Power information management platform based on data fusion

Abstract

The present application provides an electric power information management platform based on data fusion, which relates to the field of electric power information technology, including: an equipment distribution array acquisition module acquires the equipment distribution array of the target power station; a communication fitness analysis module performs communication fitness analysis on the electric power equipment; an interactive fitness analysis module performs interactive fitness analysis on the electric power equipment; a correlation identification module performs correlation identification on the combination of electric power equipment; an interactive node coefficient identification module performs interactive node coefficient identification on the electric power equipment; an interactive neighborhood construction module constructs an interactive neighborhood and configures a neighborhood processing component; a monitoring and early warning module performs asynchronous monitoring data extraction and electric power information analysis on the electric power equipment set, and performs electric power information early warning. The present application solves the technical problem that the prior art cannot fully tap the interdependence between electric power equipment, resulting in untimely and inaccurate monitoring of equipment failures, and improves the efficiency and accuracy of electric power information management.

Description

Technical Field

[0001] The present application relates to the field of electric power information technology, and specifically to an electric power information management platform based on data fusion. Background Art

[0002] In modern power systems, power information management is an important part of ensuring the stability and security of power supply, which involves multiple aspects such as power equipment status monitoring, data processing and decision support. With the continuous expansion of the power grid and the increasing number of equipment types, traditional power information management methods face many challenges.

[0003] Existing power information management technologies mainly rely on decentralized data collection and processing systems. The data of various power equipment are often stored on different independent platforms, which makes data integration difficult and makes it impossible to achieve unified management. This data isolation phenomenon makes it very difficult to fully grasp the operating status of the power system. In addition, the physical and electrical connections between devices are often regarded as isolated factors, without considering their interactions and influences in the entire power grid, which limits the system's ability to respond to dynamic changes in the power grid. When faced with large-scale data sets, traditional data analysis methods are unable to effectively mine the correlation between data and provide strong support for decision-making. Due to the failure to fully consider the interdependence between power equipment, it is impossible to quickly locate the source of the problem and take effective measures when equipment fails, which to a certain extent weakens the stability and reliability of the power system. Summary of the invention

[0004] This application provides an electric power information management platform based on data fusion, which solves the technical problem that the existing technology cannot fully explore the interdependence between power equipment when facing large-scale power stations, resulting in untimely and inaccurate monitoring of equipment failures. It achieves the technical effect of comprehensively and in-depth analysis and prediction of the operating status of the power system and improving the efficiency and accuracy of electric power information management.

[0005] In view of the above problems, the present application provides an electric power information management platform based on data fusion, the platform comprising: an equipment distribution array acquisition module, the equipment distribution array acquisition module is used to acquire the equipment distribution array of the target power station, wherein the equipment distribution array comprises K electric power devices and K electric power equipment layout positions; a communication fitness analysis module, the communication fitness analysis module is used to collect the layout positions of the control unit of the power data center of the target power station, and perform communication fitness analysis on the K electric power devices in combination with the K electric power equipment layout positions to obtain K first equipment fitnesses; an interactive fitness analysis module, the interactive fitness analysis module is used to exchange the line direction data of the target power station, perform interactive fitness analysis on the K electric power devices, and obtain K second equipment fitnesses; a correlation identification module, the correlation identification module is used to enumerate the K electric power devices in pairs to obtain multiple electric power equipment combinations, and perform correlation identification on the multiple electric power equipment combinations respectively to obtain multiple correlation indicators, wherein each a relevant indicator corresponds to a power equipment combination; an interactive node coefficient identification module, the interactive node coefficient identification module is used to identify the interactive node coefficients of the K power equipment according to the K first device fitnesses, the K second device fitnesses and the multiple relevant indicators, and obtain K interactive node coefficients, wherein the K interactive node coefficients are obtained by data analysis through the interactive node multidimensional analysis formula; an interactive neighborhood construction module, the interactive neighborhood construction module is used to screen interactive nodes based on the K interactive node coefficients and the K power equipment, construct M interactive neighborhoods, and configure the M neighborhood processing components of the power data concentration control unit according to the M interactive neighborhoods; a monitoring and early warning module, the monitoring and early warning module is used to perform asynchronous monitoring data extraction on the M power equipment sets of the M interactive neighborhoods, and transmit the M asynchronous data extraction results to the M neighborhood processing components for power information analysis, obtain M power information early warning results, and send the M power information early warning results to the staff for early warning.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The equipment distribution array acquisition module acquires the equipment distribution array of the target power station, provides the specific location and layout of the power equipment in the power station, and provides basic data for subsequent communication and interaction analysis. The communication fitness analysis module collects the layout location of the power data center control unit of the target power station, and combines the layout locations of the K power equipment to perform communication fitness analysis on the K power equipment to obtain K first equipment fitness, ensure the communication quality and stability between the equipment, and avoid system operation problems caused by poor communication. The interactive fitness analysis module interacts with the line direction data of the target power station, performs interactive fitness analysis on the K power equipment, and obtains K second equipment fitness, ensuring barrier-free interactive operation between the equipment and improving the overall stability and reliability of the system. The correlation identification module enumerates the K power equipment in pairs to obtain multiple power equipment combinations, and performs correlation identification on the multiple power equipment combinations respectively to obtain multiple related indicators, wherein each related indicator corresponds to a power equipment combination, and provides basic data for subsequent interactive node coefficient identification by evaluating the collaborative working ability and mutual influence between the equipment. The interactive node coefficient identification module identifies the interactive node coefficients of the K power devices according to the fitness of the K first devices, the fitness of the K second devices and the multiple related indicators, performs data analysis through the interactive node multidimensional analysis formula, obtains K interactive node coefficients, and ensures the accuracy and effectiveness of the identification results. The interactive neighborhood construction module screens interactive nodes based on the K interactive node coefficients and the K power devices, constructs M interactive neighborhoods, and configures the M neighborhood processing components of the power data centralized control unit according to the M interactive neighborhoods, optimizes the configuration of nodes and neighborhoods, improves resource utilization and system operation efficiency, and realizes intelligent management of the power system. The monitoring and early warning module extracts asynchronous monitoring data from the M power equipment sets of the M interactive neighborhoods, and transmits the M asynchronous data extraction results to the M neighborhood processing components for power information analysis, obtains M power information early warning results, and sends the M power information early warning results to the staff for early warning. Through real-time monitoring and early warning functions, potential risks can be discovered and early warnings can be issued in a timely manner to prevent equipment damage and system failures, and improve the safety of the system.

