Power distribution network distributed measurement synchronization method and device based on Beidou satellite time service

Abstract

The invention discloses a power distribution network distributed measurement synchronization method and device based on Beidou satellite time service, and belongs to the technical field of power distribution network measurement, and the method comprises the steps: obtaining a Beidou satellite signal, calibrating a local clock source, and generating a standard time reference; when it is continuously monitored that the Beidou satellite signal quality parameter is lower than a quality threshold value, clock working environment parameters of a local clock source are collected, and a short-term clock offset predicted value is generated in combination with a standard time reference; generating topology correction according to the topology connection relation of the power distribution network; and fusing the short-term clock offset predicted value and the topology correction quantity to generate a dynamic compensation time reference, performing time marking on the acquired electrical quantity measurement data of the power distribution network, and generating synchronous measurement data with a timestamp. According to the technical scheme, when the Beidou signal is lost, the local clock drift predicted value and the adjacent node topology correction amount are fused to generate the dynamic compensation time reference, so that the measurement data synchronization precision and robustness of the power distribution network in a complex environment can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of distribution network measurement technology, and in particular to a distribution network distributed measurement synchronization method and device based on Beidou satellite timing. Background Art

[0002] As a vital component of the power system, the safe, stable, and economical operation of the distribution network is crucial. With the widespread integration of distributed power sources and increasing user demands for power supply reliability, the operation and control of distribution networks has become increasingly complex. To achieve refined monitoring and efficient management of distribution networks, measurement devices must be deployed at key nodes along the lines to collect real-time data on electrical quantities such as voltage, current, and phase. This high-precision, time-stamped, synchronized measurement data is the foundation for advanced applications such as precise fault location, system state estimation, and power flow analysis.

[0003] To ensure temporal consistency of data collected by geographically distributed measurement devices, existing technologies typically use Global Navigation Satellite Systems (GNSS) for timing. For example, these systems utilize the pulse-per-second signal (1PPS) and time information provided by the Beidou Satellite System (BDS) or Global Positioning System (GPS) to synchronize the local clocks of each measurement device. This timing method achieves extremely high synchronization accuracy in open areas with good satellite signal coverage and no obstructions, meeting the application requirements of distribution networks.

[0004] However, in real-world distribution network environments, numerous measurement devices are installed in complex locations such as urban canyons, indoor substations, and underground cable trenches. Satellite signals in these locations are easily blocked and interfered with by tall buildings, trees, and other obstacles, leading to signal weakening or even interruption. If the satellite signal is lost, the measurement device must rely on its internal local clock source (such as a crystal oscillator) for timekeeping. However, the frequency of this local clock source can drift due to environmental factors such as temperature and voltage fluctuations. This long-term clocking error can accumulate significant time errors, leading to data loss between measurement nodes and severely impacting the accuracy and effectiveness of subsequent power grid analysis applications. Summary of the Invention

[0005] To solve the above problems, the present invention provides a distributed measurement synchronization method and device for distribution networks based on Beidou satellite timing. When the Beidou signal is lost, the method and device fuse the local clock drift prediction value with the adjacent node topology correction value to generate a dynamic compensation time reference. This technical solution can significantly improve the accuracy and robustness of measurement data synchronization of distribution networks in complex environments.

[0006] The above objectives can be achieved through the following solutions: A distributed measurement synchronization method for a distribution network based on Beidou satellite timing includes acquiring Beidou satellite signals, calibrating a local clock source based on the Beidou satellite signals, and generating a standard time reference; continuously monitoring the real-time status of the Beidou satellite signals to generate a signal quality parameter, and generating a dynamic compensation mode trigger signal when the signal quality parameter is lower than a preset quality threshold; in response to the dynamic compensation mode trigger signal, collecting clock operating environment parameters of the local clock source, and predicting drift characteristics in combination with the standard time reference and the clock operating environment parameters to generate a short-term clock offset prediction value; determining a set of topologically adjacent nodes based on a topological connection relationship of the distribution network, sending a time synchronization request to the set of topologically adjacent nodes to obtain adjacent node timestamps, calculating a theoretical time difference based on line physical parameters included in the distribution network topological connection relationship, and correcting the adjacent node timestamps using the theoretical time difference to generate a topology correction value; fusing the short-term clock offset prediction value with the topology correction value to generate a dynamic compensation time reference; collecting electrical quantity measurement data of the distribution network, and time-stamping the electrical quantity measurement data using the dynamic compensation time reference to generate synchronized measurement data with timestamps.

[0007] Optionally, generating a dynamic compensation mode trigger signal includes: collecting the number of receiving satellites, signal-to-noise ratio and position precision factor of the Beidou satellite signal to form original state data; performing a weighted comprehensive evaluation on the original state data to generate the signal quality parameter; obtaining a critical performance indicator used to characterize the transition of the signal from reliable to unreliable, and generating a quality threshold; when the signal quality parameter is lower than a preset quality threshold, generating a dynamic compensation mode trigger signal.

[0008] Optionally, the clock working environment parameters of the local clock source are collected, and drift characteristics are predicted in combination with the standard time reference and the clock working environment parameters to generate a short-term clock offset prediction value, including: obtaining the real-time operating temperature and power supply voltage of the local clock source to constitute the clock working environment parameters; establishing a drift characteristic prediction model for describing the corresponding relationship between clock frequency drift and the clock working environment parameters; inputting the clock working environment parameters collected in real time into the drift characteristic prediction model, combining the standard time reference as the initial state, and calculating and outputting the short-term clock offset prediction value.

[0009] Optionally, the establishment of a drift characteristic prediction model for describing the correspondence between clock frequency drift and the clock working environment parameters includes: when the signal quality parameter is greater than the quality threshold, continuously recording the clock working environment parameters under different working conditions; obtaining actual clock drift data corresponding to the clock working environment parameters by continuously comparing the actual time of the local clock source with the standard time reference; based on the multivariate correspondence between the clock working environment parameters and the actual clock drift data, training is performed through fitting or machine learning methods to generate a drift characteristic prediction model.

