability of data.

[0074] Current watermark embedding technologies have several limitations. They typically use fixed watermark information, embedding it in a fixed location within the data. This fixed embedding method makes the watermark lack flexibility and adaptability, resulting in poor robustness. In practical applications, standard data for power equipment often undergoes various data processing operations, such as data compression, format conversion, filtering, encryption, and decryption. These operations may damage the watermark information to varying degrees, making it easy to remove or impossible to detect, thus rendering the watermark ineffective and failing to achieve the expected data protection and authentication effects.

[0075] Therefore, embodiments of this application provide a watermark embedding method, apparatus, and electronic device for standard power grid data, which can solve the above-mentioned problems.

[0076] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0077] Figure 1 A flowchart illustrating a watermark embedding method for standard power grid data provided in this application embodiment. Figure 1 ,like Figure 1 As shown, the method includes:

[0078] S101. Obtain initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid.

[0079] For example, power equipment in a power grid refers to various devices used for the transmission, distribution, conversion, and control of electrical energy, such as transformers, circuit breakers, transmission lines, and generators.

[0080] Initial data refers to various data generated during the operation of these power devices that can reflect their operating status and performance, such as voltage, current, power, temperature, and equipment status.

[0081] Various technical means can be employed in acquiring initial data. A common approach is to utilize power grid monitoring equipment, which can monitor the operating status of power equipment in real time and transmit the data to a central processing system. For example, voltage transformers and current transformers can collect voltage and current data from various nodes in the power grid in real time; this data is crucial for assessing the voltage stability and current distribution of the power grid. Another approach is to utilize a Supervisory Control and Data Acquisition (SCADA) system, which integrates data from multiple monitoring devices, enabling centralized data management and analysis. Through SCADA systems, more comprehensive power grid operating data can be obtained, including equipment status and load characteristics, providing strong support for subsequent power grid optimization and scheduling. Furthermore, historical data can be obtained from historical databases; this data is of great significance for analyzing the long-term operating trends and fault modes of the power grid.

[0082] S102. Extract topological features from the initial data to obtain a topological feature vector; determine the watermark distribution information based on the preset watermark library and the topological feature vector; wherein, the topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data.

[0083] For example, the initial data needs to be preprocessed, such as through data cleaning and format conversion, to ensure accuracy and consistency. Then, topological analysis algorithms are used to extract topological features from the preprocessed data. This process involves constructing a topological graph of the data and calculating various topological indices, such as degree distribution and clustering coefficients, to form a topological feature vector. Next, based on the watermark information in a pre-defined watermark library and the topological feature vector, optimization or matching algorithms are used to determine the watermark embedding location and intensity distribution. This step requires consideration of the impact of watermark embedding on data quality and the robustness of the watermark. Finally, the determined watermark distribution information is applied to the initial data to complete the watermark embedding process.

[0084] S103. Based on the watermark distribution information, the initial data is embedded to obtain a dataset with embedded watermarks.

[0085] For example, a suitable watermark embedding algorithm is selected based on the watermark distribution information. If the watermark distribution information indicates embedding in the frequency domain, the initial data is first transformed in the frequency domain (e.g., discrete cosine transform) to obtain frequency domain coefficients. Then, watermark information is embedded in the selected frequency domain coefficients according to the intensity specified by the watermark distribution information. Finally, an inverse transform is performed to convert the watermarked frequency domain coefficients back to the spatial domain, obtaining the watermarked data. If the watermark distribution information indicates embedding in the spatial domain, the data is directly modified at the pixels or sampling points of the initial data according to the specified intensity to embed the watermark. After embedding, the quality of the watermarked data is evaluated to ensure that the watermark embedding does not significantly affect the data and that the watermark information can be reliably extracted.

[0086] This application provides a watermark embedding method for standard power grid data. It acquires initial data from power grid equipment, which represents the electrical data generated during equipment operation. Next, topological features are extracted from the initial data to obtain a topological feature vector characterizing its topological structure. Then, based on a pre-set watermark library and the topological feature vector, watermark distribution information is determined. The pre-set watermark library contains watermark information, while the watermark distribution information clarifies the embedding position and intensity distribution of the watermark information in the initial data, providing a basis for accurate watermark embedding. Finally, the initial data is embedded according to the watermark distribution information to obtain a watermark-embedded dataset. This method effectively solves the technical problems of unreasonable watermark embedding positions and intensities in existing power grid data watermark embedding methods, making them susceptible to attacks and tampering, thus compromising data security and integrity. It improves the accuracy and security of watermark embedding in standard power grid data, enhances the data's resistance to attacks during transmission and storage, and ensures the security and reliability of power grid data.

