Power system monitoring method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a power system monitoring method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring electrical quantity data, environmental data and video image data of power equipment in a to-be-modeled power system; performing preprocessing, feature extraction processing and summation processing on the electrical quantity data, the environment data and the video image data to obtain a target feature vector of the power equipment; inputting the target feature vector into a trained equipment state prediction model to obtain target state information corresponding to the power equipment; determining visual feature information corresponding to the power equipment according to the target state information; constructing an initial three-dimensional model of the power system, and adjusting the initial three-dimensional model according to the visual feature information corresponding to the power equipment to obtain a target three-dimensional model of the power system; and monitoring the power system through the target three-dimensional model. By adopting the method, the monitoring accuracy of the power system can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power grids, and particularly to a power system monitoring method, device, computer device, computer-readable storage medium, and computer program product. Background Art

[0002] Currently, in order to ensure the operational stability of the power system, how to accurately monitor the power system is of crucial importance.

[0003] In traditional technologies, during the process of monitoring the power system, a two-dimensional interface monitoring method is generally adopted; however, this method is difficult to intuitively display the spatial relationship and real-time changes between power equipment in the power system, resulting in a low monitoring accuracy of the power system. Summary of the Invention

[0004] Based on this, it is necessary to provide a power system monitoring method, device, computer device, computer-readable storage medium, and computer program product that can improve the monitoring accuracy of the power system for the above technical problems.

[0005] In a first aspect, the present application provides a power system monitoring method, including:

[0006] Obtaining device data of power equipment in the power system to be modeled: the device data at least includes electrical quantity data, environmental data, and video image data of the power equipment;

[0007] Preprocessing the electrical quantity data, the environmental data, and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data;

[0008] Performing feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data;

[0009] Determining a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector;

[0010] Performing a summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain a target feature vector of the power equipment;

[0011] Input the target feature vector into the trained device status prediction model to obtain the target status information corresponding to the power device;

[0012] Determine the visual feature information corresponding to the power device according to the target status information;

[0013] Construct an initial 3D model of the power system, and adjust the initial 3D model according to the visual feature information corresponding to the power device to obtain the target 3D model of the power system;

[0014] Monitor the power system through the target 3D model.

[0015] In one embodiment, the obtaining the device data of the power device in the power system to be modeled includes:

[0016] Obtain the current application scenario corresponding to the power device in the power system to be modeled;

[0017] According to the current application scenario, query the corresponding relationship between the application scenario and the data acquisition frequency, and obtain the data acquisition frequency corresponding to the current application scenario as the target data acquisition frequency corresponding to the power device;

[0018] Obtain the device data according to the target data acquisition frequency.

[0019] In one embodiment, the determining the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector includes:

[0020] Input the electrical quantity data, the environmental data, and the video image data into the trained importance prediction model respectively to obtain the first importance corresponding to the electrical quantity data, the second importance corresponding to the environmental data, and the third importance corresponding to the video image data;

[0021] Determine the first initial weight corresponding to the first feature vector, the second initial weight corresponding to the second feature vector, and the third initial weight corresponding to the third feature vector according to the first importance, the second importance, and the third importance respectively;

[0022] Perform normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector.

[0023] In one embodiment, building the initial three-dimensional model of the power system includes:

[0024] Obtain the wiring diagram of the power system;

[0025] Determine the topological information of the power system according to the wiring diagram;

[0026] Obtain the shape information and position information of the power equipment, and build the initial three-dimensional model of the power system according to the shape information and position information of the power equipment and the topological information of the power system.

[0027] In one embodiment, respectively performing feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data, includes:

[0028] Take the preprocessed electrical quantity data as the main data, and take the preprocessed environmental data and the preprocessed video image data as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain the first feature vector;

[0029] Take the preprocessed environmental data as the main data, and take the preprocessed electrical quantity data and the preprocessed video image data as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the second feature vector;

[0030] Take the preprocessed video image data as the main data, and take the preprocessed electrical quantity data and the preprocessed environmental data as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the third feature vector.

[0031] In one embodiment, the trained equipment status prediction model is trained through the following method:

[0032] Obtain the sample device data of the sample power equipment in the sample power system; the sample device data at least includes the sample electrical quantity data, sample environmental data, and sample video image data of the sample power equipment;

[0033] Respectively preprocess the sample electrical quantity data, the sample environmental data, and the sample video image data to obtain preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data;

[0034] Perform feature extraction processing on the preprocessed sample electrical quantity data, the preprocessed sample environmental data, and the preprocessed sample video image data respectively, to obtain a first sample feature vector corresponding to the preprocessed sample electrical quantity data, a second sample feature vector corresponding to the preprocessed sample environmental data, and a third sample feature vector corresponding to the preprocessed sample video image data;

[0035] Determine a first target sample weight corresponding to the first sample feature vector, a second target sample weight corresponding to the second sample feature vector, and a third target sample weight corresponding to the third sample feature vector;

[0036] According to the first target sample weight, the second target sample weight, and the third target sample weight, perform a summation process on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a target sample feature vector of the sample power equipment;

[0037] Input the target sample feature vector into the device state prediction model to be trained, and obtain the predicted state information corresponding to the sample power equipment;

[0038] Obtain the actual state information corresponding to the sample power equipment, and perform iterative training on the device state prediction model to be trained according to the difference between the predicted state information and the actual state information, to obtain the trained device state prediction model.

[0039] In a second aspect, the present application also provides a power system monitoring device, including:

[0040] A data acquisition module, configured to acquire device data of a power equipment in a power system to be modeled: the device data at least includes electrical quantity data, environmental data, and video image data of the power equipment;

[0041] A data processing module, configured to preprocess the electrical quantity data, the environmental data, and the video image data respectively, to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data;

[0042] A feature extraction module, configured to perform feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively, to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data;

[0043] A weight determination module, configured to determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector;

[0044] A vector summation module, configured to perform a summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight, to obtain a target feature vector of the power equipment;

[0045] A state prediction module, configured to input the target feature vector into a trained device state prediction model to obtain target state information corresponding to the power equipment;

[0046] An information determination module, configured to determine visual feature information corresponding to the power equipment according to the target state information;

[0047] A model processing module, configured to construct an initial three-dimensional model of the power system, and perform an adjustment process on the initial three-dimensional model according to the visual feature information corresponding to the power equipment, to obtain a target three-dimensional model of the power system;

[0048] A system monitoring module, configured to perform a monitoring process on the power system through the target three-dimensional model.

[0049] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0050] Obtain device data of power equipment in a power system to be modeled: The device data at least includes electrical quantity data, environmental data, and video image data of the power equipment;

[0051] Perform preprocessing on the electrical quantity data, the environmental data, and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data;

[0052] Perform feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data;

[0053] Determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector;

[0054] Sum the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the target feature vector of the power equipment;

[0055] Input the target feature vector into the trained equipment status prediction model to obtain the target status information corresponding to the power equipment;

[0056] Determine the visual feature information corresponding to the power equipment according to the target status information;

[0057] Construct an initial 3D model of the power system, and adjust the initial 3D model according to the visual feature information corresponding to the power equipment to obtain the target 3D model of the power system;

[0058] Monitor the power system through the target 3D model.

