Multi-node collaborative logistics data real-time synchronization method and system

Abstract

The invention discloses a multi-node collaborative logistics data real-time synchronization method and system, and particularly relates to the technical field of logistics data synchronization management. Round-trip delay between logistics nodes and a cloud platform is measured, delay jitter data is collected and analyzed, network transmission environment stability of the nodes is evaluated, when a network is unstable, initial time difference is recorded, the change trend of a clock drift value is evaluated, clock synchronization accuracy is analyzed, and the network stability and the clock synchronization accuracy are integrated. The method determines the severity of the time synchronization deviation between the logistics nodes, divides the time synchronization deviation into a slight deviation, a moderate deviation and a serious deviation, and generates a graded early warning signal to take a targeted treatment measure, thereby effectively solving the problems of data error sequence and scheduling error caused by the time synchronization deviation in the prior art, and improving the reliability of the system. The time synchronization accuracy, the network transmission stability and the overall operation efficiency of the logistics system are improved, and the accuracy and the reliability of logistics scheduling and event processing are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics data synchronization management, and in particular to a multi-node collaborative logistics data real-time synchronization method and system. Background Art

[0002] Real-time synchronization of logistics data with multi-node collaboration means that in the logistics system, multiple data nodes (such as warehouses, transport vehicles, sorting centers and distribution terminals) work together to achieve real-time sharing and updating of logistics information between different nodes. This process relies on modern information technology, such as cloud computing, big data and the Internet of Things, to efficiently transmit logistics data from each node to the system, so that each link in the logistics chain can obtain the latest data in the shortest time, thereby optimizing resource allocation and improving logistics efficiency.

[0003] In traditional logistics systems, information lag or asymmetry often leads to inefficient logistics, such as inventory backlogs, transportation delays or resource waste. However, through real-time synchronization of multi-node collaboration, transparent and dynamic information management can be achieved. For example, the transportation node can update the location and status of the goods in real time, and the warehouse node can prepare to receive the goods in advance based on the transportation information, thereby shortening the entire logistics cycle. This approach not only reduces operating costs, but also improves customer experience and response speed. By deploying sensors, smart devices and information systems at each logistics node, each node can collect and upload relevant data to the cloud platform; then, with the help of big data analysis and edge computing technology, the data will be processed in real time and distributed to other related nodes. For example, when a package is sent from a warehouse, its status will be immediately updated to the system, and the relevant sorting center and distribution terminal can obtain the latest information in real time. The entire process requires an efficient data transmission network (such as 5G) and a reliable collaboration mechanism to ensure data consistency and synchronization.

[0004] The prior art has the following deficiencies: Time synchronization deviations between logistics nodes can cause errors in the timestamps of data records, causing confusion in the system when processing the order of events. For example, when a certain cargo is just loaded from the warehouse, the system may mistakenly record that the cargo has been delivered to the destination due to the lack of synchronization of the node time. This time misorder will cause a series of problems: the sorting center may skip the processing of the cargo, the distribution terminal cannot match the corresponding cargo information, and even trigger repeated notifications or operations, resulting in waste of resources and reduced logistics efficiency. In addition, modern logistics systems rely on real-time data for dynamic scheduling (such as vehicle route optimization and sorting priority adjustment). Time deviations lead to data synchronization errors, and the scheduling system optimizes based on false information, which may dispatch the wrong vehicle or ignore high-priority orders. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-node collaborative logistics data real-time synchronization method and system to address the deficiencies in the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a multi-node collaborative logistics data real-time synchronization method, comprising the following steps: S1: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; S2: Analyze the delay jitter in the collected network delay data, determine the abnormality of the delay between the node and the cloud platform, and evaluate the stability of the network transmission environment between the node and the cloud platform; S3: When the network transmission environment between the node and the cloud platform is unstable, synchronize the time of each node through the GPS clock, record the initial time difference of each node, regularly compare the clock of each node with the standard clock, determine the change of drift value over time, and evaluate the accuracy of node clock synchronization; S4: Determine the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; S5: Classify the severity of time synchronization deviation between logistics nodes into slight deviation, moderate deviation and severe deviation, and perform corresponding early warning processing; S6: For moderate deviations, a predictive analysis is performed on the severity of the time synchronization deviation between logistics nodes within the subsequent fixed time window, and the node priority is dynamically adjusted based on the analysis results.