[0087] Figure 2 A flowchart illustrating a watermark embedding method for standard power grid data provided in this application embodiment. Figure 2 ,like Figure 2 As shown, in this embodiment... Figure 3 Based on the embodiments, a watermark embedding method for standard power grid data is described in detail. The method includes:

[0088] S201. Obtain standard data of power equipment in the power grid; wherein, standard data represents the relevant power data generated during the operation of power equipment in the power grid initially collected; perform data preprocessing on the standard data to obtain processed data; data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation; based on an encryption algorithm, perform data fingerprint encryption on the processed data to obtain initial data.

[0089] For example, various monitoring devices, sensors, and data acquisition systems in the power grid are used to collect relevant power data generated during the operation of power equipment in real time, such as voltage, current, power, equipment temperature, and operating status, forming an initial standard dataset. These devices are distributed across various key nodes and equipment in the power grid, and the collected data is transmitted to a data storage center or data processing platform via a network.

[0090] Data preprocessing is performed on standard data. Data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation.

[0091] Data cleaning and processing: The collected standard data is inspected to identify and remove noisy, erroneous, and duplicate data. For example, voltage and current values ​​that are significantly outside the normal operating range of the equipment are identified as noise and discarded; for duplicate records, only one valid record is retained.

[0092] Data standardization: Based on the characteristics of the data and the analysis requirements, select an appropriate standardization method, such as min-max standardization or Z-score standardization, to transform data with different dimensions into a unified standard range. For example, voltage data can be normalized to the [0,1] interval, and current data can be processed similarly to make data with different parameters comparable.

[0093] Data splitting and processing: Data is split according to business needs, for example, a year's continuous data is split into 12 independent data files by month.

[0094] Optionally, sensitive information in the original dataset, such as user electricity consumption data, can be protected using differential privacy techniques. A privacy budget ε is determined, typically between 0.1 and 1.0, depending on the data sensitivity. Then, random noise is generated according to a Laplace distribution and added to the sensitive data. For example, for a user's electricity consumption data at a specific moment, random noise following a Laplace distribution is added, making it impossible for attackers to accurately infer the user's actual electricity consumption from the published data.

[0095] Digital fingerprint extraction is performed on a privacy-protected original dataset using the national cryptographic algorithm SM3. The entire original dataset is used as input, and the SM3 algorithm is applied to generate a 256-bit hash value as the digital fingerprint. This digital fingerprint uniquely corresponds to the original dataset and can be used for subsequent data integrity verification.

[0096] Optionally, the original dataset and its digital fingerprint can be stored in a distributed storage blockchain network. First, a consortium-based distributed storage blockchain network is constructed, identifying key nodes within the power grid as network nodes and configuring corresponding network parameters and consensus mechanisms. Then, using an on-chain and off-chain storage model, the digital fingerprint is encapsulated into a blockchain transaction according to a specific data format, broadcast to the network for consensus verification, and stored on the blockchain after successful verification. Simultaneously, the original dataset is stored in a distributed storage system (such as IPFS), and the storage address (hash pointer) of the original data in the distributed storage system is recorded in the blockchain transaction. This pointer is used to associate the data fingerprint with the original data, facilitating subsequent data retrieval and verification.

[0097] S202. Perform data standardization on the initial data to obtain standardized data; perform similarity processing on the standardized data according to the preset similarity function to obtain the adjacency matrix; wherein, the adjacency matrix represents the similarity relationship between power equipment nodes in the standardized data.

[0098] The adjacency matrix is ​​processed to obtain the topological feature vector.

[0099] For example, the initial data is organized into a three-dimensional tensor D, with the meanings of the time dimension t, the device dimension e, and the parameter dimension p explicitly defined. For instance, suppose data on three parameters (p = 3) of five electrical devices (e = 5) in a power grid are collected over 24 hours (t = 24). Then, a 24×5×3 three-dimensional tensor D can be constructed to store this data.

