Digital twin data synchronization method and system fusing deep learning and Internet of Things

Abstract

The invention discloses a digital twin data synchronization method and system fusing deep learning and the Internet of Things, and relates to the technical field of data synchronization. The method comprises the steps of calling a timestamp and a data sequence identifier through a sliding window based on digital twinborn information of Internet of Things equipment, performing time span acquisition, calculating characteristic value variation amplitudes of adjacent time slices in a data sequence, performing transmission data sparsification processing, reducing communication burden, and obtaining a sparse characteristic set. According to the method, the sliding window technology and the time sequence analysis are adopted, data features can be efficiently extracted and compressed, redundancy in data transmission is remarkably reduced, the transmission rate and efficiency are improved, the dynamic segmentation decoding and feature mapping structure application enable data fluctuation to have higher response capacity and prediction precision, and the method is suitable for large-scale popularization and application. The data flow direction and the node connection weight are intelligently adjusted, the data synchronization path is optimized according to the real-time network state and the bandwidth allocation condition, and efficient synchronization of data between multiple nodes and multiple platforms is ensured.

Description

Technical Field

[0001] The present invention relates to the field of data synchronization technology, and in particular to a digital twin data synchronization method and system that integrates deep learning and the Internet of Things. Background Art

[0002] The field of data synchronization technology involves achieving consistency, coordination, and efficient data transmission between multiple systems, multiple devices, or multiple platforms. In the era of the Internet of Things, data synchronization is not limited to synchronization between traditional databases and applications. Its core goal is to ensure the seamless flow and consistency of data between different nodes, reduce latency, minimize data conflicts, and improve the reliability and performance of the overall system. Data synchronization is widely used in scenarios such as distributed computing, real-time monitoring, smart device interaction, cloud computing, and edge computing.

[0003] Among them, the digital twin data synchronization method that integrates deep learning and the Internet of Things refers to combining deep learning technology with the Internet of Things to optimize the data synchronization process in the digital twin system. Through the intelligent analysis capabilities of the deep learning model, it processes and predicts the data streams generated by IoT devices, and improves the real-time mapping accuracy and dynamic response capabilities of the digital twin system to real physical objects. Its purpose is to solve the complex multi-source data coordination problems in the digital twin system, optimize system performance, support applications in industrial manufacturing, smart cities, precision medicine and other fields, and achieve more efficient and intelligent digital management and control.

[0004] Existing technologies have failed to effectively solve the coordination problem of multi-source data. In complex network environments, synchronization efficiency is low and the response to device status changes is slow. Conventional synchronization cannot effectively manage high-frequency data changes under unoptimized data processing strategies, and data delays or losses are prone to occur. The lack of real-time analysis of network status and bandwidth changes limits adaptability under changing network conditions. Non-optimized synchronization paths lead to inefficient data transmission in network congestion, thereby increasing the synchronization burden and limiting performance in real-time monitoring and IoT interaction. Summary of the Invention

[0005] In order to solve the existing problems of failing to effectively solve the coordination problem of multi-source data in the existing technology, low synchronization efficiency in complex network environments, slow response to device status changes, and conventional synchronization that cannot effectively manage high-frequency data changes under unoptimized data processing strategies, which is prone to data delays or losses. Because real-time analysis of network status and bandwidth changes is not taken into account, the adaptability under changing network conditions is limited, and non-optimized synchronization paths lead to inefficient data transmission under network congestion, thereby increasing the synchronization burden and limiting the performance in real-time monitoring and Internet of Things interaction. The embodiments of the present invention provide a digital twin data synchronization method and system that integrates deep learning and the Internet of Things. The technical solution is as follows:

[0006] On the one hand, a digital twin data synchronization method integrating deep learning and the Internet of Things is provided, comprising the following steps:

[0007] S1. Based on the digital twin information of IoT devices, the timestamp and data sequence identifier are called through a sliding window to collect time spans, calculate the variation amplitude of the feature values of adjacent time slices in the data sequence, and perform transmission data sparsification processing to obtain a sparse feature set;

[0008] S2. Based on the sparse feature set and time series feature values, the difference between each feature value is calculated, the fluctuation range and change characteristics in the IoT device data synchronization are analyzed, and dynamic segmented decoding is performed based on the analysis results to obtain a fluctuation feature mapping structure;

[0009] S3. Based on the fluctuation feature mapping structure, collect data synchronization paths between IoT devices, evaluate the connection efficiency and load status of IoT device communication nodes, calculate the data flow intensity and synchronization path dependency value between nodes, and adjust the data flow direction and node connection weights to obtain spatiotemporal consistency results;

[0010] S4. Based on the spatiotemporal consistency results, collect the network status and bandwidth allocation during the synchronization process, calculate the traffic distribution of the synchronization path according to the current transmission path load status, and obtain the optimal configuration of the transmission channel based on the communication efficiency and current load of the path;

[0011] S5. Based on the optimized configuration of the transmission channel, according to the business data synchronization rate in the path, analyze the characteristic values of the synchronization rate changes in the time window, screen the data sequences with high-frequency fluctuation characteristics in the synchronization path, and obtain the data synchronization path optimization result according to the data synchronization rate in the path.

[0012] On the other hand, a digital twin data synchronization system integrating deep learning and the Internet of Things is provided. The system is applied to a digital twin data synchronization method integrating deep learning and the Internet of Things, including:

[0013] The time feature sparsification module is used for digital twin information based on IoT devices. It calls timestamps and data sequence identifiers through a sliding window to collect time spans, calculates the change amplitude of the characteristic values of adjacent time slices in the data sequence, and performs transmission data sparsification processing to obtain a sparse feature set.

[0014] A feature fluctuation decoding module is used to calculate the difference between each feature value based on the sparse feature set and the time series feature value, analyze the fluctuation range and change characteristics in the data synchronization of the Internet of Things devices, and obtain a fluctuation feature mapping structure;

[0015] A fluctuation path mapping module is used to collect data synchronization paths between IoT devices based on the fluctuation feature mapping structure, evaluate the connection efficiency and load status of IoT device communication nodes, calculate the data flow intensity and synchronization path dependency value between nodes, and obtain spatiotemporal consistency results;

[0016] A path load optimization module is used to collect the network status and bandwidth allocation during the synchronization process based on the spatiotemporal consistency results, calculate the traffic distribution of the synchronization path according to the current transmission path load status, and obtain the optimized configuration of the transmission channel according to the communication efficiency and current load of the path;

[0017] The task synchronization allocation module is used to optimize the configuration of the transmission channel, analyze the characteristic values of the synchronization rate changes in the time window according to the business data synchronization rate in the path, screen the data sequences with high-frequency fluctuation characteristics in the synchronization path, and obtain the data synchronization path optimization results according to the data synchronization rate in the path.