[0010] Optionally, generating a topology correction includes: determining a set of topologically adjacent nodes based on a topological connection relationship of the distribution network, and sending a time synchronization request to the set of topologically adjacent nodes to obtain adjacent node timestamps; extracting line physical parameters from the topological connection relationship of the distribution network, and calculating a theoretical time difference for signal propagation between nodes based on the line physical parameters; calculating the received adjacent node timestamps with the time when the local time synchronization request is sent to obtain a measured time difference; deducting the theoretical time difference and network communication delay from the measured time difference to generate a topology correction.

[0011] Optionally, sending a time synchronization request to the set of topologically adjacent nodes to obtain adjacent node timestamps includes: sending a time synchronization request to the set of topologically adjacent nodes, and receiving multiple returned adjacent node timestamps to form an initial timestamp group; evaluating the consistency and validity of each timestamp in the initial timestamp group, eliminating outliers or abnormal values ​​therein, and generating a preferred timestamp group; averaging or weighted averaging the timestamps in the preferred timestamp group to generate adjacent node timestamps.

[0012] Optionally, the fusion of the short-term clock offset prediction value and the topology correction amount to generate a dynamic compensation time base includes: while obtaining the adjacent node timestamp, obtaining a node timing quality indicator that characterizes the timing status of the node from which the adjacent node timestamp is sourced; when there is a topologically adjacent node whose node timing quality indicator is higher than a preset autonomous operation threshold, generating a dynamic fusion weight for adjusting the confidence of the topology correction amount and the short-term clock offset prediction value based on the node timing quality indicator; using the dynamic fusion weight, performing a weighted summation on the short-term clock offset prediction value and the topology correction amount to generate the dynamic compensation time base.

[0013] Optionally, the method also includes: when the node timing quality indicators of all topologically adjacent nodes are lower than the autonomous operation threshold, obtaining the adjacent node timestamps of all nodes in the topologically adjacent node set to form a regional timestamp set; using the discreteness of the adjacent node timestamps in the regional timestamp set and minimizing the discreteness as the goal, establishing a target optimization function using the discreteness of the adjacent node timestamps in the regional timestamp set, and solving to obtain the topology correction amount corresponding to the minimized discreteness; and updating the current topology correction amount using the topology correction amount corresponding to the minimized discreteness.

[0014] Optionally, after generating the dynamic compensation time base, it also includes: continuing to monitor the real-time status of the Beidou satellite signal and updating the signal quality parameters; judging whether the updated signal quality parameters within the preset time window are lower than the quality threshold; if not, terminating the use of the dynamic compensation time base, and re-calibrating the local clock source according to the Beidou satellite signal to obtain an updated standard time base; and using the updated standard time base to time-mark the electrical quantity measurement data.

[0015] Based on the same inventive concept, the present invention also provides a distributed measurement and synchronization device for a distribution network based on Beidou satellite timing, the device comprising: a Beidou calibration module for acquiring Beidou satellite signals, and calibrating a local clock source according to the Beidou satellite signals to generate a standard time reference; a signal quality monitoring module for continuously monitoring the real-time status of Beidou satellite signals, generating signal quality parameters, and generating a dynamic compensation mode trigger signal when the signal quality parameters are lower than a preset quality threshold; a drift characteristic prediction module for responding to the dynamic compensation mode trigger signal, collecting the clock working environment parameters of the local clock source, and performing drift characteristic prediction in combination with the standard time reference and the clock working environment parameters to generate a short-term A clock offset prediction value; a topology information interaction module, used to determine a set of topologically adjacent nodes based on the topological connection relationship of the distribution network, and send a time synchronization request to the topologically adjacent node set to obtain adjacent node timestamps, and then calculate the theoretical time difference in combination with the line physical parameters contained in the distribution network topological connection relationship, and use the theoretical time difference to correct the adjacent node timestamps to generate a topology correction value; a time reference fusion module, used to fuse the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference; a timestamp generation module, used to collect electrical quantity measurement data of the distribution network, and use the dynamic compensation time reference to time-mark the electrical quantity measurement data to generate synchronized measurement data with timestamps.

[0016] Compared with the prior art, the present invention has the following advantages: 1. When the Beidou satellite signal is reliable, the present invention uses high-precision satellite signals to directly calibrate the local clock, ensuring that the reference time of the measured data has extremely high accuracy. When the Beidou signal becomes unreliable due to reasons such as obstruction, the device can seamlessly switch to an innovative dynamic compensation mode. By combining the prediction of the physical characteristics of the device's own clock and network collaborative correction based on the power grid topology, time drift is effectively suppressed, thereby maintaining measurement synchronization in harsh environments and ensuring the availability and reliability of data at all times. 2. This invention learns and establishes a mathematical model of clock drift and environmental factors such as temperature and voltage in an offline state. This enables the measurement device to perform high-precision self-time calibration based on its own environmental perception after losing an external time source. This method fundamentally solves the problem of excessive frequency drift caused by environmental changes in traditional crystal oscillators in timekeeping mode, significantly improving the time maintenance accuracy of the device during autonomous operation. 3. The present invention introduces a collaborative timing mechanism based on the distribution network topology. When the timing signal of a single node is lost, it can actively communicate with its neighboring nodes and obtain a topologically corrected time reference by using the physical connection relationship and line parameters. Furthermore, in the extreme case of regional signal interruption, the present invention also designs a mechanism for collaborative optimization of the entire network to maintain the relative time consistency of the entire network by minimizing the discreteness of the node clocks in the region. This distributed and adaptive timing strategy greatly enhances the resilience and robustness of the entire measurement device.

[0017] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 The present invention is a flowchart of a method for distributing measurement synchronization in a distribution network based on BeiDou satellite timing.

[0020] Figure 2 It is a structural diagram of a distributed measurement synchronization device for a distribution network based on Beidou satellite timing according to an embodiment of the present invention.

[0021] Figure 3 This is a comparison diagram of node time errors in different synchronization modes according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0023] Reference Figure 1 One embodiment of the present invention proposes a distributed measurement synchronization method for a distribution network based on Beidou satellite timing. When the Beidou signal is lost, the method uses a technical solution that combines the local clock drift prediction value with the adjacent node topology correction value to generate a dynamic compensation time reference. This can significantly improve the accuracy and robustness of measurement data synchronization in distribution networks in complex environments.

