Intelligent waveform measurement sensing method and system for power grid end edge total element analysis

Abstract

The invention discloses an intelligent waveform measurement sensing method and system for power grid end edge total element analysis, and the method comprises the steps: carrying out the local training of a federated network model on end side equipment through the real-time synchronous waveform data collected by local waveform measurement equipment; performing multi-device collaborative optimization on the model parameters obtained by local training of all the end-side devices on the side server, and issuing the model parameters to the end-side devices; and each end-side device updates a local federated network model according to the issued model parameters and carries out local waveform fault sensing, and waveform fault sensing is carried out on the power grid parameters uploaded by all the end-side devices on the side server to obtain a global waveform fault sensing result. The invention aims to solve the problem that an existing power grid monitoring system cannot recognize and respond to power grid disturbance under the conditions of high delay, low accuracy and lack of effective cooperation, and improves the real-time situation sensing capability of fault monitoring, alarm, prediction and the like of a novel power system.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to an intelligent waveform measurement and perception method and system for full-factor analysis at a power grid end. Background Art

[0002] With the increase in the proportion of new energy, the power grid faces increasingly complex operating environments and nonlinear dynamic disturbances, which makes traditional power grid monitoring technology gradually unable to meet the current power grid's needs for safe, stable and efficient operation. The existing power grid monitoring methods have the following problems: high latency, low accuracy and limited end-edge collaboration. Traditional power grid monitoring methods have high latency when processing and responding to power grid disturbances, and cannot meet the real-time response requirements for rapid disturbances; in complex power systems, the existing systems have insufficient accuracy in detecting and locating high-frequency disturbances, especially in the case of new energy access, accurate monitoring of power grid parameters is particularly important; the existing power grid monitoring system mainly relies on centralized computing and control, lacks efficient collaboration between end-side and edge-side devices, resulting in low data utilization and inability to achieve efficient and accurate situational awareness. Therefore, there is an urgent need for a new power grid monitoring technology that can perform full-factor data collection and local computing on the end side, and improve the ability to identify and respond to power grid disturbances through end-edge collaboration. Summary of the invention

[0003] Technical problem to be solved by the present invention: In view of the above-mentioned problems of the prior art, an intelligent waveform measurement and perception method and system for full-factor analysis at the power grid end are provided. The present invention aims to solve the problems of existing power grid monitoring systems in identifying and responding to power grid disturbances under the conditions of high latency, low accuracy and lack of effective collaboration, and to enhance the real-time perception capabilities of new power systems such as fault monitoring, alarm and prediction.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: An intelligent waveform measurement and perception method for full-factor analysis of power grid terminals includes the following steps: S1, on the end-side device, local training of the federated network model is performed using the real-time synchronous waveform data collected by the local waveform measurement device, and the model parameters obtained by the local training are uploaded to the edge-side server. The input data of the federated network model is the power grid parameters extracted based on the real-time synchronous waveform data, and the output result is the local situation awareness result; S2, receiving, on the end-side device, model parameters obtained by local training of all end-side devices on the edge-side server after multi-device collaborative optimization, the model parameters sent down; S3, the end-side device updates the local federated network model according to the model parameters sent by the edge server, and uses the federated network model with updated model parameters to perform local waveform fault perception on the real-time synchronized waveform data collected by the local waveform measurement device, and uploads the power grid parameters and local situation perception results to the edge server, so that the edge server uses the edge perception model to perform global waveform fault perception on the power grid parameters uploaded by all end-side devices to obtain global waveform fault perception results.

[0005] Optionally, the function expression for multi-device collaborative optimization in step S2 is: , , in, The model parameters after multi-device collaborative optimization, is the number of devices on the end side, is the total number of data samples, For end-side devices Model parameters before multi-device collaborative optimization, For end-side devices The number of samples of real-time synchronous waveform data.

[0006] Optionally, the federated network model includes a feature extractor w0 and a collaborative sensor w deployed on the end-side device. s The feature processor w1 deployed on the edge server and the feature extractor w0 are used to extract features of power grid parameters. h 0, feature processor w1 is used to process the feature h 0Further discovered features h 1. Collaborative Perceptron w s For features h 1 Perform perception prediction to obtain the predicted waveform measurement perception results.