[0007] Preferably, in S2, a delay jitter anomaly index is generated after analyzing the delay jitter situation in the network delay data, and the method for obtaining the delay jitter anomaly index is: Setting delay jitter data From a potential distribution , where θ is the distribution parameter, n is the total number of delay jitter data, and the standard normal distribution is selected , through variational inference, the distribution q(θ) is used to approximate the posterior , the optimization goal is to minimize the KL divergence between the two, expressed as: ; q(θ) is chosen as Gaussian distribution , where μ and σ are the parameters to be optimized, and the prior distribution p(θ) is set, the expression is: ; is the prior variance, and defines the observation likelihood p(X|θ): Set the observation data It obeys normal distribution, and the expression is; ; is the mean value of jitter, indicating the stable center of network delay. is the variance of the jitter, indicating the fluctuation amplitude of the network jitter, and the variational distribution q(θ) is selected as Gaussian distribution: ; is the approximate posterior parameter obtained by optimization, optimized by gradient descent Maximize the maximum evidence lower bound ELBO: ; For Gaussian distribution, KL has an analytical solution: ; Use the optimized q(θ) to calculate each data point The abnormal probability is: ; According to the abnormal probability of all data points, the delay jitter abnormality index is defined, and the expression is: ; Where HWK is the delay jitter anomaly index.

[0008] Preferably, in S3, the drift value fluctuation index is generated after analyzing the change of the drift value over time, and the method for obtaining the drift value fluctuation index is: Set the clock drift value time series , where N is the total number of sampling points, the sampling interval is Ts, and the mean of the sequence is removed. The expression is: ; For the drift value sequence after removing the mean Perform fast Fourier transform to obtain the frequency domain signal in complex form: ; is the kth frequency component, is the amplitude and phase of the frequency domain signal. The modulus of the amplitude represents the amplitude strength of each frequency. is the kernel function of the Fourier transform; calculate the frequency amplitude: ; In the formula, yes The real part of yes The imaginary part of the frequency domain signal; calculate the total energy of the frequency domain signal , the expression is: ; Calculate the energy of high frequency components , is the high frequency dividing point, expressed as: ; Calculate the drift value fluctuation index, the expression is: ; GHQ is the drift value fluctuation index.

[0009] Preferably, in S4, the severity of the time synchronization deviation between the logistics nodes is determined according to the stability of the network transmission environment between the node and the cloud platform and the accuracy of the node clock synchronization, specifically: The delay jitter anomaly index and the drift value fluctuation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the time synchronization deviation severity value label between logistics nodes as the prediction target, and takes minimizing the sum of prediction errors of the time synchronization deviation severity value labels between all logistics nodes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The time synchronization deviation severity value between logistics nodes is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0010] Preferably, in S5, the severity of the time synchronization deviation between logistics nodes is divided into slight deviation, moderate deviation and severe deviation, and corresponding warning processing is performed; The acquired time synchronization deviation severity value between logistics nodes is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the time synchronization deviation severity value between logistics nodes is compared with the first standard threshold and the second standard threshold respectively; If the severity value of the time synchronization deviation between logistics nodes is greater than the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is high. At this time, a first-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as severe deviation; If the severity value of the time synchronization deviation between logistics nodes is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is moderate. At this time, a secondary warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as moderate deviation; If the severity value of the time synchronization deviation between logistics nodes is less than the first standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is low. At this time, a third-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as a slight deviation.

[0011] Preferably, in S6, for moderate deviations, that is, the severity values ​​of the time synchronization deviations between logistics nodes generated within a fixed time window are greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the severity values ​​of the time synchronization deviations between logistics nodes generated within subsequent fixed time windows that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after analyzing them, the node priority is dynamically adjusted according to the analysis results.

[0012] Preferably, if the mean of the time synchronization deviation severity values ​​in the data set is greater than or equal to the reference threshold of the mean of the time synchronization deviation severity values, and the standard deviation of the time synchronization deviation severity values ​​is less than the reference threshold of the standard deviation of the time synchronization deviation severity values, the severity of the time synchronization deviation is generally high, but the fluctuation is small. At this time, the medium priority should be maintained, and the clock should be recalibrated regularly or the network configuration should be adjusted to reduce the overall deviation level; If the mean value of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the mean value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the standard deviation of the time synchronization deviation severity value, the severity of the time synchronization deviation is generally high and fluctuates greatly, and the node priority should be lowered at this time; If the mean value of the time synchronization deviation severity value is less than the reference threshold value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the standard deviation of the time synchronization deviation severity value, the time synchronization deviation is low overall but fluctuates greatly. In this case, the medium priority should be maintained, and dynamic monitoring of the node should be enhanced; If the mean value of the time synchronization deviation severity value is less than the reference threshold value of the mean value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is less than the reference threshold value of the standard deviation of the time synchronization deviation severity value, the time synchronization deviation is overall low and has small fluctuations. At this time, the node priority should be increased to reduce the occupancy of its monitoring resources and reduce the system load.