[0018] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0019] The use of sliding window technology and time series analysis can efficiently extract and compress data features, significantly reduce redundancy in data transmission, reduce network burden, and improve transmission rate and efficiency. The application of dynamic segmented decoding and feature mapping structure enables stronger responsiveness and prediction accuracy to data fluctuations, intelligently adjusts data flow and node connection weights, optimizes data synchronization paths according to real-time network status and bandwidth allocation, ensures efficient synchronization of data between multiple nodes and multiple platforms, improves the real-time nature of data synchronization, reduces data conflicts, and improves overall reliability and performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 This is a flowchart of a digital twin data synchronization method that integrates deep learning and the Internet of Things, provided by an embodiment of the present invention;

[0022] Figure 2 This is a block diagram of a digital twin data synchronization system that integrates deep learning and the Internet of Things, provided by an embodiment of the present invention;

[0023] Figure 3 This is a structural diagram of a digital twin data synchronization device that integrates deep learning and the Internet of Things, provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0025] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0026] In the embodiments of the present invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same. The terms "of," "corresponding," and "corresponding" may sometimes be used interchangeably. It should be noted that, when the distinction is not emphasized, the meanings they convey are the same.

[0027] In the embodiments of the present invention, sometimes a subscript such as W1 may be written as a non-subscript such as W1. When the difference is not emphasized, the meanings to be expressed are the same.

[0028] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0029] The embodiment of the present invention provides a digital twin data synchronization method that integrates deep learning and the Internet of Things, such as Figure 1 As shown, the following steps are included:

[0030] S1. Based on the digital twin information of IoT devices, the timestamp and data sequence identifier are called through a sliding window to collect time spans, calculate the change amplitude of the eigenvalues of adjacent time slices in the data sequence, and perform transmission data sparse processing to reduce the communication burden and obtain a sparse feature set.

[0031] Optionally, the sparse feature set includes a time slice sequence, a feature value variation range, and a sparse data point set; the fluctuation feature mapping structure includes a feature fluctuation range, a time series difference distribution, and a dynamic segmentation structure; the spatiotemporal consistency results include node flow intensity values, synchronization path dependency indicators, and communication node weight distributions; the transmission channel optimization configuration includes traffic distribution results, path load distribution, and bandwidth utilization; and the data synchronization path optimization results include high-frequency fluctuation paths, synchronization task allocation tables, and path priority results.

[0032] Optionally, the specific operation steps of S1 may include the following S101-S103:

[0033] S101. Based on the digital twin information of the IoT device, call the timestamp and data sequence identifier, collect data fragments within the corresponding time period by sliding the fixed time window, and detect the data range and content integrity in the time window to obtain the time series fragments.

[0034] By sliding a fixed time window, extracting timestamps and data sequence identifiers, determining the span of the sliding window and dividing it into multiple time periods based on the timestamps, collecting the original data points in each time period, analyzing the time range and content integrity of each data segment, checking the data content for gaps or abnormal values section by section, and removing abnormal values. Then, the time series data of each window is sorted and merged in chronological order to ensure the order and integrity of the data in each time period in the collected sequence, and obtain time series fragments.

[0035] S102: Based on the time series segments, compare the eigenvalues of adjacent time segments, calculate the eigenvalue variation range in the time segments, extract the data fluctuation amplitude, and count the variation amplitude of the segments and the corresponding time points to obtain a set of eigenvalue variation amplitudes.

[0036] The eigenvalues in adjacent time segments are extracted segment by segment. By analyzing the specific numerical changes of each data segment, the eigenvalues are compared numerically, the variation range of the eigenvalues in the two data segments is calculated, and the time points corresponding to the variation values are recorded. The fluctuation amplitude of the eigenvalues in each time segment is counted according to the sequence of time segments. The time segments with significant fluctuations and their corresponding variation ranges are extracted. The mapping relationship between the eigenvalues and the time points is established. The variation amplitude data of all time segments are summarized to obtain the eigenvalue variation amplitude set.

[0037] S103: Based on the eigenvalue change amplitude set, filter the time segments with low change amplitude, remove the data content with low change amplitude, and reorganize the remaining eigenvalues according to the sparsification logic to optimize the transmission data load and obtain a sparse feature set.

[0038] Extract time segments with fluctuation amplitudes lower than the set threshold, mark the data content of the corresponding time period and eliminate segments with low fluctuation amplitudes, reorganize the time series by analyzing the eigenvalues in the remaining time segments, filter data points with low eigenvalue change frequency according to the sparsification logic, reorganize the remaining eigenvalues to reduce unnecessary data volume, compress the data load and optimize the bandwidth usage during transmission, and obtain a sparse feature set.

[0039] S2. Based on the sparse feature set and the time series eigenvalues, the difference between each eigenvalue is calculated to analyze the fluctuation range and change characteristics in the IoT device data synchronization. Dynamic segmented decoding is performed based on the analysis results to obtain the fluctuation feature mapping structure.

[0040] Optionally, the specific operation steps of S2 may include the following S201-S203:

[0041] S201 , based on the sparse feature set, arranging the feature values according to the time series, performing difference data analysis on the feature values at each time point, determining the difference between adjacent feature values, and obtaining a feature difference sequence.

[0042] Arrange the eigenvalues according to the time series, analyze the eigenvalues of each time point in the time series, calculate the difference in eigenvalues of adjacent time points point by point, extract the specific numerical difference between the eigenvalues by comparing the data change amplitude of each two consecutive time points, and record the corresponding time point information. Sorting out the characteristic difference values of all time points and arranging them in chronological order, establishing a mapping relationship between the difference values and the time points, constructing the difference change matrix in the time series, and obtaining the characteristic difference sequence.

[0043] S202: Based on the feature difference sequence, analyze the data fluctuation range of the feature value difference, count the numerical changes of the difference values, identify the time period of the numerical changes, mark the start and end time of the segments, and obtain a set of fluctuation segments.