[0024] The method of this embodiment specifically includes: Acquire BeiDou satellite signals, and calibrate a local clock source according to the BeiDou satellite signals to generate a standard time reference; Specifically, the measurement device must first be equipped with a Beidou timing receiver. This receiver, equipped with an antenna, continuously receives L-band radio frequency signals broadcast by the Beidou navigation satellite system and decodes precise time information. The key to acquiring Beidou satellite signals lies in separating two key time data types from the decoded navigation message: a serial time message containing the year, month, day, hour, minute, and second, and a high-precision pulse-per-second signal (1PPS). The 1PPS signal is a physical level signal whose rising edge is strictly aligned with the second boundary of Coordinated Universal Time (UTC), ensuring extremely high accuracy. The local clock source is typically a high-stability crystal oscillator within the measurement device, such as a temperature-compensated crystal oscillator (TCXO). This generates a stable frequency signal to drive an internal counter to maintain local time. Calibration of the local clock source involves accurately comparing and correcting local time with Beidou satellite time. Initially, by reading the serial time message, the local clock source's counter is roughly set to align with standard time at the second level. Then, the fine calibration phase begins. Using the precise second edge of the 1PPS pulse-per-second signal as an external trigger event, the local clock source's own timestamp, recorded at the moment the rising edge of the pulse signal arrives, is captured. By calculating the difference between this timestamp and the standard time second boundary represented by the 1PPS pulse-per-second signal, a precise local clock offset is obtained: , in, Represents the local clock offset that needs to be calibrated; Represents the local clock source's own timestamp recorded when the rising edge of the pulse-per-second signal 1PPS arrives. This value is obtained directly from the local clock source's counter through a hardware interrupt or a time capture unit. It represents the Coordinated Universal Time second boundary corresponding to the 1PPS pulse-per-second signal, as previously announced by the serial time message. After determining the local clock offset, a software-implemented phase-locked loop (PLL) or frequency-locked loop (FLL) algorithm continuously fine-tunes the operating frequency of the local clock source or dynamically adjusts the count value of the local time counter, with the goal of bringing the local clock offset obtained in subsequent measurements close to zero. This closed-loop control process of continuous comparison, calculation, and adjustment ultimately synchronizes the time output by the local clock source with Beidou satellite time. The resulting precisely calibrated and continuously locked local time constitutes the standard time reference for the entire measurement device.

[0025] Continuously monitor the real-time status of BeiDou satellite signals, generate signal quality parameters, and generate a dynamic compensation mode trigger signal when the signal quality parameters are lower than a preset quality threshold; In response to the dynamic compensation mode trigger signal, collecting clock working environment parameters of the local clock source, and performing drift characteristic prediction in combination with the standard time reference and the clock working environment parameters to generate a short-term clock offset prediction value; Determine a set of topologically adjacent nodes based on the topological connection relationship of the distribution network, send a time synchronization request to the set of topologically adjacent nodes to obtain adjacent node timestamps, calculate a theoretical time difference based on the line physical parameters contained in the topological connection relationship of the distribution network, and use the theoretical time difference to correct the adjacent node timestamps to generate a topological correction value; Fusion of the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference; The electrical quantity measurement data of the distribution network is collected, and the electrical quantity measurement data is time-stamped using the dynamic compensation time reference to generate synchronized measurement data with a time stamp.

[0026] Optionally, generating a dynamic compensation mode trigger signal includes: Collecting the number of receiving satellites, signal-to-noise ratio, and position precision factor of the Beidou satellite signal to form original state data; Performing a weighted comprehensive evaluation on the original state data to generate the signal quality parameter; Obtaining critical performance indicators for characterizing the transition of a signal from reliable to unreliable, and generating quality thresholds; When the signal quality parameter is lower than a preset quality threshold, a dynamic compensation mode trigger signal is generated.

[0027] Specifically, in order to accurately judge the reliability of the Beidou satellite timing signal and generate a dynamic compensation mode trigger signal, it is first necessary to collect a series of raw status data reflecting the signal quality from the Beidou timing receiver in real time, mainly including the number of received satellites, signal-to-noise ratio, and position precision factor. The number of received satellites is the number of satellites that can currently be effectively tracked and used for resolution; the signal-to-noise ratio is an indicator that measures the relative relationship between the received satellite signal strength and the background noise level, which directly affects the accuracy of information decoding; the position precision factor is a dimensionless value that characterizes the geometric configuration of the satellite in the sky. The smaller the value, the more ideal the satellite distribution and the higher the timing and positioning accuracy. These three sets of raw status data together constitute a comprehensive description of the current Beidou signal status. A weighted comprehensive evaluation is performed through these raw status data to generate a single signal quality parameter: , in, is the final signal quality parameter. , and These are preset weight coefficients corresponding to the number of received satellites, signal-to-noise ratio, and position precision factor, respectively. Their sum is 1. Their specific values ​​are determined based on experimental data or expert experience, reflecting the relative importance of each factor to timing accuracy. It represents a normalization function, which is used to map the original state data of different dimensions to a unified dimensionless interval for weighted summation. For example, for the number of received satellites and signal-to-noise ratio with better values, the normalization function maps them to larger values; and for the position precision factor with smaller values, the normalization function will perform the reverse processing and map it to a larger value as well. Next, it is necessary to obtain a quality threshold, which is a critical performance indicator that characterizes the transition from reliable to unreliable signal. This quality threshold is obtained by comparing the real-time calculated signal quality parameters with the actual timing error measured by the high-precision time reference source under different signal conditions, finding the signal quality parameter value corresponding to when the timing error exceeds the allowable range of the distribution network application, and setting it as the preset quality threshold. During operation, the current signal quality parameters will be continuously calculated and compared with the preset quality threshold. Once the signal quality parameters are found to be lower than the preset quality threshold, the Beidou signal is immediately judged to be unreliable and a dynamic compensation mode trigger signal is immediately generated. This signal will initiate the subsequent internal clock drift prediction and topology coordinated timing mechanism to ensure that the measurement device can still provide a high-precision synchronized time reference even when the satellite signal is poor.

[0028] Optionally, collecting clock operating environment parameters of the local clock source, and performing drift characteristic prediction in combination with the standard time reference and the clock operating environment parameters to generate a short-term clock offset prediction value includes: Obtaining the real-time operating temperature and supply voltage of the local clock source to form clock operating environment parameters; Establishing a drift characteristic prediction model for describing the corresponding relationship between clock frequency drift and the clock operating environment parameters; The clock working environment parameters collected in real time are input into the drift characteristic prediction model, combined with the standard time reference as an initial state, to calculate and output a short-term clock offset prediction value.