[0059] Fourthly, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0060] Obtain the device data of the power equipment in the power system to be modeled: the device data at least includes the electrical quantity data, environmental data, and video image data of the power equipment;

[0061] Preprocess the electrical quantity data, the environmental data, and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data;

[0062] Extract features from the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data;

[0063] Determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector;

[0064] Sum the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the target feature vector of the power equipment;

[0065] Input the target feature vector into the trained equipment status prediction model to obtain the target status information corresponding to the power equipment;

[0066] Determine the visual feature information corresponding to the power equipment according to the target status information;

[0067] Construct an initial three-dimensional model of the power system, and adjust the initial three-dimensional model according to the visual feature information corresponding to the power equipment to obtain the target three-dimensional model of the power system;

[0068] Monitor the power system through the target three-dimensional model.

[0069] In a fifth aspect, the present application also provides a computer program product, including a computer program, which when executed by a processor implements the following steps:

[0070] Obtain the device data of the power equipment in the power system to be modeled: the device data at least includes the electrical quantity data, environmental data, and video image data of the power equipment;

[0071] Preprocess the electrical quantity data, the environmental data, and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data;

[0072] Extract feature vectors from the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data;

[0073] Determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector;

[0074] Sum the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the target feature vector of the power equipment;

[0075] Input the target feature vector into the trained device status prediction model to obtain the target status information corresponding to the power equipment;

[0076] Determine the visual feature information corresponding to the power equipment according to the target status information;

[0077] Construct an initial three-dimensional model of the power system, and adjust the initial three-dimensional model according to the visual feature information corresponding to the power equipment to obtain the target three-dimensional model of the power system;

[0078] Monitor and process the power system through the target 3D model.

[0079] For the above power system monitoring method, device, computer device, storage medium and computer program product, first obtain the electrical quantity data, environmental data and video image data of the power equipment in the power system to be modeled, all of which are used as the device data of the power equipment in the power system to be modeled, and preprocess the electrical quantity data, environmental data and video image data respectively to obtain the preprocessed electrical quantity data, preprocessed environmental data and preprocessed video image data. Then, perform feature extraction processing on the preprocessed electrical quantity data, preprocessed environmental data and preprocessed video image data respectively to obtain the first feature vector corresponding to the preprocessed electrical quantity data, the second feature vector corresponding to the preprocessed environmental data and the third feature vector corresponding to the preprocessed video image data, and determine the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector and the third target weight corresponding to the third feature vector. Next, according to the first target weight, the second target weight and the third target weight, perform a summation process on the first feature vector, the second feature vector and the third feature vector to obtain the target feature vector of the power equipment, and input the target feature vector into the trained device state prediction model to obtain the target state information corresponding to the power equipment. Then, according to the target state information, determine the visual feature information corresponding to the power equipment, construct the initial 3D model of the power system, and adjust the initial 3D model according to the visual feature information corresponding to the power equipment to obtain the target 3D model of the power system. Finally, monitor and process the power system through the target 3D model. In this way, during the process of monitoring the power system, through a series of processes such as preprocessing, feature extraction processing, and model processing of various device data of the power equipment in the power system to be modeled, the target state information corresponding to the power equipment can be accurately obtained, so that the visual feature information corresponding to the power equipment can be more accurately determined, and combined with the initial 3D model of the power system, and then the target 3D model of the power system can be more accurately obtained, which is beneficial to improving the monitoring accuracy of the power system. Moreover, this method uses a 3D model to monitor the power system, avoiding the defect that it is difficult to intuitively display the spatial relationship and real-time changes between power equipment in the power system by using a 2D interface monitoring method, resulting in a low monitoring accuracy of the power system, and thus improving the monitoring accuracy of the power system. Description of the Drawings

[0080] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the accompanying drawings required for the description of the embodiments of the present application or related technologies. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0081] Figure 1 It is a schematic flowchart of a power system monitoring method in an embodiment;

[0082] Figure 2 It is a schematic flowchart of a power system monitoring method in another embodiment;

[0083] Figure 3 It is a schematic flowchart of a power system visual monitoring method based on a time series mapping model in an embodiment;

[0084] Figure 4 It is a structural block diagram of a power system monitoring device in an embodiment;

[0085] Figure 5 It is an internal structure diagram of a computer device in an embodiment. Detailed implementation manners

[0086] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the following further details the present application in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0087] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with relevant regulations.

[0088] In an exemplary embodiment, as Figure 1 shown, a power system monitoring method is provided. In this embodiment, it is exemplified that the method is applied to a server; it can be understood that the method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be but is not limited to various personal computers, laptop computers, smart phones and tablet computers; the server can be implemented by an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0089] Step S101: Obtain the device data of the power equipment in the power system to be modeled. The device data includes at least the electrical quantity data, environmental data, and video image data of the power equipment.

[0090] Among them, a variety of monitoring devices are deployed in the power system, such as SCADA (Supervisory Control And Data Acquisition) systems, PMU (Phasor Measurement Unit) devices, infrared thermometers, video surveillance cameras, etc.

[0091] Among them, power equipment includes power generation equipment (such as thermal power generation equipment, hydroelectric power generation equipment), transformation equipment (such as transformers, circuit breakers), transmission equipment (such as transmission lines, substations), distribution equipment (such as distribution cabinets, distribution boxes), etc.

[0092] Among them, the device data includes at least the electrical quantity data, environmental data, and video image data of the power equipment.

[0093] Among them, the electrical quantity data includes voltage, current, and power. Voltage is used to reflect the voltage levels of different nodes in the power system; current is used to monitor the load conditions of transmission lines or equipment; power includes active power and reactive power, which are used to analyze the energy efficiency of equipment and the operating stability of the power grid.

[0094] Among them, the environmental data includes temperature, humidity, and smoke concentration. Temperature and humidity are used to reflect the working environment of the equipment and help identify potential faults caused by deteriorating environmental conditions; smoke concentration is used to assist in identifying the initial signs of equipment overheating or fire.

[0095] Among them, the video image data is obtained from the real-time images captured by the monitoring cameras of key equipment (such as transformers, circuit breakers), and is used to assist in judging the equipment status and troubleshooting.

[0096] Exemplarily, the server collects in real time the electrical quantity data, environmental data, and video image data of the power equipment in the power system to be modeled through the monitoring devices deployed in the power system to be modeled, and uses the electrical quantity data, environmental data, and video image data as the device data of the power equipment in the power system to be modeled.

[0097] Step S102: Preprocess the electrical quantity data, environmental data, and video image data respectively to obtain the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data.

[0098] Among them, the preprocessing includes data cleaning, data calibration, data standardization, and data synchronization.

[0099] Among them, the preprocessed electrical quantity data refers to the electrical quantity data after preprocessing.

[0100] Among them, the preprocessed environmental data refers to the environmental data after preprocessing.

[0101] Among them, the preprocessed video image data refers to the video image data after preprocessing.

[0102] Exemplarily, the server performs preprocessing such as data cleaning, data calibration, data standardization, and data synchronization on the electrical quantity data, environmental data, and video image data respectively, to obtain the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data.

[0103] Step S103: Perform feature extraction processing on the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data respectively, to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data.

[0104] Among them, the first feature vector refers to the feature vector corresponding to the preprocessed electrical quantity data.

[0105] Among them, the second feature vector refers to the feature vector corresponding to the preprocessed environmental data.

[0106] Among them, the third feature vector refers to the feature vector corresponding to the preprocessed video image data.

[0107] Exemplarily, the server inputs the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data into the feature extraction model respectively, and through the feature extraction model, performs feature extraction processing on the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data respectively, to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data.

[0108] Step S104: Determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector.

[0109] Among them, the first target weight refers to the weight finally assigned to the first feature vector.