[0008] To summarize, this application introduces communication fitness analysis and interaction fitness analysis to accurately evaluate the communication efficiency and interaction efficiency between multiple power devices, combines correlation identification and interaction node coefficients to deeply explore the interdependence between devices, and constructs interactive neighborhoods and asynchronous monitoring to quickly locate and solve problems when equipment problems occur, respond to emergencies more quickly, and reduce the possibility of chain reactions. It significantly improves the efficiency and accuracy of power information management, enhances the safety and stability of the entire power system, and ensures the reliability of power supply.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the structure of a power information management platform based on data fusion provided in an embodiment of the present application;

[0011] Figure 2 A schematic diagram of a process for obtaining M interactive neighborhoods in a power information management platform based on data fusion provided in an embodiment of the present application;

[0012] Figure 3 A schematic diagram of a process for generating the M power information warning results in a power information management platform based on data fusion provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: device distribution array acquisition module 10 , communication fitness analysis module 20 , interaction fitness analysis module 30 , correlation identification module 40 , interaction node coefficient identification module 50 , interaction neighborhood construction module 60 , monitoring and early warning module 70 . DETAILED DESCRIPTION

[0014] The embodiment of the present application solves the technical problem that the prior art cannot fully exploit the interdependence between power equipment when facing large-scale power stations, resulting in untimely and inaccurate monitoring of equipment failures, by providing a power information management platform based on data fusion. It achieves the technical effect of comprehensively and deeply analyzing and predicting the operating status of the power system and improving the efficiency and accuracy of power information management.

[0015] like Figure 1 As shown, the embodiment of the present application provides a power information management platform based on data fusion, and the platform includes:

[0016] The device distribution array acquisition module 10 is used to acquire the device distribution array of the target power station, wherein the device distribution array includes K power devices and K power device layout positions.

[0017] Specifically, the target power station refers to the specific power site that needs to be managed and monitored, such as a substation or power plant. The equipment distribution array refers to the arrangement and distribution of all power equipment in the power station, including the location of each device and the corresponding layout point. K power equipment refers to the number of all power equipment that needs to be monitored and managed in the power station, and K is a variable for the number of equipment. The layout location refers to the specific installation and placement location of each power equipment in the power station.

[0018] First, use equipment identification technology, such as RFID tags or QR code scanning, to identify and register the power equipment in the power station. Each device will be assigned a unique identification code. Through field visits or using existing geographic information system (GIS) data, obtain the coordinates or other forms of location information of all power equipment in the power station. For example, a power station may have multiple types of equipment such as generators, transformers, switchgear, etc., and each device has its specific location. Then, digitize this location information and convert it into a data format that can be processed by a computer, such as longitude and latitude coordinates or a specific location on the power station plan. The equipment distribution array acquisition module 10 identifies and obtains the target power station equipment distribution array through information interaction, including K power equipment and K power equipment layout locations, to achieve comprehensive and accurate management of the distribution information of power equipment in the target power station. This not only improves the efficiency of power equipment management, but also provides a reliable data basis for subsequent fitness analysis and interaction analysis.

[0019] The communication fitness analysis module 20 is used to collect the layout location of the power data center control unit of the target power station, and perform communication fitness analysis on the K power devices in combination with the layout locations of the K power devices to obtain K first device fitnesses.

