Industrial real-time data transmission guarantee method and system based on TSN

Abstract

The embodiment of the invention provides an industrial real-time data transmission guarantee method and system based on a TSN. The method is applied to the field of industrial communication and intelligent scheduling, and comprises the steps of obtaining industrial field network node information, completing TSN topology deployment and time synchronization, extracting communication characteristics and historical load data of various task flows, constructing a unified time sequence, and inputting a prediction model to generate a load prediction result. And in combination with task parameters, a scheduling strategy list and a prediction result, inputting a TSN digital twin network model to carry out analogue simulation, evaluating performance indexes under multiple strategies, selecting an optimal scheduling strategy and issuing the optimal scheduling strategy to TSN switching equipment, thereby realizing accurate scheduling control configuration of various task flows. According to the scheme, intelligent evaluation and automatic optimization issuing of the scheduling strategy are realized, the real-time performance and the deterministic communication capability of an industrial network are remarkably improved, the resource utilization efficiency and the adaptability of the scheduling decision are enhanced, and the key task flow is ensured to be reliably transmitted on time.

Description

Technical Field

[0001] The present disclosure relates to the technical field of industrial communication and intelligent scheduling, and particularly to a method and system for ensuring industrial real-time data transmission based on TSN. Background Art

[0002] Traditional industrial networks are limited by the non-deterministic characteristics of Ethernet itself and are difficult to meet the stringent requirements of microsecond-level latency and nanosecond-level clock synchronization in scenarios such as high-precision motion control and multi-device collaborative operations. The emergence of Time-Sensitive Networking (TSN) technology provides a key path to solve this problem. Through mechanisms such as time synchronization, traffic scheduling, and resource reservation, it can achieve deterministic transmission of multiple service flows on the unified Ethernet physical layer. However, the industrial field environment is complex and changeable, and the device heterogeneity is strong. The traditional TSN deployment method still faces problems such as insufficient dynamic adaptation ability and difficulty in ensuring multi-service QoS. How to build an adaptive TSN network architecture to achieve efficient and reliable transmission of industrial real-time data has become the core bottleneck restricting the development of intelligent manufacturing.

[0003] Existing industrial network deployment solutions mostly adopt a statically configured TSN topology, and preset the Gate Control List (GCL) and traffic shaping parameters of each network node through a centralized management platform. In terms of time synchronization, it mainly relies on the IEEE1588v2 protocol to achieve clock alignment between nodes, and cooperates with the boundary clock or transparent clock mechanism to reduce the synchronization error. For traffic flow scheduling, static scheduling strategies based on Credit-Based Shaping (CBS) or Asynchronous Traffic Shaping (ATS) are generally adopted, and differentiated services are achieved by combining reserved bandwidth and priority queues. In terms of dynamic adaptability, some solutions introduce a simple load monitoring mechanism, and perform periodic parameter adjustment through historical data statistics, but lack in-depth coupling analysis of service characteristics and network status.

[0004] The static configuration mode of the existing technology cannot adapt to the dynamic changes in the industrial field. When devices are added or removed, task flows are changed, or network loads fluctuate, manual intervention is required for reconfiguration, resulting in a long network adjustment cycle and high operation and maintenance costs. At the same time, traditional scheduling strategies only consider single-dimensional latency or bandwidth metrics, and fail to establish a multi-parameter joint optimization model for periodicity, fault tolerance, priority, etc., and resource competition or service degradation is likely to occur in complex industrial scenarios. Moreover, the time synchronization accuracy is limited by the hardware clock source and transmission medium, and the error accumulation effect is obvious in cross-domain networking scenarios, making it difficult to meet the requirements of nanosecond-level synchronization accuracy for multi-workshop collaborative control. Summary of the Invention

[0005] In order to solve the deficiencies of the prior art, the present disclosure provides a method and system for industrial real-time data transmission based on TSN. The present disclosure solves the problem that the static configuration mode of the prior art cannot adapt to the dynamic changes of the industrial site. When the equipment increases or decreases, the task flow changes, or the network load fluctuates, manual intervention and reconfiguration are required, resulting in a long network adjustment cycle and high operation and maintenance costs; at the same time, the traditional scheduling strategy only considers the single-dimensional delay or bandwidth indicators, and fails to establish a multi-parameter joint optimization model such as periodicity, fault tolerance, and priority. In complex industrial scenarios, resource competition or service degradation is prone to occur; and the time synchronization accuracy is limited by the hardware clock source and the transmission medium. In the cross-domain networking scenario, the error accumulation effect is obvious, and it is difficult to meet the requirements of nanosecond synchronization accuracy for multi-workshop collaborative control.