[0110] Among them, the second target weight refers to the weight finally assigned to the second feature vector.

[0111] Among them, the third target weight refers to the weight finally assigned to the third feature vector.

[0112] Exemplarily, the server inputs the first feature vector, the second feature vector, and the third feature vector into the attention mechanism model respectively, and through the attention mechanism model, obtains the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector.

[0113] Step S105: According to the first target weight, the second target weight, and the third target weight, perform a summation process on the first feature vector, the second feature vector, and the third feature vector to obtain the target feature vector of the power equipment.

[0114] The target feature vector refers to the feature vector obtained by performing a summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight.

[0115] Exemplarily, the server performs a verification process on the first target weight, the second target weight, and the third target weight to obtain a verification result; when the verification result indicates that the first target weight, the second target weight, and the third target weight all meet the preset threshold, the server performs a summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the target feature vector of the power equipment.

[0116] Step S106: Input the target feature vector into the trained device state prediction model to obtain the target state information corresponding to the power equipment.

[0117] The device state prediction model refers to a network model that can use the target feature vector of the power equipment to obtain the target state information corresponding to the power equipment, such as a convolutional neural network model or a recurrent neural network model.

[0118] The target state information refers to the state information of the power equipment, including the normal operation state, the overload state, and the fault state.

[0119] Exemplarily, the server inputs the target feature vector into the trained device state prediction model to obtain the prediction probabilities of the power equipment under each preset state information; then, the server screens out the preset state information with the largest prediction probability from each preset state information as the target state information corresponding to the power equipment.

[0120] Step S107: Determine the visual feature information corresponding to the power equipment according to the target state information.

[0121] Among them, the visual feature information is used to represent the colors corresponding to the power equipment in the 3D model, such as green, red, and gray. It should be noted that different target status information corresponds to different visual feature information. For example, the normal operation status corresponds to green, the overload status corresponds to red, and the fault status corresponds to gray.

[0122] Exemplarily, the server queries the correspondence between the status information and the visual feature information according to the target status information, and obtains the visual feature information corresponding to the target status information as the visual feature information corresponding to the power equipment.

[0123] Step S108: Construct an initial 3D model of the power system, and perform adjustment processing on the initial 3D model according to the visual feature information corresponding to the power equipment to obtain the target 3D model of the power system.

[0124] Among them, the initial 3D model is used to represent the initially constructed 3D model of the power system.

[0125] Among them, the target 3D model is used to represent the finally constructed 3D model of the power system.

[0126] Exemplarily, the server constructs an initial 3D model of the power system in response to the 3D model construction instruction of the power system; then, the server performs rendering processing on the initial 3D model according to the visual feature information corresponding to the power equipment to obtain the rendered initial 3D model as the target 3D model of the power system.

[0127] Step S109: Monitor the power system through the target 3D model.

[0128] Exemplarily, the server generates video information corresponding to the target 3D model according to the monitoring instruction for the power system; then, the server monitors the power system according to the video information corresponding to the target 3D model.

[0129] In the above power system monitoring method, first, electrical quantity data, environmental data, and video image data of power equipment in the power system to be modeled are obtained, all of which are used as equipment data of the power equipment in the power system to be modeled. Then, the electrical quantity data, environmental data, and video image data are respectively preprocessed to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data. Next, feature extraction processing is respectively performed on the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data. And the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector are determined. Then, according to the first target weight, the second target weight, and the third target weight, the first feature vector, the second feature vector, and the third feature vector are summed to obtain the target feature vector of the power equipment. The target feature vector is input into the trained equipment state prediction model to obtain the target state information corresponding to the power equipment. Then, according to the target state information, the visual feature information corresponding to the power equipment is determined, and an initial three-dimensional model of the power system is constructed. And according to the visual feature information corresponding to the power equipment, the initial three-dimensional model is adjusted to obtain the target three-dimensional model of the power system. Finally, the power system is monitored through the target three-dimensional model. In this way, during the process of monitoring the power system, through a series of processes such as preprocessing, feature extraction processing, and model processing of various equipment data of the power equipment in the power system to be modeled, the target state information corresponding to the power equipment can be accurately obtained, so that the visual feature information corresponding to the power equipment can be more accurately determined, and combined with the initial three-dimensional model of the power system, the target three-dimensional model of the power system can be more accurately obtained, which is beneficial to improving the monitoring accuracy of the power system. Moreover, this method avoids the defect that the monitoring accuracy of the power system is relatively low due to the difficulty of intuitively displaying the spatial relationship and real-time changes between power equipment in the power system by using a two-dimensional interface for monitoring, and improves the monitoring accuracy of the power system by using a three-dimensional model to monitor the power system.

[0130] In an exemplary embodiment, step S101 above, obtaining the equipment data of the power equipment in the power system to be modeled specifically includes the following contents: obtaining the current application scenario corresponding to the power equipment in the power system to be modeled; according to the current application scenario, querying the corresponding relationship between the application scenario and the data acquisition frequency to obtain the data acquisition frequency corresponding to the current application scenario as the target data acquisition frequency of the power equipment; and obtaining the equipment data according to the target data acquisition frequency.

[0131] Among them, the current application scenario refers to the application scenario where the power equipment in the power system to be modeled is located at the current time. For example, the application scenarios of a wind turbine generator include stable power generation and strong wind warning.

[0132] Among them, the correspondence between the application scenario and the data collection frequency is used to represent the association information between the application scenario and the data collection frequency. For example, when the wind turbine generator is in the application scenario of stable power generation, the corresponding data collection frequency is once every 3 seconds; when the wind turbine generator is in the application scenario of strong wind warning, the corresponding data collection frequency is once per second.

[0133] Among them, the target data collection frequency refers to the data collection frequency corresponding to the power equipment.

[0134] Exemplarily, the server obtains the current application scenario corresponding to the power equipment in the power system to be modeled through the meteorological monitoring equipment of the power equipment installed in the power system to be modeled; then, the server queries the correspondence between the application scenario and the data collection frequency according to the current application scenario, obtains the data collection frequency corresponding to the current application scenario, and uses this data collection frequency as the target data collection frequency corresponding to the power equipment; according to the target data collection frequency, the server obtains the electrical quantity data, environmental data, and video image data of the power equipment as the equipment data of the power equipment.

[0135] In this embodiment, by determining the current application scenario of the power equipment and then accurately querying the corresponding ideal data collection frequency, it is possible to ensure that the collected data conforms to the actual business requirements, thereby reducing unnecessary data collection and transmission, avoiding the interference of redundant data, and being beneficial to improving the quality of data.

[0136] In an exemplary embodiment, the above step S104 of determining the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector specifically includes the following content: respectively input the electrical quantity data, environmental data, and video image data into the trained importance prediction model to obtain the first importance corresponding to the electrical quantity data, the second importance corresponding to the environmental data, and the third importance corresponding to the video image data; respectively determine the first initial weight corresponding to the first feature vector, the second initial weight corresponding to the second feature vector, and the third initial weight corresponding to the third feature vector according to the first importance, the second importance, and the third importance; perform normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector.

[0137] Among them, the importance prediction model refers to a network model used to determine the importance corresponding to the data, such as a random forest model.

[0138] Among them, the importance is used to represent the degree of influence corresponding to the data.

[0139] Among them, the first importance refers to the importance corresponding to the electrical quantity data.

[0140] Among them, the second importance refers to the importance corresponding to the environmental data.

[0141] Among them, the third importance refers to the importance corresponding to the video image data.