[0007] Optionally, before step S1, the step further includes segmenting and deploying a feature extractor w0 and a feature processor w1: S101, using knowledge distillation to reduce the number of model parameters of a feature extraction network model composed of a multi-layer neural network; S102, define model segmentation ratio : , in, is the number of parameters of feature extractor w0, is the total number of parameters of the feature extractor w0 and the feature processor w1. The optimal model segmentation ratio is determined based on the computing power and memory limitations of the end device. : , in, and are the maximum and minimum model split ratios determined according to memory limitations, and They are the computing power of the client device and the edge server respectively, and they are: , , in, represents the infimum, is the model segmentation ratio The corresponding parameter storage size of the end-side device, The maximum value of available memory on the client device. and are the number of parameters of the first and last layers of the feature extraction network model respectively; S103, based on the optimal model segmentation ratio The feature extraction model is divided into a feature extractor w0 and a feature processor w1, and the feature extractor w0 is deployed on the end-side device, and the feature processor w1 is deployed on the end-side device.

[0008] Optionally, when local training of the federated network model is performed on the end-side device using the real-time synchronized waveform data in step S1, the objective function used in the local training is: , in, For end-side devices k The objective function is For end-side devices k The model parameters, For end-side devices k The number of samples of real-time synchronous waveform data, is the loss function used in local training, The federated network model is based on i Input signal and model parameters The waveform measurement perception results, For the i Input signal The corresponding ideal waveform measurement perception results.

[0009] Optionally, the terminal side device is also deployed with a terminal sensor w e , the end sensor w e Used to detect feature errors when communication failure occurs between the client device and the edge server. h0 performs perception prediction to obtain the output result of the federated network model; the local training also includes the feature extractor w0, the end sensor w e The auxiliary model is trained to update the feature extractor w0 and the end perceptron w e The model parameters and the loss function used when training the auxiliary model The function expression is: , in, is the sample size, The end sensor w e The waveform measurement perception result predicted for the i-th sample, is the true label value of the waveform measurement perception result of the i-th sample.

[0010] Optionally, the side perception model includes a feature processor w2 and a side perception unit w P The feature processor w2 is used to h 1. Mining out features h P2 , side sensor w P Used according to characteristics h P2 Predicting and generating side perception values y p As the global waveform measurement perception result.

[0011] In addition, the present invention also provides an intelligent waveform measurement and perception system for full-factor analysis of the power grid end-side, including an edge server and multiple edge devices respectively connected to the edge server, and the edge devices are programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis of the power grid end-side.

[0012] In addition, the present invention also provides a computer-readable storage medium, which stores a computer program or instruction, and the computer program or instruction is programmed or configured to execute the intelligent waveform measurement and perception method of the full-factor analysis of the power grid end through a processor.

[0013] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis at the power grid end through a processor.

[0014] Compared with the prior art, the present invention can mainly achieve the following beneficial effects: the method of the present invention includes using the real-time synchronous waveform data collected by the local waveform measurement device on the end-side device to perform local training of the federated network model, and performing multi-device collaborative optimization of the model parameters obtained by the local training of all end-side devices on the edge-side server and sending the model parameters to the end-side devices; each end-side device updates the local federated network model according to the sent model parameters and performs local waveform fault perception, and performs waveform fault perception on the power grid parameters uploaded by all end-side devices on the edge-side server to obtain global waveform fault perception results. The method of the present invention can solve the existing The power grid monitoring system identifies and responds to power grid disturbances in the presence of high latency, low accuracy, and lack of effective collaboration. It performs full-factor data collection and local computing on the end side, improves the ability to identify and respond to power grid disturbances through end-edge collaboration, improves the real-time and accuracy of disturbance monitoring in new power systems, optimizes the collaborative efficiency between end-side and edge-side devices, and solves the problems of data transmission delay and low efficiency of computing resource utilization through the deployment of lightweight federated networks. It also effectively solves the problem of rapid detection and high-precision identification of dynamic disturbances in new power systems while ensuring user data privacy and security. It can be used for situational awareness such as monitoring, alarming, and prediction of various faults in new power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the basic flow of the method of the embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of a workflow of a terminal device and an edge server in an embodiment of the present invention.

[0017] Figure 3 It is a schematic diagram of a framework structure of a terminal side device and a waveform measurement device thereof in an embodiment of the present invention.

[0018] Figure 4 A schematic diagram of the arrangement of a federated network model and an edge perception model in an embodiment of the present invention.