[0013] The present invention also provides a multi-node collaborative logistics data real-time synchronization system, including a network delay monitoring module, a network stability analysis module, a clock synchronization and drift assessment module, a deviation severity assessment module, an early warning response module and a node priority adjustment module: Network delay monitoring module: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; Network stability analysis module: Analyzes the delay jitter in the collected network delay data, determines the abnormal degree of delay between the node and the cloud platform, and evaluates the stability of the network transmission environment between the node and the cloud platform; Clock synchronization and drift assessment module: When the network transmission environment between the node and the cloud platform is unstable, the time of each node is synchronized through the GPS clock, and the initial time difference of each node is recorded. The clock of each node is regularly compared with the standard clock to determine the change of the drift value over time and evaluate the accuracy of the node clock synchronization; Deviation severity assessment module: Determines the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; Early warning response module: classifies the severity of time synchronization deviations between logistics nodes into slight deviation, moderate deviation and severe deviation, and performs corresponding early warning processing; Node priority adjustment module: For moderate deviations, the severity of time synchronization deviations between logistics nodes within the subsequent fixed time window is predicted and analyzed, and the node priority is dynamically adjusted based on the analysis results.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention combines network monitoring tools, GPS clock synchronization and machine learning models to accurately evaluate the severity of time synchronization deviations between logistics nodes, and performs graded warnings and dynamically adjusts node priorities based on the delay jitter anomaly index and drift value fluctuation index. This method effectively solves the problems of logistics data misordering and scheduling errors caused by time synchronization deviations, improves the logistics system's correct judgment of event sequence and the accuracy of dynamic scheduling, and significantly improves the overall efficiency and reliability of the logistics system.

[0015] 2. The present invention can capture the subtle fluctuation characteristics of network transmission and clock synchronization, and realize deviation trend prediction and resource optimization configuration. The early warning mechanism has clear classification and targeted treatment measures. While ensuring the efficient operation of nodes with slight deviations, necessary intervention can be taken for nodes with moderate and severe deviations, reducing system risks and optimizing performance, and ultimately providing stable and efficient real-time data synchronization guarantee for the logistics system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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 will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 The present invention is a flow chart of the method.

[0018] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.

[0020] Example 1, please refer to Figure 1 and Figure 2 As shown, the multi-node collaborative logistics data real-time synchronization method described in this embodiment includes the following steps: S1: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; S2: Analyze the delay jitter in the collected network delay data, determine the abnormality of the delay between the node and the cloud platform, and evaluate the stability of the network transmission environment between the node and the cloud platform; S3: When the network transmission environment between the node and the cloud platform is unstable, synchronize the time of each node through the GPS clock, record the initial time difference of each node, regularly compare the clock of each node with the standard clock, determine the change of drift value over time, and evaluate the accuracy of node clock synchronization; S4: Determine the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; S5: Classify the severity of time synchronization deviation between logistics nodes into slight deviation, moderate deviation and severe deviation, and perform corresponding early warning processing; S6: For moderate deviations, a predictive analysis is performed on the severity of the time synchronization deviation between logistics nodes within the subsequent fixed time window, and the node priority is dynamically adjusted based on the analysis results.

[0021] In S1, a network monitoring tool is deployed at each logistics node to regularly measure the round-trip delay between the node and the cloud platform and collect network delay data within several time windows, specifically: Install network monitoring tools (such as Ping, Traceroute, or dedicated latency analysis tools) at each logistics node to continuously monitor network performance. These tools can run on servers or embedded devices at logistics nodes and are responsible for measuring data transmission latency between nodes and cloud platforms. Network monitoring tools run on devices that communicate with cloud platforms, usually including the node's host system or edge computing device. The tool automatically sends data packets to the cloud platform at a set frequency and measures the time from sending to receiving confirmation.

[0022] Each node periodically sends a detection data packet to the cloud platform and records the following key information: Data packet sending time: records the timestamp when the data packet is sent from the logistics node. Data packet return time: records the timestamp when the cloud platform receives the data packet and returns the confirmation packet. Round trip time (RTT): The current network delay is obtained by calculating the difference between the sending time and the return time. For example: If the sending time is 10:00:00.000 and the return time is 10:00:00.050, the RTT is 50 milliseconds.

[0023] The measurement is not just a one-time measurement, but a continuous collection of delay data within a certain time window to form a statistically significant sample set; Time window setting: Select an appropriate time window (such as 10 minutes, 1 hour, or 24 hours) to capture the dynamic changes in network delay. Sampling frequency: Set measurements every second, every minute, or at a longer time interval to ensure that the sampling density is sufficient to reflect the network status. Within a time window, a node may generate several useful data indicators: Average RTT: reflects the overall level of network delay. Maximum RTT and minimum RTT: reflect the worst and optimal states of network delay, respectively. Jitter: reflects the volatility of network delay by calculating the difference between consecutive RTTs.

[0024] The collected RTT data is stored in the node local database or transmitted to the cloud platform for subsequent analysis. The storage content includes timestamp, RTT value and network status mark (such as normal or abnormal). The data is initially aggregated and statistically processed, such as calculating the mean, standard deviation and distribution range.

[0025] During the data collection process, the network monitoring tool will also analyze in real time whether there are any abnormalities in the delay: Delay threshold setting: Set a normal range for RTT (for example, within 50ms), and values ​​outside this range are marked as abnormal. Trend monitoring: Detect whether RTT shows a trend of gradual increase or severe jitter, indicating potential network problems.

