A self-organizing network-based photovoltaic data management method and related device

Abstract

The application discloses a photovoltaic data management method based on a self-organizing network and related devices, and relates to the technical field of data processing. The method comprises the following steps: establishing a connection between each wireless communication node based on the performance interaction of a communication link and the connectivity between adjacent wireless communication nodes to form a self-organizing network; determining a target transmission path based on the time delay overhead and the signal-to-noise ratio of each communication link by using a hybrid algorithm to transmit photovoltaic data; classifying based on the feature matrix of the photovoltaic data to obtain photovoltaic data of each category; performing security requirement analysis on the photovoltaic data of each category to obtain a target requirement security level; performing service requirement degree analysis on the photovoltaic data of each category to evaluate the cold and hot degree thereof; and determining a storage strategy based on the cold and hot degree and the requirement security level of the photovoltaic data of each category in combination with storage node performance analysis to store the data into corresponding storage nodes. The application can achieve a more ideal effect for the storage management of photovoltaic data.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a photovoltaic data management method and related apparatus based on self-organizing networks. Background Technology

[0002] Ad hoc networks (ANRs) are networks composed of multiple nodes with wireless transceiver capabilities. They do not rely on pre-existing infrastructure and feature strong data transmission performance, rapid deployment, and high resilience. Many photovoltaic (PV) product companies are gradually deploying ANRs to manage PV data. However, current ANR methods often neglect the connectivity between wireless communication nodes and the performance interaction between links, resulting in insufficient organizational efficiency and search performance of the final ANR. Furthermore, current methods typically determine the required transmission path by calculating the packet loss rate of each link in the ANR, but this method cannot guarantee high stability and low latency for data transmission. After transmitting PV data to the server via an ANR, determining the storage strategy is also a crucial aspect of data management. Current storage strategies primarily consider the matching of data volume and storage nodes, neglecting the impact of data temperature and security requirements. This leads to unreliable storage strategies, susceptibility to data leaks, and impacts the response speed and overall performance of storage nodes, failing to meet the high requirements of PV data management. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a photovoltaic data management method and related device based on self-organizing network, which can improve the reliability of storage strategy analysis and achieve a more ideal effect in photovoltaic data storage management.

[0004] To address the aforementioned technical problems, this invention provides a photovoltaic data management method based on a self-organizing network, applicable to client devices, cloud servers, and a self-organizing network composed of several wireless communication nodes; the method includes:

[0005] A self-organizing network is formed by establishing connections between wireless communication nodes based on the performance interaction of the communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes.

[0006] Based on the latency overhead and signal-to-noise ratio of each communication link in the self-organizing network, a hybrid algorithm is used to determine the target transmission path, and the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path.

[0007] The cloud server classifies photovoltaic data based on the feature matrix of the photovoltaic data to obtain photovoltaic data of various categories;

[0008] Conduct security requirements analysis on various types of photovoltaic data to obtain the target security level for each type of photovoltaic data.

[0009] Analyze the business demand for each category of photovoltaic data to obtain the target business demand, and assess the popularity of each category of photovoltaic data based on the target business demand.

[0010] Based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with the performance analysis of storage nodes, a storage strategy is determined, and photovoltaic data of various categories are stored in the corresponding storage nodes based on the storage strategy.

[0011] Optionally, the step of establishing connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes to form an ad hoc network includes:

[0012] The signal strength between adjacent wireless communication nodes is calculated based on the address of each wireless communication node, and the connectivity between adjacent wireless communication nodes is determined based on the signal strength.

[0013] The target distance between adjacent wireless communication nodes is calculated based on the connectivity between them, and the neighbor list of each wireless communication node is updated based on the target distance to obtain an updated neighbor list.

[0014] The communication link exchanges performance data with nearby wireless communication nodes and performs load balancing among the wireless communication nodes based on the exchanged performance data.

[0015] A self-organizing network is formed by connecting various wireless communication nodes based on updating the neighbor list and load balancing among wireless communication nodes.

[0016] Optionally, the step of determining the target transmission path using a hybrid algorithm based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network, and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path, includes:

[0017] The link transmission rate of each communication link is determined based on the media access control address, and the latency overhead of each communication link is analyzed based on the link transmission rate and the packet loss rate.

[0018] Determine the signal-to-noise ratio of each communication link, calculate the packet reception rate of each communication link, and calculate the link quality based on the received signal power of each communication link;

[0019] The target transmission path is determined using ant colony algorithm and simulated annealing algorithm based on the delay overhead, signal-to-noise ratio, packet reception rate and link quality of each communication link.

[0020] In a self-organizing network, the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path.

[0021] Optionally, the photovoltaic data is classified based on the feature matrix of the photovoltaic data to obtain photovoltaic data of various categories, including:

[0022] A feature matrix is ​​constructed based on the node feature vectors of photovoltaic data, an adjacency matrix is ​​generated based on the feature matrix, and the target weights corresponding to the feature matrix are determined.

[0023] An objective equation is constructed based on feature fitting constraints and Laplacian regularization constraints. A graph convolutional neural network is then constructed based on the objective equation and a preset filter. The graph convolutional neural network is trained to obtain a classification model.

[0024] The target label of photovoltaic data is determined by a classification model based on the adjacency matrix, feature matrix and corresponding target weights, and the photovoltaic data is classified based on the target label to obtain photovoltaic data of each category.

[0025] Optionally, the step of performing security requirement analysis on various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data includes:

[0026] Perform information composition analysis on various types of photovoltaic data to obtain information composition parameters;

[0027] Based on the judgment matrix, the security impact is quantified using the information composition parameters to obtain information security parameters;

[0028] A hierarchical feature set and classification interval are established based on the feature set of sensitive data types, and a security hierarchical model is constructed based on the hierarchical feature set and classification interval.