[0006] According to a first aspect of the present disclosure, a TSN-based industrial real-time data transmission guarantee method is provided, comprising:

[0007] Acquire network node information of the industrial site, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure supporting the TSN protocol;

[0008] Obtain the local system time information of each TSN network node in the industrial network topology, accurately synchronize the local system time information, and obtain a unified system time benchmark for the entire network;

[0009] Acquire historical communication data and service characteristic data of various task flows in the industrial site, perform feature extraction on the historical communication data and service characteristic data, and obtain task parameters of various task flows; wherein the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level;

[0010] Obtain historical load data of various task flows, perform time alignment on the historical load data according to the system time reference, construct a unified time series, input the unified time series into a preset prediction model, and obtain load prediction results of various task flows;

[0011] Input the system time benchmark, task parameters, preset scheduling strategy list and load forecast results into the preset TSN digital twin network model, perform simulations of multiple scheduling strategies, and obtain performance indicator data under each scheduling strategy;

[0012] The optimal scheduling strategy is determined according to the performance indicator data and the task parameters, and the optimal scheduling strategy is sent to the TSN switching device to complete the scheduling control configuration of various task flows in the network.

[0013] According to a second aspect of the present disclosure, a TSN-based industrial real-time data transmission assurance system is provided, which is used to execute the method according to the first aspect, including:

[0014] A TSN network topology deployment module is used to obtain network node information of the industrial site, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure that supports the TSN protocol;

[0015] The system time synchronization module is used to obtain the local system time information of each TSN network node in the industrial network topology structure, accurately synchronize the local system time information, and obtain a unified system time benchmark for the entire network;

[0016] A task parameter extraction module is used to obtain historical communication data and service characteristic data of various task flows in the industrial site, perform feature extraction on the historical communication data and service characteristic data, and obtain task parameters of various task flows; wherein the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level;

[0017] A load prediction module is used to obtain historical load data of various task flows, time-align the historical load data according to the system time reference, construct a unified time series, input the unified time series into a preset prediction model, and obtain load prediction results of various task flows;

[0018] The TSN simulation module is used to input the system time reference, task parameters, preset scheduling strategy list and load forecast results into the preset TSN digital twin network model, perform simulations of multiple scheduling strategies, and obtain performance indicator data under each scheduling strategy;

[0019] The scheduling optimization module is used to determine the optimal scheduling strategy based on the performance indicator data and the task parameters, send the optimal scheduling strategy to the TSN switching device, and complete the scheduling control configuration of various task flows in the network.

[0020] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, the method described above is implemented.

[0021] In the TSN-based industrial real-time data transmission guarantee method and system provided above, the disclosed embodiment realizes the intelligent evaluation and automatic optimization of scheduling strategies by integrating system time synchronization, task parameter modeling, load prediction and TSN digital twin simulation, which not only significantly improves the real-time and deterministic communication capabilities of the industrial network, but also enhances the resource utilization efficiency and the adaptability of scheduling decisions, ensuring the reliable transmission of critical task flows on time. Brief Description of the Drawings

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present disclosure. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0023] Figure 1 Shows a schematic flowchart of a method for ensuring industrial real-time data transmission based on TSN according to an embodiment of the present disclosure;

[0024] Figure 2 Shows a schematic flowchart of a method for ensuring industrial real-time data transmission based on TSN according to an embodiment of the present disclosure;

[0025] Figure 3 Shows a schematic block diagram of a system for ensuring industrial real-time data transmission based on TSN according to an embodiment of the present disclosure;

[0026] Figure 4 Shows a block diagram of an exemplary electronic device according to an embodiment of the present disclosure. Detailed Description of the Embodiments

[0027] Now, various exemplary embodiments of the present disclosure will be described in detail with reference to the drawings. It should be noted that: unless otherwise specifically stated, the relative arrangements, numerical expressions, and numerical values of the components and steps described in these embodiments do not limit the scope of the present disclosure.