[0044] Analyze the fluctuation range of the eigenvalue difference, extract the fluctuation interval by calculating the maximum and minimum values of the difference value segment by segment, and perform segmented statistics on each difference value in the time series, record the time points with large value changes, extract the high fluctuation interval by comparing the fluctuation changes between time periods, gradually mark the start and end time of these time periods, sort out the time range and difference value distribution of each fluctuation segment, merge and summarize all fluctuation intervals to construct a set of fluctuation segments.

[0045] Analyze the data fluctuation range of the eigenvalue difference, and calculate the fluctuation range R of the eigenvalue difference according to the following formula (1):

[0046] (1)

[0047] The root mean square fluctuation of the eigenvalue is calculated according to the following formula (2): :

[0048] (2)

[0049] Among them, D i Represents each difference value in the feature difference sequence, max(D i) represents the maximum value of the difference value in the feature difference sequence, min(D i ) represents the minimum value of the difference value in the feature difference sequence, n represents the total number of data points in the feature difference sequence, It represents the average value of the characteristic difference sequence, and the calculation formula is:

[0050] ;

[0051] Detailed explanation of the formula and the process of formula calculation and derivation:

[0052] The difference value D in the feature difference sequence i This is obtained by comparing the time point data, that is, calculating the difference in the feature values of adjacent time points and forming a sequence. Assuming the feature difference sequence is D={2.4,3.6,1.8,4.2,3.0}, the specific steps are as follows:

[0053] Calculate the fluctuation range R;

[0054] Find the maximum and minimum differences in a sequence:

[0055] max(D i )=4.2;

[0056] min(D i )=1.8;

[0057] Substitute into the formula to calculate the fluctuation range:

[0058] ;

[0059] Calculating RMS Volatility ;

[0060] Calculates the mean of a series :

[0061] ;

[0062] Calculate the squared deviation of each difference from the mean and sum them:

[0063] ;

[0064] Substitute into the formula to calculate the root mean square fluctuation:

[0065] ;

[0066] The fluctuation range of the characteristic difference sequence is 2.4, the root mean square fluctuation is 0.849, and the fluctuation range R represents the maximum amplitude of the difference value. It reflects the stability and deviation of the overall fluctuation. These two results will be used for the annotation and analysis of subsequent time segments.

[0067] S203. Based on the set of fluctuation segments, identify the characteristic values of the time series of the fluctuation segments, calculate the numerical trend of the characteristic values changing with time, sort the characteristic values according to the time point characteristics, and obtain the fluctuation characteristic mapping structure.

[0068] Extract the corresponding time series eigenvalues in each fluctuation segment, analyze the distribution of the eigenvalues on the time axis, calculate the change amplitude of each eigenvalue at different time points, record the trend information of the eigenvalue changes over time segment by segment, arrange the eigenvalues in order of time points, construct the mapping relationship between time series and eigenvalues, and extract the trend model of the eigenvalue changes over time. Summarize and organize the trend models extracted from all segments to obtain the fluctuation feature mapping structure.

[0069] S3. Based on the fluctuation feature mapping structure, the data synchronization paths between IoT devices are collected, the connection efficiency and load status of the communication nodes of IoT devices are evaluated, the data flow intensity and synchronization path dependency value between nodes are calculated, and the data flow direction and node connection weight are adjusted at the same time to obtain the spatiotemporal consistency results.

[0070] Optionally, the specific operation steps of S3 may include the following S301-S303:

[0071] S301. Based on the fluctuation feature mapping structure, collect the data synchronization paths between IoT devices, parse the communication records between devices, identify the data flow and transmission times between nodes in the path, analyze the synchronization relationship between nodes, and obtain the data synchronization path set.

[0072] Collect data synchronization paths between IoT devices, extract data interaction information between nodes by analyzing communication records between devices, parse the source and target node information in the records one by one, count the data flow and transmission times of each node on the path, determine their synchronization relationship by comparing the transmission volume between nodes, calculate the transmission ratio of each node in the path based on the transmission times and flow distribution, mark the synchronization rules of the path in combination with the time series between nodes, merge and organize the interaction relationships of all nodes in the path to obtain a data synchronization path set.

[0073] S302. Based on the data synchronization path set, evaluate the connection efficiency and load status of the communication nodes, analyze the path traffic distribution and node load data between nodes, calculate the transmission rate and path dependency value of the nodes, identify the data flow direction and intensity between nodes, and obtain the node load evaluation result.

[0074] The connection efficiency and load status of communication nodes are analyzed. By extracting the transmission rate and path traffic distribution data of each node, the load ratio of the nodes in the path is calculated node by node. The efficiency of each node is evaluated by comparing the distribution pattern of node load and path traffic. High-load nodes are marked by analyzing the node load fluctuation and path dependency relationship. The transmission strength between nodes is calculated according to the data flow direction in the path. A path efficiency evaluation table is established by combining the load and path dependency characteristics between nodes to obtain the node load evaluation results.

[0075] S303. Based on the node load evaluation results, adjust the node connection weights in the synchronization path, calculate the load balance between nodes, redistribute the data flow direction, and adjust the node connection relationship to obtain a spatiotemporal consistency result.

[0076] Adjust the node connection weights in the synchronization path, calculate the traffic adjustment ratio between nodes by analyzing the load balancing situation of the nodes in the path, compare the connection status and data flow characteristics of the nodes one by one, re-plan the traffic distribution direction, and optimize the transmission efficiency between paths by eliminating redundant transmission tasks from the traffic of high-load nodes and reallocating them to low-load nodes. Combined with the traffic distribution characteristics formed after the adjustment of the node connection relationship, a time synchronization model is constructed to obtain spatiotemporal consistency results.

[0077] Adjust the node connection weights in the synchronization path and calculate the load balancing between nodes according to the formula:

[0078] ;

[0079] Calculate the weight between nodes ;

[0080] in, represents the data flow from node i to node j, represents the communication efficiency from node i to node j, Represents the load between node i and node j.

[0081] Detailed explanation of the formula and the process of formula calculation and derivation:

[0082] The node connection weight is calculated by combining traffic, communication efficiency, and load. The definition and calculation of each parameter are as follows:

[0083] Data traffic ;

[0084] Data flow is obtained by collecting the number of data packets transmitted between nodes, and recording the amount of data from node i to node j per unit time.

[0085] Assume that the data packet sizes from node i to node j in unit time are ,but:

[0086] ;

[0087] Assumptions (Unit: MB), then:

[0088] ;

[0089] Communication efficiency ;

[0090] Communication efficiency is obtained by measuring the success rate of transmission, and the calculation formula is:

[0091] ;

[0092] in is the amount of data successfully transferred (MB), is the total amount of data transferred (MB).