[0100] For each parameter dimension in tensor D, calculate Q1 and Q3 for the parameter value of all devices at each time point. For example, for parameter 1, sort the parameter 1 values ​​of all devices at 24 time points, and find the values ​​at the 25th and 75th positions, which are Q1 and Q3.

[0101] The IQR for each parameter dimension is calculated using the formula IQR = Q3 - Q1.

[0102] For each data point x in tensor D, determine whether it satisfies x < Q1 - 1.5×IQR or x > Q3 +1.5×IQR. If it does, mark it as an outlier.

[0103] For detected outliers, find the local region in the device and time dimensions where the outlier is located (e.g., a certain number of devices around the device where the outlier is located; a certain number of time points before and after the time point where the outlier is located), calculate the median of the local region, and replace the outlier with the median.

[0104] The min-max normalization method is used to find the minimum and maximum values ​​for each parameter dimension in tensor D. Then, the features of each dimension are normalized to the interval [-1, 1] using the formula (normalized value = (original value - minimum value) / (maximum value - minimum value) × 2 - 1). For example, for a data point x of parameter 2, the minimum and maximum values ​​of parameter 2 across all devices and time points are found, and the normalized value is calculated by substituting them into the formula.

[0105] A radial basis function (RBF) is used to define the similarity measure between data points, with the formula (similarity measure = exp(-(Euclidean distance²) / (2 × kernel width parameter²))), where the Euclidean distance is the distance between data points, and the kernel width parameter is usually set as the median of the distances between data point pairs in the dataset. First, the Euclidean distance between all data points is calculated, and then the similarity measure is calculated based on the kernel width parameter.

[0106] An adjacency matrix A is constructed based on a similarity metric function, with the size of the matrix being the number of devices multiplied by the number of devices. For each element A[i][j] in the matrix, it represents the similarity metric between device i and device j. When the similarity metric exceeds a threshold τ (usually 0.5), A[i][j] is set to 1, indicating that there is a topological connection between the two points; otherwise, it is set to 0.

[0107] The adjacency matrix is ​​processed to obtain the topological feature vector.

[0108] S203. Based on the topological invariant algorithm, feature extraction processing is performed on the adjacency matrix to obtain topological invariants; where the topological invariants represent the topological features of the adjacency matrix; the topological invariants are digitized and encoded to obtain the topological feature vector.

[0109] For example, the topological invariant algorithm is a mathematical method for analyzing the topological properties of networks or graphs. An adjacency matrix is ​​a matrix describing the connectivity between nodes in a graph, where each element indicates whether an edge exists between nodes. Topological invariants are numerical values ​​or quantities that reflect the essential characteristics of a graph structure and do not change with isomorphic transformations of the graph, such as connectivity and cycle number. Data encoding transforms topological invariants into computer-processable numerical vectors for subsequent analysis and applications. The topological feature vector, obtained after encoding, is a numerical vector that comprehensively characterizes the topological features of the adjacency matrix.

[0110] An adjacency matrix is ​​generated based on the similarity between data points (such as Euclidean distance), and a threshold τ (e.g., 0.5) is set. If the value of a matrix element exceeds τ, the corresponding two points are considered to be topologically connected; otherwise, they are considered disconnected. A simple complex is constructed based on the thresholded adjacency matrix, transforming the discrete point set into a topological structure containing high-dimensional units such as edges and faces.

[0111] Using the distance function f(x) as the filtering function, a nested complex sequence is generated. , where e is the distance threshold. As e increases, the complex gradually contains more high-dimensional units (e.g., from isolated points to connected graphs, and then to ring structures).

[0112] For each complex Calculate the p-Vibert number (e.g.) Count the number of connected components. The number of rings is counted to obtain the Betty number sequence. By tracking the life cycle of features (birth time e_birth and death time e_death), a persistent graph is constructed, with e as the horizontal axis and the Betty number as the vertical axis. The point (e_birth, e_death) represents the duration of a topological feature.

[0113] Select key points with longer lifecycles in the persistent graph (e.g., e_death - e_birth > δ), and encode their coordinates or statistics (e.g., persistence, centroid) into vectors to form the final topological feature vector.

[0114] S204. Determine watermark information from the preset watermark library and divide the watermark information into blocks to obtain watermark block data; perform perturbation processing on the topological feature vector according to the preset perturbation function to obtain the degree of change of the topological feature vector; determine the topological sensitive vector based on the degree of change; wherein, the topological sensitive vector represents the data sensitivity of the topological feature vector; determine the watermark distribution information based on the watermark block data, the topological sensitive vector, and the topological feature vector.