[0026] Analyze the network stability between the node and the cloud platform through delay data. Compensate for the transmission time in time synchronization based on delay data to reduce the impact of delay on time synchronization accuracy. Delay data can be used to optimize network configuration, such as adjusting routing paths or increasing bandwidth.

[0027] S2: Analyze the delay jitter in the collected network delay data, determine the abnormal degree of delay between the node and the cloud platform, and evaluate the stability of the network transmission environment between the node and the cloud platform.

[0028] Delay jitter refers to the degree of variation between consecutive network delays, that is, the range of fluctuations in delay instability. Jitter is usually caused by network congestion, routing changes, or hardware performance issues. Low jitter: The network transmission environment is stable and the packet transmission delay is basically consistent. High jitter: The network transmission environment is unstable and the delay changes significantly, which may affect the real-time and accuracy of logistics data.

[0029] From the collected RTT data, calculate the absolute difference between two consecutive RTTs: record all calculated jitter values, and calculate the average jitter, which reflects the overall jitter level, and the maximum jitter, which reflects the peak value of network jitter and shows the fluctuation amplitude in extreme cases.

[0030] Perform distribution statistics on the calculated jitter values ​​to determine the concentrated area of ​​jitter and the frequency of abnormal values: Normal jitter range: Set a normal range based on the network environment (for example, jitter ≤ 5ms). Abnormal jitter ratio: Count the proportion of jitter values ​​that exceed the normal range in the total samples. If the abnormal jitter ratio is high, it means that the network is unstable. Based on the statistical results of the jitter value, further delay anomaly detection is performed in combination with the average RTT and the maximum and minimum values ​​of the RTT: Stability index: Is the RTT range small (for example, the difference between the minimum RTT and the maximum RTT is ≤ 10ms). Is the average jitter within a reasonable range (for example, ≤ 5ms). Abnormal fluctuation detection: Does the RTT frequently jump significantly (for example, two consecutive RTT changes exceed 50ms). Is there an obvious delay peak (Spike), such as the RTT suddenly rising to more than 100ms in a short period of time.

[0031] After analyzing the delay jitter in the network delay data, a delay jitter anomaly index is generated to evaluate the stability of the network transmission environment between the node and the cloud platform. The delay jitter anomaly index is obtained as follows: Setting delay jitter data From a potential distribution , where θ is the distribution parameter (such as mean and variance), and n is the total number of delay jitter data. Set the prior distribution p(θ) to reflect the initial belief about the parameter, usually choose the standard normal distribution .

[0032] Through variational inference, we approximate the posterior with a tractable distribution q(θ) , the optimization goal is to minimize the KL divergence between the two, expressed as: ; q(θ) is chosen as Gaussian distribution , where μ and σ are the parameters to be optimized, and the prior distribution p(θ) is set, the expression is: ; is the prior variance, reflecting the initial belief in parameter uncertainty, and defines the observation likelihood p(X|θ): It obeys normal distribution, and the expression is; ; is the mean value of jitter, indicating the stable center of network delay. is the variance of the jitter, indicating the fluctuation amplitude of the network jitter, and the variational distribution q(θ) is selected as Gaussian distribution: ; is the approximate posterior parameter obtained by optimization. Optimization by gradient descent Maximize the maximum evidence lower bound ELBO: ; For Gaussian distribution, KL has an analytical solution: ; Use the optimized q(θ) to calculate each data point The abnormal probability is: ; According to the abnormal probability of all data points, the delay jitter abnormality index is defined, and the expression is: ; Where HWK is the delay jitter anomaly index.

[0033] When the delay jitter anomaly index is large, it indicates that there is significant instability in the network transmission environment between the node and the cloud platform. This means that the delay jitter value frequently fluctuates abnormally, or the jitter amplitude exceeds the normal range, and the volatility of network delay is high. In this case, network congestion, route hopping, insufficient hardware performance and other problems may cause inconsistent data packet transmission time, further affecting the real-time synchronization and reliability of logistics data. If the network configuration is not optimized in time, it may lead to a decline in system performance and even pose a threat to the normal operation of the logistics system.

[0034] When the delay jitter anomaly index is small, it means that the network transmission environment between the node and the cloud platform is relatively stable, the fluctuation of network delay is small, and the jitter value is basically within the normal range. This reflects that the transmission time of data packets is relatively consistent, the network status is good, and it can support efficient logistics data synchronization and real-time requirements. This stable transmission environment helps to reduce time synchronization deviation, improve the overall efficiency and response speed of the logistics system, and provide reliable data support for dynamic scheduling and event processing.

[0035] S3: When the network transmission environment between the node and the cloud platform becomes unstable, synchronize the time of each node through the GPS clock, record the initial time difference of each node, regularly compare the clock of each node with the standard clock, determine the change of drift value over time, and evaluate the accuracy of node clock synchronization.