[0019] Figure 5 Schematic diagram of the system structure in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0021] like Figure 1 As shown, the intelligent waveform measurement and perception method for full-factor analysis at the power grid end of this embodiment includes the following steps: S1, on the end-side device, local training of the federated network model is performed using the real-time synchronous waveform data collected by the local waveform measurement device, and the model parameters obtained by the local training are uploaded to the edge-side server. The input data of the federated network model is the power grid parameters extracted based on the real-time synchronous waveform data, and the output result is the local situation awareness result; S2, receiving, on the end-side device, model parameters obtained by local training of all end-side devices on the edge-side server after multi-device collaborative optimization, the model parameters sent down; S3, the end-side device updates the local federated network model according to the model parameters sent by the edge server, and uses the federated network model with updated model parameters to perform local waveform fault perception on the real-time synchronized waveform data collected by the local waveform measurement device, and uploads the power grid parameters and local situation perception results to the edge server, so that the edge server uses the edge perception model to perform global waveform fault perception on the power grid parameters uploaded by all end-side devices to obtain global waveform fault perception results.

[0022] like Figure 2 As shown, this embodiment constructs an efficient end-side collaborative architecture through end-side devices and edge-side servers. The end-side devices use the real-time synchronous waveform data collected by the local waveform measurement device to perform local training of the federated network model and local waveform fault perception; the edge-side server performs multi-device collaborative optimization, and uses the edge-side perception model to perform global waveform fault perception to obtain global waveform fault perception results, thereby forming a power grid end-side full-factor analysis system. In one or more embodiments, the work of the power grid end-side full-factor analysis system includes: using the end-side device to collect the voltage and current signals of the power grid in real time, and calculating all-factor electrical parameters such as phase angle, frequency, amplitude and frequency change rate; deploying a lightweight federated network module in the end-side device, realizing intelligent analysis of the panoramic data of the power grid through local situational awareness, and uploading the data to the edge-side device after lossless compression; the edge-side device integrates the model parameters uploaded by multiple end-side devices, adopts a synchronous aggregation strategy for model training, optimizes the federated network model parameters and sends them to the end-side device, and finally realizes the end-side device to intelligently monitor the power grid parameters, as well as various applications such as fault location, oscillation monitoring, load prediction and inertia estimation. It fully supports power grid parameter measurement, intelligent analysis and multi-scenario applications, improving the intelligence level and real-time performance of power grid situation awareness.

[0023] Figure 3This is a schematic diagram of the framework structure of the end-side device and its waveform measurement device in this embodiment. In one or more embodiments, the edge in this embodiment includes a waveform measurement device and an edge device for full-factor analysis of the power grid end. Among them, the waveform measurement device is implemented using a field programmable gate array (FPGA), and the FPGA is used to perform high-frequency synchronous sampling of voltage and current signals, and calculate key electrical parameters such as phase angle, frequency, and amplitude in real time to ensure the accuracy and timeliness of data acquisition. Through real-time synchronous sampling, the field programmable gate array FPGA can process signals up to several hundred hertz to ensure the accuracy and timeliness of data acquisition.

[0024] The end-side device can be implemented using a platform such as Raspberry Pi or STM32. It can perform local analysis by running a federated network, reduce upload latency, and optimize model performance through binarization and knowledge distillation methods. Data is uploaded to the edge device or server via Ethernet or wireless communication modules for regional power grid situation awareness. Intelligent waveform measurement and perception equipment can calculate all-factor electrical parameters such as phase angle, frequency, amplitude, frequency change rate, broadband oscillation parameters, and fault occurrence time in real time, providing millisecond-level dynamic response to support power grid situation awareness and fault location. Finally, the data is transmitted to the edge server via Ethernet (TCP / IP) or wireless communication modules (such as Wi-Fi, 4G, 5G) for local power grid situation awareness.

[0025] In step S1 of this embodiment, when local training of the federated network model is performed on the end-side device using the real-time synchronized waveform data, the objective function used in the local training is: , in, For end-side devices k The objective function is For end-side devices k The model parameters, For end-side devices k The number of samples of real-time synchronous waveform data, is the loss function used in local training, The federated network model is based on i Input signal and model parameters The waveform measurement perception results, For the i Input signal The corresponding ideal waveform measurement perception result. Among them, the loss function used in local training can use the required loss function as needed, such as cross entropy loss.