[0020] Specifically, the power data centralized control unit is the core device in the power station for centralized management and control of power data, similar to a data center. Communication fitness refers to the effectiveness of communication between each device and the data centralized control unit. This is a relative value used to measure the difficulty of communication. The first device fitness is an indicator used to measure the communication quality between each power device and the power data centralized control unit.

[0021] The module obtains the layout position of the power data centralized control unit through the GPS positioning system, sensors or manual input, etc. According to the layout position of the power data centralized control unit, combined with the layout positions of K power devices obtained by the device distribution array acquisition module 10, the communication fitness analysis is performed on the K power devices respectively, the difficulty of communication between the device and the data centralized control unit is measured, and K first device fitness is obtained, wherein each power device corresponds to a first device fitness.

[0022] Furthermore, the steps performed by the communication fitness analysis module 20 in the embodiment of the present application also include:

[0023] The communication distance is calculated according to the layout position of the power data concentration control unit and the layout positions of the K power equipment to obtain K communication distances; and the ratio of the K communication distances to the sum of the K communication distances is used as the K first device fitness.

[0024] Specifically, the communication distance refers to the physical distance between the control unit in the power data set and each power device. First, determine the location of the control unit in the data set, which is the reference point for calculating the communication distance. Using the location information of each power device, calculate the communication distance between them and the control unit. This distance can be estimated by physical distance, such as straight-line distance or shortest path, or it can be measured by signal propagation time or signal strength. Then, divide the communication distance of each device by the sum of the communication distances of all devices to obtain the first device fitness of each device. This ratio reflects the relative difficulty of communication between the device and the control unit. The larger the ratio, the easier it is for the device to communicate with the control unit, and vice versa.

[0025] The communication fitness analysis module 20 helps optimize the design of the communication network by quantitatively analyzing the communication conditions between the equipment and the control unit, ensuring that all equipment can effectively communicate with the control unit, thereby improving the efficiency and reliability of the entire power information management system.

[0026] The interactive fitness analysis module 30 is used to interact with the line direction data of the target power station, perform interactive fitness analysis on the K power devices, and obtain K second device fitnesses.

[0027] Specifically, line direction data refers to the connection lines and wiring direction data between power equipment. Interaction fitness refers to the rationality of the electrical connection between devices, that is, the efficiency of power flow and the balance of line load. The second device fitness is the result of the interaction fitness analysis of each power device.

[0028] First, the module collects the line direction data of the target power station by exchanging information with the SCADA (Supervisory Control and Data Acquisition) system or GIS (Geographic Information System) of the power system. Then, based on the line direction data, the neural network model is used to evaluate the connection strength and interaction frequency between K power devices to obtain K second device fitness. Each power device corresponds to a second device fitness.

[0029] Furthermore, the steps performed by the interaction fitness analysis module 30 in the embodiment of the present application also include:

[0030] According to the line direction data of the target power station, the connection equipment quantities of the K power devices are counted respectively to obtain K connection equipment quantities; with the line equipment quantity as the index, data is retrieved for the lines where the K power devices are located respectively to obtain K line equipment quantities; the interactive fitness analysis network layer is used to perform equipment interaction analysis on the K connection equipment quantities and the K line equipment quantities to determine the fitness of the K second devices.

[0031] Specifically, the number of connected devices refers to the number of other devices connected to each power device. The number of line devices refers to the number of power devices connected to each line. The interactive fitness analysis network layer is a network layer used to analyze the interactive fitness of power devices, taking into account factors such as the number of connected devices and the number of line devices.

[0032] According to the line direction data, the number of devices directly connected to each device is counted to obtain K connected device quantities. The more connected devices there are, the greater the role of the device in the network. Next, using the line device quantity as an index, data is retrieved for the line where each power device is located, and the total number of devices on the line where each power device is located is counted to obtain K line device quantities.

[0033] The interactive fitness analysis network layer is constructed using a neural network model, which specifically includes: first, obtaining the data on the connected equipment quantity and line equipment quantity of the power equipment from the SCADA system or GIS system of the target power station. The original data is cleaned and standardized to ensure data quality and consistency. Then, the connected equipment quantity of each power equipment is taken as part of the feature vector, and the number of equipment on each line is taken as the other part of the feature vector. The connected equipment quantity and line equipment quantity are combined to form a comprehensive feature matrix for subsequent neural network training.

[0034] Then use TensorFlow to build a neural network model, and use the feature matrix as the input of the neural network. The number of nodes in the input layer is equal to the dimension of the feature vector. Set up several hidden layers, each containing a certain number of neurons, and perform nonlinear transformations through activation functions. The number of nodes in the output layer is K, corresponding to the second device fitness of each power device. Define the loss function and select the optimization algorithm, use the feature matrix as training data, and use the training data to train the model until the model converges, and you can get a trained interactive fitness analysis network layer.