[0029] Specifically, in response to the dynamic compensation mode trigger signal, to generate a short-term clock offset prediction, the system first accurately senses the operating status of the local clock source. A temperature sensor deployed near the local clock source's physical packaging and a voltage monitoring unit built into the power supply circuit periodically and in real time collect the current operating temperature and supply voltage. These two physical quantities together constitute the clock operating environment parameters. Simultaneously, a pre-established drift characteristic prediction model is invoked. This model has been trained offline and internally captures the nonlinear relationship between the frequency drift of this specific local clock source and its clock operating environment parameters. At the start of the prediction calculation—the instantaneous receipt of the dynamic compensation mode trigger signal—the current standard time reference is used as the initial state without offset. Subsequently, the real-time collected clock operating environment parameters are continuously fed into the drift characteristic prediction model. Based on the input environmental parameters, the model calculates in real time the normalized frequency drift of the local clock source relative to its ideal rated frequency under the current operating conditions. This instantaneous frequency drift is integrated over a small time step and accumulated starting from the initial state to calculate the total accumulated time offset since the Beidou satellite signal was lost. The accumulated time deviation is the short-term clock offset prediction value, and its generation process can be expressed as follows: , in, Is the output short-term clock offset prediction value. The summation symbol Indicates that the drift calculated in each sampling period is accumulated starting from the moment the dynamic compensation mode is triggered. is the normalized frequency drift output by the drift characteristic prediction model in the i-th sampling period, which is determined by the real-time operating temperature collected in the period. and supply voltage The only certainty is that it is a dimensionless quantity. This is the fixed time step for collecting environmental parameters and performing model calculations. This allows for dynamic prediction of the timing error of the local clock source based on real-time perception of the local physical environment.

[0030] Optionally, the establishing of a drift characteristic prediction model for describing the corresponding relationship between clock frequency drift and the clock operating environment parameter includes: When the signal quality parameter is greater than the quality threshold, continuously recording clock working environment parameters under different working conditions; By continuously comparing the actual time of the local clock source with the standard time reference, actual clock drift data corresponding to the clock working environment parameters is obtained; Based on the multivariate correspondence between the clock working environment parameters and the actual clock drift data, a drift characteristic prediction model is generated by training through fitting or machine learning methods.

[0031] Specifically, in order to establish a drift characteristic prediction model for describing the corresponding relationship between the frequency drift of the local clock source and the clock working environment parameters, this method first performs an offline learning and modeling process. This process is carried out under the condition that the Beidou satellite signal quality parameters are confirmed to be continuously higher than the preset quality threshold, that is, the timing signal is absolutely reliable. During this period, the standard time base generated by high-precision Beidou satellite signal calibration is used as the true value reference. The real-time operating temperature and power supply voltage of the local clock source are continuously collected through the built-in high-precision sensors. These parameters constitute the clock working environment parameters that reflect the current operating status. In order to build a comprehensive model, this data collection process needs to be carried out extensively under various working conditions that the measuring device may experience to cover a wide range of temperature and voltage changes. While collecting environmental parameters, the actual time output by the local clock source is continuously and accurately compared with the standard time base, so as to accurately quantify the actual clock drift data caused by environmental changes. The actual clock drift data, that is, the frequency drift of the local clock source, can be calculated in the following way: , in, It represents the actual frequency drift measured in a very short time interval and is a dimensionless ratio. and In real physical time and Time reading recorded by a local clock source. and The precise time provided by a standard time reference at the same physical point in time. This differential calculation yields accurate actual clock drift data corresponding to specific clock operating environment parameters. Over a long period of operation, a large dataset of multivariate correspondences between clock operating environment parameters and actual clock drift data has been accumulated. Based on this dataset, training is performed using multivariate regression fitting analysis or machine learning methods such as support vector machines and neural networks, ultimately generating a robust drift characteristic prediction model. This model, taking newly acquired clock operating environment parameters as input, accurately outputs the corresponding clock frequency drift prediction value.

[0032] Optionally, generating a topology correction value includes: Determine a topological adjacent node set according to the topological connection relationship of the distribution network, and send a time synchronization request to the topological adjacent node set to obtain adjacent node timestamps; Extracting line physical parameters from the topological connection relationship of the distribution network, and calculating the theoretical time difference of signal propagation between nodes based on the line physical parameters; Calculate the received timestamp of the adjacent node and the time when the local time synchronization request is sent to obtain the measured time difference; The theoretical time difference and the network communication delay are deducted from the measured time difference to generate a topology correction amount.

[0033] Specifically, to generate the topology correction, first, based on the stored distribution network topology connection relationship, query and determine all nodes directly connected to the current measurement node on the physical line to form a set of topological adjacent nodes. Then, obtain their respective adjacent node timestamps. At the same time, extract the physical parameters of the line connecting the current node and each adjacent node from the distribution network topology connection relationship data, especially the precise length of the line. Based on these line physical parameters, the theoretical time difference required for the signal to propagate between two nodes along the power line can be calculated. The theoretical time difference is a time amount determined by the physical distance and the propagation medium. Subsequently, the adjacent node timestamp received from the adjacent node is compared with the local timestamp recorded by the local node when sending the corresponding time synchronization request, and the original difference between the two is calculated, that is, the measured time difference. This measured time difference is a mixture of real clock error, physical propagation delay, and communication system delay. Finally, in order to separate the pure clock deviation, that is, to generate the topology correction, the theoretical time difference and network communication delay are subtracted from the measured time difference: , in, It is the final topology correction, which directly reflects the deviation of the local clock relative to the average clock of the adjacent nodes. is the obtained timestamp of the adjacent node. It is the timestamp recorded by the local node when initiating a time synchronization request, provided by the local clock source. is the theoretical time difference, which is obtained by dividing the physical length of the line L by the signal propagation speed v. L is extracted from the topological connection relationship of the distribution network, and v is a known physical constant. The network communication delay represents the fixed processing time within a node, obtained through offline calibration. This series of operations effectively converts the combined clock information of adjacent nodes into a precise correction to the local clock, thereby achieving distributed peer-to-peer time synchronization within the region even in environments without satellite signals.