[0115] For example, select watermark information to be embedded (such as text ABC or binary sequence 01100010) from a preset watermark library.

[0116] The watermark information is divided into blocks of fixed size (or different sizes) to obtain watermark block data.

[0117] Choose the perturbation method based on the data type (such as adding Gaussian noise to image pixel values ​​or scaling graph node features), and set the perturbation size δ (usually 0.1% of the original feature value range).

[0118] The pre-defined perturbation function is a specific mathematical function that can alter the input topological feature vector to a certain extent. The topological feature vector is a vector used to describe the topological structural characteristics of the target carrier. By processing the topological feature vector through the perturbation function, we can obtain how the vector changes after being perturbed. The degree of this change can be measured by calculating the difference between the vectors before and after the perturbation, such as calculating the magnitude of the vector difference. The purpose of this step is to analyze the sensitivity of the topological feature vector to perturbations, providing a basis for subsequently determining the topologically sensitive vector.

[0119] After obtaining the degree of change in the topological feature vector, the topology-sensitive vector is determined based on this degree of change. For example, if a topological feature vector changes significantly after being perturbed, it indicates that it is more sensitive to perturbations, and the corresponding value in the topology-sensitive vector will be larger; conversely, if the degree of change is small, the corresponding value will be smaller. The topology-sensitive vector is essentially a quantitative representation of the sensitivity of the topological feature vector. It helps us more accurately understand the changing characteristics of the topological feature vector under different conditions, thereby better protecting important topological information during watermark embedding and improving the robustness of the watermark.

[0120] The topological feature vector is weighted based on the topological sensitivity vector, for example, by multiplying the corresponding elements of the topological sensitivity vector and the topological feature vector to obtain a weighted topological feature vector. This weighted topological feature vector highlights regions sensitive to disturbances, which require more careful handling during watermark embedding. Then, combined with the watermark block data, the watermark block is embedded into the region corresponding to the weighted topological feature vector according to certain rules. For example, the embedding strength of the watermark block can be determined based on the magnitude of the elements in the weighted topological feature vector; regions with larger element values ​​can have a stronger embedding strength, while regions with smaller element values ​​can have a weaker embedding strength. This method of determining the watermark distribution information allows the watermark to better adapt to the topology of the target carrier, improving its concealment and robustness, and enabling it to more effectively resist various attacks and interferences without affecting the normal use of the target carrier.

[0121] In one example, the watermark intensity vector is determined by comparing the parameters in the topology-sensitive vector with a preset threshold. The watermark intensity vector represents the watermark intensity information at different positions in the topology-sensitive vector. The watermark embedding position is determined based on the topological complexity in the topological feature vector. The watermark distribution information is determined based on the watermark block data, the watermark intensity information, and the watermark embedding position.

[0122] For example, a topology sensitivity vector is a vector containing multiple parameters that reflect the sensitivity of different locations in the topology of the target carrier.

[0123] By comparing each parameter in the topology sensitivity vector with a preset threshold, a relatively high watermark strength value is set at that location if the parameter is greater than the threshold, and a relatively low watermark strength value is set if the parameter is less than the threshold. This results in a watermark strength vector that can assign different watermark embedding intensities to different sensitive regions of the target carrier's topology, making the watermark embedding more reasonable. This ensures the robustness of the watermark while minimizing the impact on the characteristics of the target carrier itself.

[0124] Topological feature vectors describe the topological structure of a target carrier. Topological complexity is an indicator of the complexity of the topological structure, which can be calculated in various ways, such as by calculating the number of nodes and edges, connectivity, etc. Generally, regions with high topological complexity have richer and more diverse structures, providing more information hiding space. When embedding watermarks, selecting regions with high topological complexity as embedding locations can leverage the complex structure of these regions to better hide watermark information, improving the watermark's concealment and making it less likely to be detected or destroyed. Simultaneously, complex topological structures can also enhance the watermark's resistance to attacks to some extent.