[0035] Finally, the newly acquired data on the quantity of connected devices and the quantity of line devices of the power equipment are input into the trained interactive fitness analysis network layer to predict the second device fitness of each device and obtain K second device fitnesses.

[0036] The interactive fitness analysis module 30 helps ensure the stable operation and load balance of the power system and improve the efficiency and safety of power transmission by analyzing the electrical connections between power devices.

[0037] The correlation identification module 40 is used to enumerate the K power devices in pairs to obtain multiple power device combinations, and to perform correlation identification on the multiple power device combinations respectively to obtain multiple correlation indicators, wherein each correlation indicator corresponds to one power device combination.

[0038] Specifically, pairwise enumeration refers to combining K power devices in pairs to form all possible device pairs. The correlation index is a specific value used to quantify the correlation between each power device combination.

[0039] Combine K power devices in pairs to form all possible pairs of devices and obtain multiple combinations of devices. Collect the operating data of each power device, including current, voltage, temperature, frequency, etc., and perform data preprocessing. For each power device combination, calculate their correlation indexes to obtain multiple correlation indexes, where each correlation index corresponds to a power device combination. Common correlation calculation methods include Pearson correlation coefficient, Spearman correlation coefficient, etc. According to the calculated correlation index, evaluate the correlation of each device combination. A combination with high correlation indicates that the devices have strong collaborative working ability.

[0040] The correlation identification module 40 can identify the collaborative working capabilities between devices by analyzing the correlation of power equipment operation data, thereby better diagnosing problems and preventing failures, and optimizing the efficiency of the power information management system.

[0041] The interactive node coefficient identification module 50 is used to identify the interactive node coefficients of the K power devices according to the fitness of the K first devices, the fitness of the K second devices and the multiple related indicators to obtain K interactive node coefficients, wherein the K interactive node coefficients are obtained by data analysis through the interactive node multidimensional analysis formula.

[0042] Specifically, the interaction node coefficient is a quantitative indicator obtained by comprehensively considering the communication adaptability, electrical connection adaptability and mutual influence between devices, which is used to reflect the strength of the interaction between devices in the entire power system. The interaction node multidimensional analysis formula is a mathematical formula used to calculate the interaction node coefficient, which is analyzed by comprehensive multidimensional data.

[0043] The interaction node coefficient identification module 50 obtains the first device fitness, the second device fitness and related indicators of the K power devices generated by the above modules, and integrates these data into a comprehensive data set. Then a multidimensional analysis formula is constructed, the integrated data set is input into the interactive node multidimensional analysis formula, data analysis is performed, the interaction node coefficient of each power device is calculated, and K interaction node coefficients are obtained. This module integrates the communication fitness, interaction fitness and related indicators through a multidimensional analysis formula to obtain a unified quantitative indicator, namely the interaction node coefficient, so as to better understand the importance of each device in the power system and the interaction between devices, and provide a reference for the subsequent interaction neighborhood division.

[0044] Furthermore, the steps performed by the interaction node coefficient identification module 50 in the embodiment of the present application also include:

[0045] Construct an interactive node multidimensional analysis formula, wherein the interactive node multidimensional analysis formula is:

[0046]

[0047] Among them, L p is the interaction node coefficient corresponding to the pth power device, θ(x p ,x q ) is the association index between the p-th power device and the associated q-th power device, q∈n, p≤K, K is the total number of power devices in the target power station, n is the total number of power devices associated with the p-th power device, is the first fitness of the pth power device, is the second fitness of the pth power device.

[0048] The K first device fitnesses, the K second device fitnesses and the multiple related indicators are input into the interactive node multidimensional analysis formula for analysis of the K power devices to generate the K interactive node coefficients.

[0049] Specifically, a multidimensional analysis formula for an interactive node is constructed as follows:

[0050]

[0051] Among them, L p is the interaction node coefficient corresponding to the pth power device, which is used to indicate the interaction capability and importance of the device in the power system. p ,x q ) is the correlation index between the pth power device and the associated qth power device, indicating the degree of correlation between the two devices, obtained by the correlation identification module 40, q∈n, p≤K, K is the total number of power devices in the target power station, and n is the total number of power devices associated with the pth power device. 1 μp is the first fitness of the pth power device, indicating the adaptability of the device in the communication network. This indicator is obtained by the communication fitness analysis module 20.

[0052] is the second fitness of the p-th power device, indicating the interactive adaptability of the device in the power system. This indicator is obtained through the interactive fitness analysis module 30 .

[0053] The K first device fitnesses, K second device fitnesses and multiple related indicators corresponding to the K power devices obtained above are respectively input into the interactive node multidimensional analysis formula for analysis to generate corresponding K interactive node coefficients, wherein each power device corresponds to one interactive node coefficient.

[0054] The interactive neighborhood construction module 60 is used to perform interactive node screening based on the K interactive node coefficients and the K power equipment, construct M interactive neighborhoods, and configure the M neighborhood processing components of the power data center management and control unit according to the M interactive neighborhoods.