[0142] Among them, the first initial weight refers to the weight initially assigned to the first eigenvector.

[0143] Among them, the second initial weight refers to the weight initially assigned to the second eigenvector.

[0144] Among them, the third initial weight refers to the weight initially assigned to the third eigenvector.

[0145] Exemplarily, the server inputs the electrical quantity data, environmental data, and video image data into the trained importance prediction model respectively, and through the trained importance prediction model, obtains the first importance corresponding to the electrical quantity data, the second importance corresponding to the environmental data, and the third importance corresponding to the video image data; then, the server queries the corresponding relationship between the importance and the weight respectively according to the first importance, the second importance, and the third importance, and obtains the first initial weight corresponding to the first eigenvector, the second initial weight corresponding to the second eigenvector, and the third initial weight corresponding to the third eigenvector; then, the server performs normalization processing on the first initial weight, the second initial weight, and the third initial weight, and obtains the first target weight corresponding to the first eigenvector, the second target weight corresponding to the second eigenvector, and the third target weight corresponding to the third eigenvector; for example, the first initial weight, the second initial weight, and the third initial weight are 0.8, 0.6, and 0.6 respectively, and after normalization processing, the obtained first target weight, second target weight, and third target weight are 0.4, 0.3, and 0.3 respectively.

[0146] In this embodiment, through the trained importance prediction model, the key degrees of the electrical quantity, environment, and video image data can be accurately measured, so that the corresponding initial weights can be accurately determined, and combined with the normalization processing, the corresponding target weights can be more accurately determined, providing an accurate data basis for subsequent analysis.

[0147] In an exemplary embodiment, the above step S108 of constructing an initial three-dimensional model of the power system specifically includes the following: obtaining the wiring diagram of the power system; determining the topological information of the power system according to the wiring diagram; obtaining the shape information and position information of the power equipment, and constructing the initial three-dimensional model of the power system according to the shape information and position information of the power equipment and the topological information of the power system.

[0148] Among them, the wiring diagram is a graph used to represent the connection relationship, electrical path, and mutual cooperation mode of each power equipment in the power system.

[0149] Among them, the topological information is used to represent the connection relationship and network structure of each power equipment in the power system.

[0150] Among them, the shape information is used to represent the appearance contour of each power equipment in the power system.

[0151] Among them, the position information is used to represent the geographical location of each power equipment in the power system.

[0152] Exemplarily, the server obtains the identification information of the power system, and obtains the wiring diagram of the power system from the database according to the identification information of the power system; then, the server parses the wiring diagram of the power system to obtain the parsing information, and extracts the topological information of the power system from the parsing information; then, the server obtains the image information of the power equipment, and extracts the shape information of the power equipment from the image information of the power equipment; then, the server uses the global positioning system to determine the position information of the power equipment; then, the server constructs the initial three-dimensional model of the power system according to the shape information and position information of the power equipment and the topological information of the power system.

[0153] In this embodiment, the topological information is deduced from the actual wiring diagram of the power system, and then combined with the shape and position information of the equipment to construct a three-dimensional model, so that the actual physical layout of the power system can be highly restored, and then a virtual scene almost consistent with the site can be presented, avoiding the situation that it is difficult to intuitively display the spatial relationship with traditional two-dimensional drawings.

[0154] In an exemplary embodiment, in step S103 above, feature extraction processing is respectively performed on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data. The specific content is as follows: Taking the preprocessed electrical quantity data as the main data, and the preprocessed environmental data and the preprocessed video image data as auxiliary data, input them into the feature extraction model for feature extraction processing to obtain the first feature vector; taking the preprocessed environmental data as the main data, and the preprocessed electrical quantity data and the preprocessed video image data as auxiliary data, input them into the feature extraction model for feature extraction processing to obtain the second feature vector; taking the preprocessed video image data as the main data, and the preprocessed electrical quantity data and the preprocessed environmental data as auxiliary data, input them into the feature extraction model for feature extraction processing to obtain the third feature vector.

[0155] Among them, the main data can refer to the data with a larger weight.

[0156] Among them, the auxiliary data can refer to the data with a smaller weight.

[0157] Among them, the feature extraction model refers to a network model capable of performing feature extraction processing, such as a transformer model.

[0158] Exemplarily, the server takes the preprocessed electrical quantity data as the main data, and the preprocessed environmental data and the preprocessed video image data as auxiliary data, inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed electrical quantity data as the first feature vector; then, the server takes the preprocessed environmental data as the main data, and the preprocessed electrical quantity data and the preprocessed video image data as auxiliary data, inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed environmental data as the second feature vector; then, the server takes the preprocessed video image data as the main data, and the preprocessed electrical quantity data and the preprocessed environmental data as auxiliary data, inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed video image data as the third feature vector.

[0159] In this embodiment, by taking different types of data as the main data and the remaining data as auxiliary data for corresponding feature extraction processing, key features can be deeply explored from each dimension, and further, the first feature vector, the second feature vector, and the third feature vector are made more accurate, providing a basis for subsequent data processing.

[0160] In an exemplary embodiment, the power system monitoring method provided by this application further includes the training step of the trained equipment status prediction model, which specifically includes the following content: obtaining sample equipment data of sample power equipment in a sample power system; the sample equipment data at least includes sample electrical quantity data, sample environmental data, and sample video image data of the sample power equipment; respectively preprocessing the sample electrical quantity data, sample environmental data, and sample video image data to obtain preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data; respectively performing feature extraction processing on the preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data to obtain a first sample feature vector corresponding to the preprocessed sample electrical quantity data, a second sample feature vector corresponding to the preprocessed sample environmental data, and a third sample feature vector corresponding to the preprocessed sample video image data; determining a first target sample weight corresponding to the first sample feature vector, a second target sample weight corresponding to the second sample feature vector, and a third target sample weight corresponding to the third sample feature vector; according to the first target sample weight, the second target sample weight, and the third target sample weight, performing a summation process on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a target sample feature vector of the sample power equipment; inputting the target sample feature vector into the equipment status prediction model to be trained to obtain predicted status information corresponding to the sample power equipment; obtaining the actual status information corresponding to the sample power equipment, and iteratively training the equipment status prediction model to be trained according to the difference between the predicted status information and the actual status information to obtain the trained equipment status prediction model.

[0161] Among them, the sample power system refers to the power system used to train the equipment status prediction model to be trained.

[0162] Among them, the sample power equipment refers to the power equipment in the sample power system.

[0163] Among them, the sample equipment data at least includes sample electrical quantity data, sample environmental data, and sample video image data of the sample power equipment.

[0164] Among them, the sample electrical quantity data refers to the electrical quantity data of the sample power equipment.

[0165] Among them, the sample environmental data refers to the environmental data of the sample power equipment.

[0166] Among them, the sample video image data refers to the video image data of the sample power equipment.

[0167] Among them, the preprocessed sample electrical quantity data refers to the sample electrical quantity data after preprocessing.

[0168] Among them, the preprocessed sample environmental data refers to the sample environmental data after preprocessing.

[0169] Among them, the preprocessed sample video image data refers to the sample video image data after preprocessing.

[0170] Among them, the first sample feature vector refers to the feature vector corresponding to the preprocessed sample electrical quantity data.

[0171] Among them, the second sample feature vector refers to the feature vector corresponding to the preprocessed sample environmental data.

[0172] Among them, the third sample feature vector refers to the feature vector corresponding to the preprocessed sample video image data.

[0173] Among them, the first target sample weight refers to the weight finally assigned to the first sample feature vector.