[0125] The watermark block data consists of small segments of the actual watermark information to be embedded. Watermark strength information is provided by the watermark strength vector, specifying the intensity of the watermark embedding at each location. The watermark embedding location determines the specific coordinates or area of ​​the watermark on the target carrier. The watermark block data is embedded into the defined locations according to the watermark strength information. For example, in locations with high watermark strength, the embedding depth or number of repetitions can be appropriately increased; in locations with low watermark strength, the embedding is relatively gentler. This method allows the watermark to form a reasonable distribution on the target carrier, ensuring effective embedding while maximizing adaptation to the characteristics of the target carrier, thus improving the overall performance of the watermark.

[0126] S205. Perform time-domain processing and wavelet transform processing on the initial data in sequence to obtain frequency-domain data; perform embedding processing on the frequency-domain data according to the watermark distribution information to obtain a dataset with embedded watermark.

[0127] For example, the initial data characterizes the relevant power data generated during the operation of the power equipment in the power grid. These data cover a variety of types, such as voltage data, which reflects the magnitude of the potential difference between the two ends of the power equipment and is one of the important indicators for measuring power quality; current data, which reflects the directional movement of charge in the conductor and is closely related to the load and power of the equipment; and power data, including active power and reactive power, which are directly related to the work capacity and energy utilization efficiency of the power equipment.

[0128] Time-domain processing involves the analysis and manipulation of these power data along the time dimension. For example, for continuously collected voltage data, time-domain processing may include data sampling, converting continuous analog signals into discrete digital signals for computer processing; and data filtering, removing high-frequency noise and interference from the data to make it smoother. For instance, a moving average filtering algorithm can be used to replace the original data by calculating the average value of the data within a certain time window, thereby reducing the impact of random noise.

[0129] Wavelet transform can convert signals from the time domain to the frequency domain. Compared with the traditional Fourier transform, wavelet transform has the characteristic of multi-resolution analysis, allowing signal analysis at different scales. For power data, wavelet transform processing can decompose the data into sub-bands with different frequency components. These sub-bands contain information about the power data in different frequency ranges. For example, low-frequency sub-bands may reflect the basic trends and long-term changes in power data, while high-frequency sub-bands may contain abrupt changes and transient information. This transformation from the time domain to the frequency domain can better reveal the intrinsic characteristics of power data, providing more favorable conditions for subsequent watermark embedding, because watermark embedding in the frequency domain often achieves better concealment and robustness, especially in the face of various interferences and attacks that may exist in the power system.

[0130] The watermark distribution information, determined in previous steps, details the watermark's specific location and strength within the frequency domain data. In power data scenarios, watermarks are used to identify data origin, copyright information, or for data integrity verification. Embedding is the process of integrating the watermark information into the frequency domain data according to the watermark distribution information. For example, the watermark distribution information might specify embedding the watermark in certain frequency sub-bands or locations within the frequency domain data, and specify the embedding strength. During embedding, corresponding operations, such as addition, multiplication, or more complex modulation operations, are performed on the watermark data and frequency domain data according to these requirements, organically combining the watermark information with the frequency domain data. Through this embedding process, the resulting dataset contains the watermark information. Because the embedding occurs in the frequency domain, the watermark is better adapted to the characteristics of power data, exhibiting stronger resistance to various signal processing operations in power systems (such as data compression and filtering) and potential malicious attacks (such as data tampering). This ensures the effectiveness and reliability of the watermark, guaranteeing the security and integrity of the power data.

[0131] This application provides a method for embedding watermarks into standard power grid data. The method involves acquiring standard data from power equipment, initially collected during equipment operation. Preprocessing of the standard data includes data cleaning, standardization, and segmentation to remove noise, unify the format, and rationally divide the data. Data fingerprinting is then performed using an encryption algorithm to generate initial data, ensuring data security. The initial data is then standardized again, and an adjacency matrix representing the similarity relationships between power equipment nodes is obtained using a preset similarity function. Topological invariants of the adjacency matrix are extracted using a topological invariant algorithm and digitized to obtain topological feature vectors. Watermark information is selected from a preset watermark library and divided into blocks. The topological feature vectors are perturbed using a preset perturbation function to determine the topological sensitivity vector representing the data sensitivity, which is then combined with the watermark block data to determine the watermark distribution information. The initial data is then processed sequentially in the time domain and subjected to wavelet transform to obtain frequency domain data. Finally, the watermark is embedded in the frequency domain data based on the watermark distribution information. These methods effectively solve the problems of inaccurate watermark embedding and susceptibility to damage in existing methods, which threaten the security and integrity of power grid data. They achieve the beneficial effects of improving the accuracy and robustness of watermark embedding and ensuring the safe and reliable transmission and storage of power grid standard data.