[0029] Based on the security classification model, the initial demand security level is analyzed for each category of photovoltaic data to obtain the initial demand security level.

[0030] Based on the aforementioned information security parameters and initial security requirements, the target security requirements level corresponding to each category of photovoltaic data is determined.

[0031] Optionally, the step of analyzing the business demand level of various categories of photovoltaic data to obtain the target business demand level, and assessing the popularity of various categories of photovoltaic data based on the target business demand level, includes:

[0032] Establish a development relationship between each category of photovoltaic data and its corresponding target business based on access information corresponding to several categories of photovoltaic data;

[0033] Based on the aforementioned development context, business demand analysis is performed on various categories of photovoltaic data to obtain the target business demand level.

[0034] Based on the target business demand level, combined with the hot and cold degree assessment elements and data temperature analysis model, the hot and cold degree corresponding to various types of photovoltaic data is assessed.

[0035] Optionally, the step of determining a storage strategy based on the hotness / coldness of photovoltaic data for each category and the target security level, combined with storage node performance analysis, and storing photovoltaic data for each category into the corresponding storage node based on the storage strategy, includes:

[0036] Partial regression coefficients are calculated using the least squares method based on log information from each storage node in the cloud storage center of the cloud server.

[0037] The node performance values ​​of each storage node are analyzed based on the partial regression coefficients.

[0038] An initial strategy is determined based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with the node performance values ​​of each storage node. The initial strategy is then tested, and the initial strategy is adjusted based on the test results to determine the storage strategy.

[0039] Based on the storage strategy, photovoltaic data of various categories are labeled with volume identifiers, and the photovoltaic data of various categories are stored in the corresponding storage nodes based on the volume identifiers.

[0040] In addition, the present invention also provides a photovoltaic data management device based on a self-organizing network, applicable to client devices, cloud servers, and a self-organizing network composed of several wireless communication nodes; the device includes:

[0041] Ad hoc network forming module: used to establish connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thus forming an ad hoc network;

[0042] Data transmission module: used to determine the target transmission path based on the delay overhead and signal-to-noise ratio of each communication link in the self-organizing network using a hybrid algorithm, and to transmit the photovoltaic data recorded by the client device to the cloud server based on the target transmission path;

[0043] Data classification module: Used by the cloud server to classify photovoltaic data based on the feature matrix of photovoltaic data to obtain photovoltaic data of various categories;

[0044] Security Requirements Analysis Module: Used to perform security requirements analysis on various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data;

[0045] Hot and cold assessment module: used to analyze the business demand level of various types of photovoltaic data, obtain the target business demand level, and assess the hot and cold status of various types of photovoltaic data based on the target business demand level;

[0046] Storage module: Used to determine storage strategies based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with storage node performance analysis, and to store photovoltaic data of various categories into the corresponding storage nodes based on the storage strategies.

[0047] In addition, the present invention also provides a photovoltaic data management system based on a self-organizing network. The system includes a client device, a cloud server, and a self-organizing network composed of several wireless communication nodes. The system is configured to execute the above-described photovoltaic data management method based on a self-organizing network.

[0048] In addition, the present invention also provides a computer-readable storage medium that stores computer instructions, which, when executed on an electronic device, cause the electronic device to perform the above-described photovoltaic data management method based on a self-organizing network.

[0049] In this embodiment of the invention, connections between wireless communication nodes are established based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby effectively improving the organizational efficiency and search performance of the formed ad hoc network. A hybrid algorithm is used to determine the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network to transmit photovoltaic data recorded by the client device to the cloud server, improving data transmission stability and reducing latency. A classification model is used to classify the photovoltaic data based on the adjacency matrix, feature matrix, and corresponding target weights, improving the accuracy of photovoltaic data classification. The target demand security level for each category of photovoltaic data is determined based on the information security parameters obtained from the security impact quantification of photovoltaic data and the initial demand security level obtained from the security grading model, improving the reliability of data demand security level analysis. Storage strategies are determined based on the hotness / coldness and demand security level of each category of photovoltaic data, combined with storage node performance analysis, enabling a more comprehensive and specific analysis of storage strategies. Storing photovoltaic data using this strategy avoids data leakage and prevents poor node response speed and overall performance caused by inappropriate storage node selection. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating the photovoltaic data management method based on a self-organizing network in an embodiment of the present invention.

[0052] Figure 2 This is a flowchart illustrating a photovoltaic data management method based on a self-organizing network according to another embodiment of the present invention;

[0053] Figure 3 This is a schematic diagram of the structural composition of a photovoltaic data management system based on a self-organizing network in an embodiment of the present invention;

[0054] Figure 4 This is a schematic diagram of the structural composition of a photovoltaic data management device based on a self-organizing network in an embodiment of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1

[0057] Please see Figure 1 , Figure 1 This is a flowchart illustrating a photovoltaic data management method based on a self-organizing network according to an embodiment of the present invention. The method is applied to a client device, a cloud server, and a self-organizing network composed of several wireless communication nodes. The method includes:

[0058] S11: Based on the performance interaction of the communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, a connection is established between each wireless communication node to form an ad hoc network;

[0059] In a specific implementation of this invention, the step of establishing connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes to form an ad hoc network includes: calculating the signal strength between adjacent wireless communication nodes based on the addresses of each wireless communication node, and determining the connectivity between adjacent wireless communication nodes based on the signal strength; calculating the target distance between adjacent wireless communication nodes based on the connectivity between adjacent wireless communication nodes, and updating the neighbor list of each wireless communication node based on the target distance to obtain an updated neighbor list; exchanging performance data between the communication links and neighboring wireless communication nodes, and performing load balancing between wireless communication nodes based on the exchanged performance data; and establishing connections between each wireless communication node based on the updated neighbor list and the load balancing between wireless communication nodes to form an ad hoc network.

