Power equipment state monitoring method based on Beidou high-precision time service and related equipment

Abstract

The invention discloses a power equipment state monitoring method based on Beidou high-precision time service and related equipment, and relates to the technical field of Beidou satellites.The power equipment state monitoring method based on Beidou high-precision time service comprises the steps that Beidou satellite signals are obtained, and real-time ionosphere data are collected; based on the Beidou satellite signal and the ionosphere real-time data, ionosphere delay compensation is carried out according to a preset dynamic compensation model, a high-precision Beidou time service signal is determined, and accurate synchronization of the multi-source data of the power equipment is realized according to the high-precision Beidou time service signal; the method comprises the steps of extracting feature information of real-time operation parameters of power equipment, obtaining global model knowledge of preset edge nodes in combination with a federal learning mechanism, analyzing and judging the state of the power equipment in real time according to a pre-trained lightweight neural network model, and obtaining real-time monitoring information of the health degree of the power equipment, the lightweight neural network model is obtained by performing parameter compression training through channel pruning and quantitative perception training.

Description

Technical Field

[0001] The present application relates to the field of Beidou satellite technology, and in particular to a method and related equipment for monitoring the status of electric power equipment based on Beidou high-precision timing. Background Art

[0002] In recent years, with the increase in the load of power grids and power equipment, the real-time monitoring and maintenance of power equipment status has become increasingly important. Traditional power equipment status monitoring methods often rely on ground monitoring stations or local sensors, with low data acquisition and transmission efficiency, making it difficult to achieve real-time status tracking of large-scale power equipment. In addition, since power equipment operates in complex environments (such as ionospheric delay, multipath effects, etc.), traditional timing methods have insufficient accuracy, which limits the accuracy of monitoring results.

[0003] Therefore, how to accurately obtain the real-time operating status information of power equipment and realize real-time synchronization, transmission and analysis of data has become a key technical problem in the current power system operation and maintenance and intelligent scheduling. Summary of the invention

[0004] The main purpose of this application is to provide a method and related equipment for monitoring the status of power equipment based on Beidou high-precision timing, aiming to solve the technical problem of difficulty in accurately obtaining real-time operating status information of power equipment.

[0005] To achieve the above purpose, the present application proposes a method for monitoring the status of power equipment based on Beidou high-precision timing, the method comprising: Obtain Beidou satellite signals and collect real-time ionosphere data; Based on the Beidou satellite signal and the real-time ionospheric data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved based on the high-precision Beidou timing signal; The characteristic information of the real-time operating parameters of the power equipment is extracted, and the global model knowledge of the preset edge nodes is obtained in combination with the federated learning mechanism. The state of the power equipment is analyzed and judged in real time according to the pre-trained lightweight neural network model to obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

[0006] In one embodiment, the step of performing ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model to determine a high-precision Beidou timing signal includes: Analyze the Beidou satellite signal to determine the Beidou satellite elevation angle; Analyzing the real-time ionospheric data to determine the ionospheric electron density profile; The ionospheric height range is divided into a plurality of discrete height layers, the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius are input into a preset receiver vertical ionospheric delay field model, and the delay compensation amount corresponding to each height layer is discretized and calculated; Accumulate the delay compensation values ​​of all altitude layers to obtain the real-time delay compensation value of the ionosphere; Based on the real-time ionospheric delay compensation amount, ionospheric delay compensation is performed on the Beidou satellite signal to determine a high-precision Beidou timing signal.

[0007] In one embodiment, the step of dividing the ionospheric height range into a plurality of discrete height layers, inputting the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius into a preset receiver vertical ionospheric delay field model, and discretizing and calculating the delay compensation amount corresponding to each height layer comprises: Obtain historical ionospheric data, corresponding historical satellite elevation angles, and corresponding timing error data to construct a training data set; Inputting the training data set into the time series feature extraction network to be trained for training, and extracting key time series features based on the nonlinear mapping relationship between the spatiotemporal variation of the ionosphere and the timing error; According to the key timing characteristics, a correction coefficient is determined by the network weight obtained through training; Based on the correction coefficient, the vertical ionospheric delay field model of the receiver is adaptively calibrated.

[0008] In one embodiment, the step of extracting the characteristic information of the real-time operating parameters of the power equipment, acquiring the global model knowledge of the preset edge node in combination with the federated learning mechanism, and analyzing and judging the power equipment state in real time according to the pre-trained lightweight neural network model to obtain the real-time monitoring information of the equipment health includes: Obtain historical status data of power equipment and corresponding fault records; The historical status data of the power equipment is input into the neural network model to be trained to analyze and judge the equipment status, and obtain equipment health prediction information, wherein the high-precision floating-point weights are converted into low-precision fixed-point numbers through quantization perception training during the equipment status analysis and judgment; Calculate the contribution of each channel of the neural network model according to the activation value and gradient in the process of analyzing and judging the status of the device, and prune the channels whose contribution is lower than the preset warning threshold; Comparing the equipment health prediction information with the fault record to determine a prediction difference value; If the predicted difference value does not meet the preset prediction allowable error range, then return to the step of inputting the historical status data of the power equipment into the neural network model to be trained for equipment status analysis and judgment; if the predicted difference value meets the preset prediction allowable error range, then output the trained lightweight neural network model.

[0009] In one embodiment, the step of extracting characteristic information of the real-time operating parameters of the power equipment, acquiring the global model knowledge of the preset edge node in combination with the federated learning mechanism, and analyzing and judging the power equipment state in real time according to the pre-trained lightweight neural network model to obtain the real-time monitoring information of the equipment health includes: Extracting characteristic information of real-time operating parameters of the electric power equipment and constructing a real-time characteristic vector; Encrypt the parameters of the local lightweight neural network model and upload them to the aggregation server, and obtain the global model parameters returned by the aggregation server; The real-time feature vector is input into a lightweight neural network model using the global model parameters to analyze and judge the state of the power equipment in real time and determine the real-time monitoring information of the health of the equipment.

[0010] In one embodiment, the step of obtaining real-time monitoring information of the health of the equipment includes: Extract key alarm data from real-time monitoring information and compress it into short message data packets; Send short message data packets to the preset operation and maintenance warning equipment in priority preemptive mode through the preset Beidou channel; If the transmission fails, the short message data packet is automatically retransmitted within a preset time interval based on an exponential backoff algorithm.

[0011] In addition, to achieve the above purpose, the present application also proposes a power equipment status monitoring device based on Beidou high-precision timing, and the power equipment status monitoring device based on Beidou high-precision timing includes: Acquisition module, used to acquire Beidou satellite signals and collect real-time ionospheric data; A synchronization module, for performing ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model, determining a high-precision Beidou timing signal, and realizing accurate synchronization of multi-source data of power equipment according to the high-precision Beidou timing signal; The monitoring module is used to extract the characteristic information of the real-time operating parameters of the power equipment, obtain the global model knowledge of the preset edge nodes in combination with the federated learning mechanism, analyze and judge the status of the power equipment in real time according to the pre-trained lightweight neural network model, and obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

[0012] In addition, to achieve the above-mentioned purpose, the present application also proposes a power equipment status monitoring device based on Beidou high-precision timing, the device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the power equipment status monitoring method based on Beidou high-precision timing as described above.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the power equipment status monitoring method based on Beidou high-precision timing are implemented as described above.