[0174] Among them, the second target sample weight refers to the weight finally assigned to the second sample feature vector.

[0175] Among them, the third target sample weight refers to the weight finally assigned to the third sample feature vector.

[0176] Among them, the target sample feature vector refers to the feature vector obtained by summing the first sample feature vector, the second sample feature vector, and the third sample feature vector according to the first target sample weight, the second target sample weight, and the third target sample weight.

[0177] Among them, the predicted status information corresponding to the sample power equipment refers to the predicted value of the status information corresponding to the sample power equipment.

[0178] Among them, the actual status information corresponding to the sample power equipment refers to the actual value of the status information corresponding to the sample power equipment.

[0179] Exemplarily, in response to a model training instruction for a device state prediction model to be trained, the server obtains sample electrical quantity data, sample environmental data, and sample video image data of sample power devices in a sample power system from a database; then, the server preprocesses the sample electrical quantity data, sample environmental data, and sample video image data respectively to obtain preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data; then, the server performs feature extraction processing on the preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data respectively to obtain a first sample feature vector corresponding to the preprocessed sample electrical quantity data, a second sample feature vector corresponding to the preprocessed sample environmental data, and a third sample feature vector corresponding to the preprocessed sample video image data; then, the server determines a first target sample weight corresponding to the first sample feature vector, a second target sample weight corresponding to the second sample feature vector, and a third target sample weight corresponding to the third sample feature vector; then, the server performs a summation process on the first sample feature vector, the second sample feature vector, and the third sample feature vector according to the first target sample weight, the second target sample weight, and the third target sample weight to obtain a target sample feature vector of the sample power device; then, the server inputs the target sample feature vector into the device state prediction model to be trained to obtain predicted state information corresponding to the sample power device; then, the server obtains the actual state information corresponding to the sample power device from the database and obtains a loss value according to the difference between the predicted state information and the actual state information; then, the server adjusts the model parameters of the device state prediction model to be trained according to the loss value; then, the server retrains the device state prediction model with adjusted model parameters until the loss value obtained by the trained device state prediction model is less than the loss value threshold, then stops training, and uses the trained device state prediction model as the device state prediction model that has been trained.

[0180] In this embodiment, by pre-training the device state prediction model, it is convenient to predict the target state information corresponding to the power device after obtaining the target feature vector of the power device in actual application; moreover, the device state prediction model receives new data in each iteration and performs internal improvement and optimization of the model, which is convenient for more effective prediction and helps to improve the prediction accuracy of the device state prediction model.

[0181] In an exemplary embodiment, as Figure 2 shown, another power system monitoring method is provided. Taking the application of this method to the server side as an example, it includes the following steps:

[0182] Step S201, obtain the current application scenario corresponding to the power equipment in the power system to be modeled; according to the current application scenario, query the corresponding relationship between the application scenario and the data acquisition frequency, and obtain the data acquisition frequency corresponding to the current application scenario as the target data acquisition frequency corresponding to the power equipment.

[0183] Step S202, obtain device data according to the target data acquisition frequency: the device data at least includes electrical quantity data, environmental data, and video image data of the power equipment.

[0184] Step S203, respectively preprocess the electrical quantity data, environmental data, and video image data to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data.

[0185] Step S204, take the preprocessed electrical quantity data as the main data, and take the preprocessed environmental data and preprocessed video image data as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the first feature vector.

[0186] Step S205, take the preprocessed environmental data as the main data, and take the preprocessed electrical quantity data and preprocessed video image data as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the second feature vector.

[0187] Step S206, take the preprocessed video image data as the main data, and take the preprocessed electrical quantity data and preprocessed environmental data as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the third feature vector.

[0188] Step S207, respectively input the electrical quantity data, environmental data, and video image data into the trained importance prediction model to obtain the first importance corresponding to the electrical quantity data, the second importance corresponding to the environmental data, and the third importance corresponding to the video image data.

[0189] Step S208, respectively determine the first initial weight corresponding to the first feature vector, the second initial weight corresponding to the second feature vector, and the third initial weight corresponding to the third feature vector according to the first importance, the second importance, and the third importance.

[0190] Step S209, perform normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector.

[0191] Step S210: Sum the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the target feature vector of the power equipment.

[0192] Step S211: Input the target feature vector into the trained equipment status prediction model to obtain the target status information corresponding to the power equipment.

[0193] Step S212: Determine the visual feature information corresponding to the power equipment according to the target status information.

[0194] Step S213: Obtain the wiring diagram of the power system; determine the topological information of the power system according to the wiring diagram; obtain the shape information and position information of the power equipment, and construct the initial 3D model of the power system according to the shape information and position information of the power equipment and the topological information of the power system.

[0195] Step S214: Adjust the initial 3D model according to the visual feature information corresponding to the power equipment to obtain the target 3D model of the power system.

[0196] Step S215: Monitor the power system through the target 3D model.

[0197] In the above power system monitoring method, during the process of monitoring the power system, through a series of processes such as preprocessing, feature extraction processing, and model processing of various device data of the power equipment in the power system to be modeled, the target status information corresponding to the power equipment can be accurately obtained, so that the visual feature information corresponding to the power equipment can be more accurately determined, and combined with the initial 3D model of the power system, and then the target 3D model of the power system can be more accurately obtained, which is beneficial to improving the monitoring accuracy of the power system; moreover, this method avoids the defect that it is difficult to intuitively display the spatial relationship and real-time changes between power equipment in the power system by using a 2D interface for monitoring, resulting in low monitoring accuracy of the power system, and thus improves the monitoring accuracy of the power system.

[0198] In an exemplary embodiment, in order to more clearly illustrate the power system monitoring method provided by the embodiments of the present application, the following uses a specific embodiment to specifically describe the power system monitoring method. In one embodiment, as Figure 3As shown in the figure, the present application also provides a visualization monitoring method for a power system based on a time series mapping model. During the monitoring of the power system, first, real-time operation data is collected from the monitoring devices deployed in the power system, including electrical quantity data, environmental data, video images, and other data. Then, the collected multi-source data is preprocessed, including data cleaning, data calibration, and data synchronization, to ensure the accuracy and consistency of the data. Next, based on the geographical wiring diagram and equipment layout diagram of the power system, a three-dimensional model of the power system is constructed. Finally, the time series mapping model is used to map the fused comprehensive data set into the three-dimensional model, thereby realizing the three-dimensional visualization display of dynamic data. The specific contents are as follows:

[0199] Step 1, data collection. Collect real-time operation data from the monitoring devices deployed in the power system, including electrical quantity data, environmental data, video images, and other data.

[0200] Step 2, data preprocessing. Preprocess the collected multi-source data, including data cleaning, data calibration, and data synchronization, to ensure the accuracy and consistency of the data.

[0201] Step 3, three-dimensional modeling. Based on the geographical wiring diagram and equipment layout diagram of the power system, construct a three-dimensional model of the power system. The three-dimensional model contains information such as the spatial position, connection relationship, size, and shape of power equipment, and can truly reflect the physical structure of the power system.

[0202] Step 4, data mapping and visualization. Use the time series mapping model to map the fused comprehensive data set into the three-dimensional model, thereby realizing the three-dimensional visualization display of dynamic data.

[0203] Further, Step 1 specifically includes:

[0204] Collect real-time operation data from various monitoring devices (such as SCADA systems, PMU devices, infrared thermometers, video surveillance cameras, etc.) deployed in the power system. The data that can be collected includes but is not limited to the following categories:

[0205] (1) Electrical quantity data:

[0206] Voltage: Reflects the voltage levels at different nodes in the power system.