[0034] Optionally, sending a time synchronization request to the topological adjacent node set to obtain adjacent node timestamps includes: Sending a time synchronization request to the topological adjacent node set, and receiving multiple returned adjacent node timestamps to form an initial timestamp group; Evaluate the consistency and validity of each timestamp in the initial timestamp group, remove outliers or abnormal values ​​therein, and generate an optimal timestamp group; The timestamps in the preferred timestamp group are averaged or weighted averaged to generate adjacent node timestamps.

[0035] Specifically, to accurately obtain neighboring node timestamps for time synchronization comparison, this method first sends a time synchronization request to all nodes within a defined set of topologically adjacent nodes via the communication network. In response, each topologically adjacent node returns its current time information. These time data returned by multiple different nodes together constitute the initial timestamp group. Due to potential jitter in network transmission or temporary significant deviations in the clocks of individual neighboring nodes, the data in the initial timestamp group may not be completely consistent or reliable. Therefore, this initial timestamp group must be screened to improve the accuracy of the final reference time. This process involves evaluating the consistency and validity of each timestamp in the initial timestamp group. This step can be performed using statistical methods, such as calculating the mean and standard deviation of the timestamp group. Based on these statistics, a reasonable confidence interval is set, and any values ​​outside this interval are identified as outliers. These outliers, which may be caused by excessive latency due to network anomalies or significant deviation from the group due to source node clock failures, should be considered outliers and removed from the initial timestamp group. This screening step results in a preferred timestamp group with higher data quality and better consistency. Finally, to generate a unique and representative neighboring node timestamp from the preferred timestamp group, the remaining valid timestamps in the group need to be fused. The most direct approach is to perform an arithmetic average, which adds up all the timestamp values ​​in the preferred timestamp group and divides the sum by the number of timestamps to obtain an average value. In more sophisticated strategies, a weighted average can be used. The weights here can be based on the reliability information of each neighboring node timestamp source, such as the quality of the Beidou signal of the node itself or the degree of synchronization consistency with more other nodes. By averaging or weighted averaging the preferred timestamp group, a robust and accurate neighboring node timestamp is ultimately generated, which integrates the time information of multiple reliable nodes in the area.

[0036] Optionally, the fusing the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference includes: While obtaining the adjacent node timestamp, obtaining a node timing quality indicator that characterizes the timing status of the node from which the adjacent node timestamp is derived; When a node timing quality index of a topologically adjacent node is higher than a preset autonomous operation threshold, a dynamic fusion weight for adjusting the topology correction amount and the confidence level of the short-term clock offset prediction value is generated based on the node timing quality index; The dynamic fusion weight is used to perform a weighted summation on the short-term clock offset prediction value and the topology correction amount to generate a dynamic compensation time reference.

[0037] Specifically, when Beidou satellite signals become unreliable and dynamic compensation mode is entered, generating an accurate dynamically compensated time base requires intelligently fusing two different types of time correction information: a short-term clock offset prediction based on local physical environment perception and a topological correction based on network collaboration. This fusion process is not a simple addition, but rather employs an adaptive dynamic weighting strategy. Its core focus is on evaluating and leveraging the timing reliability of topologically adjacent nodes. Specifically, when a node sends a time synchronization request to a set of topologically adjacent nodes and obtains a timestamp from the neighboring node, in addition to the timestamp itself, it must also synchronously obtain a key parameter: the node timing quality indicator (NTQI). The NTQI is a quantitative value that characterizes the current status and reliability of the timing of the neighboring node providing the timestamp. For example, this indicator can comprehensively reflect whether the neighboring node is receiving high-quality Beidou satellite signals or the stability of its internal clock. This indicator is sent to the node by the neighboring node in its response to the time synchronization request.

[0038] Next, the obtained node timing quality indicators of each topologically adjacent node are compared with a preset autonomous operation threshold. The autonomous operation threshold is a pre-set critical value used to determine whether the time base of a neighboring node is sufficiently reliable to serve as a timing reference for the region. If the node timing quality indicator of at least one node in the set of topologically adjacent nodes is found to be significantly higher than this autonomous operation threshold, this indicates that there are one or more "master" nodes in the network with excellent timing quality. At this time, a pair of dynamic fusion weights are generated based on these node timing quality indicators that are above the threshold. The higher the node timing quality indicator, the more reliable the time information it provides. Therefore, a higher weight is assigned to the topology correction, while the weight of the local self-predicted short-term clock offset prediction value is correspondingly reduced.

[0039] Finally, using this pair of dynamic fusion weights, the short-term clock offset prediction value and the topology correction amount are weighted and summed to calculate the final dynamic compensation time base: , in, is the generated dynamically compensated time base, representing the total correction to the local clock. and are the dynamic fusion weights of the short-term clock offset prediction value and the topology correction value, respectively. They are dimensionless confidence coefficients and their sum is 1. This is the clock offset predicted by the local clock source based on its own environment parameters, that is, the short-term clock offset prediction value. The clock deviation correction is calculated by comparing the time with the topologically adjacent nodes, i.e. the topological correction. Through this refined adaptive fusion mechanism, a more robust and accurate dynamic compensation time base can be generated.

[0040] Optionally, the method further includes: When the node timing quality indicators of all topologically adjacent nodes are lower than the autonomous operation threshold, obtaining the adjacent node timestamps of all nodes in the topologically adjacent node set to form a regional timestamp set; Taking the discreteness of adjacent node timestamps in the regional timestamp set and minimizing the discreteness as the goal, a target optimization function is established using the discreteness of adjacent node timestamps in the regional timestamp set, and a topology correction amount corresponding to the minimized discreteness is obtained by solving the problem; The current topology correction amount is updated using the topology correction amount corresponding to the minimized discreteness.

[0041] Specifically, when the node timing quality indicators of all topologically adjacent nodes fall below a preset autonomous operation threshold, this indicates that all measurement nodes in the entire region have lost a reliable external time source, such as the BeiDou satellite signal, and are unable to find a node with significantly better timing quality to serve as a dominant reference. In this level-of-stake autonomous operation mode, the goal of time synchronization shifts from pursuing absolute time accuracy to maintaining relative time consistency across all nodes in the region. To this end, an optimization strategy aimed at minimizing regional time dispersion is employed to generate topological corrections.