[0060] Specifically, the signal strength between adjacent wireless communication nodes is calculated based on the addresses of each node. The signal strength between adjacent nodes is obtained by combining the coordinates and addresses of each node with their broadcast detection signals. The connectivity between adjacent wireless communication nodes is then determined based on this signal strength; that is, communication connections between adjacent wireless communication nodes are judged based on the signal strength. If the signal strength is greater than or equal to a preset strength threshold, the adjacent wireless communication nodes can communicate; if the signal strength is less than the preset strength threshold, they cannot communicate. The target distance between adjacent wireless communication nodes is calculated based on the connectivity, i.e., the distance between communicably connected adjacent wireless communication nodes. This can be calculated using the corresponding coordinates of the wireless communication nodes. The neighbor list of each wireless communication node is updated based on the target distance. Each wireless communication node selects only the few nearest wireless communication nodes as its neighbors and adds these selected neighbor nodes to the neighbor list to update the neighbor list, maintaining the sparsity and efficiency of the network. The communication link exchanges performance data with neighboring wireless communication nodes. The communication link is formed by updating the neighbor list of wireless communication nodes and neighboring communication nodes. This performance data includes available service bandwidth, available connections, and available file transfer protocol bandwidth. Load balancing among the wireless communication nodes is performed based on the exchanged performance data. The system determines whether the service should be provided by the communication node on this link or whether the resources required for the service should be distributed among its neighboring wireless communication nodes. Connections between wireless communication nodes are established based on the updated neighbor list and the load balancing among them. A dynamic resource adjustment strategy is formed based on the load balancing and deployed in the communication links formed by the connections of wireless communication nodes, creating an ad hoc network. This ad hoc network has higher organizational efficiency and search performance, can meet the needs of larger-scale and more demanding network processing, achieves load balancing among nodes in the ad hoc network, improves the processing performance of the ad hoc network, and reduces the risk of failure.

[0061] S12: Based on the latency overhead and signal-to-noise ratio of each communication link in the self-organizing network, a hybrid algorithm is used to determine the target transmission path, and the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path;

[0062] In the specific implementation of this invention, the step of determining the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network using a hybrid algorithm, and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path, includes: determining the link transmission rate of each communication link based on the media access control address, and analyzing the latency overhead of each communication link based on the link transmission rate and packet loss rate; determining the signal-to-noise ratio of each communication link, calculating the corresponding packet reception rate of each communication link, and calculating the link quality based on the received signal power of each communication link; determining the target transmission path using an ant colony algorithm and a simulated annealing algorithm based on the latency overhead, signal-to-noise ratio, packet reception rate, and link quality of each communication link; and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path in the ad hoc network.

[0063] Specifically, the link transmission rate of each communication link is determined based on the Media Access Control (MAC) address. The link transmission rate is dynamically scanned at preset periods by the underlying driver and stored in a database along with its MAC address. This allows for the lookup of the corresponding link transmission rate in the database based on the MAC address. The latency overhead of each communication link is analyzed based on the link transmission rate and packet loss rate. The reciprocal of the link transmission rate is used to obtain the bit time, and the latency overhead is determined based on the bit time and packet loss rate combined with corresponding weights. The signal-to-noise ratio (SNR) of each communication link is determined based on the signal strength and noise intensity of each link. The packet reception rate for each communication link is calculated based on the number of successfully received sampled data packets and the total number of sampled data packets. The link quality is calculated based on the received signal power of each communication link, the received signal strength, and the received signal strength. This link quality reflects the quality of data reception. Based on the latency, signal-to-noise ratio (SNR), packet reception rate, and link quality of each communication link, the target transmission path is determined using ant colony optimization (ACO) and simulated annealing (SAM) algorithms. These factors are then incorporated into the path selection probability calculation of a hybrid algorithm. The hybrid algorithm, composed of ant colony optimization and SAM, leverages distributed computing, strong dynamic adaptability, and suitability for discrete combinatorial optimization problems. Compared to genetic algorithms and particle swarm optimization (PSO), it is more likely to discover optimal paths. However, ant colony optimization has a slower convergence speed in large-scale problems and may get stuck in local optima. Therefore, incorporating SAM improves performance, enabling the selection of the optimal path. The hybrid algorithm calculates the transmission path selection probability using the latency, SNR, packet reception rate, and link quality of each communication link, and the path with the highest selection probability is chosen as the final target transmission path. In the self-organizing network, the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path. The photovoltaic data recorded by the client device includes photovoltaic equipment data, photovoltaic equipment address data, photovoltaic power generation revenue data, customer information, etc. The client device transmits the photovoltaic data to the cloud server through the target transmission path in the self-organizing network.

[0064] S13: The cloud server classifies photovoltaic data based on the feature matrix of photovoltaic data to obtain photovoltaic data of various categories;

[0065] In the specific implementation of this invention, the classification of photovoltaic data based on the feature matrix of photovoltaic data to obtain photovoltaic data of various categories includes: constructing a feature matrix based on the node feature vectors of photovoltaic data; generating an adjacency matrix based on the feature matrix and determining the target weights corresponding to the feature matrix; constructing an objective equation based on feature fitting constraints and Laplace regularization constraints; constructing a graph convolutional neural network based on the objective equation and a preset filter; training the graph convolutional neural network to obtain a classification model; determining the target label of the photovoltaic data based on the adjacency matrix, the feature matrix, and the corresponding target weights using the classification model; and classifying the photovoltaic data based on the target label to obtain photovoltaic data of various categories.