[0014] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the power equipment status monitoring method based on Beidou high-precision timing as described above.

[0015] One or more technical solutions proposed in this application have at least the following technical effects: Compared with the related technologies, the traditional power equipment status monitoring method often relies on ground monitoring stations or local sensors, and the data acquisition efficiency and transmission efficiency are low, making it difficult to achieve real-time status tracking of large-scale power equipment. In addition, since the power equipment operates in a complex environment (such as ionospheric delay, multipath effect, etc.), the traditional timing method has the problem of insufficient accuracy, which limits the accuracy of the monitoring results. In comparison, the present application obtains Beidou satellite signals and collects real-time ionospheric data; based on the Beidou satellite signals and the real-time ionospheric data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved based on the high-precision Beidou timing signal; the characteristic information of the real-time operating parameters of the power equipment is extracted, and the global model knowledge of the preset edge nodes is obtained by combining the federated learning mechanism, and the power equipment status is analyzed and judged in real time according to the pre-trained lightweight neural network model, and the real-time monitoring information of the health of the power equipment is obtained, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization perception training. It is understandable that this application solves the problem of insufficient timing accuracy in traditional power equipment status monitoring methods, ensuring accurate data synchronization and high reliability. Beidou timing synchronization can efficiently integrate multi-source data from different devices, realize real-time transmission and analysis, and help the power system quickly discover and respond to problems. The lightweight neural network model can perform real-time analysis based on remote data, and combined with the mechanism of federated learning, it can quickly extract key feature information to provide intelligent decision-making support for power maintenance and scheduling. Real-time monitoring and analysis of equipment status helps to optimize energy scheduling and improve power system efficiency and reliability. By predicting equipment failures and performing repairs in advance, the cost of overhaul and replacement of power equipment can be reduced and asset life can be increased. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 A flowchart of the first embodiment of the method for monitoring the state of electric power equipment based on Beidou high-precision timing is provided in this application; Figure 2 A flow chart of Embodiment 2 of the method for monitoring the state of electric power equipment based on Beidou high-precision timing is provided for this application; Figure 3 A brief flow chart of the power equipment status monitoring method based on Beidou high-precision timing provided in this application; Figure 4 This is a schematic diagram of the module structure of the power equipment status monitoring device based on Beidou high-precision timing according to an embodiment of the present application; Figure 5 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the method for monitoring the status of electric power equipment based on Beidou high-precision timing in an embodiment of the present application.

[0019] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0021] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0022] The main solutions of the embodiments of this application are: Obtain Beidou satellite signals and collect real-time ionosphere data; Based on the Beidou satellite signal and the real-time ionospheric data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved based on the high-precision Beidou timing signal; The characteristic information of the real-time operating parameters of the power equipment is extracted, and the global model knowledge of the preset edge nodes is obtained in combination with the federated learning mechanism. The state of the power equipment is analyzed and judged in real time according to the pre-trained lightweight neural network model to obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

[0023] In this embodiment, the present application takes the power equipment status monitoring device based on Beidou high-precision timing as the execution subject. For the convenience of expression, it is specifically described below as "device".

[0024] Since existing technologies often rely on ground monitoring stations or local sensors, data acquisition and transmission efficiency are low, making it difficult to achieve real-time status tracking of large-scale power equipment. In addition, since power equipment operates in complex environments (such as ionospheric delay, multipath effects, etc.), traditional timing methods have insufficient accuracy, which limits the accuracy of monitoring results.

[0025] This application provides a solution to the problem of insufficient timing accuracy in traditional power equipment status monitoring methods, ensuring accurate data synchronization and high reliability. Beidou timing synchronization can efficiently integrate multi-source data from different devices, realize real-time transmission and analysis, and help the power system quickly discover and respond to problems. The lightweight neural network model can perform real-time analysis based on remote data, and combined with the mechanism of federated learning, it can quickly extract key feature information to provide intelligent decision-making support for power maintenance and scheduling. Real-time monitoring and analysis of equipment status helps to optimize energy scheduling and improve the efficiency and reliability of the power system. By predicting equipment failures and performing repairs in advance, the cost of power equipment maintenance and replacement can be reduced and asset life can be increased.

[0026] Based on this, the embodiment of the present application provides a method for monitoring the state of power equipment based on Beidou high-precision timing. Figure 1 , Figure 1 This is a flow chart of the first embodiment of the method for monitoring the status of electric power equipment based on Beidou high-precision timing.

[0027] In this embodiment, the power equipment status monitoring method based on Beidou high-precision timing includes steps S10 to S30: Step S10, acquiring Beidou satellite signals and collecting real-time ionospheric data; It should be noted that the Beidou satellite signal refers to the signal emitted by the Beidou Satellite Navigation System (BDS) for positioning, navigation and timing. The Beidou Satellite Navigation System is a global satellite navigation system independently developed by China. Its signals include but are not limited to civil signals in the B1 and B2 frequency bands and possible military or special frequency band signals. In the present invention, the Beidou satellite signal is the basic data source for achieving high-precision timing and ionospheric delay compensation. Its protection scope should cover all Beidou satellite signals that can be parsed by the receiver and used for timing, including but not limited to existing frequency bands and frequency bands that may be expanded in the future.

[0028] The ionospheric real-time data refers to data related to the ionospheric state collected in real time during the monitoring process. The ionosphere is part of the earth's atmosphere, and its characteristics (such as electron density) will affect the propagation time of satellite signals, thereby affecting the timing accuracy. In the present invention, the ionospheric real-time data includes but is not limited to ionospheric electron density profile, total electron content (TEC), satellite elevation angle and other information. These data can be obtained through the built-in sensors or external sensors of the Beidou receiver and used in the dynamic compensation model to improve the timing accuracy. Its protection scope should include all data types related to the ionospheric state and can be used for timing compensation.

[0029] Exemplarily, this embodiment uses an integrated Beidou receiver system to obtain Beidou satellite signals and collect real-time ionospheric data. The system mainly includes the following components: Beidou receiver: A high-performance Beidou-3 compatible receiver is used to support B1I and B2I band signal reception. The receiver has a built-in domestic chip and has a high-precision timing function and ionospheric data acquisition module.

[0030] Ionospheric data acquisition module: This module receives navigation messages and pseudo-range measurement data from Beidou satellite signals and analyzes the ionospheric total electron content (TEC) change information in real time. At the same time, it combines the receiver's geographical location and satellite elevation information to calculate the impact of ionospheric delay on timing signals.

[0031] Data processing unit: The data processing unit inside the receiver adopts a real-time operating system, which can quickly process Beidou satellite signals and ionospheric data, and perform ionospheric delay compensation through a built-in dynamic compensation model, and finally output a high-precision timing signal.

[0032] It is understandable that after the receiver is started, it automatically searches for and locks the Beidou satellite signal, and obtains the satellite's pseudorange, Doppler frequency shift and navigation message information. By parsing the navigation message, the ionospheric delay correction parameters (such as the ionospheric delay model coefficient) are extracted. At the same time, combined with the high-precision clock built into the receiver, the ionospheric TEC changes are measured in real time. The collected ionospheric data is input into the preset dynamic compensation model, the ionospheric delay compensation amount at the current moment is calculated, and the Beidou timing signal is corrected. The accuracy of the compensated timing signal reaches the sub-nanosecond level, which can meet the needs of high-precision power equipment monitoring systems.