[0093] Assumptions , ,but:

[0094] ;

[0095] Load ;

[0096] The load between nodes is obtained by collecting the number of data packets per unit time and the node response time. The load formula is:

[0097] ;

[0098] in Indicates the node response time (seconds), Indicates the number of data packets per unit time;

[0099] when , ,but:

[0100] ;

[0101] Calculate node connection weights ;

[0102] Substitute the above calculation results into the formula:

[0103] ;

[0104] The connection weight between node i and node j is 1.045. A higher weight value indicates higher communication efficiency and lower load between nodes. This weight value will be used to adjust the node connection relationship in the synchronization path, thereby optimizing the direction of data flow and ultimately obtaining a spatiotemporal consistency result.

[0105] S4. Based on the spatiotemporal consistency results, the network status and bandwidth allocation during the synchronization process are collected. According to the current transmission path load status, the traffic distribution of the synchronization path is calculated. According to the communication efficiency and current load of the path, the data transmission channel is reallocated to obtain the optimized configuration of the transmission channel.

[0106] Optionally, the specific operation steps of S4 may include the following S401-S403:

[0107] S401. Based on the spatiotemporal consistency results, collect the network status and bandwidth allocation data of multiple nodes during the synchronization process, analyze the bandwidth occupancy between nodes, extract the traffic distribution characteristics in the synchronization path, and classify the multi-node transmission rate and path data traffic to obtain the synchronization path load distribution.

[0108] The network status and bandwidth allocation data of multiple nodes are collected during the synchronization process. The bandwidth usage of each node is extracted by analyzing the communication traffic between nodes. The bandwidth usage is correlated with the time period and the resource usage ratio of each node in different time periods is determined. The nodes are classified path by path according to their actual transmission rate and path data traffic. The average traffic distribution of each path is calculated. The load characteristics of high-traffic nodes and low-traffic nodes in the path are compared to organize the relationship between node and path load to obtain the load distribution of the synchronization path.

[0109] S402. Based on the load distribution of the synchronization path, calculate the traffic distribution characteristics of the synchronization path, analyze the communication efficiency and load between multiple paths, locate the load path, organize the traffic distribution and load status between the paths, and obtain the path communication efficiency evaluation result.

[0110] Calculate the traffic distribution characteristics between paths, analyze the traffic transmission ratio path by path, extract the traffic occupancy differences between multiple paths, compare the data transmission efficiency of high-load paths and low-load paths, and mark the load balance status between paths in combination with the node load. Determine the path dependency by analyzing the fluctuation amplitude and transmission direction of the path traffic data. Organize the traffic distribution and node load of each path, mark the paths with relatively high load one by one, and merge them to generate a complete evaluation table of the communication efficiency between paths to obtain the path communication efficiency evaluation results.

[0111] Calculate the traffic distribution characteristics of the synchronization path, analyze the communication efficiency and load between multiple paths, and locate the load path according to the formula;

[0112] ;

[0113] Calculate the traffic distribution characteristics of the path ;

[0114] in, represents the communication time of path i, represents the data load of path i, and n represents the total number of paths.

[0115] Detailed explanation of the formula and the process of formula calculation and derivation:

[0116] : Communication time;

[0117] The communication time is calculated by monitoring the time points at which data transmission between nodes starts and ends, using the formula:

[0118] ;

[0119] in is the transmission end time of path i, is the transmission start time of path i;

[0120] : Data payload;

[0121] Data payload is the total amount of data transmitted by path i within a specified time, in MB, and is calculated by summing the packet sizes on the path using the following formula:

[0122] ;

[0123] in is the size of the jth packet on path i, is the total number of packets on path i;

[0124] Formula derivation process:

[0125] According to the formula;

[0126] ;

[0127] Communication time for each path and load Calculate the weighted average of the path traffic distribution through monitoring and collection , reflects the comprehensive characteristics of all paths in terms of load and time;

[0128] The example is introduced:

[0129] Assume there are three paths and the collected parameters are as follows:

[0130] Path 1: Second, MB;

[0131] Path 2: Second, MB;

[0132] Path 3: Second, MB;

[0133] Calculate the traffic distribution characteristics of each path:

[0134] Molecular computing:

[0135] ;

[0136] Denominator calculation:

[0137] ;

[0138] Flow distribution characteristics calculation:

[0139] ;

[0140] Indicates the traffic distribution characteristics of the path The value is 66.67 MB / s, indicating that the combined performance of traffic and time across all paths is a data flow capacity of approximately 66.67 MB per second. This result is used for path load and communication efficiency analysis in subsequent steps.

[0141] S403. Based on the results of the path communication efficiency evaluation and in combination with the path load status, the data traffic distribution is adjusted to select the traffic distribution of low-load paths and balanced-load paths, optimize the traffic transmission direction, improve the synchronization path load structure, and obtain the optimized configuration of the transmission channel.

[0142] Adjust the data traffic distribution based on the path load status, determine the priority distribution path for optimizing traffic transmission by screening the paths with lower loads, adjust the traffic distribution ratio according to the connection relationship of the nodes, eliminate redundant data traffic path by path and redistribute it to the low-load path, balance the data pressure in the high-load path, optimize the transmission direction between nodes and form a balanced load distribution, organize the traffic distribution data after the path adjustment, and obtain the optimized configuration of the transmission channel.

[0143] S5. Based on the optimized configuration of the transmission channel and the business data synchronization rate in the path, the characteristic values of the synchronization rate changes within the time window are analyzed, and the data sequences with high-frequency fluctuation characteristics in the synchronization path are screened. Based on the data synchronization rate in the path, the path allocation of the synchronization task is optimized to obtain the data synchronization path optimization result.

[0144] Optionally, the specific operation steps of S5 may include the following S501-S503:

[0145] S501. Based on the optimized configuration of the transmission channel, the synchronization rate of the business data in the path is collected, the rate change within the time window is analyzed, and based on the key characteristic values of the synchronization rate, the time points and the corresponding rate data are compared to construct a rate characteristic sequence to obtain the synchronization rate characteristic sequence.

[0146] The synchronization rate of business data in the acquisition path is analyzed by recording the transmission rate data at different time points, analyzing the rate changes within the time window, comparing the key characteristic values of the synchronization rate at each time point, extracting the matching relationship between the time point and the corresponding rate data, marking the time period with obvious changes according to the rate fluctuation range, and organizing the rate data within the time period into a sequence in chronological order. Combining the correlation structure between the characteristic value and time, a complete rate change model is generated to obtain the synchronization rate characteristic sequence.