[0066] Specifically, the cloud server receives photovoltaic data and constructs a feature matrix based on the node feature vectors of the photovoltaic data. The photovoltaic data may include data in formats such as images, videos, and text. The photovoltaic data includes several feature nodes, and the feature matrix is ​​constructed from the node feature vectors of each feature node. For example, if the photovoltaic data is in image format, its feature nodes are pixels, and the feature information contained therein is the pixel value. An adjacency matrix is ​​generated based on the feature matrix. The adjacency matrix is ​​generated according to the relationship information between the feature information in the feature matrix, and the target weight corresponding to the feature matrix is ​​determined. That is, the target weight of the photovoltaic data of a certain type is determined based on the feature matrix using a bidirectional long short-term memory network. For example, a certain target weight may correspond to power grid equipment data, and a certain target weight may correspond to customer information, etc. The objective equation is constructed based on feature fitting constraints and Laplacian regularization constraints. Feature fitting constraints aim to establish a relationship between node representations and original node features, essentially modifying the node features to consider a certain relationship between the desired learned representation and existing node features, and then controlling this relationship. This is achieved through flexible convolution defined by these two constraints. The Laplacian regularization constraint, specifically the graph Laplacian regularization constraint, essentially requires that the feature representations of two connected points on the adjacency matrix remain similar, thus describing the smoothing effect of the topology and achieving feature smoothing along the network topology. The first constraint term (composed of feature fitting constraints) and the second constraint term (composed of Laplacian regularization constraints) are used to construct the objective equation using a pre-defined optimization objective framework. Based on this objective equation and a pre-defined filter, a graph convolutional neural network (GNN) is constructed. The pre-defined filter can provide high-frequency and low-frequency filtering for the GNN. The GNN is then trained to obtain a classification model. Based on the adjacency matrix, feature matrix, and corresponding target weights, a classification model is used to determine the target labels for photovoltaic (PV) data. The adjacency matrix, feature matrix, and corresponding target weights are input into the classification model, which outputs several target labels corresponding to the PV data. The PV data is then classified based on these target labels. According to the mapping relationship between target labels and data categories, the category of each PV data type is determined, thus obtaining the PV data for each category. For example, if the label value of a target label is 1, the PV data belongs to the category of PV equipment-related data. This reduces the time spent on data classification while improving the accuracy of the classification results.

[0067] S14: Conduct security requirement analysis on various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data;

[0068] In the specific implementation of this invention, the step of performing security requirement analysis on various types of photovoltaic data to obtain the target security requirement level corresponding to each type of photovoltaic data includes: performing information composition analysis on various types of photovoltaic data to obtain information composition parameters; quantifying the security impact using the information composition parameters based on a judgment matrix to obtain information security parameters; establishing a hierarchical feature set and classification interval based on a sensitive data type feature set, and constructing a security classification model based on the hierarchical feature set and classification interval; performing initial security requirement level analysis on various types of photovoltaic data based on the security classification model to obtain an initial security requirement level; and determining the target security requirement level corresponding to each type of photovoltaic data based on the information security parameters and the initial security requirement level.

[0069] Specifically, information composition analysis is performed on photovoltaic data of various categories, and data type analysis is performed on photovoltaic data of various categories. Data types include specific data types in photovoltaic equipment data, such as equipment composition, etc. The data type of photovoltaic data of various categories is determined according to a preset type tree. The type tree includes multiple nodes, and each node corresponds to a judgment rule. The judgment rule is used to determine whether the data belongs to the data type represented by each node. The multiple nodes include leaf nodes without child nodes. The data type is determined by the data type represented by the leaf node. That is, the type tree is traversed from the starting node until the data type of photovoltaic data is determined. Data volume analysis is performed on photovoltaic data of various categories to obtain data volume information. The data volume information and data type information constitute information composition parameters. Based on the judgment matrix, the security impact is quantified using the information composition parameters. Parameter attribute analysis is performed on photovoltaic data of various categories to obtain parameter attribute information. Parameter attributes include common parameters and characteristic parameters. Common parameters refer to data with low data security requirements, such as customer names and quantities. Characteristic parameters refer to data where customers have special security needs, such as photovoltaic power generation revenue, address, and customer photos. The parameter attribute information is composed of common and characteristic parameters. Risk event evaluation indicators are set based on the frequency of historical risk events, and a judgment matrix is ​​constructed based on historical risk events and evaluation indicators. Historical risk events include information leakage or theft. The risk parameters of historical risk events are determined based on the judgment matrix. The correlation between risk parameters and information composition parameters and parameter attribute information is analyzed. That is, importance analysis is performed on information composition parameters to obtain the first importance corresponding to the information composition parameters. Importance analysis is performed on parameter attribute information to obtain the corresponding second importance. Correlation analysis is performed on risk parameters and their corresponding importance. Based on the correlation, the security impact is quantified to obtain information security parameters. Information security parameters include security impact quantification coefficients for photovoltaic data of various categories. Information security parameters can reflect the degree of risk encountered by photovoltaic data of various categories. A hierarchical feature set and classification intervals are established based on the feature set of sensitive data types. Sensitive data type features are defined and established, and these features are set by administrators. Classification intervals for sensitivity levels are set, including non-sensitive, internal sensitive, general commercial secret, core commercial secret, and core confidential. For example, some sensitive data type features may be classified as internal sensitive. Different sensitivity levels correspond to different storage security levels. A security classification model is constructed based on the hierarchical feature set and classification intervals, and this model can employ deep learning algorithms. Based on this security classification model, an initial security level analysis is performed on various categories of photovoltaic data. The photovoltaic data of each category is input into the security classification model to analyze its required initial storage security level and obtain the initial required security level.Based on the information security parameters and the initial security requirement level, the target security requirement level corresponding to each category of photovoltaic data is determined. The corresponding correction coefficient is matched according to the information security parameters, and the initial security requirement level is adjusted according to the corresponding correction coefficient to obtain the target security requirement level corresponding to each category of photovoltaic data, that is, to obtain the storage security requirement level required for each category of photovoltaic data.