[0026] As an optional implementation, the input grid parameters of the federated network model in this embodiment include: phase angle refers to the phase difference between voltage and current, which is used to judge the change of grid load and power factor; frequency refers to the change of grid frequency, and excessive frequency fluctuation indicates abnormal grid operation; amplitude refers to the amplitude of voltage and current, and voltage exceeding ±0.05pu is regarded as voltage over-limit; frequency change rate refers to the rate of frequency change, reflecting the change trend of grid disturbance. As an optional implementation, the local situation awareness result of the federated network model in this embodiment can be one of short-term frequency fluctuation, load, disturbance, broadband oscillation parameter, and fault information. Among them, broadband oscillation parameter refers to the frequency, damping rate and modal number of broadband oscillation of the power grid; fault information includes fault occurrence time and fault type, fault occurrence time refers to the start and end time of different fault events, and fault types include new energy power supply switching / high frequency switch / line short circuit / load shedding faults in the distribution network under distributed new energy access. The intelligent waveform measurement and perception equipment for full-factor analysis at the power grid end can provide dynamic parameters of the power grid at the millisecond level through real-time calculation of the end-side synchronous phasor measurement unit FPGA module. These parameters will be used for subsequent situational awareness and fault location. Federated learning enables devices to perform real-time model updates and adjustments based on local data through distributed learning and local training, without uploading all data to the cloud, which can reduce network bandwidth consumption and improve real-time performance.

[0027] In this embodiment, a device-side parallel collaborative processing mechanism is implemented through the device-side and the edge-side server. By optimizing the distributed training and parallel computing of the equipment, the efficiency and real-time performance of the power grid situation awareness are achieved. The device-side uses real-time power grid measurement data for feature extraction and local model training. The edge-side device completes the optimization and update of the global model by receiving the model parameters uploaded by the device-side, and ensures the parameter consistency and convergence of the global model through the synchronization aggregation strategy. This method significantly optimizes the utilization rate of distributed computing resources, has the characteristics of strong computing performance and resource friendliness, and provides strong technical support for real-time status perception and fault location in dynamic power grid environments.

[0028] On the edge server, all data uploaded by the end-side devices will be integrated, summarized and used for local training. On the edge device, multi-device collaborative optimization is performed based on the uploaded end-side data, and the optimized model parameters are sent to the end-side device. Through this end-edge collaborative processing mechanism, the system can maintain efficient state perception capabilities and ensure that the equipment always maintains the best parameter detection and fault location capabilities in a dynamic power grid environment. In order to ensure the consistency and global convergence of the end-side model update, a synchronous aggregation strategy is adopted for the end-side model and the edge model. Synchronous aggregation is a global model update method that requires all devices to complete local training before performing global model parameter aggregation. The synchronous aggregation strategy used in the multi-device collaborative optimization in step S2 of this embodiment has a function expression as follows: , , in, The model parameters after multi-device collaborative optimization, is the number of devices on the end side, is the total number of data samples, For end-side devices Model parameters before multi-device collaborative optimization, For end-side devices The number of samples of real-time synchronous waveform data.

[0029] like Figure 4 As shown, the federated network model in this embodiment includes a feature extractor w0 and a collaborative sensor w deployed on the end-side device. s The feature processor w1 deployed on the edge server and the feature extractor w0 are used to extract features of power grid parameters. h 0, feature processor w1 is used to process the feature h 0Further discovered features h 1. Collaborative Perceptron w s For features h 1. Perform perception prediction to obtain the predicted waveform measurement perception result. Figure 4 As shown, in this embodiment, an end sensor w is also deployed on the end-side device. e , the end sensor w e Used to detect feature errors when communication failure occurs between the client device and the edge server. h 0 performs perception prediction to obtain the output result of the federated network model; the local training also includes the feature extractor w0, the end sensor w e The auxiliary model is trained to update the feature extractor w0 and the end perceptron w e The model parameters and the loss function used when training the auxiliary model The function expression is: , in, is the sample size, The end sensor w e For The waveform measurement perception result obtained by sample prediction is For the The true label value of the waveform measurement perception result of each sample.