[0055] Specifically, interaction node screening is the process of dividing devices into different categories or groups based on the interaction node coefficient of each device. M interaction neighborhoods divide the devices into M different groups or areas based on the interaction node coefficients between devices. The power data centralized management and control unit is the central unit for data collection, processing and control in the power system. M neighborhood processing components are processing units configured according to M interaction neighborhoods, which are used to process and manage power equipment in each neighborhood. Wherein, M is a positive integer, and M<K.

[0056] First, the module uses K interaction node coefficients to screen power devices. These coefficients can help determine the importance and interaction capabilities of each device in the power system. Based on the screening results, the power devices are divided into M different neighborhoods, each of which contains a group of devices with high interactivity. Then, according to the characteristics of each neighborhood, the neighborhood processing components of the corresponding power data center control unit are configured. Finally, the module outputs M interactive neighborhoods and corresponding processing component configuration information. This information can be used for subsequent power information management and optimization.

[0057] Further, such as Figure 2 As shown, the steps performed by the interactive neighborhood construction module 60 in the embodiment of the present application also include:

[0058] The K interaction node coefficients are sorted in descending order, the M interaction node coefficients in the first M positions are extracted, and the M power equipment corresponding to the M interaction node coefficients are used as M initial interaction nodes; M initial topological structures are constructed according to the layout positions of the M power equipment of the M initial interaction nodes; the similarity between the M initial topological structures and the device topological structures corresponding to the device distribution array is calculated using the twin network model to obtain a first initial topological similarity; it is determined whether the first initial topological similarity meets the preset similarity requirement, if not, the M interaction coefficients corresponding to the M initial interaction nodes are updated and iterated according to the K interaction node coefficients until M interaction nodes are obtained; if so, the M initial interaction nodes are used as M interaction nodes, and the remaining N power equipment are added to the neighborhood corresponding to the interaction node with the largest relevant index to obtain M interaction neighborhoods, where N=KM.

[0059] Specifically, the initial interaction node is the most interactive power equipment obtained through screening. The initial topology is a network topology built based on the initial interaction nodes for further analysis and optimization. The twin network model is a machine learning model used to compare and calculate the similarity of two topologies. The first initial topology similarity refers to the similarity between the initial topology and the actual device topology. The preset similarity requirement refers to a pre-set similarity threshold used to determine whether the topology is reasonable.

[0060] According to the interaction node coefficient, K power devices are sorted and the first M interaction nodes are selected as the initial interaction nodes. Where M is a positive integer and M < K, representing the number of interaction domains planned to be constructed. The value of M can be set according to the actual accuracy requirements and platform computing resources. According to the position and connection relationship of the initial interaction nodes, M initial topological structures are constructed using network analysis tools, each topological structure is represented as a graph, and the graph is converted into a form suitable for neural network processing, such as an adjacency matrix or a node feature matrix.

[0061] Build a twin network model. The twin network consists of two identical sub-networks, which process two input graphs respectively and calculate their feature representations. The similarity of the graphs is calculated by comparing the similarity of the two feature representations. Exemplarily, a twin network model framework is built using TensorFlow and Keras, and trained using labeled similar and dissimilar topological structures. For each pair of inputs, calculate their feature representations and compare their similarities, then adjust the model parameters according to the true labels (similar or dissimilar), and finally obtain a trained twin network model.

[0062] The M initial topological structures and the topological structures in the device distribution array are respectively input into the twin network model, and the similarity of each pair of topological structures is calculated to obtain the first initial topological similarity. The first initial topological similarity is compared with the preset similarity to determine whether it meets the preset similarity requirements. If the similarity does not meet the requirements, the interaction node coefficient is updated and iterated, that is, the interaction node coefficient at the M+1 position is iterated against the interaction node coefficient at the Mth position, and the above steps are repeated until an interaction node that meets the preset similarity requirements is obtained. If the similarity meets the requirements, the M initial interaction nodes are used as M interaction nodes, and the remaining N power devices are added to the neighborhood corresponding to the interaction node with the largest relevant index to obtain M interaction neighborhoods, where N=KM.

[0063] The interactive neighborhood construction module 60 constructs the interactive neighborhood and configures the processing components through the interactive node coefficients, twin models and equipment topology structure, which can efficiently complete data processing and analysis, ensure the reasonable construction and resource allocation of the interactive neighborhood, and thus realize the efficient management of large-scale power equipment.

[0064] Furthermore, the steps performed by the interactive neighborhood construction module 60 in the embodiment of the present application also include:

[0065] Taking the number of power equipment as an index, the M interactive neighborhoods are statistically counted to obtain the number of power equipment in the M neighborhoods; the server resource information of the power data center management and control unit is collected, and component resource allocation is performed on the server resource information according to the number of power equipment in the M neighborhoods to generate the M neighborhood processing components.