[0036] Deploy GPS clock synchronization equipment and install GPS clock modules at each logistics node as time reference equipment. GPS clocks provide high-precision global standard time (UTC) by receiving satellite signals, and their error is usually at the nanosecond level.

[0037] Record the initial time difference ΔTinit between each node device clock and GPS standard time: ΔTinit=TGPS-Tnode; where: TGPS is the time of the GPS clock, and Tnode is the time of the local clock of the node device.

[0038] Adjust the node device clock to align it with the GPS time and eliminate the initial time difference: T after synchronization = T node + ΔTinit; ensure that the time of each node is consistent with the GPS clock and establish a unified time reference.

[0039] During the operation of the node device, the time difference between the node clock and the GPS clock is regularly compared. , the expression is: ΔT(t)=TGPS(t)-Tnode(t); t is the time point of comparison.

[0040] Record the time difference at different time points and calculate the time drift value ΔR: in, Separately for the moment and time difference. The time drift value ΔR represents the clock offset per unit time and is used to evaluate the stability of the clock.

[0041] After analyzing the change of drift value over time, the drift value fluctuation index is generated to evaluate the accuracy of node clock synchronization. The method for obtaining the drift value fluctuation index is as follows: Input data: time series of clock drift values , where N is the total number of sampling points and the sampling interval is Ts (seconds). To eliminate the DC component (zero-frequency component) of the drift value, the sequence is removed by the mean, and the expression is: ; For the drift value sequence after removing the mean Perform fast Fourier transform to obtain the frequency domain signal in complex form: ; is the kth frequency component, is the amplitude and phase of the frequency domain signal. The modulus of the amplitude represents the amplitude strength of each frequency. is the kernel function of the Fourier transform; calculate the frequency amplitude: ; In the formula, yes The real part of yes The imaginary part of the frequency domain signal; calculate the total energy of the frequency domain signal , the expression is: ; Only the first N / 2 points are used because the FFT output is symmetrical. Calculate the energy of the high frequency components , is the high frequency dividing point, expressed as: ; Calculate the drift value fluctuation index, the expression is: ; GHQ is the drift value fluctuation index.

[0042] When the drift value fluctuation index is large, it means that the high-frequency fluctuation component in the drift value of the node clock is dominant, and the changes are frequent and large. This indicates that the clock stability of the node is poor, and the time difference with the standard clock is significantly irregular. The larger the drift value fluctuation index, the lower the accuracy of clock synchronization, which may lead to the accumulation of time synchronization errors and affect the system's accurate judgment of event sequence and time-sensitive tasks.

[0043] When the drift value fluctuation index is small, it means that the drift value of the node clock changes relatively smoothly, the low-frequency component is dominant, and the fluctuation amplitude is small. This shows that the node clock runs relatively stably, the time difference change trend is consistent with the standard clock, and the amplitude is within the controllable range. The smaller the drift value fluctuation index, the higher the accuracy of the node clock synchronization, and it can more reliably support the real-time synchronization and scheduling operations of logistics data.

[0044] S4: Determine the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization.

[0045] The delay jitter anomaly index and the drift value fluctuation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the time synchronization deviation severity value label between logistics nodes as the prediction target, and takes minimizing the sum of prediction errors of the time synchronization deviation severity value labels between all logistics nodes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The time synchronization deviation severity value between logistics nodes is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0046] The method for obtaining the severity value of the time synchronization deviation between logistics nodes is to obtain the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: ; In the formula, is the output function of the model, HWK is the delay jitter anomaly index, GHQ is the drift value fluctuation index, It is the severity value of time synchronization deviation between logistics nodes.

[0047] S5: Classify the severity of time synchronization deviation between logistics nodes into slight deviation, moderate deviation and severe deviation, and perform corresponding early warning processing; The acquired time synchronization deviation severity value between logistics nodes is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the time synchronization deviation severity value between logistics nodes is compared with the first standard threshold and the second standard threshold respectively; If the severity value of the time synchronization deviation between logistics nodes is greater than the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is high. At this time, a first-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as severe deviation; If the severity value of the time synchronization deviation between logistics nodes is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is moderate. At this time, a secondary warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as moderate deviation; If the severity value of the time synchronization deviation between logistics nodes is less than the first standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is low. At this time, a third-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as a slight deviation.

[0048] It should be noted here that the importance of the first-level warning signal is greater than that of the second-level warning signal, and the importance of the second-level warning signal is greater than that of the third-level warning signal. Relevant personnel can take corresponding handling measures according to different warning signal levels.

[0049] Level 1 warning signal (serious deviation) Importance: Highest priority, indicating that the time synchronization deviation of the logistics node seriously affects the normal operation of the system, which may cause the failure of real-time synchronization of logistics data, confusion of event sequence or scheduling task errors.