[0028] Those skilled in the art can understand that terms such as "first" and "second" in the embodiments of the present disclosure are only used to distinguish different steps, devices, or modules, etc., and neither represent any specific technical meaning nor indicate an inevitable logical order between them. It should also be understood that in the embodiments of the present disclosure, "a plurality of" may refer to two or more, and "at least one" may refer to one, two, or more. It should also be understood that for any component, data, or structure mentioned in the embodiments of the present disclosure, without clear definition or contrary indication in the context, it can generally be understood as one or more. In addition, the term "and / or" in the present disclosure is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present disclosure generally represents an "or" relationship between the associated objects before and after. It should also be understood that the present disclosure emphasizes the differences between the various embodiments, and their similarities or similarities can be referred to each other. For the sake of brevity, they will not be repeated one by one.

[0029] Meanwhile, it should be understood that, for the convenience of description, the dimensions of the various parts shown in the drawings are not drawn according to the actual proportional relationship. The following description of at least one exemplary embodiment is actually merely illustrative and in no way limits the present disclosure, its application, or its use. Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the said technologies, methods, and devices should be regarded as part of the specification. It should be noted that: like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0030] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure. Apparently, the described embodiments are some, but not all, of the embodiments of the present disclosure. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present disclosure without creative efforts shall fall within the scope of protection of the present disclosure.

[0031] Figure 1 It is a schematic flow diagram of a method for ensuring industrial real-time data transmission based on TSN provided by an embodiment of the present disclosure. The method of the embodiment of the present disclosure aims to achieve precise detection of large and small targets in images.

[0032] S101, obtain the network node information of the industrial site, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure that supports the TSN protocol.

[0033] The industrial site may refer to the physical environment where an industrial control system (ICS) is actually deployed, including a field control network composed of sensors, actuators, PLCs (programmable logic controllers), industrial robots, edge computing devices, data acquisition devices, control terminals, etc.

[0034] The network node information may refer to the identification and attribute data of each communication participating unit in the industrial network, including the node ID or MAC address; the ports and topological positions to which the node is connected; the supported communication protocol stack (such as whether it supports TSN); the node type (such as controller node, sensor node, actuator node, switch node); the network interface parameters (IP address, bandwidth capacity, latency characteristics); and the link status information with other nodes.

[0035] The TSN protocol can be a set of Ethernet real-time transmission extension standards developed by the IEEE 802.1 TSN working group to support deterministic communication at the industrial level. Its key capabilities include time synchronization (IEEE 802.1AS): ensuring that all network devices share a unified clock; time-aware scheduling (IEEE 802.1Qbv): arranging data transmission windows according to a periodic table; traffic shaping and queuing (such as IEEE 802.1Qav, Qch); path redundancy and seamless handover (such as 802.1CB); flow identification and management mechanism (802.1Qci, etc.).

[0036] The industrial network topology structure can refer to a communication network structure diagram that supports the TSN protocol constructed based on the physical connection relationship and logical transmission strategy in the industrial field, including the physical and logical connection relationships between nodes (chain, star, ring, hybrid, etc.); the deployment location and connection strategy of TSN switching devices; path planning and redundancy design (for fault tolerance); the mapping path of communication task flows; parameters such as bandwidth, delay, and synchronization characteristics of each link.

[0037] By means of network scanning, device reporting, and configuration parsing, detailed information of each network node in the industrial field is obtained. This information includes the unique identifier of the node (such as MAC address, device serial number), network interface parameters (such as IP address, number of ports, supported bandwidth and protocol type), device type (such as TSN switch, PLC, sensor, edge gateway, etc.), and the physical connection relationship between nodes. The industrial Ethernet management protocol (such as SNMP, LLDP, OPCUA) can be used to communicate with field devices, collect and parse the link information and capability description provided by the devices. The key network devices with TSN capabilities, such as TSN switches, TSN-supported terminal devices, and edge controllers, are identified and screened out using the collected network node information and marked as TSN nodes. According to the function support information of these TSN nodes, function verification and role division are performed on them, for example, specifying the Grandmaster Clock in IEEE 802.1AS time synchronization, configuring boundary clocks and slave nodes.