[0070] S15: Analyze the business demand for each category of photovoltaic data, obtain the target business demand, and assess the popularity of each category of photovoltaic data based on the target business demand.

[0071] In the specific implementation of this invention, the step of analyzing the business demand level of various categories of photovoltaic data to obtain the target business demand level, and evaluating the popularity of various categories of photovoltaic data based on the target business demand level, includes: establishing a development relationship between each category of photovoltaic data and the corresponding target business based on the access information corresponding to several categories of photovoltaic data; analyzing the business demand level of various categories of photovoltaic data based on the development relationship to obtain the target business demand level; and evaluating the popularity of various categories of photovoltaic data based on the target business demand level in combination with popularity evaluation elements and data temperature analysis models.

[0072] Specifically, based on the access information corresponding to several categories of photovoltaic (PV) data, a developmental relationship is established between each category of PV data and its corresponding target business. Access information for each category of PV data is obtained, including access users and access methods. For example, the access user for data in the category of PV power generation revenue is the administrator. Different categories of PV data correspond to different target businesses, such as revenue analysis and evaluation. The developmental relationship is represented by a data table formed from the processed PV power generation revenue data, which is then used in the revenue analysis of the target business. The developmental relationship extends from the PV power generation revenue data to its application. Based on this developmental relationship, a business demand analysis is performed on each category of PV data. The frequency of business applications for each category of PV data is analyzed according to the developmental relationship, thus obtaining the target business demand level. Based on the target business demand level, combined with the hot / cold rating assessment factors and the data temperature analysis model, the hot / cold rating of various categories of photovoltaic data is evaluated. The hot / cold rating assessment factors include program application frequency, application introduction frequency, and data distribution frequency. The target business demand level and the hot / cold rating assessment factors are input into the data temperature analysis model to obtain the hot / cold rating of various categories of photovoltaic data. The data temperature analysis model adopts a long short-term memory network and uses the temperature attribute of the data as the basis for identifying the hot / cold rating of the data. The model assumes that data with a lower current temperature has a lower probability of being accessed in the future, so it is considered cold data, while data with a higher current temperature has a higher probability of being accessed in the future, so it is considered hot data. The hot / cold rating of data is defined as the level of data temperature: the hot / cold rating of data is equivalent to the level of data temperature.

[0073] S16: Based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with the performance analysis of storage nodes, determine the storage strategy, and store photovoltaic data of various categories into the corresponding storage nodes based on the storage strategy.

[0074] In the specific implementation of this invention, the process of determining a storage strategy based on the hotness / coldness of photovoltaic data of various categories and the target security level combined with storage node performance analysis, and storing photovoltaic data of various categories into corresponding storage nodes based on the storage strategy, includes: calculating partial regression coefficients using the least squares method based on log information of each storage node in the cloud storage center of the cloud server; analyzing the node performance value of each storage node based on the partial regression coefficients; determining an initial strategy based on the hotness / coldness of photovoltaic data of various categories and the security level of the target security level combined with the node performance value of each storage node, testing the initial strategy, adjusting the initial strategy based on the test results, and determining the storage strategy; marking the photovoltaic data of various categories with volume identifiers based on the storage strategy, and storing the photovoltaic data of various categories into corresponding storage nodes based on the volume identifiers.

[0075] Specifically, partial regression coefficients are calculated using the least squares method based on log information from each storage node in the cloud storage center within the cloud server. The cloud server deploys a cloud storage center, and the log information for each storage node includes memory performance values, CPU performance values, etc. These performance values ​​from the log information are used as sample data for a multiple regression model, and partial regression coefficients are calculated using the least squares method. The node performance values ​​of each storage node are analyzed based on these partial regression coefficients, and these coefficients are then substituted into a linear regression model. The node performance values ​​of the storage nodes are calculated in conjunction with the performance values ​​from the log information. An initial strategy is determined based on the hotness / coldness of photovoltaic data for each category, the target security level, and the node performance values ​​of each storage node. Specifically, the corresponding storage nodes and storage encryption levels are determined based on the hotness / coldness of photovoltaic data for each category, the target security level, and the node performance values ​​of each storage node. This initial strategy is then tested on a cloud server testing platform. The initial strategy is adjusted based on the test results; for example, if the test results show insufficient capacity, the capacity can be increased to finalize the storage strategy. Based on the storage strategy, photovoltaic data of various categories are marked with volume identifiers. The volume identifiers can accurately and quickly determine the storage nodes, and the photovoltaic data of various categories are stored in the corresponding storage nodes based on the volume identifiers. The photovoltaic data stored in each storage node is encrypted to an appropriate degree according to the storage strategy.

[0076] In this embodiment of the invention, connections between wireless communication nodes are established based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby effectively improving the organizational efficiency and search performance of the formed ad hoc network. A hybrid algorithm is used to determine the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network to transmit photovoltaic data recorded by the client device to the cloud server, improving data transmission stability and reducing latency. A classification model is used to classify the photovoltaic data based on the adjacency matrix, feature matrix, and corresponding target weights, improving the accuracy of photovoltaic data classification. The target demand security level for each category of photovoltaic data is determined based on the information security parameters obtained from the security impact quantification of photovoltaic data and the initial demand security level obtained from the security grading model, improving the reliability of data demand security level analysis. Storage strategies are determined based on the hotness / coldness and demand security level of each category of photovoltaic data, combined with storage node performance analysis, enabling a more comprehensive and specific analysis of storage strategies. Storing photovoltaic data using this strategy avoids data leakage and prevents poor node response speed and overall performance caused by inappropriate storage node selection.