[0050] Isolate seriously deviated logistics nodes from the real-time synchronization network to prevent their erroneous data from causing chain reactions to other nodes. Switch to backup nodes to ensure the continuous operation of the logistics system. Perform deep time calibration on the problem node through GPS or NTP to reset its clock. Check and optimize the network configuration of the problem node, such as adjusting the route, increasing bandwidth, or replacing hardware. Notify relevant technicians to immediately troubleshoot the problem and repair the equipment or network failure.

[0051] Level 2 warning signal (moderate deviation) Importance: Medium priority, indicating that the time synchronization deviation between nodes may affect some functions of the system, such as scheduling delays or event sequence deviations.

[0052] Monitor nodes with moderate deviations in real time to see if their time synchronization deviations continue to increase. During data synchronization, introduce a time deviation compensation mechanism for nodes with moderate deviations to reduce the impact of deviations on the system. Schedule regular calibration of node clocks to reduce drift errors and optimize their time synchronization mechanism with the cloud platform. Check for network delay jitter issues, optimize transmission lines or replace unstable hardware. Send intermediate warning notifications to the technical support team to track the node operation status.

[0053] Level 3 warning signal (slight deviation) Importance: Lowest priority, indicating that the time synchronization deviation between nodes is low and has limited impact on system operation, but care needs to be taken to prevent problems from accumulating.

[0054] Add nodes with slight deviations to the watch list and record their time synchronization status regularly to ensure that the deviations are maintained within a controllable range. During the next system maintenance or upgrade, optimize the clock synchronization accuracy and network transmission quality of nodes with slight deviations. Back up the real-time synchronization data of nodes with slight deviations so that they can be quickly restored when problems arise. Send low-priority notifications to relevant managers, only prompting them to pay attention to the operation of the nodes.

[0055] S6: For moderate deviations, a predictive analysis is performed on the severity of the time synchronization deviation between logistics nodes within the subsequent fixed time window, and the node priority is dynamically adjusted based on the analysis results.

[0056] For moderate deviations, that is, the severity values ​​of the time synchronization deviations between logistics nodes generated within a fixed time window are greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the severity values ​​of the time synchronization deviations between logistics nodes generated within subsequent fixed time windows that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established. The mean and standard deviation of the data set are calculated, and after analyzing them, the node priority is dynamically adjusted according to the analysis results.

[0057] If the mean of the time synchronization deviation severity values ​​in the data set is greater than or equal to the reference threshold of the mean of the time synchronization deviation severity values, and the standard deviation of the time synchronization deviation severity values ​​is less than the reference threshold of the standard deviation of the time synchronization deviation severity values, the severity of the time synchronization deviation is generally high, but the fluctuation is small, indicating that the deviation value is relatively stable, which may be a systemic problem or inherent characteristic. At this time, a medium priority should be maintained, and deep optimization of the nodes should be planned, such as regularly recalibrating the clock or adjusting the network configuration to reduce the overall deviation level.

[0058] If the mean value of the time synchronization deviation severity value is greater than or equal to the reference threshold of the mean value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold of the standard deviation of the time synchronization deviation severity value, the overall severity of the time synchronization deviation is high and fluctuates greatly, indicating that the node time synchronization status is unstable and easily affected by external factors (such as network fluctuations). At this time, the node priority should be lowered to reduce its interference with the real-time synchronization of the system, and the deviation change trend of the node should be monitored. If necessary, more stringent network and clock calibration measures should be taken.

[0059] If the mean of the time synchronization deviation severity value is less than the reference threshold of the mean of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold of the standard deviation of the time synchronization deviation severity value, the time synchronization deviation is low overall but fluctuates greatly, indicating that the node runs normally most of the time, but occasionally there will be significant deviations. At this time, the medium priority should be maintained, and the dynamic monitoring of the node should be strengthened. When the deviation fluctuates significantly, temporary optimization measures should be triggered, such as restarting the synchronization process or short-term isolation.

[0060] If the mean value of the time synchronization deviation severity value is less than the reference threshold value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is less than the reference threshold value of the time synchronization deviation severity value, the time synchronization deviation is overall low and has small fluctuations, indicating that the node operation status is relatively stable. At this time, the node priority should be increased to allow it to participate more in the system real-time synchronization tasks, while reducing the occupation of its monitoring resources and reducing the system load.

[0061] In this embodiment, first, a network monitoring tool is deployed at each logistics node, the round-trip delay between the node and the cloud platform is measured regularly, and network delay data within multiple time windows are collected; then, the collected network delay data is subjected to delay jitter analysis to evaluate the stability of the network transmission environment between the node and the cloud platform; when it is detected that the network transmission environment is unstable, the time of each node is synchronized through the GPS clock, and the initial time difference is recorded, and the clock drift value is regularly compared to evaluate the accuracy of the node clock synchronization; then, the severity of the time synchronization deviation between the logistics nodes is determined in combination with the stability of the network transmission environment and the accuracy of the clock synchronization; according to the severity of the deviation, it is divided into slight deviation, moderate deviation and severe deviation, and a warning signal of the corresponding level is generated, and corresponding processing measures are taken at the same time; for moderate deviations, the changing trend of the severity of the time synchronization deviation is predicted in the subsequent fixed time window, the node priority is dynamically adjusted, and the system resource allocation and operation efficiency are optimized.