[0147] S502: Based on the synchronization rate feature sequence, analyze the rate fluctuation frequency and amplitude in the time window, calculate the frequency of rate change in the time period, filter data paths with abnormal frequencies, mark abnormal fluctuation paths, and obtain an abnormal fluctuation path feature set.

[0148] Analyze the rate fluctuation frequency and amplitude within the time window, calculate the frequency data of rate change segment by segment, classify the fluctuation amplitude of rate data within the time period, filter out data segments with significantly higher fluctuation frequency than other paths, mark data paths with abnormal frequency, analyze the rate change characteristics of the time period in the abnormal path by comparing the time distribution of the fluctuation amplitude and frequency of each path, summarize the abnormal data path and the rate fluctuation characteristics of the corresponding time period, and obtain the abnormal fluctuation path feature set.

[0149] Analyze the rate fluctuation frequency and amplitude within the time window and calculate the frequency of rate change within the time period according to the formula:

[0150] ;

[0151] Calculate the frequency fluctuation characteristic F;

[0152] in, Represents the total length of the time window in seconds. Represents the data rate change in the i-th time period, in M bps , Represents the time span of the i-th time period in seconds, and n represents the total number of segments in the time window.

[0153] Detailed explanation of the formula and the process of formula calculation and derivation:

[0154] Parameter acquisition method:

[0155] : The total length of the time window is directly recorded by the actual transmission time period of the monitoring device to ensure that the complete synchronization period is covered, for example, the acquisition window is 10 seconds;

[0156] : It is calculated by collecting the instantaneous rate difference of the equipment in each time period. The calculation formula is:

[0157] ;

[0158] in and Time periods and Data rate;

[0159] :The time length of each segment, which can be directly calculated through time records, such as time period The start and end time difference;

[0160] n: total number of time segments, based on the time window and the length of time between segments The ratio of is obtained, the formula is;

[0161] ;

[0162] Specific numerical examples:

[0163] Second;

[0164] Second;

[0165] Data rate acquisition value R:

[0166] ;

[0167] M bps ;

[0168] Segment calculation process:

[0169] ;

[0170] Calculate the rate change for each segment :

[0171] Paragraph 1: M bps ;

[0172] Paragraph 2: M bps ;

[0173] Paragraph 3: M bps;

[0174] Paragraph 4: M bps ;

[0175] Calculate the rate change frequency for each segment:

[0176] Paragraph 1: ;

[0177] Paragraph 2: ;

[0178] Paragraph 3: ;

[0179] Paragraph 4: ;

[0180] Calculate the square value of each segment:

[0181] Paragraph 1: ;

[0182] Paragraph 2: ;

[0183] Paragraph 3: ;

[0184] Paragraph 4: ;

[0185] Final calculation:

[0186] ;

[0187] The frequency characteristic value of rate fluctuation is 6.25, which represents the intensity of rate fluctuation in the time window. By comparing with the set frequency threshold, we can filter out paths with abnormal frequencies and mark them as abnormal fluctuation paths, thus providing support for the next step of optimizing the data synchronization path.

[0188] S503. Based on the abnormal fluctuation path feature set, combined with the synchronization rate data and node load status in the path, the synchronization task is reallocated to the path with lower fluctuation amplitude, and the synchronization configuration between the paths is optimized to obtain the data synchronization path optimization result.

[0189] Combining the synchronization rate data and node load status in the path, the rate variation range of the node is analyzed path by path. The path with low fluctuation amplitude is selected according to the path load data, the synchronization tasks of the high fluctuation amplitude path are eliminated, and the tasks are reallocated to the low fluctuation amplitude path. The overall synchronization configuration is optimized by adjusting the task distribution between paths, and the load relationship and task distribution of nodes and paths are updated to obtain the data synchronization path optimization result.

[0190] In the embodiment of the present invention, sliding window technology and time series analysis are used to efficiently extract and compress data features, significantly reduce redundancy in data transmission, reduce network burden, and improve transmission rate and efficiency. The application of dynamic segmented decoding and feature mapping structure enables stronger responsiveness and prediction accuracy to data fluctuations, intelligently adjusts data flow and node connection weights, optimizes data synchronization paths according to real-time network status and bandwidth allocation, ensures efficient synchronization of data between multiple nodes and multiple platforms, improves the real-time nature of data synchronization, reduces data conflicts, and improves overall reliability and performance.

[0191] Figure 2 This is a block diagram of a digital twin data synchronization system that integrates deep learning and the Internet of Things according to an embodiment of the present invention. The system is used for a digital twin data synchronization method that integrates deep learning and the Internet of Things. Figure 2 The system includes a temporal feature sparsification module 210, a feature fluctuation decoding module 220, a fluctuation path mapping module 230, a path load optimization module 240, and a task synchronization allocation module 250, wherein:

[0192] The time feature sparsification module 210 is used to collect time spans based on the digital twin information of IoT devices, call timestamps and data sequence identifiers through a sliding window, calculate the variation range of feature values of adjacent time slices in the data sequence, and perform transmission data sparsification processing to obtain a sparse feature set;

[0193] The feature fluctuation decoding module 220 is used to calculate the difference between each feature value based on the sparse feature set and the time series feature value, analyze the fluctuation range and change characteristics in the IoT device data synchronization, and perform dynamic segmented decoding based on the analysis results to obtain a fluctuation feature mapping structure;

[0194] The fluctuation path mapping module 230 is used to collect data synchronization paths between IoT devices based on the fluctuation feature mapping structure, evaluate the connection efficiency and load status of IoT device communication nodes, calculate the data flow intensity and synchronization path dependency value between nodes, and adjust the data flow direction and node connection weight to obtain spatiotemporal consistency results;

[0195] The path load optimization module 240 is used to collect the network status and bandwidth allocation during the synchronization process based on the spatiotemporal consistency results, calculate the traffic distribution of the synchronization path according to the current transmission path load status, and obtain the optimized transmission channel configuration based on the communication efficiency and current load of the path;

[0196] The task synchronization allocation module 250 is used to optimize the configuration of the transmission channel, analyze the characteristic values of the synchronization rate changes in the time window according to the business data synchronization rate in the path, screen the data sequences with high-frequency fluctuation characteristics in the synchronization path, and obtain the data synchronization path optimization results based on the data synchronization rate in the path.