[0033] It is understandable that the data synchronization accuracy of the power equipment monitoring system is significantly improved by real-time collection of ionospheric data and dynamic compensation. By utilizing the anti-spoofing technology and high-precision timing characteristics of Beidou satellite signals, the risk of timing signals being tampered with is effectively avoided, and the reliability of the system is improved. The system can operate stably in complex electromagnetic environments, especially in areas with frequent ionospheric disturbances, and can still maintain high-precision timing, providing a reliable timing benchmark for real-time monitoring of power equipment. Through the integrated receiver design and the application of domestically produced chips, the hardware cost of the system is reduced, while reducing the dependence on external high-precision timing equipment, and improving the economy of the system.

[0034] In summary, this embodiment achieves high-precision timing and data synchronization by acquiring Beidou satellite signals and collecting real-time ionosphere data, thereby significantly improving the performance and reliability of the power equipment monitoring system.

[0035] Step S20, based on the Beidou satellite signal and the ionospheric real-time data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved according to the high-precision Beidou timing signal; It should be noted that the dynamic compensation model refers to a mathematical model or algorithm for real-time correction of the influence of ionospheric delay. The model is based on the characteristics of Beidou satellite signals and the temporal and spatial variation of the ionosphere. By analyzing parameters such as the ionospheric electron density profile and satellite elevation angle, the ionospheric delay compensation is calculated to improve the accuracy of the timing signal.

[0036] The ionospheric delay compensation refers to the process of correcting the signal propagation delay caused by the ionosphere through a dynamic compensation model. Ionospheric delay is one of the main factors affecting the accuracy of satellite timing. By collecting ionospheric data (such as TEC value) in real time and combining it with the compensation model, the impact of delay on timing accuracy can be significantly reduced. The high-precision Beidou timing signal refers to the Beidou satellite timing signal after ionospheric delay compensation. Its accuracy is usually better than traditional timing methods (such as GPS timing), and can reach sub-nanosecond level or even higher. The precise synchronization of multi-source data of power equipment refers to the process of accurately aligning multi-source data from different sensors and monitoring equipment in time using high-precision Beidou timing signals. Multi-source data usually includes different types of data such as current, voltage, temperature, vibration, etc. Accurate synchronization can ensure the consistency of these data in the time dimension, thereby providing a reliable data basis for status monitoring and fault diagnosis of power equipment.

[0037] For example, based on the signal characteristics of BeiDou-3 satellite, a vertical ionospheric delay field model of the receiver is designed, and the delay compensation formula is as follows: Where Ne(h) is the ionospheric electron density profile, θ is the satellite elevation angle, R is the radius of the earth, and k is the correction factor.

[0038] For example, refer to Figure 2 , the Beidou receiver receives Beidou satellite signals in real time, analyzes satellite elevation angles and navigation messages, and collects ionospheric TEC data. The sensor network collects multi-source data such as current, voltage, temperature and vibration of power equipment. The ionospheric TEC data, satellite elevation angle and receiver geographic location are input into the dynamic compensation model to calculate the ionospheric delay compensation. The Beidou satellite signal is delayed and a high-precision timing signal is generated. Based on the high-precision Beidou timing signal, the multi-source data collected by the sensor is timestamped to achieve precise synchronization.

[0039] In this implementation, the ionospheric delay compensation amount is calculated to perform delay compensation on the Beidou satellite signal to generate a high-precision timing signal, thereby achieving time stamping of multi-source data collected by the sensor based on the high-precision Beidou timing signal.

[0040] Step S30, extract the characteristic information of the real-time operating parameters of the power equipment, combine the federated learning mechanism to obtain the global model knowledge of the preset edge nodes, analyze and judge the status of the power equipment in real time according to the pre-trained lightweight neural network model, and obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization perception training.

[0041] It should be noted that the characteristic information of the real-time operating parameters refers to the key features that can reflect the status of the equipment extracted from the real-time operating data of the power equipment. These characteristic information include but are not limited to the statistical values, change rates, spectral characteristics, etc. of parameters such as current, voltage, temperature, vibration, and partial discharge intensity. The federated learning mechanism is a distributed machine learning method that allows multiple edge nodes (devices) to collaboratively train a global model by sharing model parameters without sharing the original data. The lightweight neural network model refers to a neural network model optimized by a specific technology, which significantly reduces the number of model parameters and computational complexity while maintaining a high accuracy rate to adapt to the resource-constrained edge computing environment.

[0042] In addition, it should be noted that the channel pruning is a model compression technology that analyzes the contribution of each channel in the neural network and removes channels with lower contribution, thereby reducing the number of model parameters and the amount of calculation. The quantization-aware training is a model optimization technology that converts high-precision floating-point weights into low-precision fixed-point numbers during the training phase, and adjusts the model structure and parameters to reduce the storage space and computational complexity of the model while maintaining the accuracy of the model as much as possible. The real-time monitoring information of the health of the power equipment refers to the information reflecting the current health status of the equipment obtained by analyzing the real-time operating data of the power equipment, combined with a lightweight neural network model and a federated learning mechanism. This information includes the health score of the equipment, the probability of failure, the remaining life prediction, etc.

[0043] Exemplarily, the sensor network collects the operating parameters of the power equipment in real time, including current, voltage, temperature, vibration and partial discharge signals. The feature extraction module preprocesses the collected data, extracts key feature information, and forms a feature vector. Each edge node uses the locally collected data to train a lightweight neural network model, extracts the model parameters and encrypts and uploads them to the aggregation server. The aggregation server updates the global model parameters based on the federated learning algorithm and the parameters of each edge node, and broadcasts the updated parameters back to each edge node. The edge node uses the updated global model parameters to analyze and judge the real-time feature vector and evaluate the health status of the power equipment. According to the health score and failure probability output by the model, real-time monitoring information of the health of the power equipment is generated. If the health score is lower than the preset threshold or the failure probability exceeds the warning value, the system will extract the key alarm data and compress it into a short message data packet. The alarm information is transmitted to the operation and maintenance center in real time through Beidou short message communication or 4G / 5G network.

[0044] To sum up, this embodiment realizes efficient and accurate real-time monitoring of the health of power equipment by extracting the real-time operating parameter characteristic information of power equipment and combining the federated learning mechanism and lightweight neural network model, thereby significantly improving the operation and maintenance efficiency and reliability of the power system.

[0045] This embodiment provides a method for monitoring the status of power equipment based on Beidou high-precision timing, which solves the problem of insufficient timing accuracy in traditional power equipment status monitoring methods, and ensures accurate data synchronization and high reliability. Beidou timing synchronization can efficiently integrate multi-source data from different devices, realize real-time transmission and analysis, and help the power system quickly discover and respond to problems. The lightweight neural network model can perform real-time analysis based on remote data, and combined with the mechanism of federated learning, it can quickly extract key feature information and provide intelligent decision-making support for power maintenance and scheduling. Real-time monitoring and analysis of equipment status helps to optimize energy scheduling and improve the efficiency and reliability of the power system. By predicting equipment failures and performing repairs in advance, the cost of overhaul and replacement of power equipment can be reduced, and the life of assets can be increased.