[0030] In this embodiment, the federated network model is divided into a main model (feature extractor w0 deployed on the end-side device, collaborative sensor w s and feature processor w1 deployed on the edge server) and auxiliary models (feature extractor w0 and end sensor w deployed on the end device) e ), and run them on the edge server and the end device respectively, which not only realizes the end-edge collaborative training, but also significantly optimizes the utilization of distributed computing resources. The end device first uses the power grid data x As input, the initial characteristic representation is obtained by phasor calculation method h 0= f ( w 0, x ). Then, the feature processor in the side server w 1 pair means further digging and getting h 1= f ( w 1, h 0). Based on this, the main model generates a predicted output y s = f (w s , h 1) At the same time, the auxiliary model generates local prediction perception directly on the edge device y e = f ( w e , h 0). At the same time, the side server can target the feature processor w 1 parameters for further feature perception w 2 and side sensor w P Get the side model perception y p Compared with the traditional model segmentation training method, the edge-to-edge parallel coordinated processing design has the following two major innovations: 1) Reduced communication overhead: By combining gradient calculation with w1. Decoupling: The edge server does not need to return the back-propagation gradient to the end-side device, which significantly reduces the communication volume and saves about a quarter of the communication overhead. 2) Parallel computing acceleration: The edge and end parallel coordination processing allows the feature extractor w0 of the end-side device and the gradient of the edge server w1 to be calculated simultaneously, thus shortening the time required for back-propagation by nearly half.

[0031] like Figure 4 As shown, the side perception model in this embodiment includes a feature processor w2 and a side perception unit w P The feature processor w2 is used to h 1. Mining out features h P2 , side sensor w P Used according to characteristics h P2 Predicting and generating side perception values y p Taking the global waveform measurement perception result as the training result, the side perception model is trained for the feature extractor w0, feature processor w1, feature processor w2 and side perception. w P The end-to-end network model is trained, and the specific loss function and training method can be used as required, which will not be described in detail here. e , collaborative sensor w s , side sensor w P All of them are existing publicly known network models.

[0032] The implementation of the federated network model is mainly divided into four parts: model construction and initialization, local training, global optimization, and model distribution. Model construction and initialization: Build a lightweight federated network suitable for the computing power of the end-side, use binarization and knowledge distillation methods to reduce the number of end-side model parameters; divide the end-side and edge-side models according to the end-side device resource conditions, and determine the segmentation ratio , to achieve model compression. Local training: The end-side device uses local data to train the feature extractor and uploads the intermediate parameters to the edge server. Global optimization: The edge server receives the update results of all end-side devices and updates the global model through synchronous aggregation. Model delivery: The edge server delivers the optimized model to the end-side device to start a new round of training. According to the computing power, memory limitations and communication conditions of the end-side device, the complete deep learning model needs to be split into two parts for model segmentation to initialize the model. The model is divided into two parts: the end-side model part and the edge model part. End-side model part: deployed on the end-side device for feature extraction and preliminary calculation; edge model part: deployed on the edge server for reasoning tasks with high computational complexity.

[0033] In this embodiment, before step S1, the process further includes segmenting and deploying a feature extractor w0 and a feature processor w1: S101, using knowledge distillation to reduce the number of model parameters of a feature extraction network model composed of a multi-layer neural network; S102, define model segmentation ratio : , in, is the number of parameters of feature extractor w0, is the total number of parameters of the feature extractor w0 and the feature processor w1. The optimal model segmentation ratio is determined based on the computing power and memory limitations of the end device. : , in, and are the maximum and minimum model split ratios determined according to memory limitations, and They are the computing power of the client device and the edge server respectively, and they are: , , in, represents the infimum, is the model segmentation ratio The corresponding parameter storage size of the end-side device, The maximum value of available memory on the client device. and are the number of parameters of the first and last layers of the feature extraction network model respectively; the model segmentation ratio The parameter storage size of the corresponding end-side device is the product of the storage size of the model parameters and the number of parameters. The number of layers of the feature extraction network model; S103, based on the optimal model segmentation ratio The feature extraction model is divided into a feature extractor w0 and a feature processor w1, and the feature extractor w0 is deployed on the end-side device, and the feature processor w1 is deployed on the end-side device.