[0062] Embodiment 2, a multi-node collaborative logistics data real-time synchronization system described in this embodiment includes a network delay monitoring module, a network stability analysis module, a clock synchronization and drift assessment module, a deviation severity assessment module, an early warning response module and a node priority adjustment module: Network delay monitoring module: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; Network stability analysis module: Analyzes the delay jitter in the collected network delay data, determines the abnormal degree of delay between the node and the cloud platform, and evaluates the stability of the network transmission environment between the node and the cloud platform; Clock synchronization and drift assessment module: When the network transmission environment between the node and the cloud platform is unstable, the time of each node is synchronized through the GPS clock, and the initial time difference of each node is recorded. The clock of each node is regularly compared with the standard clock to determine the change of the drift value over time and evaluate the accuracy of the node clock synchronization; Deviation severity assessment module: Determines the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; Early warning response module: classifies the severity of time synchronization deviations between logistics nodes into slight deviation, moderate deviation and severe deviation, and performs corresponding early warning processing; Node priority adjustment module: For moderate deviations, the severity of time synchronization deviations between logistics nodes within the subsequent fixed time window is predicted and analyzed, and the node priority is dynamically adjusted based on the analysis results.

[0063] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0064] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0065] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0066] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the 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.

Claims

1. A multi-node collaborative logistics data real-time synchronization method, characterized by: The following steps are involved: S1: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; S2: Analyze the delay jitter in the collected network delay data, determine the abnormality of the delay between the node and the cloud platform, and evaluate the stability of the network transmission environment between the node and the cloud platform; S3: When the network transmission environment between the node and the cloud platform is unstable, synchronize the time of each node through the GPS clock, record the initial time difference of each node, regularly compare the clock of each node with the standard clock, determine the change of drift value over time, and evaluate the accuracy of node clock synchronization; S4: Determine the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; S5: Classify the severity of time synchronization deviation between logistics nodes into slight deviation, moderate deviation and severe deviation, and perform corresponding early warning processing; S6: For moderate deviations, a predictive analysis is performed on the severity of the time synchronization deviation between logistics nodes within the subsequent fixed time window, and the node priority is dynamically adjusted based on the analysis results.

2. According to claim 1, a multi-node collaborative logistics data real-time synchronization method is characterized by: In S2, the delay jitter in the network delay data is analyzed to generate a delay jitter anomaly index. The delay jitter anomaly index is obtained as follows: Setting delay jitter data From a potential distribution , where θ is the distribution parameter, n is the total number of delay jitter data, and the standard normal distribution is selected , through variational inference, the distribution q(θ) is used to approximate the posterior , the optimization goal is to minimize the KL divergence between the two, expressed as: ; q(θ) is chosen as Gaussian distribution , where μ and σ are the parameters to be optimized, and the prior distribution p(θ) is set, the expression is: ; is the prior variance, and defines the observation likelihood p(X|θ): Set the observation data It obeys normal distribution, and the expression is; ; is the mean value of jitter, indicating the stable center of network delay. is the variance of the jitter, indicating the fluctuation amplitude of the network jitter, and the variational distribution q(θ) is selected as Gaussian distribution: ; is the approximate posterior parameter obtained by optimization, optimized by gradient descent Maximize the maximum evidence lower bound ELBO: ; For Gaussian distribution, KL has an analytical solution: ; Use the optimized q(θ) to calculate each data point The abnormal probability is: ; According to the abnormal probability of all data points, the delay jitter abnormality index is defined, and the expression is: ; Where HWK is the delay jitter anomaly index.

3. The multi-node collaborative logistics data real-time synchronization method according to claim 2, characterized in that: In S3, the drift value fluctuation index is generated after analyzing the change of the drift value over time. The method for obtaining the drift value fluctuation index is: Set the clock drift value time series , where N is the total number of sampling points, the sampling interval is Ts, and the mean of the sequence is removed. The expression is: ; For the drift value sequence after removing the mean Perform fast Fourier transform to obtain the frequency domain signal in complex form: ; is the kth frequency component, is the amplitude and phase of the frequency domain signal. The modulus of the amplitude represents the amplitude strength of each frequency. is the kernel function of Fourier transform; Calculate the frequency amplitude: ; In the formula, yes The real part of yes The imaginary part of the frequency domain signal; calculate the total energy of the frequency domain signal , the expression is: ; Calculate the energy of high frequency components , is the high frequency dividing point, expressed as: ; Calculate the drift value fluctuation index, the expression is: ; GHQ is the drift value fluctuation index.