[0046] In a feasible implementation manner, the step of performing ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model to determine a high-precision Beidou timing signal includes: Analyze the Beidou satellite signal to determine the Beidou satellite elevation angle; Analyzing the real-time ionospheric data to determine the ionospheric electron density profile; The ionospheric height range is divided into a plurality of discrete height layers, the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius are input into a preset receiver vertical ionospheric delay field model, and the delay compensation amount corresponding to each height layer is discretized and calculated; Accumulate the delay compensation values ​​of all altitude layers to obtain the real-time delay compensation value of the ionosphere; Based on the real-time ionospheric delay compensation amount, ionospheric delay compensation is performed on the Beidou satellite signal to determine a high-precision Beidou timing signal.

[0047] It should be noted that the Beidou satellite elevation angle refers to the angle between the Beidou satellite signal transmission direction and the receiver horizontal plane. It is an important parameter used in satellite navigation systems to describe the position of the satellite relative to the receiver. The size of the elevation angle affects the length of the signal propagation path and the amount of signal delay in the ionosphere. The ionospheric electron density profile refers to the distribution curve of the electron density in the ionosphere as it changes with altitude. It reflects the degree of influence of the ionosphere on the propagation of satellite signals and is a key parameter for ionospheric delay compensation.

[0048] In addition, it should be noted that dividing the ionospheric height range into multiple discrete height layers means that in order to simplify the calculation, the continuous height interval of the ionosphere is divided into several discrete height layers. The electron density of each height layer can be estimated by interpolation or fitting methods, so as to calculate the ionospheric delay. The vertical ionospheric delay field model of the receiver is a mathematical model for calculating the ionospheric delay. It is based on the parameters such as the ionospheric electron density profile, satellite elevation angle, receiver geographic coordinates and earth radius, and calculates the delay of the signal in the ionosphere by analytical or numerical methods. Discrete calculation means that after the ionosphere is divided into multiple discrete height layers, the contribution of each height layer to the satellite signal delay is calculated separately, and the total delay is obtained by accumulation. This method can effectively simplify the calculation complexity and improve the accuracy of delay compensation. The real-time ionospheric delay compensation refers to the real-time correction value of the ionospheric delay obtained by the above model and calculation method. This compensation is used to correct the delay error of the Beidou satellite signal during the propagation of the ionosphere, thereby improving the timing accuracy.

[0049] For example, in high-precision timing applications, ionospheric delay is one of the main factors affecting the timing accuracy of satellite signals. When Beidou satellite signals pass through the ionosphere, they are delayed due to changes in the electron density of the ionosphere. In order to improve the timing accuracy, it is necessary to accurately compensate for the ionospheric delay. This embodiment achieves high-precision timing signal output by parsing Beidou satellite signals and ionospheric real-time data, combined with a vertical ionospheric delay field model.

[0050] The Beidou receiver receives Beidou satellite signals in real time and parses the satellite elevation angle and ionospheric delay correction parameters in the navigation message.

[0051] The ionospheric monitoring module collects ionospheric TEC data and estimates the ionospheric electron density profile by combining the receiver's geographic coordinates (and the earth's radius). The ionospheric altitude range is divided into multiple discrete altitude layers.

[0052] For each altitude layer, the ionospheric electron density profile, satellite elevation angle, receiver geographic coordinates and earth radius are input into the receiver vertical ionospheric delay field model. For each discrete altitude layer, the corresponding delay compensation is calculated. The delay compensation amounts of all discrete altitude layers are accumulated to obtain the real-time ionospheric delay compensation amount. According to the real-time ionospheric delay compensation amount, the Beidou satellite signal is corrected to obtain a high-precision Beidou timing signal, and the corrected timing signal that meets the high-precision timing requirements is output.

[0053] It is understandable that through ionospheric delay compensation, the timing accuracy is improved from the microsecond level of traditional methods to the sub-nanosecond level, meeting the needs of high-precision applications such as power system synchronization and financial transaction timestamps. By adopting a discrete calculation method, it can adapt to ionospheric changes in different geographical locations and atmospheric conditions, and improve the stability of the system in complex environments. By dividing the ionosphere into discrete altitude layers for calculation, the calculation process of delay compensation is simplified, the consumption of hardware resources is reduced, and it is suitable for running on resource-constrained edge devices. Combined with the anti-interference capability and high-precision timing characteristics of the Beidou system, the system can still maintain high-precision timing in complex electromagnetic environments and areas with frequent ionospheric disturbances, and the reliability is significantly improved. By updating the ionospheric delay compensation in real time, the timing error caused by ionospheric changes is reduced, the dependence on redundant equipment is reduced, and the utilization of system resources is optimized.

[0054] In summary, this embodiment achieves high-precision timing signal output by parsing Beidou satellite signals and real-time ionospheric data, combined with the vertical ionospheric delay field model, significantly improves the timing accuracy and system reliability, and is suitable for a variety of high-precision timing application scenarios.

[0055] In a feasible implementation manner, the step of dividing the ionospheric height range into a plurality of discrete height layers, inputting the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius into a preset receiver vertical ionospheric delay field model, and discretizing and calculating the delay compensation amount corresponding to each height layer includes: Obtain historical ionospheric data, corresponding historical satellite elevation angles, and corresponding timing error data to construct a training data set; Inputting the training data set into the time series feature extraction network to be trained for training, and extracting key time series features based on the nonlinear mapping relationship between the spatiotemporal variation of the ionosphere and the timing error; According to the key timing characteristics, a correction coefficient is determined by the network weight obtained through training; Based on the correction coefficient, the vertical ionospheric delay field model of the receiver is adaptively calibrated.

[0056] It should be noted that the historical ionospheric data refers to the ionospheric characteristic data recorded in the past period of time, including information such as ionospheric electron density, total electron content, and ionospheric height distribution. These data usually come from satellite observations, ground monitoring stations, or other atmospheric detection methods. The historical satellite elevation angle refers to the elevation angle data of the satellite signal relative to the receiver in the past period of time. These data record the elevation angle changes of the satellite at different times and different positions, reflecting the geometric characteristics of the signal propagation path. The timing error data refers to the deviation data between the satellite timing signal and the real time in practical applications. These errors are caused by various factors such as ionospheric delay, tropospheric delay, satellite clock error, and receiver noise. The time series feature extraction network is a neural network architecture for processing time series data, which can extract time-related feature information from historical data. The key time series feature refers to the core feature extracted from historical data that can reflect the relationship between the spatiotemporal changes of the ionosphere and the timing error. These features may include statistical features of the time series (such as mean, variance), spectral features, and rate of change. The correction coefficient refers to the parameter used to calibrate the model calculated based on the extracted key time series features and network weights. It reflects the actual impact of the spatiotemporal changes of the ionosphere on the timing error and is used to adjust the accuracy of the ionosphere delay model.

[0057] For example, in a satellite timing system, ionospheric delay is one of the main factors affecting timing accuracy. Traditional ionospheric delay models are usually based on empirical formulas and are difficult to adapt to complex ionospheric temporal and spatial changes, especially in areas with intense ionospheric activity. In order to improve timing accuracy, this embodiment dynamically calibrates the ionospheric delay model by acquiring historical data and using machine learning methods, thereby achieving high-precision timing signal output.

[0058] It is understandable that historical ionospheric TEC data, satellite elevation data, and timing error data obtained by comparison with high-precision time references (such as atomic clocks) are obtained from Beidou receivers.