[0034] In summary, the method of this embodiment includes the following features: Lightweight federated network design and optimization: A lightweight federated network (Federated Learning) module is deployed for local data analysis and situational awareness. On the device side, the collected power grid panoramic information (phase angle, frequency, amplitude, frequency change rate, broadband oscillation parameters, fault occurrence time, etc.) is input into the federated network module for local situational awareness. Federated learning enables the device to perform real-time model updates and adjustments based on local data through distributed learning and local training, without uploading all data to the cloud, which can reduce network bandwidth consumption and improve real-time performance. End-side collaborative processing and optimization: After the end-side device performs lossless compression on the data, it uploads the data to the edge device through the wireless communication network. The edge device aggregates the data from multiple end-side devices and trains and optimizes the local model based on these data. Through local training, the edge device can update the parameters of the federated network module and send these optimized models to the end-side device, thereby improving the power grid status monitoring and fault location capabilities of the entire system. Data security and sharing mechanism: In this system, each end-side device can independently collect and process data and upload part of the data to the edge-side device. The edge-side device trains the model based on the uploaded data and optimizes the data transmission efficiency through lossless data compression technology. All end-side devices in the system share the updated federal network model, but the data of each device is retained locally to ensure data privacy and security. The full-factor analysis method and system of the power grid end-edge can dynamically respond to changes in the power grid state through full-factor power grid information collection and real-time analysis, improve the level of intelligent monitoring of the power grid, and realize applications such as power grid disturbance detection, load forecasting, inertia estimation, fault location and oscillation monitoring.

[0035] like Figure 5As shown, this embodiment also provides an intelligent waveform measurement and perception system for full-factor analysis of power grid end-side, including an edge server and multiple edge devices connected to the edge server respectively, and the edge devices are programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis of power grid end-side. Among them, the end-side device: the end-side is designed with an intelligent waveform measurement and perception device, and the device includes an end-side synchronous phasor measurement unit (FPGA) and an end-side device (Raspberry Pi, microcontroller STM32 and other series of high-performance edge computing modules), which is used to collect information such as voltage and current of the power grid in real time, calculate all-factor electrical parameters such as phase angle, frequency, amplitude, and upload the processed data to the edge device through a wireless network. The built-in lightweight federated network module of the end-side device can perform local situation awareness and model update. Edge server: The edge device is responsible for receiving data from multiple end-side devices, integrating and summarizing the data, and training local models. The edge device sends the optimized lightweight federated network parameters to the end-side device to ensure the coordinated operation of the entire system. On the edge server, all data uploaded by the end-side devices will be integrated, summarized and used for local training. On the edge device, based on the uploaded end-side data, the edge system optimizes the model through multiple rounds of training and sends the updated lightweight federated network parameters to the end-side device. Through this end-edge collaborative processing mechanism, the system can maintain efficient state perception capabilities and ensure that the device always maintains the best parameter detection and fault location capabilities in a dynamic power grid environment. The end-side device deploys a feature extraction module to collect voltage, current and other signals of the power grid in real time and complete feature extraction. The end-side device uses local data for local training and updates the parameters of the end-side model once a day or a week. The end-side device uploads the extracted intermediate parameters and some model update parameters to the edge device. On the edge device, a classification or perception module is deployed to receive the parameters uploaded by the end-side device and complete the remaining computing tasks. At the same time, the model updates from multiple end-side devices are aggregated to optimize the global model and send the updated model parameters through the synchronization strategy. The intelligent waveform measurement and perception system for full-factor analysis at the power grid end of this embodiment uses synchronous information to collect panoramic information of power grid voltage and current at the end; locally calculates all-factor electrical parameters including phase angle, frequency, amplitude and frequency change rate; deploys a lightweight federated network in the equipment data management module; inputs the panoramic information into the lightweight federated network module and performs local situation awareness; the end-side device losslessly compresses the data and uploads it to the edge device or edge server; the edge device collects all the end-side information, performs local training and updates the parameters of the lightweight federated network module and sends it to the end-side device for new applications such as power grid parameter detection, fault location, oscillation monitoring, load forecasting, inertia estimation, etc. The present invention aims to enhance the real-time situation perception capabilities of new power system disturbance monitoring, alarm and forecasting, and solve the problem of high latency and low accuracy in identifying new power grid disturbances under new power systems.

[0036] In addition, this embodiment also provides a computer-readable storage medium, in which a computer program or instruction is stored. The computer program or instruction is programmed or configured to execute the intelligent waveform measurement and perception method of the full-factor analysis of the power grid end through a processor.

[0037] In addition, this embodiment also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis at the power grid end through a processor.