[0042] Specifically, the node first needs to proactively initiate a full time information exchange with its entire set of topologically adjacent nodes. This not only allows the node to obtain the timestamps of each neighboring node, but more importantly, through this exchange, each node can learn the timestamps of all its neighbors. This allows the node to construct a regional timestamp set. This regional timestamp set includes the timestamps of the node itself and all of its topologically adjacent nodes, essentially aggregating the clock information of all nodes in the adjacent area centered on the node.

[0043] Next, calculate the dispersion within the timestamp set of the region in the current state. Dispersion is a statistical indicator used to quantify the degree of deviation of a set of data points around its central value. Commonly used indicators are variance or standard deviation. In order to calculate the dispersion, it is necessary to first calculate the average value of all timestamps in the regional timestamp set, then calculate the sum of the squares of the difference between each timestamp and this average value, and finally find the average value, that is, the variance. After that, enter an optimization calculation process, whose goal is to find an optimal topology correction. After this topology correction is applied to the clock of this node, it can minimize the dispersion of the new set composed of the corrected timestamp of this node and the timestamps of other nodes in the region. That is, find a correction value , so that after applying this correction, the time of the entire area reaches the maximum degree of coordination. This process can be expressed as minimizing the objective function: , in, Represents the minimum value search operation. is the variance of the regional timestamp set, i.e., the dispersion. These are the timestamps of other nodes in the region. These values ​​are obtained from the region timestamp collection. Is the current uncorrected time of this node. is the topological correction quantity to be determined that can minimize the time discreteness of the entire region. By solving this optimization problem, we can obtain This is the final output topology correction, which aligns the time of this node to the overall "center" of the regional time. This topology correction is used to calculate the dynamic compensation time base.

[0044] Optionally, after generating the dynamic compensation time reference, the method further includes: Continuing to monitor the real-time status of the BeiDou satellite signal and updating signal quality parameters; Determining whether the updated signal quality parameter within the preset time window is lower than the quality threshold; If not, terminating the use of the dynamic compensation time reference, and recalibrating the local clock source according to the BeiDou satellite signal to obtain an updated standard time reference; The electrical quantity measurement data is time-stamped using the updated standard time reference.

[0045] Specifically, after switching to dynamic compensation mode due to poor Beidou satellite signal quality and beginning to use the dynamic compensation time base to timestamp electrical quantity measurement data, external signal monitoring does not cease. Instead, a continuous, dynamic recovery monitoring and switching mechanism is initiated to ensure rapid restoration to optimal synchronization once external timing conditions improve. Specifically, even in dynamic compensation mode, the signal quality monitoring module continues to run uninterrupted in the background, continuously collecting and processing raw Beidou satellite signal data to update signal quality parameters in real time. This process is identical to the monitoring behavior before dynamic compensation mode is triggered. Next, a preset time window is introduced, defining an observation period for determining whether signal quality has stabilized. Within this time window, each newly generated signal quality parameter is continuously evaluated to see if it has returned to a preset quality threshold. The key to this step is that switching is not based on a single recovery in signal quality parameters; rather, it requires that signal quality remain stable and reliable over a sustained period of time. If all updated signal quality parameters fail to consistently exceed the quality threshold throughout the preset time window, the Beidou satellite signal is deemed unrecoverable and the current dynamic compensation mode remains in effect, with the dynamic compensation time base used for time-stamping electrical measurement data. Conversely, if the updated signal quality parameters consistently and consistently exceed the preset quality threshold within the preset time window, this indicates that the Beidou satellite signal timing has recovered from an unreliable state to a high-quality state. The system then determines that the conditions for exiting dynamic compensation mode have been met, and immediately terminates the use of the dynamic compensation time base generated through internal prediction and topology collaboration. The initial Beidou calibration process is then re-executed, utilizing the current high-quality Beidou satellite signal to perform a new, precise calibration of the local clock source. This eliminates any minor errors that may have accumulated during dynamic compensation, generating an updated standard time base that is highly synchronized with Coordinated Universal Time (UTC). Upon obtaining this updated standard time base, a seamless switch occurs, immediately using this most accurate standard time base for time-stamping newly collected distribution network electrical measurement data.

[0046] Based on the same inventive concept, the present invention also provides a distributed measurement synchronization device for a distribution network based on Beidou satellite timing, the device comprising: A BeiDou calibration module is used to obtain BeiDou satellite signals and calibrate the local clock source according to the BeiDou satellite signals to generate a standard time reference; A signal quality monitoring module is used to continuously monitor the real-time status of Beidou satellite signals, generate signal quality parameters, and generate a dynamic compensation mode trigger signal when the signal quality parameters are lower than a preset quality threshold; a drift characteristic prediction module, configured to respond to the dynamic compensation mode trigger signal, collect clock operating environment parameters of the local clock source, and perform drift characteristic prediction in combination with the standard time reference and the clock operating environment parameters to generate a short-term clock offset prediction value; A topology information interaction module is configured to determine a set of topologically adjacent nodes based on a topological connection relationship of the distribution network, send a time synchronization request to the set of topologically adjacent nodes to obtain timestamps of adjacent nodes, calculate a theoretical time difference based on the line physical parameters contained in the topological connection relationship of the distribution network, and use the theoretical time difference to correct the timestamps of the adjacent nodes to generate a topology correction value; A time reference fusion module, configured to fuse the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference; The timestamp generation module is used to collect electrical quantity measurement data of the distribution network, and use the dynamic compensation time reference to time-mark the electrical quantity measurement data to generate synchronous measurement data with a timestamp.

[0047] To verify the feasibility and effectiveness of this invention in practical applications, the invention was applied to a distributed synchronous measurement project for a regional distribution network. This project aims to deploy high-precision synchronous measurement devices at multiple distribution network nodes, such as distributed power access points, intelligent switches, and distribution terminal units (DTUs). This allows for high-precision synchronous acquisition of electrical quantities such as voltage and current across the entire network, supporting advanced distribution automation applications such as precise fault location, state estimation, and new energy consumption analysis.

[0048] The synchronization method and device of the present invention were integrated and deployed in a 20-node 10kV simulated distribution network environment. This network simulated a complex urban scenario with a mix of overhead lines and cables, as well as building obstructions, to test the device's synchronization performance under both ideal and non-ideal Beidou satellite signal conditions.