[0059] Organize these data into time series format and construct a training dataset. The dataset includes the following: Input features: ionospheric TEC value, satellite elevation angle, receiver geographic location, timestamp.

[0060] Output label: timing error.

[0061] The training data set is input into the LSTM neural network for training. The LSTM network can capture the long-term dependencies in time series data and is suitable for processing the complex nonlinear relationship between the spatiotemporal changes of the ionosphere and the timing error.

[0062] During the training process, the network learns the mapping relationship between ionospheric TEC changes, satellite elevation angle changes and timing errors, and extracts key timing features, such as TEC change rate, error fluctuation rate, etc. The correction coefficient is calculated based on the key timing features extracted by the LSTM network and the network weights obtained through training. The correction coefficient reflects the adjustment range of the ionospheric delay model under different conditions. The calculated correction coefficient is applied to the vertical ionospheric delay field model of the receiver to dynamically adjust the model parameters. The calibrated model can more accurately reflect the current ionospheric state, thereby improving the accuracy of the timing signal.

[0063] It can be understood that this embodiment significantly improves the timing accuracy and system reliability by acquiring historical data and using machine learning methods to dynamically calibrate the ionospheric delay model, and is suitable for a variety of high-precision timing application scenarios.

[0064] In a feasible implementation manner, the step of extracting the characteristic information of the real-time operating parameters of the power equipment, combining the federated learning mechanism to obtain the global model knowledge of the preset edge node, and analyzing and judging the power equipment state in real time according to the pre-trained lightweight neural network model to obtain the real-time monitoring information of the equipment health includes: Obtain historical status data of power equipment and corresponding fault records; The historical status data of the power equipment is input into the neural network model to be trained to analyze and judge the equipment status, and obtain equipment health prediction information, wherein the high-precision floating-point weights are converted into low-precision fixed-point numbers through quantization perception training during the equipment status analysis and judgment; Calculate the contribution of each channel of the neural network model according to the activation value and gradient in the process of analyzing and judging the status of the device, and prune the channels whose contribution is lower than the preset warning threshold; Comparing the equipment health prediction information with the fault record to determine a prediction difference value; If the predicted difference value does not meet the preset prediction allowable error range, then return to the step of inputting the historical status data of the power equipment into the neural network model to be trained for equipment status analysis and judgment; if the predicted difference value meets the preset prediction allowable error range, then output the trained lightweight neural network model.

[0065] It should be noted that the historical status data of the power equipment refers to various data generated by the power equipment during operation, including but not limited to operating parameters such as current, voltage, temperature, vibration, partial discharge intensity, and maintenance records of the equipment, operating time and other information. These data reflect the operating status of the equipment at different time points and are the basis for equipment health analysis. Quantization-aware training is a neural network training technology that converts high-precision floating-point weights into low-precision fixed-point numbers during the training phase, and adjusts the network structure and parameters to reduce the storage space and computational complexity of the model while maintaining the accuracy of the model as much as possible. Channel contribution refers to the contribution of each channel (or feature map) in the neural network to the model output. By analyzing the activation value and gradient, the importance of each channel can be quantified. Channels with low contribution have less impact on the model output and can be pruned to optimize the model structure. Pruning is a model optimization technology that reduces the number of parameters and computational complexity of the model by removing channels or weights with low contribution in the neural network, while maintaining the accuracy of the model as much as possible. Pruning can be divided into structured pruning (removing the entire channel) and unstructured pruning (removing a single weight). The prediction difference value refers to the difference between the output of the neural network model (equipment health prediction information) and the true value (fault record). By calculating the prediction difference value, the accuracy and reliability of the model can be evaluated. The prediction allowable error range refers to a pre-set error threshold range used to determine whether the prediction results of the model meet the requirements of actual applications. If the prediction difference value is within the allowable error range, the model's prediction is considered reliable; otherwise, the model needs to be further optimized. A lightweight neural network model refers to a neural network model optimized through technologies such as quantization-aware training and channel pruning. It can significantly reduce the number of model parameters and computational complexity while maintaining a high accuracy rate, and is suitable for running in resource-constrained edge computing environments.

[0066] For example, the health status monitoring of power equipment is crucial to ensure the stable operation of the power system. Traditional methods rely on manual inspections and empirical judgments, which are inefficient and difficult to monitor in real time. This embodiment obtains the historical status data and fault records of power equipment, uses a neural network model to perform equipment status analysis and health prediction, and optimizes the model through quantitative perception training and channel pruning technology, and finally obtains a lightweight neural network model to achieve efficient and accurate health monitoring of power equipment.

[0067] It is understandable that the historical status data of the power equipment is obtained from the sensor network, including parameters such as current, voltage, temperature, vibration, etc. The historical status data is associated with the corresponding fault records to build a training data set. The historical status data is input into the neural network model for equipment status analysis and health prediction.

[0068] During the training process, quantization-aware training technology is used to gradually convert high-precision floating-point weights into low-precision fixed-point numbers, reducing the storage space and computational complexity of the model. During the model training process, the activation value and gradient information of each channel are recorded. The contribution of each channel is calculated based on the activation value and gradient, and channels with contributions lower than the preset warning threshold are removed to optimize the model structure. The device health prediction information is compared with the fault record to calculate the prediction difference value. If the prediction difference value does not meet the preset prediction allowable error range, return to retraining and continue to optimize the model.

[0069] If the predicted difference value meets the preset error range, the trained lightweight neural network model is output.

[0070] It is understandable that the neural network model can be used to analyze the status of power equipment and predict its health, which significantly improves the monitoring accuracy, detects potential faults in advance, and reduces equipment downtime. By adopting quantitative perception training and channel pruning technology, the number of model parameters is reduced, the reasoning speed is improved, and the computing power consumption is reduced, making it suitable for running in resource-constrained edge computing environments. Through historical data training and iterative optimization, the model can adapt to different types of power equipment and operating environments and has good generalization capabilities.

[0071] In summary, this embodiment achieves efficient and accurate health monitoring of power equipment by acquiring historical status data and fault records of power equipment and optimizing the neural network model using quantization-aware training and channel pruning technology, thereby significantly improving the operation and maintenance efficiency and reliability of the power system.

[0072] In a feasible implementation manner, the steps of extracting characteristic information of real-time operating parameters of the power equipment, acquiring global model knowledge of preset edge nodes in combination with a federated learning mechanism, and analyzing and judging the state of the power equipment in real time according to a pre-trained lightweight neural network model to obtain real-time monitoring information of the health of the equipment include: Extracting characteristic information of real-time operating parameters of the electric power equipment and constructing a real-time characteristic vector; Encrypt the parameters of the local lightweight neural network model and upload them to the aggregation server, and obtain the global model parameters returned by the aggregation server; The real-time feature vector is input into a lightweight neural network model using the global model parameters to analyze and judge the state of the power equipment in real time and determine the real-time monitoring information of the health of the equipment.