[0066] Specifically, first, count the number of each of the M interactive neighborhoods and calculate how many power devices are in each neighborhood. This is to ensure that each neighborhood has enough resources to process the device data and control requirements within it. Then, collect the server resource information of the power data center control unit, including CPU usage, memory size, storage space, etc. This information will be used to determine how to effectively allocate resources to each neighborhood processing component. Next, allocate server resources based on the number of power devices in each neighborhood. This process can be done using resource management tools, such as the resource scheduler of the cloud computing platform, which can dynamically allocate resources to different tasks. Finally, generate M neighborhood processing components. These components will be specifically used to process the device data within each interactive neighborhood to ensure that each neighborhood can operate efficiently. For example, if neighborhood A has 20 devices and neighborhood B has 30 devices, then the processing component of neighborhood B may be allocated more computing resources because it has more devices and greater processing requirements.

[0067] In this way, the interactive neighborhood building module 60 can ensure that each interactive neighborhood can be supported by appropriate resources, thereby improving the efficiency and response speed of the entire power system.

[0068] The monitoring and early warning module 70 is used to perform asynchronous monitoring data extraction on the M sets of power equipment in the M interactive neighborhoods, and transmit the M asynchronous data extraction results to the M neighborhood processing components for power information analysis, obtain M power information early warning results, and send the M power information early warning results to the staff for early warning.

[0069] Specifically, asynchronous monitoring data extraction refers to collecting data from various devices asynchronously, which means that data collection is not continuous, but is performed as needed or at predetermined time intervals. M asynchronous data extraction results refer to the data results extracted from each interactive neighborhood. Power information analysis refers to the analysis of power equipment data to detect potential problems or anomalies. Power information early warning results refer to the results after analysis, and if a problem is detected, a corresponding warning will be generated.

[0070] The monitoring and early warning module 70 performs asynchronous monitoring data extraction on the set of power equipment in each interactive neighborhood, that is, collects data at different time points to obtain M asynchronous data extraction results. The collected data is then transmitted to the corresponding neighborhood processing component for analysis to find abnormal patterns or trends, such as excessive equipment temperature, voltage fluctuations, etc. Next, the M neighborhood processing components will generate corresponding M power information early warning results based on the analysis results. Each early warning result corresponds to an interactive neighborhood. If a potential problem is detected, an early warning will be generated. For example, if the neighborhood processing component finds that the temperature of a certain device continues to rise, it may trigger an overheating early warning. Finally, these early warning results will be sent to staff to remind them to take appropriate actions based on these early warnings, which can be achieved through an alarm system, email, or mobile application.

[0071] In this way, the monitoring and early warning module 70 can monitor the operating status of the power system in real time, discover and warn potential problems in a timely manner, thereby improving the safety and reliability of the power system.

[0072] Further, such as Figure 3 As shown, the steps performed by the monitoring and early warning module 70 in the embodiment of the present application also include:

[0073] Collect monitoring data of preset monitoring windows for M sets of electric power equipment in the M interactive neighborhoods to obtain M sets of monitoring data; traverse the M sets of monitoring data to perform stability identification to determine M sets of stability factors; determine M first monitoring steps and M second monitoring steps according to the M sets of stability factors, wherein the M second monitoring steps are smaller than the M first monitoring steps; perform asynchronous monitoring data extraction on the M sets of electric power equipment based on the M first monitoring steps and the M second monitoring steps to obtain the M asynchronous data extraction results, wherein the M asynchronous data extraction results include M first monitoring data extraction sets and M second monitoring data extraction sets; transmit the M asynchronous data extraction results to the M neighborhood processing components for analysis to generate the M electric power information early warning results.

[0074] Preferably, the steps performed by the monitoring and early warning module 70 in the embodiment of the present application further include:

[0075] M first monitoring step lengths are configured based on the maximum values ​​in the M stability factor sets respectively; and M second monitoring step lengths are configured based on the minimum values ​​in the M stability factor sets respectively.

[0076] Specifically, the preset monitoring window refers to a fixed time period used to monitor the operating data of power equipment, and data is collected during this time period. The monitoring data set is a data set obtained by collecting monitoring data for each power equipment. Stability identification is a process used to evaluate data stability. The stability factor set refers to a set of indicators reflecting data stability obtained through stability identification, which reflects the stability level of the monitoring data. The first monitoring step and the second monitoring step are time intervals used to determine the frequency of data collection. The first monitoring step is usually larger and is used for routine monitoring; the second monitoring step is smaller and is used for more frequent monitoring, especially when stability is low. The first monitoring data extraction set and the second monitoring data extraction set are data sets collected according to different monitoring steps.

[0077] First, the monitoring and early warning module 70 collects data from the power equipment set of each interactive neighborhood within the preset monitoring window to obtain M monitoring data sets. Then, the M monitoring data sets are traversed, stability identification is performed on each monitoring data set, data fluctuations or abnormal values ​​are analyzed, and the stability factor set of each interactive neighborhood is determined to obtain M stability factor sets.