[0207] Current: Monitors the load conditions of transmission lines or equipment.

[0208] Power: Includes active power and reactive power, and is used to analyze the energy efficiency of equipment and the operation stability of the power grid.

[0209] (2) Environmental data:

[0210] Temperature and humidity: Reflect the working environment of the equipment and help identify potential faults caused by deteriorating environmental conditions.

[0211] Smoke concentration: Assist in identifying early signs of overheating or fire in the device.

[0212] (3) Video image data:

[0213] Obtain real-time images from the monitoring cameras of key devices (such as transformers, circuit breakers) for assisting in judging the device status and troubleshooting faults.

[0214] It should be noted that the data acquisition frequency of each monitoring device can be optimized according to the application scenario (for example, the PMU device samples 50 times per second, and the SCADA device samples once per second).

[0215] Furthermore, step two specifically includes:

[0216] The collected data usually contains problems such as noise, missing values, or asynchronous timestamps, and needs to be preprocessed to ensure the accuracy of subsequent analysis. The specific steps of preprocessing include:

[0217] (1) Data cleaning:

[0218] Remove or mark abnormal data based on threshold rules or anomaly detection algorithms, and delete duplicate data and data beyond the reasonable range (such as abnormally high or negative voltage values).

[0219] Fill in the missing data using interpolation or fitting methods.

[0220] (2) Data calibration and standardization:

[0221] Calibration: Unify the dimensions of multi-source data (such as unifying the temperature unit to degrees Celsius and the voltage to kilovolts).

[0222] Standardization: Normalize data with different units and magnitudes to make them have consistent dimensions.

[0223] (3) Data feature extraction:

[0224] Extract features from the original data to improve the data's expressive ability:

[0225] Among them, electrical quantity features:

[0226] First-order difference: Reflects the change rate of the electrical quantity.

[0227] Δx(t) = x(t) - x(t - Δt), Equation (2)

[0228] Fluctuation characteristics: Calculate the variance within a short-time window.

[0229] Among them, environmental quantity features:

[0230] Moving window average: Represents the short-term trend.

[0231] Extreme value detection: Record the maximum and minimum values of temperature and humidity.

[0232] Among them, video features:

[0233] Use image processing algorithms to extract features such as brightness and edge density of video frames.

[0234] (4) Data synchronization:

[0235] Align the data of different devices based on timestamps to ensure that all data sources are fused on the same time basis.

[0236] Since different data sources may have different sampling frequencies and timestamps, time and space synchronization are required:

[0237] Among them, time synchronization is to use interpolation or alignment methods to align the data with a unified time basis; for high-frequency data such as PMU data, direct sampling can be used; for low-frequency data such as environmental data, linear interpolation can be used.

[0238] Among them, space synchronization is that if data acquisition devices are distributed at different locations, their spatial information can be mapped to a unified power system topology through coordinate mapping methods.

[0239] Furthermore, step (4) specifically includes:

[0240] Process the data of multiple data sources collected and map them as the time series of the data sources of the time series data mapping model:

[0241] X(t) = [X 1 (t), X 2 (t),..., X n (t)], Equation (3)

[0242] Where X i (t) is the i-th eigenvalue at time t.

[0243] The real-time data provided by each sensor or monitoring device can form a time series. For example:

[0244] X 1 (t): Voltage time series, representing the voltage value at time t.

[0245] X 2 (t): Current time series, representing the current value at time t.

[0246] X 3 (t): Temperature time series, representing the temperature value at time t.

[0247] These data will be determined according to the actually collected monitoring data.

[0248] Further, Step 3 specifically includes:

[0249] Based on the equipment layout and wiring diagram of the power system, construct a 3D model to display the physical structure and connection relationship of the system:

[0250] (1) Model elements:

[0251] Simulate the real shapes (such as transformers, high-voltage circuit breakers) and spatial positions of the equipment.

[0252] Construct the line routes and topological structures of the power grid.

[0253] (2) Dynamic update:

[0254] Use real-time data to dynamically adjust the states in the model, such as the color, brightness, and transparency of the equipment.

[0255] (3) Model optimization:

[0256] Adopt lightweight modeling technology to reduce the computational load and improve the display performance.

[0257] Add a background map and geographical information to make the 3D model more practical.

[0258] Introduce a hierarchical display function (such as only displaying equipment in a specific area or of a specific type) to improve the operability of the system.

[0259] Further, Step 4 specifically includes:

[0260] (1) Multidimensional data mapping:

[0261] Perform multidimensional data mapping in the 3D model to achieve dynamic visualization display.

[0262] Through time series analysis of electrical quantity data (such as voltage, current, power, etc.) and environmental data (such as temperature, humidity, smoke concentration, etc.), establish a time series mapping function:

[0263]

[0264] where, a i is the weight coefficient, X i (t) is the time series of the i-th data source, t is the time variable, and n is the number of data sources.

[0265] This mapping function converts the time series information of various data sources into a unified comprehensive time series, thereby dynamically adjusting the states of the equipment in the 3D model.

[0266] The comprehensive data f(t) is the result obtained through weighted summation, which reflects the comprehensive situation of all input data sources. In 3D visualization, this value will be used to control the state of the device.

[0267] Electrical quantity data: The comprehensive value calculated based on parameters such as voltage and current determines whether the device is in a normal, overloaded, or faulty state.

[0268] Environmental data: Affects the operating state of the device according to environmental parameters such as temperature and humidity, and may adjust the appearance of the device (such as changing color or transparency).

[0269] Among them, a i The weight coefficient determines the influence degree of each data source on the comprehensive mapping result f(t). If voltage and current are the main monitoring factors, the weight coefficients a 1 and a 2 for voltage and current can be set to higher values; if environmental data (such as temperature and humidity) has less impact on the state of power equipment, the weight coefficients a 3 for temperature and humidity can be set to lower values.

[0270] The weight coefficient a i can be determined by the following methods:

[0271] Empirical values: Based on the analysis of historical data, infer the importance of different data sources to the system operating state.

[0272] Data analysis: Use statistical methods (such as correlation analysis, regression analysis, etc.) to calculate the influence degree of each data source on the target variable (such as device state, system security) to obtain the weight coefficient.

[0273] Expert knowledge: Adjust the weight coefficient by experts or engineers in the power system according to actual experience.

[0274] Among them, n refers to the number of data sources, and this value is the total number of all data sources in the model. The specific number depends on the number of devices deployed in the power system. For example, there may be multiple sensor data such as voltage, current, temperature, and humidity, and each sensor is used as a data source i.

[0275] Exemplarily, assume there are 3 data sources, namely voltage, current, and temperature:

[0276] X 1 (t): Voltage time series, representing the voltage value at time t.

[0277] X 2 (t): Current time series, representing the current value at time t.

[0278] X 3(t): Temperature time series, representing the temperature value at time t.

[0279] If we want to construct a comprehensive mapping model, we can assume:

[0280] Voltage (X 1 (t)) and current (X 2 (t)) are more important for the device state, so higher weight coefficients are assigned to them, such as a 1 = 0.4, a 2 = 0.4;

[0281] Temperature (X 3 (t)) has less impact on the device state, so a lower weight coefficient is assigned to it, such as a 3 = 0.2.