[0197] Optionally, the sparse feature set includes a time slice sequence, a feature value variation range, and a sparse data point set; the fluctuation feature mapping structure includes a feature fluctuation range, a time series difference distribution, and a dynamic segmentation structure; the spatiotemporal consistency results include node flow intensity values, synchronization path dependency indicators, and communication node weight distributions; the transmission channel optimization configuration includes traffic distribution results, path load distribution, and bandwidth utilization; the data synchronization path optimization results include high-frequency fluctuation paths, synchronization task allocation tables, and path priority results.

[0198] Optionally, the temporal feature sparsification module 210 is configured to:

[0199] S101. Based on the digital twin information of the IoT device, call the timestamp and data sequence identifier, collect data fragments within the corresponding time period by sliding a fixed time window, and detect the data range and content integrity in the time window to obtain time series fragments;

[0200] S102: Based on the time series segments, compare the eigenvalues of adjacent time segments, calculate the eigenvalue variation range in the time segments, extract the data fluctuation amplitude, and count the variation amplitude of the segments and the corresponding time points to obtain a set of eigenvalue variation amplitudes;

[0201] S103: Based on the eigenvalue change amplitude set, filter the time segments with low change amplitude, remove the data content with low change amplitude, and reorganize the remaining eigenvalues according to the sparsification logic to optimize the transmission data load and obtain a sparse feature set.

[0202] Optionally, the characteristic fluctuation decoding module 220 is configured to:

[0203] S201, based on the sparse feature set, arranging feature values according to time series, performing difference data analysis on the feature values at each time point, determining the difference between adjacent feature values, and obtaining a feature difference sequence;

[0204] S202: Based on the characteristic difference sequence, analyze the data fluctuation range of the characteristic value difference, count the numerical changes of the difference values, identify the time period of the numerical changes, mark the start and end time of the segments, and obtain a set of fluctuation segments;

[0205] S203: Based on the set of fluctuation segments, identify the characteristic values of the time series of the fluctuation segments, calculate the numerical trend of the characteristic values changing with time, sort the characteristic values according to the time point characteristics, and obtain the fluctuation characteristic mapping structure.

[0206] Optionally, the characteristic fluctuation decoding module 220 is configured to:

[0207] The fluctuation range R of the characteristic value difference is calculated according to the following formula (1):

[0208] (1)

[0209] The root mean square fluctuation of the eigenvalue is calculated according to the following formula (2): :

[0210] (2)

[0211] Among them, D i Represents each difference value in the feature difference sequence, max(D i ) represents the maximum value of the difference value in the feature difference sequence, min(D i ) represents the minimum value of the difference value in the feature difference sequence, n represents the total number of data points in the feature difference sequence, Represents the mean of the feature difference series.

[0212] Optionally, the fluctuation path mapping module 230 is configured to:

[0213] S301. Based on the fluctuation characteristic mapping structure, collect data synchronization paths between IoT devices, analyze communication records between devices, identify data flow and transmission times between nodes in the path, analyze synchronization relationships between nodes, and obtain a data synchronization path set;

[0214] S302: Based on the data synchronization path set, evaluate the connection efficiency and load status of the communication nodes, analyze the path traffic distribution and node load data between the nodes, calculate the transmission rate and path dependency value of the nodes, identify the data flow direction and intensity between the nodes, and obtain a node load evaluation result;

[0215] S303. Based on the node load evaluation result, adjust the node connection weights in the synchronization path, calculate the load balance between nodes, redistribute the data flow direction, and adjust the node connection relationship to obtain a spatiotemporal consistency result.

[0216] Optionally, the path load optimization module 240 is configured to:

[0217] S401. Based on the spatiotemporal consistency results, collect network status and bandwidth allocation data of multiple nodes during the synchronization process, analyze bandwidth usage between nodes, extract traffic distribution characteristics in the synchronization path, and classify multi-node transmission rates and path data traffic to obtain synchronization path load distribution;

[0218] S402: Based on the synchronization path load distribution, calculate the traffic distribution characteristics of the synchronization path, analyze the communication efficiency and load between multiple paths to locate the load path, organize the traffic distribution and load status between the paths, and obtain a path communication efficiency evaluation result;

[0219] S403. Based on the path communication efficiency evaluation results and in combination with the path load status, the data traffic distribution is adjusted to select the traffic distribution of low-load paths and balanced-load paths, optimize the traffic transmission direction, improve the synchronization path load structure, and obtain the optimized configuration of the transmission channel.

[0220] Optionally, the task synchronization allocation module 250 is configured to:

[0221] S501: Based on the optimized configuration of the transmission channel, collect the synchronization rate of the service data in the path, analyze the rate change within the time window, and compare the time points and the corresponding rate data based on the key characteristic values of the synchronization rate to construct a rate characteristic sequence to obtain the synchronization rate characteristic sequence;

[0222] S502: Based on the synchronization rate feature sequence, analyze the rate fluctuation frequency and amplitude within the time window, calculate the frequency of rate change within the time period, filter data paths with abnormal frequencies, mark abnormal fluctuation paths, and obtain an abnormal fluctuation path feature set;

[0223] S503. Based on the abnormal fluctuation path feature set, combined with the synchronization rate data and node load status in the path, the synchronization task is reallocated to the path with lower fluctuation amplitude, and the synchronization configuration between the paths is optimized to obtain the data synchronization path optimization result.

[0224] Optionally, the task synchronization allocation module 250 is configured to:

[0225] The rate fluctuation frequency and amplitude within the analysis time window are calculated, and the frequency fluctuation characteristic F is calculated according to the following formula (3):

[0226] (3)

[0227] in, Represents the total length of the time window, Represents the data rate change in the i-th time period, represents the time span of the i-th time period, and n represents the total number of segments in the time window.

[0228] In the embodiment of the present invention, sliding window technology and time series analysis are used to efficiently extract and compress data features, significantly reduce redundancy in data transmission, reduce network burden, and improve transmission rate and efficiency. The application of dynamic segmented decoding and feature mapping structure enables stronger responsiveness and prediction accuracy to data fluctuations, intelligently adjusts data flow and node connection weights, optimizes data synchronization paths according to real-time network status and bandwidth allocation, ensures efficient synchronization of data between multiple nodes and multiple platforms, improves the real-time nature of data synchronization, reduces data conflicts, and improves overall reliability and performance.