[0038] Subsequently, the TSN topology configuration process is executed, which specifically includes the following operations: Time synchronization configuration: Based on the clock support capabilities and connection relationships of each network node, according to the IEEE802.1AS protocol, precise time protocol parameters are configured for all TSN nodes, a synchronous message forwarding path is established, and the clocks of all network nodes are aligned to a unified system time reference; Path and resource configuration: Combining the forwarding capabilities, resource status, and link topology information of network nodes, according to the IEEE802.1Qcc standard, using a centralized network configuration mechanism, the task flow transmission path is planned, and forwarding rules, queue resources, traffic scheduling policies, etc. in the switching nodes are configured for each path; Scheduling policy distribution: Based on the support of network nodes for scheduling protocols (such as IEEE802.1Qbv), a GCL (Gate Control List) containing the scheduling period and transmission time slots is generated and distributed to each switching node to achieve precise timing control of task flow transmission. TSN function enabling and verification: Relying on the known network node distribution, function characteristics, and connection relationships, the TSN function of each node is enabled and connectivity is verified to ensure that the scheduling table is successfully loaded, the time synchronization status is stable, and the path planning is correct. Through the above steps, the acquisition and screening of industrial field network node information are successfully completed, the network topology deployment and configuration based on the TSN protocol are realized, and finally an industrial network topology structure supporting the TSN protocol is obtained, providing a basic guarantee for the determinism of industrial real-time data transmission.

[0039] S102, Obtain the local system time information of each TSN network node in the industrial network topology structure, perform precise synchronization processing on the local system time information, and obtain a unified system time reference for the entire network.

[0040] A TSN network node can refer to a key node that is selected from all network nodes and has TSN capabilities and plays a role in TSN deployment, including TSN switching devices; terminal devices supporting the TSN protocol stack; time synchronization master and slave nodes; devices participating in scheduling configuration and task forwarding. It is a result subset after processes such as functional identification, protocol capability filtering, and task adaptation of "industrial field network node information".

[0041] The local system time information can refer to the current time value of the system clock maintained inside each TSN network node, which represents the node's understanding of the "current moment". Since there are slight differences in the clock hardware of each node, these local times may have deviations. It can include the current local timestamp; clock frequency and drift conditions; time synchronization protocol support information (such as synchronization status, slave / master clock role); data such as the delay and offset of time synchronization messages (PTP, IEEE1588, or 802.1AS).

[0042] The system time reference can refer to a unified standard time reference established among all network nodes through the TSN time synchronization protocol (such as IEEE802.1AS). It ensures that the local clocks of all TSN network nodes are highly consistent, enabling cross-device scheduling and control operations to be based on a unified time axis.

[0043] After completing the deployment of the TSN industrial network topology and identifying the nodes supporting the TSN protocol, to achieve time synchronization, first start the network-wide time synchronization process based on IEEE802.1AS (General Precision Time Protocol, gPTP). Specifically, by enabling the gPTP protocol stack on each TSN network node, read and extract the local system time information of the node, that is, the current timing value of its internal high-precision clock, which is usually provided by the hardware timestamp unit integrated in the device or the system counter driven by the crystal oscillator. Subsequently, according to the Best Master Clock Election algorithm defined by IEEE802.1AS, compare elements such as the clock accuracy level, stability parameter, and priority configuration of each TSN node, and automatically elect the node with the best performance as the master clock. The elected master clock node will periodically broadcast Sync messages and Follow_Up messages to spread its system time information throughout the network.

[0044] When other nodes receive these synchronization messages, use the local timestamp capture mechanism to record the reception time, and cooperate with the PeerDelay measurement mechanism in IEEE802.1AS (using Pdelay_Req and Pdelay_Resp messages) to estimate the link transmission delay between it and the master clock. Based on the master clock time carried in the Sync and Follow_Up messages, as well as the local reception time and link delay, perform clock offset calculation. After completion, adjust the local timer frequency or step to align the current time to make its own system time consistent with the master clock.

[0045] In scenarios with a complex topology or multiple subnets, some TSN switch nodes can be configured as Boundary Clocks. These nodes act as both time slave nodes on a certain link and time master nodes on another link, thus achieving time coordination for cross-domain synchronization links and avoiding error accumulation caused by clock hierarchy transmission.

[0046] To ensure that the time synchronization accuracy meets the TSN requirements (generally within ±1 microsecond or even more precise), the system will set the broadcast period of the synchronization message (such as every 125ms or shorter), and enable the moving average, synchronization jitter filtering, and deviation monitoring mechanisms during the synchronization process to dynamically evaluate the synchronization accuracy. When the time deviation between all nodes is less than the preset threshold value and the synchronization status has been stable for a certain period, it can be confirmed that the establishment of the network-wide unified system time reference is completed.