[0077] Example 2

[0078] Please see Figure 2 , Figure 2 This is a flowchart illustrating a photovoltaic data management method based on a self-organizing network according to another embodiment of the present invention. The method is applied to a client device, a cloud server, and a self-organizing network composed of several wireless communication nodes. The method includes:

[0079] S201: Based on the performance interaction of the communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, a connection is established between each wireless communication node to form an ad hoc network;

[0080] S202: Based on the latency overhead and signal-to-noise ratio of each communication link in the self-organizing network, a hybrid algorithm is used to determine the target transmission path, and the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path;

[0081] S203: The cloud server constructs a feature matrix based on the node feature vectors of photovoltaic data, generates an adjacency matrix based on the feature matrix, and determines the target weights corresponding to the feature matrix;

[0082] S204: Construct an objective equation based on feature fitting constraints and Laplacian regularization constraints, and construct a graph convolutional neural network based on the objective equation and a preset filter, and train the graph convolutional neural network to obtain a classification model;

[0083] S205: Based on the adjacency matrix, feature matrix and corresponding target weights, a classification model is used to determine the target label of the photovoltaic data, and the photovoltaic data is classified based on the target label to obtain photovoltaic data of each category;

[0084] S206: Conduct security requirement analysis on various types of photovoltaic data to obtain the target security level for each type of photovoltaic data;

[0085] In the specific implementation of this invention, the step of performing security requirement analysis on various types of photovoltaic data to obtain the target security requirement level corresponding to each type of photovoltaic data includes: performing information composition analysis on various types of photovoltaic data to obtain information composition parameters; quantifying the security impact using the information composition parameters based on a judgment matrix to obtain information security parameters; establishing a hierarchical feature set and classification interval based on a sensitive data type feature set, and constructing a security classification model based on the hierarchical feature set and classification interval; performing initial security requirement level analysis on various types of photovoltaic data based on the security classification model to obtain an initial security requirement level; and determining the target security requirement level corresponding to each type of photovoltaic data based on the information security parameters and the initial security requirement level.

[0086] S207: Analyze the business demand for each type of photovoltaic data, obtain the target business demand, and assess the popularity of each type of photovoltaic data based on the target business demand.

[0087] S208: Based on the hotness and coldness of photovoltaic data of various categories and the required security level, combined with the performance analysis of storage nodes, determine the storage strategy, and store photovoltaic data of various categories into the corresponding storage nodes based on the storage strategy.

[0088] In this embodiment of the invention, connections between wireless communication nodes are established based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby effectively improving the organizational efficiency and search performance of the formed ad hoc network. A hybrid algorithm is used to determine the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network to transmit photovoltaic data recorded by the client device to the cloud server, improving data transmission stability and reducing latency. A classification model is used to classify the photovoltaic data based on the adjacency matrix, feature matrix, and corresponding target weights, improving the accuracy of photovoltaic data classification. The target demand security level for each category of photovoltaic data is determined based on the information security parameters obtained from the security impact quantification of photovoltaic data and the initial demand security level obtained from the security grading model, improving the reliability of data demand security level analysis. Storage strategies are determined based on the hotness / coldness and demand security level of each category of photovoltaic data, combined with storage node performance analysis, enabling a more comprehensive and specific analysis of storage strategies. Storing photovoltaic data using this strategy avoids data leakage and prevents poor node response speed and overall performance caused by inappropriate storage node selection.

[0089] Example 3

[0090] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a photovoltaic data management system based on a self-organizing network according to an embodiment of the present invention. The system includes a client device 31, a cloud server 33, and a self-organizing network 32 composed of several wireless communication nodes. The system is configured to execute the photovoltaic data management method based on a self-organizing network described in the above embodiment.

[0091] In the specific implementation of this invention, the client device 31 can be a user's smart terminal for inputting photovoltaic data, and the cloud server 33 is used to receive, analyze, and store the photovoltaic data. Each wireless communication node in the self-organizing network 32 includes a wireless radio frequency module, which includes a wireless radio frequency transceiver. The radio frequency communication technology of the wireless radio frequency module can ensure the reliability of the self-organizing network, realize long-distance communication, and has a stronger cost-effectiveness. The client device 31 transmits photovoltaic data to the cloud server 33 through the self-organizing network 32. Figure 3 The system shown does not constitute a limitation on all components and may include more or fewer components than shown, or combine certain components.

[0092] In the specific implementation of this invention, the specific implementation methods of the system items can be referred to the above embodiments, and will not be repeated here.

[0093] In this embodiment of the invention, connections between wireless communication nodes are established based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby effectively improving the organizational efficiency and search performance of the formed ad hoc network. A hybrid algorithm is used to determine the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network to transmit photovoltaic data recorded by the client device to the cloud server, improving data transmission stability and reducing latency. A classification model is used to classify the photovoltaic data based on the adjacency matrix, feature matrix, and corresponding target weights, improving the accuracy of photovoltaic data classification. The target demand security level for each category of photovoltaic data is determined based on the information security parameters obtained from the security impact quantification of photovoltaic data and the initial demand security level obtained from the security grading model, improving the reliability of data demand security level analysis. Storage strategies are determined based on the hotness / coldness and demand security level of each category of photovoltaic data, combined with storage node performance analysis, enabling a more comprehensive and specific analysis of storage strategies. Storing photovoltaic data using this strategy avoids data leakage and prevents poor node response speed and overall performance caused by inappropriate storage node selection.