[0132] This application provides a method for embedding watermarks into standard power grid data. The method further includes quality inspection of the watermarked data; the quality inspection mainly includes calculating topology preservation rate, data distortion rate, and watermark undetectability.

[0133] Topology retention rate refers to the degree to which the topology structure is preserved before and after watermarking:

[0134]

[0135] Wherein, TRP is the topology preservation rate, which represents the degree to which the topology structure is preserved before and after watermark embedding. The value range is [0,1]. The larger the value, the higher the degree of preservation. The number of matched topology features refers to the number of topology features that remain unchanged before and after watermarking. The total number of topology features refers to the total number of topology features in the original data.

[0136] Data distortion rate: The degree of data distortion caused by watermarking.

[0137]

[0138] Where DDR is the data distortion rate, representing the degree of data distortion caused by watermark embedding, with a value range of [0,1]. The smaller the value, the lower the degree of distortion. D is the original data. For the data after embedding the watermark, The norm difference between two numbers is usually expressed using Euclidean distance or the Frobenius norm. Represents the norm of the original data.

[0139] Watermark Undetectability (WUD): The difficulty of detecting a watermark.

[0140]

[0141] Wherein, WUD is the watermark undetectability, which represents the difficulty for an unauthorized detector to discover the watermark. The value range is [0,1]. The larger the value, the more difficult the watermark is to be detected. The number of successful watermark detections refers to the number of attempts to successfully detect the watermark when the watermark parameters are unknown. The total number of detections refers to the total number of detection attempts.

[0142] Typically, TPR > 0.95, DDR < 0.03, and WUD > 0.9 are required. When the metrics exceed the preset thresholds, an adjustment mechanism is triggered to ensure that watermark embedding does not affect data quality and subsequent model training.

[0143] This application provides a method for embedding watermarks into standard power grid data. The method also includes verifying the data integrity of the watermarked standard power grid data. First, based on the feature extraction step in the preceding steps, the system reconstructs the data topology, generates a simple complex, calculates persistent homology features, compares them with the original feature vector, and recovers the topological marker points. Specifically, this unit first reconstructs the topology of the received dataset, generating a simple complex K'; then it calculates persistent homology features, generating a feature vector T'; next, it compares it with the original feature vector T to identify matching topological features; finally, it generates a sequence of topological marker points to guide the watermark extraction process.

[0144] Then, the system adopts a hierarchical adaptive strategy to extract watermark information from the core layer, permission layer, and verification layer sequentially from the topologically stable region to reconstruct the complete watermark. The watermark extraction process is divided into three stages: the first stage extracts the core layer watermark information from the topologically stable region; the second stage extracts the permission layer watermark based on the core information; the third stage combines the information from the first two layers to extract the verification layer watermark; finally, the three layers of information are integrated to reconstruct the complete watermark.

[0145] Secondly, the system identifies the types of interference the data may experience and applies corresponding recovery strategies, using redundant information to repair the damaged watermark. First, common interference types are defined, including noise, compression, and resampling; then, compensation mechanisms are designed for each type of interference; next, an interference detector identifies the interference the data may experience; finally, based on the detection results, an appropriate recovery strategy is selected to repair the damaged watermark.

[0146] Finally, the system compares the extracted watermark with the original record, identifies inconsistencies, analyzes topological changes, and locates potential tampering areas. Specifically, this unit first compares the extracted watermark W' with the original watermark W stored on the blockchain; then it identifies inconsistencies, records their locations and degrees of change; next, it analyzes the topological changes corresponding to these locations; finally, based on the changes, it locates potential tampering areas, assesses the scope of tampering impact, and generates a chain of evidence to support subsequent accountability.

[0147] This application provides a watermark embedding method for standard power grid data. The method also includes data sharing of the standard power grid data. The system responds to a data sharing request; the data sharing request includes requester information, a description of the requested data, the purpose of use, and the time limit for use. The system performs format verification and preliminary legality checks on the request to ensure that it meets the basic requirements.