[0047] This time reference will serve as the key basis for time-sensitive operations such as subsequent scheduling simulation, schedule generation (such as GCL), and task sending control, ensuring that all real-time tasks in the entire industrial control network have a unified and highly accurate time coordinate system.

[0048] S103, obtain the historical communication data and service characteristic data of various task flows in the industrial field, perform feature extraction on the historical communication data and service characteristic data, and obtain the task parameters of various task flows; wherein, the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level.

[0049] Various task flows can refer to data communication flows generated by different business systems or devices in the industrial field, with specific transmission characteristics and real-time requirements. It can include control instruction flows (such as the control signals of PLCs to actuators), status monitoring flows (such as data uploaded by temperature, pressure, and position sensors), alarm information flows (such as alarm data triggered by fault detection), video monitoring flows (such as high-definition video real-time transmission flows), and system configuration flows (such as system parameter updates and remote configuration information). Each type of task flow usually corresponds to a specific business scenario and has different quality of service requirements.

[0050] Historical communication data can refer to the communication records of task flows collected during the network operation in the past period of time, which can include link quality indicators such as packet arrival timestamps, packet sizes (in bytes), packet intervals (periods), actual delay and jitter data, packet loss rate, and retransmission times.

[0051] Service characteristic data can refer to information related to the business background, importance, and operation logic of task flows, such as task type (periodic or event-triggered), device role (master / slave / sensor / actuator), data generation mechanism (timed, edge-triggered, etc.), and the required levels of real-time and reliability.

[0052] Task parameters can be a set of key indicators describing the scheduling requirements of task flows generated after feature extraction of historical communication data and service characteristic data, mainly used for TSN scheduling simulation and decision optimization, including periodicity: indicating whether the task flow is a periodic task and its fixed or variable sending period, such as 10ms, 100ms, etc. The shorter the period, the higher the real-time requirement.

[0053] Maximum delay tolerance: the maximum transmission delay allowed for a task flow from the source node to the target node. For example, for servo control signals, this value may be less than 1ms.

[0054] Bandwidth requirement: It represents the amount of data that a task flow needs to transmit per unit time, usually expressed in Mbps. For example, a video stream may require more than 5 Mbps, while sensor data only needs dozens of Kbps.

[0055] Jitter tolerance: It measures the stability requirement of the arrival interval of a task flow, referring to the maximum acceptable range of transmission time fluctuations. Such as ±20 μs.

[0056] Priority level: It is used to describe the priority scheduling order of a task flow during scheduling. Usually, it is mapped to TC (Traffic Class) based on the priority of IEEE802.1Q. For example, real-time control flow: highest priority, alarm flow: higher priority, ordinary status reporting: medium priority, file update or non-real-time data: lowest priority.

[0057] By deploying industrial communication data acquisition units (such as industrial firewalls, mirror port listeners, industrial protocol gateways, etc.), deep packet capture and protocol identification can be performed on the data packets on key network links. Through the parsing module that supports common protocols in industrial fields (such as Profinet, EtherCAT, OPCUA), protocol layer decoding and task flow identification are performed on the network packets collected in real time. The communication behavior is classified and sorted according to the triple form of data source - data destination - communication content, and a communication flow identifier with business semantics is obtained. For each identified task flow, its historical communication data is recorded, including message send / receive timestamps (used to calculate intervals and delays), message sizes and total bytes (used to calculate bandwidth), actual path delays and hop counts (used for network performance evaluation), whether retransmission occurs, whether congestion packet loss occurs, and other communication status information. These data are stored in local edge nodes or industrial cloud platforms to form a historical communication log database with time continuity.

[0058] In parallel, business characteristic data is collected from the device side. By reading PLC / industrial controller configuration files, task scheduling tables, OPCUA server node structures, etc., meta-information of each type of task such as trigger mechanism (period / event), control logic priority, real-time requirement, and fault level can be obtained. In addition, engineering design drawings of industrial systems (such as BOM, IO list) and BIM integrated systems can also provide data sources at the business layer.