[0073] It should be noted that the real-time feature vector refers to a vector composed of feature information extracted from the real-time operating parameters of the power equipment, which is used to characterize the state of the equipment at the current moment. These feature information may include statistical values, change rates or spectral characteristics of parameters such as current, voltage, temperature, vibration, and partial discharge intensity. The aggregation server refers to a central server used to collect and process data from multiple edge nodes in the network. In federated learning, the aggregation server is responsible for receiving model parameters uploaded by each edge node, updating the global model parameters through a specific algorithm (such as weighted average), and broadcasting the updated parameters back to each edge node. The global model parameters refer to the parameters of the neural network model updated through the federated learning mechanism. It integrates the local model knowledge of multiple edge nodes and can better adapt to the characteristics of different equipment and environments. The real-time monitoring information of equipment health refers to the information reflecting the current health status of the power equipment obtained through model analysis and judgment. This information may include health scores, failure probability, remaining service life prediction, etc., which are used to help operation and maintenance personnel understand the equipment status in a timely manner and take corresponding measures.

[0074] For example, in the power system, real-time monitoring of the health status of power equipment is crucial to prevent failures and ensure the safe operation of the power grid. However, traditional centralized monitoring methods have problems such as high data transmission delay and high computing resource consumption, and are difficult to adapt to large-scale distributed monitoring scenarios. This embodiment extracts the real-time operating parameter feature information of power equipment, combines the federated learning mechanism and the lightweight neural network model, and realizes efficient and accurate real-time monitoring of the health of power equipment, while protecting data privacy and reducing computing resource consumption.

[0075] It can be understood that the sensor network collects the operating parameters of the power equipment in real time, including current, voltage, temperature, vibration and partial discharge signals.

[0076] The collected data is preprocessed through the feature extraction module to extract key feature information and form a real-time feature vector. Each edge computing node uses the locally collected data to train a lightweight neural network model, extract the model parameters and encrypt and upload them to the aggregation server.

[0077] The aggregation server receives model parameters from multiple edge nodes, updates the global model parameters through a federated learning algorithm (such as FedAvg), and broadcasts the updated parameters back to each edge node. The edge node inputs the real-time feature vector into a lightweight neural network model that uses the global model parameters for real-time analysis and judgment.

[0078] The model outputs real-time monitoring information on the health of the equipment. If the health score is lower than the preset threshold or the failure probability exceeds the warning value, the system will trigger an alarm and transmit the key information to the operation and maintenance center via Beidou short message communication or 4G / 5G network.

[0079] In this embodiment, through the lightweight neural network model and federated learning mechanism, the system can analyze the status of power equipment in real time at the edge node, significantly improving the monitoring efficiency and response speed. The real-time output of health scores and failure probabilities provides accurate decision support for operation and maintenance personnel.

[0080] To sum up, this embodiment realizes efficient and accurate real-time monitoring of the health of power equipment by extracting real-time operating parameter characteristic information of power equipment and combining it with a federated learning mechanism and a lightweight neural network model, which significantly improves the operation and maintenance efficiency and reliability of the power system while protecting data privacy and reducing computing resource consumption.

[0081] In a feasible implementation manner, the step of obtaining real-time monitoring information of equipment health includes: Extract key alarm data from real-time monitoring information and compress it into short message data packets; Send short message data packets to the preset operation and maintenance warning equipment in priority preemptive mode through the preset Beidou channel; If the transmission fails, the short message data packet is automatically retransmitted within a preset time interval based on an exponential backoff algorithm.

[0082] It should be noted that critical alarm data refers to data related to equipment failure, abnormal status or other emergency situations extracted from real-time monitoring information. These data usually have high priority and need to be transmitted in time to notify operation and maintenance personnel to take measures. Short message data packets refer to the compression and encapsulation of critical alarm data into a data format suitable for short-distance, low-bandwidth transmission. In the Beidou system, short message communication is a function that supports text message transmission, which is usually used in emergency situations or scenarios with small data volumes. Preset Beidou channels refer to Beidou communication channels pre-configured for specific applications (such as power equipment monitoring). The Beidou system supports multiple communication modes, including short message communication, satellite communication, etc. Preset channels are usually optimized according to the needs of application scenarios to ensure the reliability and priority of data transmission. Priority preemption mode refers to the allocation of transmission priority according to the importance and urgency of data during communication. In an emergency, critical alarm data can preempt channel resources and be transmitted first to ensure timely delivery. The exponential backoff algorithm is an algorithm that delays retries according to an exponential law after a communication transmission fails. The algorithm increases the success rate of data transmission by adding a delay time to each retry to avoid channel congestion caused by frequent retries. The preset time interval refers to the time range set for each retry in the exponential backoff algorithm. This interval can be adjusted according to the needs of the application scenario to balance the retry frequency and transmission success rate.

[0083] For example, in the operation and maintenance of power systems, timely transmission of key alarm information is crucial for rapid response to equipment failures. However, traditional communication methods may cause delays or failures in alarm information transmission due to network congestion or signal interruption. This embodiment extracts key alarm data from real-time monitoring information and uses the priority preemptive mode and exponential backoff algorithm of Beidou short message communication to ensure that alarm information can be efficiently and reliably transmitted to the operation and maintenance alarm equipment.

[0084] It is understandable that the monitoring equipment collects the operating status data of the power equipment in real time and transmits it to the data processing unit. The data processing unit analyzes the equipment status through a preset health model and extracts key alarm data. The alarm data is compressed and encapsulated into a short message data packet. The data processing unit sends the short message data packet to the Beidou communication module. The Beidou communication module sends the alarm data packet to the operation and maintenance alarm device according to the preset priority preemption mode. For example, if there are other non-alarm data transmissions at the same time, the alarm data packet will take priority to occupy the channel resources. If the operation and maintenance alarm device does not receive the alarm data packet within the preset time (for example, the transmission fails due to signal interruption or network congestion), the Beidou communication module will start the exponential backoff algorithm.

[0085] According to the exponential backoff algorithm, the first retransmission delay is 2 seconds, the second retransmission delay is 4 seconds, the third retransmission delay is 8 seconds, and so on. The preset time interval is to complete the retransmission within 60 seconds.

[0086] After successful retransmission, the operation and maintenance alarm device receives and parses the alarm data packet and triggers the corresponding alarm prompt.

[0087] It can be understood that the priority preemption mode ensures that key alarm data can be transmitted first in a complex network environment, which significantly improves the transmission reliability of alarm information.

[0088] The exponential backoff algorithm automatically retransmits when transmission fails, further improving the transmission success rate of alarm information and ensuring that operation and maintenance personnel can receive alarm information in a timely manner.

[0089] In summary, this embodiment achieves efficient and reliable transmission of alarm information by extracting key alarm data and utilizing the priority preemption mode and exponential backoff algorithm of Beidou short message communication, thereby significantly improving the operation and maintenance efficiency and reliability of the power system.

[0090] For example, in order to help understand the implementation process of the power equipment status monitoring method based on Beidou high-precision timing obtained by combining this embodiment with the above-mentioned embodiment 1, please refer to Figure 3 , Figure 3 A brief flow chart of a method for monitoring the status of power equipment based on Beidou high-precision timing is provided. Specifically: This embodiment comprehensively applies Beidou high-precision timing, ionospheric delay compensation, federated learning, a lightweight neural network model, and a reliable alarm transmission mechanism to achieve high-precision monitoring and real-time alarm of the status of power equipment.

[0091] The Beidou receiver acquires Beidou satellite signals in real time and analyzes satellite elevation angles and ionospheric TEC data.

[0092] The ionospheric height range is divided into multiple discrete height layers. The ionospheric electron density profile, the receiver geographic coordinates and the earth radius are combined and input into the preset vertical ionospheric delay field model to discretize and calculate the delay compensation amount of each height layer.