[0094] Example 4

[0095] Please see Figure 4 , Figure 4 This is a schematic diagram of a photovoltaic data management device based on a self-organizing network according to an embodiment of the present invention. The device is applied to a client device, a cloud server, and a self-organizing network composed of several wireless communication nodes; the device includes:

[0096] Ad hoc network forming module 41: used to establish connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby forming an ad hoc network;

[0097] Data transmission module 42: used to determine the target transmission path based on the delay overhead and signal-to-noise ratio of each communication link in the self-organizing network using a hybrid algorithm, and to transmit the photovoltaic data recorded by the client device to the cloud server based on the target transmission path;

[0098] Data classification module 43: Used by the cloud server to classify photovoltaic data based on the feature matrix of photovoltaic data to obtain photovoltaic data of various categories;

[0099] Security Requirements Analysis Module 44: Used to perform security requirements analysis on various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data;

[0100] Hot and cold assessment module 45: used to analyze the business demand level of various types of photovoltaic data, obtain the target business demand level, and assess the hot and cold status of various types of photovoltaic data based on the target business demand level;

[0101] Storage module 46: is used to determine the storage strategy based on the hotness and coldness of photovoltaic data of various categories and the required security level, combined with the performance analysis of storage nodes, and to store photovoltaic data of various categories into the corresponding storage nodes based on the storage strategy.

[0102] In the specific implementation of this invention, the specific implementation methods of the device can be referred to the above embodiments, and will not be repeated here.

[0103] In this embodiment of the invention, connections between wireless communication nodes are established based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thereby effectively improving the organizational efficiency and search performance of the formed ad hoc network. A hybrid algorithm is used to determine the target transmission path based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network to transmit photovoltaic data recorded by the client device to the cloud server, improving data transmission stability and reducing latency. A classification model is used to classify the photovoltaic data based on the adjacency matrix, feature matrix, and corresponding target weights, improving the accuracy of photovoltaic data classification. The target demand security level for each category of photovoltaic data is determined based on the information security parameters obtained from the security impact quantification of photovoltaic data and the initial demand security level obtained from the security grading model, improving the reliability of data demand security level analysis. Storage strategies are determined based on the hotness / coldness and demand security level of each category of photovoltaic data, combined with storage node performance analysis, enabling a more comprehensive and specific analysis of storage strategies. Storing photovoltaic data using this strategy avoids data leakage and prevents poor node response speed and overall performance caused by inappropriate storage node selection.

[0104] This invention provides a computer-readable storage medium storing a computer program. When executed by a processor, this program implements the photovoltaic data management method based on a self-organizing network, as described in any of the above embodiments. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. In other words, the storage device includes any medium that stores or transmits information in a readable form by a device (e.g., a computer, a mobile phone), and can be a read-only memory, a disk, or an optical disk, etc.

[0105] Furthermore, the above provides a detailed description of a photovoltaic data management method and related apparatus based on a self-organizing network provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A photovoltaic data management method based on self-organizing networks, characterized in that, The method is applicable to client devices, cloud servers, and self-organizing networks composed of several wireless communication nodes; the method includes: A self-organizing network is formed by establishing connections between wireless communication nodes based on the performance interaction of the communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes. Based on the latency overhead and signal-to-noise ratio of each communication link in the self-organizing network, a hybrid algorithm is used to determine the target transmission path, and the photovoltaic data recorded by the client device is transmitted to the cloud server based on the target transmission path. The cloud server classifies photovoltaic data based on the feature matrix of the photovoltaic data to obtain photovoltaic data of various categories; Conduct security requirements analysis on various types of photovoltaic data to obtain the target security level for each type of photovoltaic data. Analyze the business demand for each category of photovoltaic data to obtain the target business demand, and assess the popularity of each category of photovoltaic data based on the target business demand. Based on the hotness or coldness of photovoltaic data of various categories and the security level of target requirements, combined with the performance analysis of storage nodes, a storage strategy is determined, and photovoltaic data of various categories are stored in the corresponding storage nodes based on the storage strategy. The process of determining the target transmission path using a hybrid algorithm based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network, and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path, includes: determining the link transmission rate of each communication link based on the media access control address, and analyzing the latency overhead of each communication link based on the link transmission rate and packet loss rate; determining the signal-to-noise ratio of each communication link, calculating the corresponding packet reception rate of each communication link, and calculating the link quality based on the received signal power of each communication link; determining the target transmission path using an ant colony algorithm and a simulated annealing algorithm based on the latency overhead, signal-to-noise ratio, packet reception rate, and link quality of each communication link; and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path in the ad hoc network. The method of classifying photovoltaic data based on feature matrices to obtain photovoltaic data of various categories includes: constructing a feature matrix based on the node feature vectors of photovoltaic data; generating an adjacency matrix based on the feature matrix and determining the target weights corresponding to the feature matrix; constructing an objective equation based on feature fitting constraints and Laplacian regularization constraints; constructing a graph convolutional neural network based on the objective equation and a preset filter; training the graph convolutional neural network to obtain a classification model; determining the target label of the photovoltaic data using the classification model based on the adjacency matrix, feature matrix, and corresponding target weights; and classifying the photovoltaic data based on the target label to obtain photovoltaic data of various categories.