[0059] Next, run the task parameter extraction algorithm on the edge computing node or data analysis platform. For the extraction of periodic parameters, first analyze the timestamp sequence of the task flow through the sliding window algorithm. If the standard deviation of the adjacent message intervals is less than the preset threshold, it is determined as a periodic task, and the average interval is taken as its period value; if there is no stable interval, it is identified as an event-triggered task. The extraction of the maximum delay tolerance can be combined with the control response time requirements set in the service description document, or estimated by finding the maximum delay value before the anomaly in the historical communication data and corrected in combination with the fault tolerance of the service system. The bandwidth requirement forms two indicators, average bandwidth and peak bandwidth, by statistically calculating the total data volume transmitted by the task flow per unit time, which is used to guide the resource allocation of TSN. The jitter tolerance is obtained by statistically analyzing the deviation range between the time interval difference of the periodic task and the theoretical period, and the maximum allowable jitter range is obtained, which reflects its dependence on time stability. The priority level is comprehensively determined by combining the security level, control level, alarm level, etc. of the device to which the task belongs, and can refer to the PCP field in the VLAN tag, the industrial QoS policy configuration, or the priority rules mapped manually by the system engineer.

[0060] Finally, all the extracted task parameters are encapsulated into a unified description format, such as a structured table or a standardized JSON object, and associated with the identification information of the task flow to form a task parameter library. This task parameter library can be directly called by the subsequent TSN digital twin simulation platform or network configuration module for operations such as network scheduling simulation, path planning, bandwidth allocation, and priority mapping, to achieve the real-time performance guarantee of TSN for business requirements.

[0061] S104. Obtain the historical load data of various task flows, align the historical load data according to the system time reference to construct a unified time series, and input the unified time series into a preset prediction model to obtain the load prediction results of various task flows.

[0062] The historical load data can refer to the network load characteristics such as bandwidth usage, data frame sending frequency, and data peak traffic generated during the actual operation of various task flows in the industrial field in the past period. It includes the total number of bytes of data transmitted by a task flow in each time slice (such as 1s, 100ms); the statistical distribution of the number of messages and the size of messages; peak and average bandwidth; the time density of message arrival (i.e., the load burstiness in a short period); the change in network link utilization; the length or delay index of the forwarding queue in each hop path; whether there are packet loss, retransmission, etc. in the task.

[0063] A unified time series can refer to the standardized timeline data structure formed by aligning and resampling the historical load data collected by each network node according to the unified system time reference of the entire network obtained in the previous steps. Due to the deviation or asynchrony in the data collection time of different devices, all node data can be aligned in strict chronological order through a unified time reference (such as the synchronous time provided by IEEE802.1AS), constructing a data structure in the following form:

[0064] Time points: t0, t1, t2,... (with a fixed interval, such as every 10 ms)

[0065] The load values of each task flow corresponding to each time point: For example, Task A sent 128 bytes at time point t0, Task B sent 512 bytes, and so on.

[0066] A preset prediction model can refer to a machine learning model or statistical model that is pre-trained or dynamically loaded in a data analysis platform or edge node for load trend prediction. Common model types include time series models: such as ARIMA, Prophet, used for modeling and predicting periodic and trend-based loads; neural network models: such as recurrent neural networks like LSTM, GRU, suitable for predicting task flow loads with long-term dependencies; temporal convolutional networks (TCN): suitable for capturing local bursts and multi-scale patterns in loads; hybrid models: integrating periodic rules and data-driven algorithms, applicable to industrial flows with partially deterministic business characteristics; lightweight edge models: such as approximate prediction methods based on weighted averages of sliding windows, applicable to resource-constrained edge devices.

[0067] The load prediction result can refer to the result output after inputting the unified time series into the preset prediction model, used to reflect the load change trends of various task flows within a certain period in the future (such as the next 10 seconds). It can include the expected bandwidth requirements of each task flow at several future time points (such as the number of bytes sent per second predicted for the next 5 seconds); the possible moments of bandwidth peaks (predicting the traffic peaks that will occur); whether there are burst loads (such as out-of-cycle control commands, alarm data); the load overlap or conflict time windows between task flows. Prediction communication cycle: refers to the expected trigger cycle of the task flow within a future period, with the unit of millisecond (ms). Predicted bandwidth requirement: represents the expected bandwidth occupancy of the task flow per unit time, with the unit of Mbps or kB / s. Predicted maximum latency: refers to the maximum end-to-end transmission delay expected after network scheduling, with the unit of ms. Predicted jitter range: Jitter is the fluctuation range of latency, reflecting the stability of scheduling. Predicted packet loss rate: For task flows with high real-time reliability (such as remote control signals), the packet loss rate must be extremely low.