[0038] Those skilled in the art should understand that the technical solutions provided by the embodiments of the present invention may be in the form of methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that instructions executed by the processor of a computer or other programmable data processing device generate instructions for implementing the functions in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0039] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent waveform measurement and perception method for full-factor analysis at the power grid end, characterized in that: The steps include: S1, on the end-side device, local training of the federated network model is performed using the real-time synchronous waveform data collected by the local waveform measurement device, and the model parameters obtained by the local training are uploaded to the edge-side server. The input data of the federated network model is the power grid parameters extracted based on the real-time synchronous waveform data, and the output result is the local situation awareness result; S2, receiving, on the end-side device, model parameters obtained by local training of all end-side devices on the edge-side server after multi-device collaborative optimization, the model parameters sent down; S3, the end-side device updates the local federated network model according to the model parameters sent by the edge server, and uses the federated network model with updated model parameters to perform local waveform fault perception on the real-time synchronized waveform data collected by the local waveform measurement device, and uploads the power grid parameters and local situation perception results to the edge server, so that the edge server uses the edge perception model to perform global waveform fault perception on the power grid parameters uploaded by all end-side devices to obtain global waveform fault perception results.

2. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 1 is characterized in that: The function expression for multi-device collaborative optimization in step S2 is: ， ， in, The model parameters after multi-device collaborative optimization, is the number of devices on the end side, is the total number of data samples, For end-side devices Model parameters before multi-device collaborative optimization, For end-side devices The number of samples of real-time synchronous waveform data.

3. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 1 is characterized in that: The federated network model includes a feature extractor w0 and a collaborative sensor w deployed on the end-side device. s The feature processor w1 deployed on the edge server and the feature extractor w0 are used to extract features of power grid parameters. h 0, feature processor w1 is used to process the feature h 0Further discovered features h 1. Collaborative Perceptron w s For features h 1 Perform perception prediction to obtain the predicted waveform measurement perception results.

4. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 3 is characterized in that: Step S1 also includes the segmentation deployment feature extractor w0 and feature processor w1: S101, using knowledge distillation to reduce the number of model parameters of a feature extraction network model composed of a multi-layer neural network; S102, define model segmentation ratio : ， in, is the number of parameters of feature extractor w0, is the total number of parameters of the feature extractor w0 and the feature processor w1. The optimal model segmentation ratio is determined based on the computing power and memory limitations of the end device. : , in, and are the maximum and minimum model split ratios determined according to memory limitations, and They are the computing power of the client device and the edge server respectively, and they are: ， ， in, represents the infimum, is the model segmentation ratio The corresponding parameter storage size of the end-side device, The maximum value of available memory on the client device. and are the number of parameters of the first and last layers of the feature extraction network model respectively; S103, based on the optimal model segmentation ratio The feature extraction model is divided into a feature extractor w0 and a feature processor w1, and the feature extractor w0 is deployed on the end-side device, and the feature processor w1 is deployed on the end-side device.

5. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 3 is characterized in that: When local training of the federated network model is performed on the end-side device using real-time synchronized waveform data in step S1, the objective function used in the local training is: ， in, For end-side devices k The objective function is For end-side devices k The model parameters, For end-side devices k The number of samples of real-time synchronous waveform data, is the loss function used in local training, The federated network model is based on i Input signal and model parameters The waveform measurement perception results, For the i Input signal The corresponding ideal waveform measurement perception results.

6. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 5 is characterized in that: The terminal side device is also deployed with a terminal sensor w e , the end sensor w e Used to detect feature errors when communication failure occurs between the client device and the edge server. h 0 performs perception prediction to obtain the output result of the federated network model; the local training also includes the feature extractor w0, the end sensor w e The auxiliary model is trained to update the feature extractor w0 and the end perceptron w e The model parameters and the loss function used when training the auxiliary model The function expression is: ， in, is the sample size, The end sensor w e The waveform measurement perception result predicted for the i-th sample, is the true label value of the waveform measurement perception result of the i-th sample.

7. The intelligent waveform measurement and perception method for full-factor analysis of power grid end edges according to claim 1 is characterized in that: The side perception model includes a feature processor w2 and a side perception w P The feature processor w2 is used to h 1. Mining out features h P2 , side sensor w P Used according to characteristics h P2 Predicting and generating side perception values y p As the global waveform measurement perception result.

8. An intelligent waveform measurement and perception system for full-factor analysis of power grid edge, comprising an edge server and a plurality of edge devices respectively connected to the edge server, characterized in that: The end-side device is programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis of the power grid end-side as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program or instruction stored therein, characterized in that: The computer program or instruction is programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis at the power grid end as described in any one of claims 1 to 7 through a processor.

10. A computer program product comprising a computer program or instructions, characterized in that The computer program or instruction is programmed or configured to execute the intelligent waveform measurement and perception method for full-factor analysis at the power grid end as described in any one of claims 1 to 7 through a processor.

Images