[0049] In this embodiment, each distribution terminal unit equipped with the present invention first acquires Beidou satellite signals through its built-in Beidou calibration module. The Beidou timing receiver within this module decodes the satellite signals into a serial time message and a high-precision pulse-per-second signal (1PPS). The serial time message provides year, month, day, hour, minute, and second information. By aligning the serial message with an internal clock counter and triggering hardware time capture using the precise rising edge of the 1PPS signal, the device calculates the precise offset between the local clock source (temperature-compensated crystal oscillator (TCXO)) and standard UTC time. Subsequently, a software phase-locked loop (PLL) algorithm continuously fine-tunes the local clock to maintain high synchronization with Beidou time, generating a standard time reference. Under good Beidou signal conditions, the synchronization accuracy of each node is measured to be better than 100 nanoseconds, fully meeting the requirements for synchronized phasor measurement in distribution networks.

[0050] To verify the effectiveness of the dynamic compensation mechanism, experiments simulated scenarios where Beidou signal loss could occur, such as using a GPS signal jammer or placing the device deep inside a building. The signal quality monitoring module continuously evaluated parameters such as the number of received satellites, signal-to-noise ratio (SNR), and position dilution of precision (PDOP). The experiment was designed so that when the weighted, comprehensively evaluated signal quality parameter value fell below a preset threshold of 0.6, the device automatically generated a trigger signal for dynamic compensation mode. The preset threshold of 0.6 corresponds to the critical point at which synchronization error could exceed 1 microsecond.

[0051] Once dynamic compensation mode is entered, the drift characteristic prediction module immediately activates. This module first calls a drift characteristic prediction model pre-trained through machine learning. This model was developed by recording actual local clock drift data at different operating temperatures and supply voltages when the Beidou signal is strong. After signal loss, the module inputs real-time local clock operating environment parameters, such as a temperature increase from 25°C to 40°C, into the prediction model to calculate a short-term clock offset prediction. Data shows that within the first 30 minutes without any external reference, compensation based solely on this prediction keeps clock drift error within 5 microseconds, a significant improvement compared to no compensation.

[0052] At the same time, the topology information exchange module is activated. Based on the pre-stored distribution network topology, the device sends time synchronization requests to adjacent nodes directly connected to the physical line, such as upstream switches and downstream DTUs. Through multiple requests and responses, a reliable, outlier-free adjacent node timestamp is obtained. After losing the Beidou signal, a node requests time from its three adjacent nodes. If one of the adjacent nodes still receives a good Beidou signal, the timestamp of this adjacent node is assigned a higher weight. The device combines the physical parameters of the line, namely a line length of 500 meters, to calculate a theoretical propagation delay of approximately 2.5 microseconds, and then calculates the topology correction.

[0053] The time reference fusion module dynamically weights and fuses these two corrections. In this scenario, because a neighboring node is still providing high-quality Beidou timing—that is, its timing quality index exceeds the preset autonomous operation threshold—the device assigns a higher weight of 0.7 to the topology correction and a weight of 0.3 to the local short-term clock offset prediction. The resulting dynamically compensated time reference ensures that the node's synchronization error remains within 2 microseconds during a one-hour signal loss, far exceeding the performance of relying on either compensation method alone.

[0054] In another test, a regional Beidou signal outage was simulated, and the timing quality indicators of all adjacent nodes fell below the autonomous operation threshold. At this point, the system switched to fully autonomous operation mode. Each node, aiming to minimize regional time dispersion, calculated topological corrections using an optimization algorithm, achieving high relative consistency of time across all nodes in the region. The experimental results showed that despite slight drift in absolute time, the synchronization error between any two nodes in the region remained within 5 microseconds, effectively ensuring the accuracy of applications such as fault location based on phase difference calculations.

[0055] When the external signal interference is removed and the signal quality monitoring module detects that the signal quality parameter remains stable above 0.8 for 5 minutes, the device determines that the Beidou signal has been restored. It then automatically exits dynamic compensation mode and re-executes the Beidou signal calibration process to seamlessly align the local clock with the standard time.

[0056] Table 1 Comparison of node time errors under different synchronization modes

[0057] Table 2 Device mode switching response time data table

[0058] In order to more intuitively demonstrate the technical effects of the present invention, Figure 3 As shown in the figure, the variation of node time error with signal loss time under different synchronization modes is compared. From the data in Tables 1 and 2 above, it can be seen that when the Beidou satellite signal is unreliable or completely interrupted, the method and device of the present invention can significantly improve the synchronization accuracy and robustness of the device through a dynamic compensation mechanism that combines internal clock drift prediction with external topology collaborative timing. Compared with the uncompensated free-running mode, the present invention reduces the synchronization error of up to 1 hour from hundreds of microseconds to less than 5 microseconds, ensuring the normal operation of advanced applications in the distribution network.

[0059] It should be noted that the electrical connections between the above-mentioned units do not necessarily mean direct connections of lines. Indirect connections are applicable to the embodiments of the present invention as long as the purpose of the present invention is achieved. The above description is only an exemplary embodiment of the present invention and is not intended to limit the scope of the present invention.

[0060] That is, any equivalent changes and modifications made according to the teachings of the present invention are still within the scope of the present invention. Those skilled in the art will readily conceive of other embodiments of the present invention after considering the disclosure of the specification and practical truths. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary technical means in the art not described herein.

Claims

1. A distributed measurement synchronization method for distribution network based on Beidou satellite timing, characterized in that: The method comprises: Acquire BeiDou satellite signals, and calibrate a local clock source according to the BeiDou satellite signals to generate a standard time reference; Continuously monitor the real-time status of BeiDou satellite signals, generate signal quality parameters, and generate a dynamic compensation mode trigger signal when the signal quality parameters are lower than a preset quality threshold; In response to the dynamic compensation mode trigger signal, collecting clock working environment parameters of the local clock source, and performing drift characteristic prediction in combination with the standard time reference and the clock working environment parameters to generate a short-term clock offset prediction value; Determine a set of topologically adjacent nodes based on the topological connection relationship of the distribution network, send a time synchronization request to the set of topologically adjacent nodes to obtain adjacent node timestamps, calculate a theoretical time difference based on the line physical parameters contained in the topological connection relationship of the distribution network, and use the theoretical time difference to correct the adjacent node timestamps to generate a topological correction value; Fusion of the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference; The electrical quantity measurement data of the distribution network is collected, and the electrical quantity measurement data is time-stamped using the dynamic compensation time reference to generate synchronized measurement data with a time stamp.

2. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 1, characterized in that: Generating a dynamic compensation mode trigger signal includes: Collecting the number of receiving satellites, signal-to-noise ratio, and position precision factor of the Beidou satellite signal to form original state data; Performing a weighted comprehensive evaluation on the original state data to generate the signal quality parameter; Obtaining critical performance indicators for characterizing the transition of a signal from reliable to unreliable, and generating quality thresholds; When the signal quality parameter is lower than a preset quality threshold, a dynamic compensation mode trigger signal is generated.

3. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 2, characterized in that: The collecting of clock operating environment parameters of the local clock source and performing drift characteristic prediction in combination with the standard time reference and the clock operating environment parameters to generate a short-term clock offset prediction value includes: Obtaining the real-time operating temperature and supply voltage of the local clock source to form clock operating environment parameters; Establishing a drift characteristic prediction model for describing the corresponding relationship between clock frequency drift and the clock operating environment parameters; The clock working environment parameters collected in real time are input into the drift characteristic prediction model, combined with the standard time reference as an initial state, to calculate and output a short-term clock offset prediction value.

4. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 3, characterized in that: The step of establishing a drift characteristic prediction model for describing the corresponding relationship between clock frequency drift and the clock operating environment parameters includes: When the signal quality parameter is greater than the quality threshold, continuously recording clock working environment parameters under different working conditions; By continuously comparing the actual time of the local clock source with the standard time reference, actual clock drift data corresponding to the clock working environment parameters is obtained; Based on the multivariate correspondence between the clock working environment parameters and the actual clock drift data, a drift characteristic prediction model is generated by training through fitting or machine learning methods.

5. The method for distributed measurement synchronization of distribution network based on BeiDou satellite timing according to claim 1, characterized in that: Generating a topology correction comprises: Determine a topological adjacent node set according to the topological connection relationship of the distribution network, and send a time synchronization request to the topological adjacent node set to obtain adjacent node timestamps; Extracting line physical parameters from the topological connection relationship of the distribution network, and calculating the theoretical time difference of signal propagation between nodes based on the line physical parameters; Calculate the received timestamp of the adjacent node and the time when the local time synchronization request is sent to obtain the measured time difference; The theoretical time difference and the network communication delay are deducted from the measured time difference to generate a topology correction amount.

6. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 5, characterized in that: The sending of a time synchronization request to the topological adjacent node set to obtain adjacent node timestamps includes: Sending a time synchronization request to the topological adjacent node set, and receiving multiple returned adjacent node timestamps to form an initial timestamp group; Evaluate the consistency and validity of each timestamp in the initial timestamp group, remove outliers or abnormal values ​​therein, and generate an optimal timestamp group; The timestamps in the preferred timestamp group are averaged or weighted averaged to generate adjacent node timestamps.

7. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 5, characterized in that: The fusing of the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference includes: While obtaining the adjacent node timestamp, obtaining a node timing quality indicator that characterizes the timing status of the node from which the adjacent node timestamp is derived; When a node timing quality index of a topologically adjacent node is higher than a preset autonomous operation threshold, a dynamic fusion weight for adjusting the topology correction amount and the confidence level of the short-term clock offset prediction value is generated based on the node timing quality index; The dynamic fusion weight is used to perform a weighted summation on the short-term clock offset prediction value and the topology correction amount to generate the dynamic compensation time reference.

8. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 7, characterized in that: The method further comprises: When the node timing quality indicators of all topologically adjacent nodes are lower than the autonomous operation threshold, obtaining the adjacent node timestamps of all nodes in the topologically adjacent node set to form a regional timestamp set; Taking the discreteness of adjacent node timestamps in the regional timestamp set and minimizing the discreteness as the goal, a target optimization function is established using the discreteness of adjacent node timestamps in the regional timestamp set, and a topology correction amount corresponding to the minimized discreteness is obtained by solving the problem; The current topology correction amount is updated using the topology correction amount corresponding to the minimized discreteness.

9. The method for distributed measurement synchronization of a distribution network based on BeiDou satellite timing according to claim 1, characterized in that: After generating the dynamic compensation time reference, the method further includes: Continuing to monitor the real-time status of the BeiDou satellite signal and updating signal quality parameters; Determining whether the updated signal quality parameter within the preset time window is lower than the quality threshold; If not, terminating the use of the dynamic compensation time reference, and recalibrating the local clock source according to the BeiDou satellite signal to obtain an updated standard time reference; The electrical quantity measurement data is time-stamped using the updated standard time reference.

10. A distributed measurement synchronization device for a distribution network based on BeiDou satellite timing, applied to a distributed measurement synchronization method for a distribution network based on BeiDou satellite timing according to any one of claims 1 to 9, characterized in that: The device comprises: A BeiDou calibration module is used to obtain BeiDou satellite signals and calibrate the local clock source according to the BeiDou satellite signals to generate a standard time reference; A signal quality monitoring module is used to continuously monitor the real-time status of Beidou satellite signals, generate signal quality parameters, and generate a dynamic compensation mode trigger signal when the signal quality parameters are lower than a preset quality threshold; a drift characteristic prediction module, configured to respond to the dynamic compensation mode trigger signal, collect clock operating environment parameters of the local clock source, and perform drift characteristic prediction in combination with the standard time reference and the clock operating environment parameters to generate a short-term clock offset prediction value; A topology information interaction module is configured to determine a set of topologically adjacent nodes based on a topological connection relationship of the distribution network, send a time synchronization request to the set of topologically adjacent nodes to obtain timestamps of adjacent nodes, calculate a theoretical time difference based on the line physical parameters contained in the topological connection relationship of the distribution network, and use the theoretical time difference to correct the timestamps of the adjacent nodes to generate a topology correction value; A time reference fusion module, configured to fuse the short-term clock offset prediction value and the topology correction value to generate a dynamic compensation time reference; The timestamp generation module is used to collect electrical quantity measurement data of the distribution network, and use the dynamic compensation time reference to time-mark the electrical quantity measurement data to generate synchronous measurement data with a timestamp.

Images