[0068] In the process of industrial network operation, in order to obtain the historical load data of various task flows, firstly, in the network architecture supporting TSN protocol, by deploying edge traffic perception and analysis technology, combined with the port traffic collection capability of TSN switching equipment, the identified task flows are continuously monitored. Based on the previously identified communication triples (data source, data purpose, communication content), this process continuously collects the load-related indicators such as the number of messages, total message size, sending time, and receiving time of various tasks in each time period in the network. In order to ensure that multi-source data has a unified time reference, the industrial network uses IEEE1588 precise time synchronization technology to establish a time synchronization system with the system time base as the unified standard, and maps the load data of all task flows to the unified time axis of the whole network. The specific approach is: according to the system time base, the original data of each load collection point is reconstructed and aligned with the timestamp, and a standard time slice sequence with a fixed time granularity (for example, 10 milliseconds) is constructed, and the instantaneous bandwidth occupancy, total number of transmitted bytes or communication density of the corresponding task flow are counted in each time slice, thereby generating a structured unified time series. The time series is organized according to the task flow dimension and forms a multi-dimensional time series data structure as the next prediction input. Subsequently, the unified time series is input into the pre-trained load prediction model, which is based on the characteristics of industrial load behavior and is built in combination with technologies such as long short-term memory neural network (LSTM), gated recurrent unit (GRU) or time series regression analysis. The model uses historical bandwidth occupancy patterns, fluctuation patterns, communication cycles and other time series characteristics to output the load prediction results within the prediction period, that is, the expected bandwidth requirements, peak communication loads and load trend changes of various task flows at different time points in the future time window. The prediction results participate in the TSN digital twin simulation and scheduling optimization process as a key input to improve the adaptability of industrial networks to the dynamic changes of real-time task flows and the accuracy of resource scheduling.

[0069] The training process of the preset prediction model is:

[0070] The training process of the preset prediction model mainly relies on a large amount of historical load data collected from industrial sites as training samples, and combines the unified system time base for data alignment processing to construct a standardized unified time series data set as the input feature of the model. First, in the data collection stage, it is necessary to continuously monitor the network communication behavior of various task flows in a fine-grained manner, and extract the key load indicators of each task flow at the unit time granularity (such as every 10 milliseconds, every 100 milliseconds), including the number of messages, the total number of bytes of messages, the instantaneous bandwidth occupancy, the peak rate, the average interval time, the number of congestion occurrences, etc. All collected data must use the unified high-precision time synchronization mechanism of the industrial network (such as IEEE1588PTP) for timestamp alignment to ensure that data from different collection points are under the same time base, forming a time-continuous and ordered load time series data set.

[0071] In the sample construction phase, the system performs sliding window processing on the time series data of each communication flow according to the task flow granularity, sets a fixed window length (such as 30 time steps) as the input sequence of the model, and each time step contains multiple feature dimensions at that time point (such as bandwidth, number of packets, average delay, etc.), thus forming the input matrix of the model. The corresponding actual load result is the key load indicators within several time steps after the prediction window (such as the instantaneous bandwidth occupancy or total load in the next 5 time steps) to achieve the prediction of future load trends. Therefore, the actual load result constitutes the prediction target within this time period, which can be set as single-point prediction (predicting the next time point) or sequence prediction (predicting a sequence of future time points) to guide the supervised training process of the model.

[0072] In the model design and training phase, considering that the industrial task flow load data has significant time dependence and periodic fluctuation characteristics, a recurrent neural network structure in deep learning, especially the long short-term memory network (LSTM) or gated recurrent unit (GRU), is selected to model the input load time series. Such structures can effectively capture potential periodic and sudden changes in the historical sequence and have the ability to jointly predict short-term and medium- to long-term trends. During the training process, the mean squared error (MSE) or mean absolute error (MAE) is used as the loss function to calculate the difference between the model output and the corresponding actual load result, guiding the gradient update of network parameters to continuously improve the prediction accuracy.

[0073] To enhance the generalization ability and stability of the model, the training dataset should cover different network operating state scenarios, such as normal communication state, slight congestion state, link jitter state, and sudden high load state, and perform normalization processing on all feature data to eliminate the influence of dimension differences on the training effect. In addition, an early stopping mechanism, anti-overfitting regularization terms, and cross-validation methods can be introduced in the post-training stage to optimize the model hyperparameter combination to ensure that the model has good adaptability in actual deployment.