[0229] Figure 3 This is a structural diagram of a digital twin data synchronization device that integrates deep learning and the Internet of Things, provided by an embodiment of the present invention. Figure 3 As shown, the digital twin data synchronization device integrating deep learning and IoT can include the above Figure 2 The digital twin data synchronization system integrating deep learning and the Internet of Things is shown. Optionally, the digital twin data synchronization device 310 integrating deep learning and the Internet of Things may include a first processor 2001.

[0230] Optionally, the digital twin data synchronization device 310 integrating deep learning and the Internet of Things may also include a memory 2002 and a transceiver 2003 .

[0231] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0232] The following combination Figure 3 The following describes in detail the components of the digital twin data synchronization device 310 that integrates deep learning and the Internet of Things:

[0233] The first processor 2001 is the control center of the digital twin data synchronization device 310 that integrates deep learning and the Internet of Things. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs) or one or more field programmable gate arrays (FPGAs).

[0234] Optionally, the first processor 2001 can perform various functions of the digital twin data synchronization device 310 that integrates deep learning and the Internet of Things by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0235] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 3 CPU0 and CPU1 are shown in FIG.

[0236] In a specific implementation, as an embodiment, the digital twin data synchronization device 310 integrating deep learning and the Internet of Things may also include multiple processors, such as Figure 3 1 and 2. The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). A processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0237] The memory 2002 is used to store the software program for executing the solution of the present invention, and is controlled by the first processor 2001 for execution. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0238] Optionally, the memory 2002 may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001 or exist independently and be synchronized with the first processor 2001 through the interface circuit ( Figure 3 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0239] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0240] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 3 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0241] Optionally, the transceiver 2003 can be integrated with the first processor 2001 or can exist independently and integrate deep learning with the interface circuit of the digital twin data synchronization device 310 of the Internet of Things ( Figure 3 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0242] It should be noted that Figure 3 The structure of the digital twin data synchronization device 310 that integrates deep learning and the Internet of Things shown in the figure does not constitute a limitation on the router. The actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0243] In addition, the technical effects of the digital twin data synchronization device 310 that integrates deep learning and the Internet of Things can refer to the technical effects of the digital twin data synchronization method that integrates deep learning and the Internet of Things described in the above method embodiment, and will not be repeated here.

[0244] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0245] It should also be understood that the memory in the embodiments of the present invention may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0246] The above embodiments can be implemented in whole or in part via software, hardware (e.g., circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When loaded or executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are fully or partially performed. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. The available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0247] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. A and B can be singular or plural. Furthermore, the character " / " as used herein generally indicates an "or" relationship between the associated objects, but it may also indicate an "and / or" relationship. For specific understanding, please refer to the context.

[0248] In this disclosure, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, "at least one of a, b, or c" can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or plural.

[0249] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0250] Those skilled 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. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0251] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0252] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the system or unit, which can be electrical, mechanical or other forms.

[0253] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0254] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0255] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or the portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical disks.

[0256] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A digital twin data synchronization method integrating deep learning and the Internet of Things, characterized by: The method comprises: S1. Based on the digital twin information of IoT devices, the timestamp and data sequence identifier are called through a sliding window to collect time spans, calculate the variation range of the feature values of adjacent time slices in the data sequence, and perform transmission data sparsification processing to obtain a sparse feature set; S2. Based on the sparse feature set and time series feature values, the difference between each feature value is calculated, the fluctuation range and change characteristics in the IoT device data synchronization are analyzed, and dynamic segmented decoding is performed based on the analysis results to obtain a fluctuation feature mapping structure; S3. Based on the fluctuation feature mapping structure, collect data synchronization paths between IoT devices, evaluate the connection efficiency and load status of IoT device communication nodes, calculate the data flow intensity and synchronization path dependency value between nodes, and adjust the data flow direction and node connection weights to obtain spatiotemporal consistency results; S4. Based on the spatiotemporal consistency results, collect the network status and bandwidth allocation during the synchronization process, calculate the traffic distribution of the synchronization path according to the current transmission path load status, and obtain the optimal configuration of the transmission channel based on the communication efficiency and current load of the path; S5. Based on the optimized configuration of the transmission channel, according to the business data synchronization rate in the path, analyze the characteristic values of the synchronization rate changes in the time window, screen the data sequences with high-frequency fluctuation characteristics in the synchronization path, and obtain the data synchronization path optimization result according to the data synchronization rate in the path.

2. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: The sparse feature set includes a time slice sequence, a feature value variation range, and a sparse data point set; the fluctuation feature mapping structure includes a feature fluctuation range, a time series difference distribution, and a dynamic segmentation structure; the spatiotemporal consistency results include node flow intensity values, synchronization path dependency indicators, and communication node weight distributions; the transmission channel optimization configuration includes traffic distribution results, path load distribution, and bandwidth utilization; the data synchronization path optimization results include high-frequency fluctuation paths, synchronization task allocation tables, and path priority results.

3. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: The digital twin information of S1 based on the IoT device calls the timestamp and data sequence identifier through a sliding window, collects the time span, calculates the variation amplitude of the characteristic values of adjacent time slices in the data sequence, and performs transmission data sparse processing to obtain a sparse feature set, including: S101. Based on the digital twin information of the IoT device, call the timestamp and data sequence identifier, collect data fragments within the corresponding time period by sliding a fixed time window, and detect the data range and content integrity in the time window to obtain time series fragments; S102: Based on the time series segments, compare the eigenvalues of adjacent time segments, calculate the eigenvalue variation range in the time segments, extract the data fluctuation amplitude, and count the variation amplitude of the segments and the corresponding time points to obtain a set of eigenvalue variation amplitudes; S103: Based on the eigenvalue change amplitude set, filter the time segments with low change amplitude, remove the data content with low change amplitude, and reorganize the remaining eigenvalues according to the sparsification logic to optimize the transmission data load and obtain a sparse feature set.

4. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: The S2 is based on the sparse feature set and the time series feature value, calculates the difference between each feature value, analyzes the fluctuation range and change characteristics in the IoT device data synchronization, and performs dynamic segmented decoding based on the analysis results to obtain a fluctuation feature mapping structure, including: S201, based on the sparse feature set, arranging feature values according to time series, performing difference data analysis on the feature values at each time point, determining the difference between adjacent feature values, and obtaining a feature difference sequence; S202: Based on the characteristic difference sequence, analyze the data fluctuation range of the characteristic value difference, count the numerical changes of the difference values, identify the time period of the numerical changes, mark the start and end time of the segments, and obtain a set of fluctuation segments; S203: Based on the set of fluctuation segments, identify the characteristic values of the time series of the fluctuation segments, calculate the numerical trend of the characteristic values changing with time, sort the characteristic values according to the time point characteristics, and obtain the fluctuation characteristic mapping structure.

5. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 4 is characterized in that: The data fluctuation range of the analysis of the characteristic value difference in S202 and the numerical change of the statistical difference value include: The fluctuation range R of the characteristic value difference is calculated according to the following formula (1): （1） The root mean square fluctuation of the eigenvalue is calculated according to the following formula (2): : （2） Among them, D i Represents each difference value in the feature difference sequence, max(D i ) represents the maximum value of the difference value in the feature difference sequence, min(D i ) represents the minimum value of the difference value in the feature difference sequence, n represents the total number of data points in the feature difference sequence, Represents the mean of the feature difference series.

6. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: Based on the fluctuation feature mapping structure, S3 collects data synchronization paths between IoT devices, evaluates the connection efficiency and load status of IoT device communication nodes, calculates the data flow intensity and synchronization path dependency value between nodes, and adjusts the data flow direction and node connection weight to obtain spatiotemporal consistency results, including: S301. Based on the fluctuation characteristic mapping structure, collect data synchronization paths between IoT devices, analyze communication records between devices, identify data flow and transmission times between nodes in the path, analyze synchronization relationships between nodes, and obtain a data synchronization path set; S302: Based on the data synchronization path set, evaluate the connection efficiency and load status of the communication nodes, analyze the path traffic distribution and node load data between the nodes, calculate the transmission rate and path dependency value of the nodes, identify the data flow direction and intensity between the nodes, and obtain a node load evaluation result; S303. Based on the node load evaluation result, adjust the node connection weights in the synchronization path, calculate the load balance between nodes, redistribute the data flow direction, and adjust the node connection relationship to obtain a spatiotemporal consistency result.

7. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: The S4 collects the network status and bandwidth allocation during the synchronization process based on the spatiotemporal consistency result, calculates the traffic distribution of the synchronization path according to the current transmission path load status, and obtains the optimized configuration of the transmission channel according to the communication efficiency and current load of the path, including: S401. Based on the spatiotemporal consistency results, collect network status and bandwidth allocation data of multiple nodes during the synchronization process, analyze bandwidth usage between nodes, extract traffic distribution characteristics in the synchronization path, and classify multi-node transmission rates and path data traffic to obtain synchronization path load distribution; S402: Based on the synchronization path load distribution, calculate the traffic distribution characteristics of the synchronization path, analyze the communication efficiency and load between multiple paths to locate the load path, organize the traffic distribution and load status between the paths, and obtain a path communication efficiency evaluation result; S403. Based on the path communication efficiency evaluation results and in combination with the path load status, the data traffic distribution is adjusted to select the traffic distribution of low-load paths and balanced-load paths, optimize the traffic transmission direction, improve the synchronization path load structure, and obtain the optimized configuration of the transmission channel.

8. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 1 is characterized in that: The S5 is based on the transmission channel optimization configuration, according to the service data synchronization rate in the path, analyzing the characteristic value of the synchronization rate change in the time window, screening the data sequence with high-frequency fluctuation characteristics in the synchronization path, and obtaining the data synchronization path optimization result according to the data synchronization rate in the path, including: S501: Based on the optimized configuration of the transmission channel, collect the synchronization rate of the service data in the path, analyze the rate change within the time window, and compare the time points and the corresponding rate data based on the key characteristic values of the synchronization rate to construct a rate characteristic sequence to obtain the synchronization rate characteristic sequence; S502: Based on the synchronization rate feature sequence, analyze the rate fluctuation frequency and amplitude within the time window, calculate the frequency of rate change within the time period, filter data paths with abnormal frequencies, mark abnormal fluctuation paths, and obtain an abnormal fluctuation path feature set; S503. Based on the abnormal fluctuation path feature set, combined with the synchronization rate data and node load status in the path, the synchronization task is reallocated to the path with lower fluctuation amplitude, and the synchronization configuration between the paths is optimized to obtain the data synchronization path optimization result.

9. The digital twin data synchronization method integrating deep learning and the Internet of Things according to claim 8 is characterized in that: The analysis of the rate fluctuation frequency and amplitude within the time window and the calculation of the frequency of rate change within the time period include: The rate fluctuation frequency and amplitude within the analysis time window are calculated, and the frequency fluctuation characteristic F is calculated according to the following formula (3): （3） in, Represents the total length of the time window, Represents the data rate change in the i-th time period, represents the time span of the i-th time period, and n represents the total number of segments in the time window.

10. A digital twin data synchronization system integrating deep learning and the Internet of Things, wherein the digital twin data synchronization system integrating deep learning and the Internet of Things is used to implement the digital twin data synchronization method integrating deep learning and the Internet of Things as described in any one of claims 1 to 9, characterized in that: The system comprises: The time feature sparsification module is used for digital twin information based on IoT devices. It uses a sliding window to call timestamps and data sequence identifiers to collect time spans, calculate the change amplitude of the characteristic values of adjacent time slices in the data sequence, and perform transmission data sparsification processing to obtain a sparse feature set. The feature fluctuation decoding module is used to calculate the difference between each feature value based on the sparse feature set and the time series feature value, analyze the fluctuation range and change characteristics in the IoT device data synchronization, and perform dynamic segmented decoding based on the analysis results to obtain the fluctuation feature mapping structure; The fluctuation path mapping module is used to collect data synchronization paths between IoT devices based on the fluctuation feature mapping structure, evaluate the connection efficiency and load status of IoT device communication nodes, calculate the data flow intensity and synchronization path dependency value between nodes, and adjust the data flow direction and node connection weight to obtain spatiotemporal consistency results; The path load optimization module is used to collect network status and bandwidth allocation during the synchronization process based on the spatiotemporal consistency results. It calculates the traffic distribution of the synchronization path according to the current transmission path load status and obtains the optimal configuration of the transmission channel based on the communication efficiency and current load of the path. The task synchronization allocation module is used to optimize the configuration of the transmission channel, analyze the characteristic values of the synchronization rate changes within the time window according to the business data synchronization rate in the path, screen the data sequences with high-frequency fluctuation characteristics in the synchronization path, and obtain the data synchronization path optimization results based on the data synchronization rate in the path.

Images