[0282] Then, the comprehensive mapping function f(t) can be expressed as:

[0283] f(t) = 0.4·X 1 (t) + 0.4·X 2 (t) + 0.2·X 3 (t), Equation (4)

[0284] In this formula:

[0285] X 1 (t), X 2 (t), X 3 (t) are the data collected from the sensors in real time.

[0286] The comprehensive value f(t) at each time point can be used to adjust the state of the power equipment in the 3D model (for example, the health status of the equipment is displayed through color or brightness changes). For example, if the value has a small difference from the standard value, the equipment is displayed in green, indicating normal operation; if the value has a large difference from the standard value, the equipment is displayed in red, indicating overload or failure.

[0287] (2) 3D model dynamic mapping:

[0288] According to the output results of the above time series mapping function, dynamically adjust the visual features such as color, brightness, and transparency of each power equipment in the 3D model. The color or brightness of the equipment corresponds to its electrical quantity data. For example:

[0289] Normal operation state: Green;

[0290] Overload state: Red;

[0291] Fault state: Gray.

[0292] Meanwhile, environmental data (such as temperature, humidity, etc.) is superimposed on the 3D model through color layers, enabling the monitoring personnel to observe the operating status of the power system and the surrounding environmental conditions simultaneously.

[0293] In the above-mentioned embodiment, during the process of monitoring the power system, through a series of processes such as preprocessing, feature extraction processing, and model processing of various device data of the power equipment in the power system to be modeled, the target status information corresponding to the power equipment can be accurately obtained. Thus, the visual feature information corresponding to the power equipment can be determined more accurately, and combined with the initial 3D model of the power system, the target 3D model of the power system can be obtained more accurately, which is beneficial to improving the monitoring accuracy of the power system. Moreover, this method uses the 3D model to monitor the power system, avoiding the defect that it is difficult to intuitively display the spatial relationship and real-time changes between power equipment in the power system by using the 2D interface monitoring method, resulting in a lower monitoring accuracy of the power system, and thus improving the monitoring accuracy of the power system. At the same time, a time series mapping model is introduced to perform spatio-temporal correlation analysis on real-time data from multiple data sources (electrical quantities, environmental data, etc.) and map it into a comprehensive dynamic data set. Through this model, the operating status of power equipment, environmental changes, and video monitoring can be dynamically and accurately visualized in 3D, providing more accurate monitoring information. Through the time series mapping model, based on multi-dimensional data such as real-time collected electrical quantities and environmental data, by methods such as dynamic weight adjustment and time series analysis, it can not only reflect the health status of the equipment in real time, but also automatically adapt and adjust the display effect according to the changes in the system state, enabling the system to quickly respond when the operating status of the power equipment changes and dynamically adjust the display parameters in the 3D model, such as equipment color, brightness, etc., so as to provide more accurate and timely feedback to the monitoring personnel. By introducing the time series mapping model, through the time series analysis of various data of the power system, multi-source information such as electrical quantity data, environmental data, and video monitoring data is fused to form a dynamic and real-time comprehensive mapping result, providing more accurate dynamic updates for 3D visualization. Through the integration of 3D visualization and multi-dimensional data, not only the status of power equipment is displayed, but also the equipment operating environment and video monitoring information can be displayed, providing a comprehensive perception of the system state. Through real-time data feedback and anomaly detection, using the time series mapping model, combined with anomaly detection rules and early warning mechanisms, it is ensured that the system can respond to the changes in the state of power equipment in real time, and potential faults can be detected and warned in time.

[0294] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0295] Based on the same inventive concept, an embodiment of the present application further provides a power system monitoring device for implementing the above-mentioned power system monitoring method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the power system monitoring device provided below can refer to the limitations on the power system monitoring method in the above text, and will not be repeated here.

[0296] In an exemplary embodiment, as Figure 4 shown, a power system monitoring device is provided, including: a data acquisition module 401, a data processing module 402, a feature extraction module 403, a weight determination module 404, a vector summation module 405, a state prediction module 406, an information determination module 407, a model processing module 408, and a system monitoring module 409, where:

[0297] The data acquisition module 401 is used to acquire device data of power equipment in the power system to be modeled: the device data at least includes electrical quantity data, environmental data, and video image data of the power equipment.

[0298] The data processing module 402 is used to preprocess the electrical quantity data, environmental data, and video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data.

[0299] The feature extraction module 403 is used to perform feature extraction processing on the preprocessed electrical quantity data, preprocessed environmental data, and preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data.

[0300] The weight determination module 404 is used to determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector.

[0301] The vector summation module 405 is configured to perform a summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight, so as to obtain the target feature vector of the power equipment.

[0302] The state prediction module 406 is configured to input the target feature vector into the trained equipment state prediction model to obtain the target state information corresponding to the power equipment.

[0303] The information determination module 407 is configured to determine the visual feature information corresponding to the power equipment according to the target state information.

[0304] The model processing module 408 is configured to construct an initial three-dimensional model of the power system, and perform an adjustment process on the initial three-dimensional model according to the visual feature information corresponding to the power equipment, so as to obtain the target three-dimensional model of the power system.

[0305] The system monitoring module 409 is configured to perform a monitoring process on the power system through the target three-dimensional model.

[0306] In an exemplary embodiment, the data acquisition module 401 is further configured to acquire the current application scenario corresponding to the power equipment in the power system to be modeled; query the corresponding relationship between the application scenario and the data acquisition frequency according to the current application scenario, and obtain the data acquisition frequency corresponding to the current application scenario as the target data acquisition frequency corresponding to the power equipment; acquire the equipment data according to the target data acquisition frequency.

[0307] In an exemplary embodiment, the weight determination module 404 is further configured to input the electrical quantity data, the environmental data, and the video image data into the trained importance prediction model respectively to obtain the first importance corresponding to the electrical quantity data, the second importance corresponding to the environmental data, and the third importance corresponding to the video image data; determine the first initial weight corresponding to the first feature vector, the second initial weight corresponding to the second feature vector, and the third initial weight corresponding to the third feature vector according to the first importance, the second importance, and the third importance respectively; perform a normalization process on the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the first feature vector, the second target weight corresponding to the second feature vector, and the third target weight corresponding to the third feature vector.

[0308] In an exemplary embodiment, the model processing module 408 is further configured to acquire the wiring diagram of the power system; determine the topological information of the power system according to the wiring diagram; acquire the shape information and the position information of the power equipment, and construct the initial three-dimensional model of the power system according to the shape information and the position information of the power equipment, and the topological information of the power system.

[0309] In an exemplary embodiment, the feature extraction module 403 is further configured to use the preprocessed electrical quantity data as the main data, and the preprocessed environmental data and the preprocessed video image data as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a first feature vector; use the preprocessed environmental data as the main data, and the preprocessed electrical quantity data and the preprocessed video image data as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a second feature vector; use the preprocessed video image data as the main data, and the preprocessed electrical quantity data and the preprocessed environmental data as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a third feature vector.

[0310] In an exemplary embodiment, the power system monitoring device further includes a model training module, configured to obtain sample device data of sample power devices in a sample power system; the sample device data at least includes sample electrical quantity data, sample environmental data, and sample video image data of the sample power devices; respectively preprocess the sample electrical quantity data, the sample environmental data, and the sample video image data to obtain preprocessed sample electrical quantity data, preprocessed sample environmental data, and preprocessed sample video image data; respectively perform feature extraction processing on the preprocessed sample electrical quantity data, the preprocessed sample environmental data, and the preprocessed sample video image data to obtain a first sample feature vector corresponding to the preprocessed sample electrical quantity data, a second sample feature vector corresponding to the preprocessed sample environmental data, and a third sample feature vector corresponding to the preprocessed sample video image data; determine a first target sample weight corresponding to the first sample feature vector, a second target sample weight corresponding to the second sample feature vector, and a third target sample weight corresponding to the third sample feature vector; according to the first target sample weight, the second target sample weight, and the third target sample weight, perform a summation process on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a target sample feature vector of the sample power device; input the target sample feature vector into a device state prediction model to be trained to obtain predicted state information corresponding to the sample power device; obtain actual state information corresponding to the sample power device, and perform iterative training on the device state prediction model to be trained according to the difference between the predicted state information and the actual state information to obtain a trained device state prediction model.