[0078] Next, the first monitoring step and the second monitoring step are determined according to the stability factor set. The second monitoring step is smaller than the first monitoring step. Generally, the higher the stability factor, the larger the monitoring step, and vice versa. Based on the maximum value in the M stability factor sets, M first monitoring step are configured; based on the minimum value in the M stability factor sets, M second monitoring step are configured.

[0079] Then, the first monitoring step and the second monitoring step are used to perform asynchronous monitoring data extraction on the power equipment set to obtain M asynchronous data extraction results. Each asynchronous data extraction result includes a first monitoring data extraction set and a second monitoring data extraction set. The asynchronous data extraction results are transmitted to the neighborhood processing component for analysis to generate a power information early warning result.

[0080] In this way, the monitoring and early warning module 70 can flexibly adjust the monitoring frequency according to the actual operating conditions of the equipment, which can not only ensure efficiency but also detect potential problems in a timely manner, ensuring the accuracy and reliability of monitoring and early warning.

[0081] In summary, the power information management platform based on data fusion provided by the embodiment of the present application has the following technical effects:

[0082] The equipment distribution array acquisition module 10 acquires the equipment distribution array of the target power station, provides the specific location and layout of the power equipment in the power station, and provides basic data for subsequent communication and interaction analysis. The communication fitness analysis module 20 collects the layout location of the power data center control unit of the target power station, and performs communication fitness analysis on the K power equipment in combination with the layout locations of the K power equipment to obtain K first equipment fitnesses, ensure the communication quality and stability between the equipment, and avoid system operation problems caused by poor communication. The interactive fitness analysis module 30 interacts with the line direction data of the target power station, performs interactive fitness analysis on the K power equipment, obtains K second equipment fitnesses, ensures barrier-free interactive operations between the equipment, and improves the overall stability and reliability of the system. The correlation identification module 40 enumerates the K power equipment in pairs to obtain multiple power equipment combinations, and performs correlation identification on the multiple power equipment combinations respectively to obtain multiple related indicators, wherein each related indicator corresponds to a power equipment combination, and provides basic data for subsequent interactive node coefficient identification by evaluating the collaborative working ability and mutual influence between the equipment. The interactive node coefficient identification module 50 identifies the interactive node coefficients of the K power devices according to the fitness of the K first devices, the fitness of the K second devices and the multiple related indicators, performs data analysis through the interactive node multidimensional analysis formula, obtains K interactive node coefficients, and ensures the accuracy and effectiveness of the identification results. The interactive neighborhood construction module 60 screens interactive nodes based on the K interactive node coefficients and the K power devices, constructs M interactive neighborhoods, and configures the M neighborhood processing components of the power data centralized control unit according to the M interactive neighborhoods, optimizes the configuration of nodes and neighborhoods, improves resource utilization and system operation efficiency, and realizes intelligent management of the power system. The monitoring and early warning module 70 extracts asynchronous monitoring data from the M power equipment sets of the M interactive neighborhoods, and transmits the M asynchronous data extraction results to the M neighborhood processing components for power information analysis, obtains M power information early warning results, and sends the M power information early warning results to the staff for early warning. Through real-time monitoring and early warning functions, potential risks are discovered and early warnings are issued in a timely manner to prevent equipment damage and system failures, and improve the safety of the system.

[0083] Overall, the embodiments of the present application introduce communication fitness analysis and interaction fitness analysis to accurately evaluate the communication efficiency and interaction efficiency between multiple power devices, combine correlation identification and interaction node coefficients to deeply explore the interdependence between devices, and build interactive neighborhoods and asynchronous monitoring to quickly locate and solve problems when equipment problems occur, respond to emergencies more quickly, and reduce the possibility of chain reactions. This significantly improves the efficiency and accuracy of power information management, enhances the safety and stability of the entire power system, and ensures the reliability of power supply.