[0093] Accumulate the delay compensation of all altitude layers to obtain the real-time delay compensation of the ionosphere, compensate the Beidou satellite signal, and output a high-precision timing signal.

[0094] The sensor network collects the operating parameters of the power equipment in real time, extracts the characteristic information and constructs the real-time characteristic vector.

[0095] Encrypt the parameters of the local lightweight neural network model and upload them to the aggregation server to obtain the global model parameters.

[0096] The real-time feature vector is input into the global model, the power equipment status is analyzed in real time, and the real-time monitoring information of the equipment health is output.

[0097] Extract key alarm data from real-time monitoring information and compress it into short message data packets.

[0098] Through the preset Beidou channel, short message data packets are sent to the operation and maintenance alarm equipment in priority preemptive mode.

[0099] If the transmission fails, the data packet is automatically retransmitted within a preset time interval based on the exponential backoff algorithm.

[0100] This embodiment achieves high-precision monitoring and real-time alarm of the status of power equipment by comprehensively applying Beidou high-precision time synchronization, ionospheric delay compensation, federated learning, lightweight neural network model and reliable alarm transmission mechanism, which significantly improves the operation and maintenance efficiency and reliability of the power system.

[0101] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the power equipment status monitoring method based on Beidou high-precision timing. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0102] This application also provides a power equipment status monitoring device based on Beidou high-precision timing, please refer to Figure 4 , the power equipment status monitoring device based on Beidou high-precision timing includes: The acquisition module 10 is used to acquire Beidou satellite signals and collect real-time ionospheric data; A synchronization module 20 is used to perform ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model, determine a high-precision Beidou timing signal, and realize accurate synchronization of multi-source data of power equipment according to the high-precision Beidou timing signal; The monitoring module 30 is used to extract the characteristic information of the real-time operating parameters of the power equipment, obtain the global model knowledge of the preset edge nodes in combination with the federated learning mechanism, analyze and judge the status of the power equipment in real time according to the pre-trained lightweight neural network model, and obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization perception training.

[0103] And / or, the synchronization module 20 includes: A first analysis module is used to analyze the Beidou satellite signal and determine the Beidou satellite elevation angle; A second analysis module is used to analyze the real-time ionospheric data to determine the ionospheric electron density profile; The first discrete module is used to divide the ionospheric height range into a plurality of discrete height layers, input the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius into a preset receiver vertical ionospheric delay field model, and discretize and calculate the delay compensation amount corresponding to each height layer; The first accumulation module is used to accumulate the delay compensation amounts of all altitude layers to obtain the real-time delay compensation amount of the ionosphere; The first compensation module is used to perform ionospheric delay compensation on the Beidou satellite signal based on the real-time ionospheric delay compensation amount to determine a high-precision Beidou timing signal.

[0104] And / or, the first discrete module includes: The first acquisition module acquires historical ionospheric data, corresponding historical satellite elevation angles and corresponding timing error data to construct a training data set; A first extraction module is used to input the training data set into a time series feature extraction network to be trained for training, and extract key time series features based on the nonlinear mapping relationship between the spatiotemporal variation of the ionosphere and the timing error; A first determination module is used to determine a correction coefficient based on the key timing characteristics through the network weights obtained through training; The first calibration module is used to adaptively calibrate the vertical ionospheric delay field model of the receiver based on the correction coefficient.

[0105] And / or, the power equipment status monitoring device based on Beidou high-precision timing includes: The second acquisition module is used to acquire the historical status data of the power equipment and the corresponding fault records; A first judgment module is used to input the historical status data of the power equipment into the neural network model to be trained to analyze and judge the equipment status, and obtain equipment health prediction information, wherein when analyzing and judging the equipment status, high-precision floating-point weights are converted into low-precision fixed-point numbers through quantization perception training; A first calculation module is used to calculate the contribution of each channel of the neural network model according to the activation value and gradient in the device state analysis and judgment process, and prune the channels whose contribution is lower than a preset warning threshold; A first comparison module, used to compare the equipment health prediction information with the fault record to determine a prediction difference value; The first loop module is used to return to the step of inputting the historical status data of the power equipment into the neural network model to be trained for equipment status analysis and judgment if the predicted difference value does not meet the preset prediction allowable error range; if the predicted difference value meets the preset prediction allowable error range, the trained lightweight neural network model is output.

[0106] And / or, the monitoring module 30 includes: A second extraction module is used to extract characteristic information of real-time operating parameters of the electric power equipment and construct a real-time characteristic vector; A first uploading module is used to encrypt and upload the parameters of the local lightweight neural network model to the aggregation server, and obtain the global model parameters returned by the aggregation server; The first analysis module is used to input the real-time feature vector into a lightweight neural network model using the global model parameters, analyze and judge the state of the power equipment in real time, and determine the real-time monitoring information of the health of the equipment.

[0107] And / or, the monitoring module 30 includes: The third extraction module is used to extract key alarm data from the real-time monitoring information and compress it into short message data packets; A first sending module is used to send a short message data packet to a preset operation and maintenance alarm device in a priority preemptive mode through a preset Beidou channel; The first transmission module is used to automatically retransmit the short message data packet within a preset time interval based on an exponential backoff algorithm if the transmission fails.

[0108] The power equipment status monitoring device based on Beidou high-precision timing provided by the present application adopts the power equipment status monitoring method based on Beidou high-precision timing in the above-mentioned embodiment, which can solve the technical problem of difficulty in accurately obtaining the real-time operating status information of the power equipment. Compared with the prior art, the beneficial effects of the power equipment status monitoring device based on Beidou high-precision timing provided by the present application are the same as the beneficial effects of the power equipment status monitoring method based on Beidou high-precision timing provided by the above-mentioned embodiment, and the other technical features of the power equipment status monitoring device based on Beidou high-precision timing are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0109] The present application provides an electric power equipment status monitoring device based on Beidou high-precision timing, and the electric power equipment status monitoring device based on Beidou high-precision timing includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the electric power equipment status monitoring method based on Beidou high-precision timing in the above-mentioned embodiment one.

[0110] Reference below Figure 5 , which shows a schematic diagram of the structure of the power equipment status monitoring device based on Beidou high-precision timing suitable for implementing the embodiment of the present application. The power equipment status monitoring device based on Beidou high-precision timing in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, tablet computers, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electric power equipment status monitoring device based on Beidou high-precision timing is shown as just an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0111] like Figure 5As shown, the power equipment status monitoring device based on Beidou high-precision timing may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the power equipment status monitoring device based on Beidou high-precision timing are also stored in RAM1004. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the power equipment status monitoring device based on Beidou high-precision timing to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a power equipment status monitoring device based on Beidou high-precision timing with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or provided alternatively.

[0112] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0113] The power equipment status monitoring device based on Beidou high-precision timing provided by this application adopts the power equipment status monitoring method based on Beidou high-precision timing in the above-mentioned embodiment, which can solve the technical problem of difficulty in accurately obtaining the real-time operating status information of power equipment. Compared with the prior art, the beneficial effects of the power equipment status monitoring device based on Beidou high-precision timing provided by this application are the same as the beneficial effects of the power equipment status monitoring method based on Beidou high-precision timing provided by the above-mentioned embodiment, and the other technical features of the power equipment status monitoring device based on Beidou high-precision timing are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0114] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0115] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0116] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the power equipment status monitoring method based on Beidou high-precision timing in the above-mentioned embodiment.