[0311] Each module in the above power system monitoring device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of a processor in a computer device in the form of hardware, or stored in a memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0312] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structural diagram may be as shown in Figure 5 . The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store electrical quantity data, environmental data, video image data, etc. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a power system monitoring method.

[0313] Those skilled in the art can understand that Figure 5 the structure shown in is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0314] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, it implements the steps in the above method embodiments.

[0315] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0316] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, it implements the steps in the above method embodiments.

[0317] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0318] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0319] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A power system monitoring method, characterized in that: The method comprises: Acquire equipment data of electric power equipment in the electric power system to be modeled: the equipment data at least includes electrical quantity data, environmental data and video image data of the electric power equipment; Preprocessing the electrical quantity data, the environmental data and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data and preprocessed video image data; Performing feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data, and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data, and a third feature vector corresponding to the preprocessed video image data; Determine a first target weight corresponding to the first eigenvector, a second target weight corresponding to the second eigenvector, and a third target weight corresponding to the third eigenvector; According to the first target weight, the second target weight and the third target weight, the first eigenvector, the second eigenvector and the third eigenvector are summed to obtain a target eigenvector of the electric power equipment; Inputting the target feature vector into the trained equipment state prediction model to obtain the target state information corresponding to the power equipment; Determining visual feature information corresponding to the electric power equipment according to the target state information; Constructing an initial three-dimensional model of the power system, and adjusting the initial three-dimensional model according to visual feature information corresponding to the power equipment to obtain a target three-dimensional model of the power system; The power system is monitored and processed through the target three-dimensional model.

2. The method according to claim 1, characterized in that The step of obtaining equipment data of electric equipment in the electric power system to be modeled includes: Obtaining the current application scenario corresponding to the power equipment in the power system to be modeled; According to the current application scenario, query the correspondence between the application scenario and the data acquisition frequency, and obtain the data acquisition frequency corresponding to the current application scenario as the target data acquisition frequency corresponding to the power equipment; The device data is acquired according to the target data collection frequency.

3. The method according to claim 1, characterized in that The determining of a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector includes: Inputting the electrical quantity data, the environmental data and the video image data into the trained importance prediction model respectively, to obtain a first importance corresponding to the electrical quantity data, a second importance corresponding to the environmental data and a third importance corresponding to the video image data; Determining a first initial weight corresponding to the first eigenvector, a second initial weight corresponding to the second eigenvector, and a third initial weight corresponding to the third eigenvector according to the first importance, the second importance, and the third importance respectively; The first initial weight, the second initial weight, and the third initial weight are normalized to obtain a first target weight corresponding to the first eigenvector, a second target weight corresponding to the second eigenvector, and a third target weight corresponding to the third eigenvector.

4. The method according to claim 1, characterized in that: The constructing of the initial three-dimensional model of the power system comprises: Obtaining a wiring diagram of the power system; Determining topological information of the power system according to the wiring diagram; The shape information and the position information of the electric power equipment are acquired, and an initial three-dimensional model of the electric power system is constructed according to the shape information and the position information of the electric power equipment and the topology information of the electric power system.

5. The method according to claim 1, characterized in that The performing feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data and the preprocessed video image data respectively to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data and a third feature vector corresponding to the preprocessed video image data comprises: The preprocessed electrical quantity data is used as main data, and the preprocessed environmental data and the preprocessed video image data are used as auxiliary data, and are input into a feature extraction model for feature extraction processing to obtain the first feature vector; The preprocessed environmental data is used as main data, and the preprocessed electrical quantity data and the preprocessed video image data are used as auxiliary data, and are input into the feature extraction model for feature extraction processing to obtain the second feature vector; The preprocessed video image data is used as main data, and the preprocessed electrical quantity data and the preprocessed environmental data are used as auxiliary data, which are input into the feature extraction model for feature extraction processing to obtain the third feature vector.

6. The method according to any one of claims 1 to 5, characterized in that: The trained device state prediction model is obtained by training in the following way: Acquire sample equipment data of sample power equipment in a sample power system; the sample equipment data at least includes sample electrical quantity data, sample environment data and sample video image data of the sample power equipment; Preprocessing the sample electrical quantity data, the sample environmental data and the sample video image data respectively to obtain preprocessed sample electrical quantity data, preprocessed sample environmental data and preprocessed sample video image data; Performing feature extraction processing on the preprocessed sample electrical quantity data, the preprocessed sample environmental data, and the preprocessed sample video image data respectively to obtain a first sample feature vector corresponding to the preprocessed sample electrical quantity data, a second sample feature vector corresponding to the preprocessed sample environmental data, and a third sample feature vector corresponding to the preprocessed sample video image data; Determine a first target sample weight corresponding to the first sample feature vector, a second target sample weight corresponding to the second sample feature vector, and a third target sample weight corresponding to the third sample feature vector; According to the first target sample weight, the second target sample weight and the third target sample weight, the first sample feature vector, the second sample feature vector and the third sample feature vector are summed to obtain a target sample feature vector of the sample power equipment; Inputting the target sample feature vector into the equipment state prediction model to be trained to obtain the prediction state information corresponding to the sample power equipment; The actual state information corresponding to the sample power equipment is obtained, and according to the difference between the predicted state information and the actual state information, the device state prediction model to be trained is iteratively trained to obtain the trained device state prediction model.

7. A power system monitoring device, characterized in that: The device comprises: A data acquisition module, used to acquire equipment data of electric equipment in the electric power system to be modeled: the equipment data at least includes electrical quantity data, environmental data and video image data of the electric equipment; A data processing module, used to preprocess the electrical quantity data, the environmental data and the video image data respectively to obtain preprocessed electrical quantity data, preprocessed environmental data and preprocessed video image data; a feature extraction module, used to perform feature extraction processing on the preprocessed electrical quantity data, the preprocessed environmental data and the preprocessed video image data, respectively, to obtain a first feature vector corresponding to the preprocessed electrical quantity data, a second feature vector corresponding to the preprocessed environmental data and a third feature vector corresponding to the preprocessed video image data; A weight determination module, used to determine a first target weight corresponding to the first feature vector, a second target weight corresponding to the second feature vector, and a third target weight corresponding to the third feature vector; a vector summing module, configured to sum the first eigenvector, the second eigenvector and the third eigenvector according to the first target weight, the second target weight and the third target weight, so as to obtain a target eigenvector of the electric power equipment; A state prediction module, used for inputting the target feature vector into the trained equipment state prediction model to obtain the target state information corresponding to the power equipment; An information determination module, used to determine visual feature information corresponding to the electric power equipment according to the target state information; A model processing module, used for constructing an initial three-dimensional model of the power system, and adjusting the initial three-dimensional model according to the visual feature information corresponding to the power equipment to obtain a target three-dimensional model of the power system; The system monitoring module is used to monitor the power system through the target three-dimensional model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images