2. The photovoltaic data management method based on self-organizing networks according to claim 1, characterized in that, The process of establishing connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes to form an ad hoc network includes: The signal strength between adjacent wireless communication nodes is calculated based on the address of each wireless communication node, and the connectivity between adjacent wireless communication nodes is determined based on the signal strength. The target distance between adjacent wireless communication nodes is calculated based on the connectivity between them, and the neighbor list of each wireless communication node is updated based on the target distance to obtain an updated neighbor list. The communication link exchanges performance data with nearby wireless communication nodes and performs load balancing among the wireless communication nodes based on the exchanged performance data. A self-organizing network is formed by connecting various wireless communication nodes based on updating the neighbor list and load balancing among wireless communication nodes.

3. The photovoltaic data management method based on self-organizing networks according to claim 1, characterized in that, The aforementioned security requirement analysis of various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data includes: Perform information composition analysis on various types of photovoltaic data to obtain information composition parameters; Based on the judgment matrix, the security impact is quantified using the information composition parameters to obtain information security parameters; A hierarchical feature set and classification interval are established based on the feature set of sensitive data types, and a security hierarchical model is constructed based on the hierarchical feature set and classification interval. Based on the security classification model, the initial demand security level is analyzed for each category of photovoltaic data to obtain the initial demand security level. Based on the aforementioned information security parameters and initial security requirements, the target security requirements level corresponding to each category of photovoltaic data is determined.

4. The photovoltaic data management method based on self-organizing network according to claim 1, characterized in that, The process of analyzing the business demand for various categories of photovoltaic data to obtain the target business demand level, and assessing the popularity of different categories of photovoltaic data based on the target business demand level, includes: Establish a development relationship between each category of photovoltaic data and its corresponding target business based on access information corresponding to several categories of photovoltaic data; Based on the aforementioned development context, business demand analysis is performed on various categories of photovoltaic data to obtain the target business demand level. Based on the target business demand level, combined with the hot and cold degree assessment elements and data temperature analysis model, the hot and cold degree corresponding to various types of photovoltaic data is assessed.

5. The photovoltaic data management method based on self-organizing networks according to claim 1, characterized in that, The process involves determining a storage strategy based on the hotness / coldness of photovoltaic data for each category, the target security level, and the performance analysis of storage nodes. Then, based on this strategy, photovoltaic data for each category is stored in its corresponding storage node. This includes: Partial regression coefficients are calculated using the least squares method based on log information from each storage node in the cloud storage center of the cloud server. The node performance values ​​of each storage node are analyzed based on the partial regression coefficients. An initial strategy is determined based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with the node performance values ​​of each storage node. The initial strategy is then tested, and the initial strategy is adjusted based on the test results to determine the storage strategy. Based on the storage strategy, photovoltaic data of various categories are labeled with volume identifiers, and the photovoltaic data of various categories are stored in the corresponding storage nodes based on the volume identifiers.

6. A photovoltaic data management device based on a self-organizing network, characterized in that, The device is applicable to client devices, cloud servers, and ad hoc networks composed of several wireless communication nodes; the device includes: Ad hoc network forming module: used to establish connections between wireless communication nodes based on the performance interaction of communication links established by each wireless communication node and the connectivity between adjacent wireless communication nodes, thus forming an ad hoc network; Data transmission module: used to determine the target transmission path based on the delay overhead and signal-to-noise ratio of each communication link in the self-organizing network using a hybrid algorithm, and to transmit the photovoltaic data recorded by the client device to the cloud server based on the target transmission path; Data classification module: Used by the cloud server to classify photovoltaic data based on the feature matrix of photovoltaic data to obtain photovoltaic data of various categories; Security Requirements Analysis Module: Used to perform security requirements analysis on various types of photovoltaic data to obtain the target security level corresponding to each type of photovoltaic data; Hot and cold assessment module: used to analyze the business demand level of various types of photovoltaic data, obtain the target business demand level, and assess the hot and cold status of various types of photovoltaic data based on the target business demand level; Storage module: used to determine storage strategies based on the hotness or coldness of photovoltaic data of various categories and the target security level, combined with storage node performance analysis, and to store photovoltaic data of various categories into the corresponding storage nodes based on the storage strategies; The process of determining the target transmission path using a hybrid algorithm based on the latency overhead and signal-to-noise ratio of each communication link in the ad hoc network, and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path, includes: determining the link transmission rate of each communication link based on the media access control address, and analyzing the latency overhead of each communication link based on the link transmission rate and packet loss rate; determining the signal-to-noise ratio of each communication link, calculating the corresponding packet reception rate of each communication link, and calculating the link quality based on the received signal power of each communication link; determining the target transmission path using an ant colony algorithm and a simulated annealing algorithm based on the latency overhead, signal-to-noise ratio, packet reception rate, and link quality of each communication link; and transmitting the photovoltaic data recorded by the client device to the cloud server based on the target transmission path in the ad hoc network. The method of classifying photovoltaic data based on feature matrices to obtain photovoltaic data of various categories includes: constructing a feature matrix based on the node feature vectors of photovoltaic data; generating an adjacency matrix based on the feature matrix and determining the target weights corresponding to the feature matrix; constructing an objective equation based on feature fitting constraints and Laplacian regularization constraints; constructing a graph convolutional neural network based on the objective equation and a preset filter; training the graph convolutional neural network to obtain a classification model; determining the target label of the photovoltaic data using the classification model based on the adjacency matrix, feature matrix, and corresponding target weights; and classifying the photovoltaic data based on the target label to obtain photovoltaic data of various categories.

7. A photovoltaic data management system based on a self-organizing network, characterized in that, The system includes a client device, a cloud server, and an ad hoc network consisting of several wireless communication nodes. The system is configured to execute the photovoltaic data management method based on the ad hoc network as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the photovoltaic data management method based on a self-organizing network as described in any one of claims 1 to 5.

Images