[0117] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0118] The above-mentioned computer-readable storage medium may be included in the power equipment status monitoring device based on Beidou high-precision timing; or it may exist independently without being assembled into the power equipment status monitoring device based on Beidou high-precision timing.

[0119] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the power equipment status monitoring device based on Beidou high-precision timing, the power equipment status monitoring device based on Beidou high-precision timing: obtains Beidou satellite signals and collects real-time ionospheric data; Based on the Beidou satellite signal and the real-time ionospheric data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved based on the high-precision Beidou timing signal; The characteristic information of the real-time operating parameters of the power equipment is extracted, and the global model knowledge of the preset edge nodes is obtained in combination with the federated learning mechanism. The state of the power equipment is analyzed and judged in real time according to the pre-trained lightweight neural network model to obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

[0120] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0121] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0122] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0123] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned method for monitoring the state of power equipment based on Beidou high-precision timing, and can solve the technical problem of difficulty in accurately obtaining real-time operating status information of power equipment. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the method for monitoring the state of power equipment based on Beidou high-precision timing provided by the above-mentioned embodiment, and will not be repeated here.

[0124] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for monitoring the status of power equipment based on Beidou high-precision timing.

[0125] The computer program product provided by this application can solve the technical problem of difficulty in accurately obtaining real-time operating status information of power equipment. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the power equipment status monitoring method based on Beidou high-precision timing provided by the above embodiment, which will not be repeated here.

[0126] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for monitoring the state of power equipment based on Beidou high-precision timing, characterized in that: The method includes: Obtain Beidou satellite signals and collect real-time ionosphere data; Based on the Beidou satellite signal and the real-time ionospheric data, ionospheric delay compensation is performed according to a preset dynamic compensation model, a high-precision Beidou timing signal is determined, and accurate synchronization of multi-source data of power equipment is achieved based on the high-precision Beidou timing signal; The characteristic information of the real-time operating parameters of the power equipment is extracted, and the global model knowledge of the preset edge nodes is obtained in combination with the federated learning mechanism. The state of the power equipment is analyzed and judged in real time according to the pre-trained lightweight neural network model to obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

2. The method according to claim 1, characterized in that The step of performing ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model to determine a high-precision Beidou timing signal comprises: Analyze the Beidou satellite signal to determine the Beidou satellite elevation angle; Analyzing the real-time ionospheric data to determine the ionospheric electron density profile; The ionospheric height range is divided into a plurality of discrete height layers, the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius are input into a preset receiver vertical ionospheric delay field model, and the delay compensation amount corresponding to each height layer is discretized and calculated; Accumulate the delay compensation values ​​of all altitude layers to obtain the real-time delay compensation value of the ionosphere; Based on the real-time ionospheric delay compensation amount, ionospheric delay compensation is performed on the Beidou satellite signal to determine a high-precision Beidou timing signal.

3. The method according to claim 2, characterized in that The step of dividing the ionospheric height range into a plurality of discrete height layers, inputting the ionospheric electron density profile, the Beidou satellite elevation angle, the preset receiver geographic coordinates and the earth radius into a preset receiver vertical ionospheric delay field model, and discretizing and calculating the delay compensation amount corresponding to each height layer comprises: Obtain historical ionospheric data, corresponding historical satellite elevation angles, and corresponding timing error data to construct a training data set; Inputting the training data set into the time series feature extraction network to be trained for training, and extracting key time series features based on the nonlinear mapping relationship between the spatiotemporal variation of the ionosphere and the timing error; According to the key timing characteristics, a correction coefficient is determined by the network weight obtained through training; Based on the correction coefficient, the vertical ionospheric delay field model of the receiver is adaptively calibrated.

4. The method according to claim 1, characterized in that The step of extracting the characteristic information of the real-time operating parameters of the power equipment, acquiring the global model knowledge of the preset edge node in combination with the federated learning mechanism, and analyzing and judging the state of the power equipment in real time according to the pre-trained lightweight neural network model to obtain the real-time monitoring information of the health of the equipment includes: Obtain historical status data of power equipment and corresponding fault records; The historical status data of the power equipment is input into the neural network model to be trained to analyze and judge the equipment status, and obtain equipment health prediction information, wherein the high-precision floating-point weights are converted into low-precision fixed-point numbers through quantization perception training during the equipment status analysis and judgment; Calculate the contribution of each channel of the neural network model according to the activation value and gradient in the process of analyzing and judging the status of the device, and prune the channels whose contribution is lower than the preset warning threshold; Comparing the equipment health prediction information with the fault record to determine a prediction difference value; If the predicted difference value does not meet the preset prediction allowable error range, then return to the step of inputting the historical status data of the power equipment into the neural network model to be trained for equipment status analysis and judgment; if the predicted difference value meets the preset prediction allowable error range, then output the trained lightweight neural network model.

5. The method according to claim 1, characterized in that The steps of extracting the characteristic information of the real-time operating parameters of the power equipment, acquiring the global model knowledge of the preset edge nodes in combination with the federated learning mechanism, and analyzing and judging the state of the power equipment in real time according to the pre-trained lightweight neural network model to obtain the real-time monitoring information of the health of the equipment include: Extracting characteristic information of real-time operating parameters of the electric power equipment and constructing a real-time characteristic vector; Encrypt the parameters of the local lightweight neural network model and upload them to the aggregation server, and obtain the global model parameters returned by the aggregation server; The real-time feature vector is input into a lightweight neural network model using the global model parameters to analyze and judge the state of the power equipment in real time and determine the real-time monitoring information of the health of the equipment.

6. The method according to claim 1, characterized in that The step of obtaining real-time monitoring information of equipment health includes: Extract key alarm data from real-time monitoring information and compress it into short message data packets; Send short message data packets to the preset operation and maintenance warning equipment in priority preemptive mode through the preset Beidou channel; If the transmission fails, the short message data packet is automatically retransmitted within a preset time interval based on an exponential backoff algorithm.

7. A power equipment status monitoring device based on Beidou high-precision timing, characterized in that: The device comprises: Acquisition module, used to acquire Beidou satellite signals and collect real-time ionospheric data; A synchronization module, for performing ionospheric delay compensation based on the Beidou satellite signal and the ionospheric real-time data according to a preset dynamic compensation model, determining a high-precision Beidou timing signal, and realizing accurate synchronization of multi-source data of power equipment according to the high-precision Beidou timing signal; The monitoring module is used to extract the characteristic information of the real-time operating parameters of the power equipment, obtain the global model knowledge of the preset edge nodes in combination with the federated learning mechanism, analyze and judge the status of the power equipment in real time according to the pre-trained lightweight neural network model, and obtain real-time monitoring information on the health of the power equipment, wherein the lightweight neural network model is obtained by parameter compression training through channel pruning and quantization-aware training.

8. A power equipment status monitoring device based on Beidou high-precision timing, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the method for monitoring the state of electric power equipment based on Beidou high-precision timing as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the power equipment status monitoring method based on Beidou high-precision timing are implemented as described in any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method for monitoring the state of electric power equipment based on Beidou high-precision timing are implemented as claimed in any one of claims 1 to 6.

Images