4. The method for real-time synchronization of logistics data with multi-node collaboration according to claim 3 is characterized in that: In S4, the severity of the time synchronization deviation between logistics nodes is determined based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of the node clock synchronization, specifically: The delay jitter anomaly index and the drift value fluctuation index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each group of comprehensive feature vectors to predict the time synchronization deviation severity value label between logistics nodes as the prediction target, and takes minimizing the sum of prediction errors of the time synchronization deviation severity value labels between all logistics nodes as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The time synchronization deviation severity value between logistics nodes is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. A multi-node collaborative logistics data real-time synchronization method according to claim 4, characterized in that: In S5, the severity of the time synchronization deviation between logistics nodes is divided into slight deviation, moderate deviation and severe deviation, and corresponding warning processing is carried out; The acquired time synchronization deviation severity value between logistics nodes is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the time synchronization deviation severity value between logistics nodes is compared with the first standard threshold and the second standard threshold respectively; If the severity value of the time synchronization deviation between logistics nodes is greater than the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is high. At this time, a first-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as severe deviation; If the severity value of the time synchronization deviation between logistics nodes is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is moderate. At this time, a secondary warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as moderate deviation; If the severity value of the time synchronization deviation between logistics nodes is less than the first standard threshold, it means that the severity of the time synchronization deviation between logistics nodes is low. At this time, a third-level warning signal is generated, and the severity of the time synchronization deviation between logistics nodes is classified as a slight deviation.

6. A multi-node collaborative logistics data real-time synchronization method according to claim 5, characterized in that: In S6, for moderate deviations, that is, the severity values ​​of the time synchronization deviations between logistics nodes generated within a fixed time window are greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the severity values ​​of the time synchronization deviations between logistics nodes generated within subsequent fixed time windows that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold are collected, and a corresponding data set is established, the mean and standard deviation of the data set are calculated, and after analyzing them, the node priority is dynamically adjusted according to the analysis results.

7. A multi-node collaborative logistics data real-time synchronization method according to claim 6, characterized in that: If the mean of the time synchronization deviation severity values ​​in the data set is greater than or equal to the reference threshold of the mean of the time synchronization deviation severity values, and the standard deviation of the time synchronization deviation severity values ​​is less than the reference threshold of the standard deviation of the time synchronization deviation severity values, the severity of the time synchronization deviation is generally high, but the fluctuation is small. At this time, the medium priority should be maintained, and the clock should be recalibrated regularly or the network configuration should be adjusted to reduce the overall deviation level; If the mean value of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the mean value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the standard deviation of the time synchronization deviation severity value, the severity of the time synchronization deviation is generally high and fluctuates greatly, and the node priority should be lowered at this time; If the mean value of the time synchronization deviation severity value is less than the reference threshold value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is greater than or equal to the reference threshold value of the standard deviation of the time synchronization deviation severity value, the time synchronization deviation is low overall but fluctuates greatly. In this case, the medium priority should be maintained, and dynamic monitoring of the node should be enhanced; If the mean value of the time synchronization deviation severity value is less than the reference threshold value of the mean value of the time synchronization deviation severity value, and the standard deviation of the time synchronization deviation severity value is less than the reference threshold value of the standard deviation of the time synchronization deviation severity value, the time synchronization deviation is overall low and has small fluctuations. At this time, the node priority should be increased to reduce the occupancy of its monitoring resources and reduce the system load.

8. A multi-node collaborative logistics data real-time synchronization system, used to implement a multi-node collaborative logistics data real-time synchronization method according to any one of claims 1 to 7, characterized in that: It includes network delay monitoring module, network stability analysis module, clock synchronization and drift assessment module, deviation severity assessment module, early warning response module and node priority adjustment module: Network delay monitoring module: Deploy network monitoring tools at each logistics node, regularly measure the round-trip delay between the node and the cloud platform, and collect network delay data within several time windows; Network stability analysis module: Analyzes the delay jitter in the collected network delay data, determines the abnormal degree of delay between the node and the cloud platform, and evaluates the stability of the network transmission environment between the node and the cloud platform; Clock synchronization and drift assessment module: When the network transmission environment between the node and the cloud platform is unstable, the time of each node is synchronized through the GPS clock, and the initial time difference of each node is recorded. The clock of each node is regularly compared with the standard clock to determine the change of the drift value over time and evaluate the accuracy of the node clock synchronization; Deviation severity assessment module: Determines the severity of time synchronization deviation between logistics nodes based on the stability of the network transmission environment between the node and the cloud platform and the accuracy of node clock synchronization; Early warning response module: classifies the severity of time synchronization deviations between logistics nodes into slight deviation, moderate deviation and severe deviation, and performs corresponding early warning processing; Node priority adjustment module: For moderate deviations, the severity of time synchronization deviations between logistics nodes within the subsequent fixed time window is predicted and analyzed, and the node priority is dynamically adjusted based on the analysis results.

Images