[0148] Furthermore, the system verifies the applicant's permissions based on the permission information embedded in the watermark, combined with access control policies. Specifically, the system first extracts permission information from the data watermark; then queries the access control policy library to obtain the corresponding permission rules; next, it performs permission matching verification based on the applicant's identity and the content of the application; finally, it generates a permission verification report to guide the subsequent authorization process.

[0149] The system assigns access permissions to verified applicants and generates authorization credentials. Access permissions are divided into four levels: read, analyze, train, and redistribute, with different levels assigned based on the applicant's legitimate needs and level of trust. The authorization credentials are generated using digital signature technology and contain information such as the permission content, validity period, and usage conditions.

[0150] The system records data access operations and updates the usage records in the watermark, supporting subsequent auditing. The system monitors data access and usage in real time, recording information such as the accessor, access time, access content, and operation type. These records are updated in the data watermark and stored on the blockchain, forming a complete access tracking chain for easy auditing and accountability.

[0151] Figure 1 A schematic diagram of the structure of a watermark embedding device for standard power grid data provided in this application embodiment. Figure 3 ,like Figure 4 As shown, the watermark embedding device 30 for standard power grid data provided in this embodiment includes:

[0152] The acquisition module 301 is used to acquire initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid;

[0153] The determination module 302 is used to extract topological features from the initial data to obtain a topological feature vector; and to determine the watermark distribution information based on the preset watermark library and the topological feature vector. The topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data.

[0154] The processing module 303 is used to perform embedding processing on the initial data according to the watermark distribution information to obtain a dataset with embedded watermarks.

[0155] This embodiment provides a watermark embedding device for standard power grid data, which can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0156] Figure 2 A schematic diagram of the structure of a watermark embedding device for standard power grid data provided in this application embodiment. Figure 4 ,like Figure 5 As shown, the watermark embedding device 40 for standard power grid data provided in this embodiment includes:

[0157] The acquisition module 401 is used to acquire initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid;

[0158] The determination module 402 is used to extract topological features from the initial data to obtain a topological feature vector; and to determine the watermark distribution information based on the preset watermark library and the topological feature vector; wherein, the topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data.

[0159] The processing module 403 is used to perform embedding processing on the initial data according to the watermark distribution information to obtain a dataset with embedded watermarks.

[0160] In one possible implementation, the determining module 402 includes:

[0161] The standardization processing module 4021 is used to perform data standardization processing on the initial data to obtain standardized data;

[0162] The similarity processing module 4022 is used to perform similarity processing on the standardized data according to a preset similarity function to obtain an adjacency matrix; wherein, the adjacency matrix represents the similarity relationship between power equipment nodes in the standardized data;

[0163] The feature processing module 4023 is used to process the adjacency matrix to obtain the topological feature vector.

[0164] In one possible implementation, the feature processing module 4023 includes:

[0165] Based on the topological invariant algorithm, feature extraction is performed on the adjacency matrix to obtain topological invariants; where the topological invariants represent the topological features of the adjacency matrix.

[0166] The topological invariants are digitized and encoded to obtain the topological feature vector.

[0167] In one possible implementation, the determining module 402 includes:

[0168] The block processing module 4024 is used to determine the watermark information from the preset watermark library and to process the watermark information into blocks to obtain watermark block data.

[0169] The perturbation processing module 4025 is used to perturb the topological feature vector according to a preset perturbation function to obtain the degree of change of the topological feature vector; and to determine the topological sensitive vector based on the degree of change; wherein the topological sensitive vector characterizes the data sensitivity of the topological feature vector.

[0170] The confirmation submodule 4026 is used to determine the watermark distribution information based on the watermark block data, the topology sensitivity vector, and the topology feature vector.

[0171] In one possible implementation, the confirmation submodule 4026 includes:

[0172] The watermark intensity vector is determined by comparing the parameters in the topology-sensitive vector with a preset threshold; the watermark intensity vector represents the watermark intensity information at different positions in the topology-sensitive vector.

[0173] The watermark embedding position is determined based on the topological complexity in the topological feature vector;

[0174] Based on the watermark block data, watermark intensity information, and watermark embedding location, the watermark distribution information is determined.

[0175] In one possible implementation, the processing module 403 includes:

[0176] The initial data is processed sequentially in the time domain and then by wavelet transform to obtain the frequency domain data.

[0177] Based on the watermark distribution information, the frequency domain data is embedded to obtain a dataset with embedded watermarks.