[0084] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The power information management platform based on data fusion is characterized by: The platform includes: A device distribution array acquisition module, wherein the device distribution array acquisition module is used to acquire a device distribution array of a target power station, wherein the device distribution array includes K power devices and K power device layout positions; A communication fitness analysis module, the communication fitness analysis module is used to collect the layout location of the power data center control unit of the target power station, and perform communication fitness analysis on the K power devices in combination with the layout locations of the K power devices to obtain K first device fitnesses, wherein the communication fitness refers to the effectiveness of communication between each device and the data center control unit, which is used to measure the difficulty of communication; An interactive fitness analysis module, which is used to interact with the line direction data of the target power station, perform interactive fitness analysis on the K power devices, and obtain K second device fitnesses, wherein the interactive fitness refers to the rationality of the electrical connection between the devices, that is, the efficiency of the power flow and the balance of the line load; A correlation identification module, wherein the correlation identification module is used to enumerate the K power devices in pairs to obtain multiple power device combinations, and respectively perform correlation identification on the multiple power device combinations to obtain multiple correlation indicators, wherein each correlation indicator corresponds to one power device combination; An interaction node coefficient identification module, the interaction node coefficient identification module is used to identify the interaction node coefficients of the K power devices according to the fitness of the K first devices, the fitness of the K second devices and the multiple related indicators, and obtain K interaction node coefficients, wherein the K interaction node coefficients are obtained by data analysis through an interaction node multidimensional analysis formula, and the interaction node coefficient is a quantitative indicator obtained by comprehensively considering the communication fitness, electrical connection fitness and mutual influence between devices, and is used to reflect the strength of the interaction of the devices in the entire power system; An interactive neighborhood construction module, the interactive neighborhood construction module is used to perform interactive node screening based on the K interactive node coefficients and the K power devices, construct M interactive neighborhoods, and configure M neighborhood processing components of the power data centralized management and control unit according to the M interactive neighborhoods; A monitoring and early warning module, the monitoring and early warning module is used to extract asynchronous monitoring data from the M sets of power equipment in the M interactive neighborhoods, and transmit the M asynchronous data extraction results to the M neighborhood processing components for power information analysis, obtain M power information early warning results, and send the M power information early warning results to the staff for early warning; The steps performed by the communication fitness analysis module further include: Calculate the communication distance according to the layout position of the power data centralized control unit and the layout positions of the K power devices to obtain K communication distances; respectively taking the ratios of the K communication distances to the sum of the K communication distances as K first device fitnesses; The steps performed by the interaction fitness analysis module further include: According to the line direction data of the target power station, respectively counting the number of connected devices of the K power devices to obtain K number of connected devices; Using the line equipment quantity as an index, respectively search for data on the lines where the K power equipment are located to obtain the K line equipment quantities; Performing device interaction analysis on the K connected device quantities and the K line device quantities by using an interactive fitness analysis network layer to determine the fitness of the K second devices; The steps performed by the interaction node coefficient identification module further include: Construct an interactive node multidimensional analysis formula, wherein the interactive node multidimensional analysis formula is: Among them, L p is the interaction node coefficient corresponding to the pth power device, θ ( x p ,x q ) is the association index between the p-th power device and the associated q-th power device, q∈n, p≤K, K is the total number of power devices in the target power station, n is the total number of power devices associated with the p-th power device, is the first fitness of the pth power device, is the second fitness of the pth power device; The K first device fitnesses, the K second device fitnesses and the multiple related indicators are input into the interactive node multidimensional analysis formula for analysis of the K power devices to generate the K interactive node coefficients.

2. The platform according to claim 1, characterized in that The steps performed by the interactive neighborhood building module also include: The K interaction node coefficients are sorted in descending order, the M interaction node coefficients located in the first M positions are extracted, and the M power devices corresponding to the M interaction node coefficients are used as M initial interaction nodes; Constructing M initial topological structures according to the M power equipment layout positions of the M initial interaction nodes; Calculating the similarity between the M initial topological structures and the device topological structure corresponding to the device distribution array using a twin network model to obtain a first initial topological similarity; Determine whether the first initial topological similarity meets the preset similarity requirement, and if not, iterate and update the M interaction coefficients corresponding to the M initial interaction nodes according to the K interaction node coefficients until M interaction nodes are obtained; If so, the M initial interaction nodes are taken as M interaction nodes, and the remaining N power devices are added into the neighborhood corresponding to the interaction node with the largest relevant index to obtain M interaction neighborhoods, where N=KM.

3. The platform according to claim 1, characterized in that The steps performed by the interactive neighborhood building module also include: Taking the number of power equipment as an index, performing quantity statistics on the M interactive neighborhoods to obtain the quantity of power equipment in the M neighborhoods; The server resource information of the power data centralized management and control unit is collected, component resource allocation is performed on the server resource information according to the M neighborhood power equipment quantities, and the M neighborhood processing components are generated.

4. The platform according to claim 1, characterized in that The steps performed by the monitoring and early warning module also include: Collecting monitoring data of a preset monitoring window for the M sets of power equipment in the M interactive neighborhoods to obtain M sets of monitoring data; Traversing the M monitoring data sets to perform stability identification and determine M stability factor sets; Determining M first monitoring step lengths and M second monitoring step lengths according to the M stability factor sets, wherein the M second monitoring step lengths are smaller than the M first monitoring step lengths; Performing asynchronous monitoring data extraction on the M sets of power equipment based on the M first monitoring step lengths and the M second monitoring step lengths to obtain the M asynchronous data extraction results, wherein the M asynchronous data extraction results include M first monitoring data extraction sets and M second monitoring data extraction sets; The M asynchronous data extraction results are transmitted to the M neighborhood processing components for analysis to generate the M power information warning results.

5. The platform according to claim 4, characterized in that The steps performed by the monitoring and early warning module also include: configuring M first monitoring step lengths based on the maximum values ​​of the M stability factor sets respectively; M second monitoring step sizes are configured based on the minimum values ​​in the M stability factor sets respectively.

Images