[0178] In one possible implementation, the acquisition module 401 includes:

[0179] Acquire standard data of power equipment in the power grid; wherein, standard data represents the relevant power data generated during the operation of power equipment in the power grid initially collected;

[0180] Standard data is preprocessed to obtain processed data; data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation.

[0181] Based on the encryption algorithm, the processed data is encrypted using a data fingerprint to obtain the initial data.

[0182] This embodiment provides a watermark embedding device for standard power grid data, which can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0183] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. ​ As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0184] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0185] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0186] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0187] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0188] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0189] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0190] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0191] The aforementioned readable 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. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0192] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0193] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0194] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0195] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0196] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0197] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0198] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for embedding watermarks into standard power grid data, characterized in that, include: Acquire initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid; The initial data is subjected to topological feature extraction to obtain a topological feature vector; watermark distribution information is determined based on a preset watermark library and the topological feature vector; wherein, the topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data. Based on the watermark distribution information, the initial data is embedded to obtain a dataset with embedded watermarks.

2. The method according to claim 1, characterized in that, Topological feature extraction is performed on the initial data to obtain a topological feature vector, including: The initial data is subjected to data standardization processing to obtain standardized data; The standardized data is processed according to a preset similarity function to obtain an adjacency matrix; wherein the adjacency matrix represents the similarity relationship between power equipment nodes in the standardized data; The adjacency matrix is ​​processed to obtain the topological feature vector.

3. The method according to claim 2, characterized in that, The adjacency matrix is ​​processed to obtain the topological feature vector, including: According to the topological invariant algorithm, the adjacency matrix is ​​subjected to feature extraction processing to obtain topological invariants; wherein, the topological invariants represent the topological features of the adjacency matrix; The topological invariants are digitized and encoded to obtain the topological feature vector.

4. The method according to claim 1, characterized in that, Based on the preset watermark library and the topological feature vector, the watermark distribution information is determined, including: Watermark information is determined from the preset watermark library, and the watermark information is divided into blocks to obtain watermark block data. The topological feature vector is perturbed according to a preset perturbation function to obtain the degree of change of the topological feature vector; based on the degree of change, a topological sensitivity vector is determined; wherein, the topological sensitivity vector characterizes the data sensitivity of the topological feature vector; The watermark distribution information is determined based on the watermark block data, the topology sensitivity vector, and the topology feature vector.

5. The method according to claim 4, characterized in that, The watermark distribution information is determined based on the watermark block data, the topology sensitivity vector, and the topology feature vector, including: The watermark intensity vector is determined by comparing the parameters in the topology-sensitive vector with a preset threshold; wherein the watermark intensity vector represents the watermark intensity information at different positions in the topology-sensitive vector. The watermark embedding position is determined based on the topological complexity in the topological feature vector. The watermark distribution information is determined based on the watermark block data, the watermark intensity information, and the watermark embedding position.

6. The method according to claim 1, characterized in that, Based on the watermark distribution information, the initial data is embedded to obtain a dataset with embedded watermarks, including: The initial data is processed sequentially in the time domain and then in the wavelet transform domain to obtain frequency domain data. Based on the watermark distribution information, the frequency domain data is embedded to obtain a dataset with embedded watermarks.

7. The method according to any one of claims 1-6, characterized in that, The acquisition of initial data from power equipment in the power grid includes: Acquire standard data of power equipment in the power grid; wherein, the standard data represents the relevant power data generated during the operation of the power equipment in the power grid initially collected; The standard data is preprocessed to obtain processed data; the data preprocessing includes one or more of the following: data cleaning, data standardization, and data segmentation. Based on the encryption algorithm, the processed data is subjected to data fingerprint encryption to obtain the initial data.

8. A watermark embedding device for standard data of a power grid, characterized in that, include: An acquisition module is used to acquire initial data of the power equipment in the power grid; wherein, the initial data represents the relevant power data generated during the operation of the power equipment in the power grid; The determination module is used to extract topological features from the initial data to obtain a topological feature vector; and to determine watermark distribution information based on a preset watermark library and the topological feature vector; wherein, the topological feature vector represents the topological structure features of the initial data; the preset watermark library includes watermark information; and the watermark distribution information represents the embedding position and intensity distribution of the watermark information in the initial data. The processing module is used to perform embedding processing on the initial data according to the watermark distribution information to obtain a dataset with embedded watermarks.

9. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images