[0074] Finally, the trained prediction model has the ability to quickly predict the change trends of various task flow load indicators within a short future period according to the input unified time series. The prediction output is the approximate value of the future actual load result, such as the bandwidth requirement and transmission load size in the next 5 seconds. After the model is deployed, it can run in real time on edge computing nodes or industrial cloud platforms, providing accurate and forward-looking load prediction results for subsequent TSN scheduling simulation and policy decision-making. At the same time, the system can regularly introduce newly collected data for fine-tuning and retraining to continuously adapt to task flow evolution, business changes, and network state changes, maintaining high-quality prediction performance and scheduling decision support capabilities.

[0075] S105, input the system time benchmark, task parameters, preset scheduling strategy list and load forecast results into the preset TSN digital twin network model, perform simulation of multiple scheduling strategies, and obtain performance indicator data under each scheduling strategy.

[0076] The preset TSN digital twin network model can be a simulation system or digital replica, which is used to simulate the operation behavior of the real TSN industrial network, including the scheduling, forwarding, delay, packet loss and other behaviors of the communication flow. The model has the ability of "virtual network runtime environment".

[0077] The preset scheduling strategy list can be a strategy set, which lists the scheduling mode combinations that can be used for TSN network simulation and optimization in advance. Each scheduling strategy corresponds to a set of scheduling algorithms and parameter configurations, and the simulation system will simulate and execute each strategy in turn.

[0078] Scheduling strategy refers to a set of rules and algorithms used by TSN switching equipment to determine when, in what order, and how to send messages of different task flows. The core goal is to ensure indicators such as latency, bandwidth, real-time performance, and fairness.

[0079] Performance indicator data can refer to the quantitative evaluation data of the operation results of each scheduling strategy in the digital twin network model, which is used to analyze whether the network meets the industrial real-time communication requirements under the strategy.

[0080] First, the system obtains a unified system time base for the entire network through the completed time synchronization process, and uses it as a unified time reference for all simulation behaviors. Subsequently, the system inputs various task parameters extracted from historical business data into the simulation model, including the periodicity of the task, maximum delay tolerance, bandwidth requirements, jitter tolerance, and priority level. These parameters are used to define different types of task flow models and configure the generation behavior of virtual flows in the simulation. Next, the load forecast results based on historical load data and a unified timeline are input into the simulation system to generate a simulation of business traffic intensity in a specific time window in the future cycle, ensuring that the simulation can accurately reflect the impact of dynamic load changes on the performance of the scheduling strategy.

[0081] Then, the system loads the preset scheduling strategy list, which includes a variety of typical TSN scheduling methods (such as TAS, CBS, hybrid scheduling, strict priority, etc.) and their scheduling parameter combinations. The simulation system calls the scheduling configuration module in sequence according to the strategy order in the list, and loads the queue management logic, time slot configuration, gate control list (GCL) or credit calculation rules under each scheduling strategy into the simulation scheduling engine.

[0082] During the simulation running phase, the TSN digital twin network model drives the event scheduler based on the input system time reference, and according to the network topology structure and clock synchronization configuration, simulates the forwarding behavior of data packets from the source node to the target node, and records the end-to-end transmission process of each task flow under each scheduling policy. During the whole process, the simulation engine monitors multiple performance metrics in real time, and outputs the performance metric data corresponding to each policy after the simulation ends, including the maximum end-to-end delay, jitter value, packet loss rate, link utilization rate, number of time window misalignments, and priority violation rate, etc.

[0083] The training process of the preset TSN digital twin network model is as follows:

[0084] Based on the real industrial network topology structure, TSN protocol specifications and device behavior standards, a basic simulation framework is constructed, which should include virtual switching nodes, terminal nodes, link bandwidth configuration, clock synchronization logic (such as IEEE802.1AS), scheduler models (such as time-aware scheduler TAS, credit-based scheduler CBS), and queuing and forwarding mechanisms, etc. In order to enable the model to have generalization ability, when constructing the training data set, it is necessary to collect the operation data under various actual industrial network configuration scenarios, including different types of task flows (periodic, bursty), and scheduling configurations and execution results under different bandwidth and delay constraints.

[0085] In the data generation phase, using an industrial simulation platform (such as OMNeT++, NS-3 with NeSTiNg extension) or an actual deployment platform, a large number of training samples are constructed. Each sample consists of four parts: the input features include the system time reference (used to unify the scheduling trigger time point), task parameters (such as period, delay tolerance, bandwidth requirements, etc.), scheduling policy parameters (such as GCL period, time slot length, priority allocation rules), and load prediction results (the bandwidth usage trend of the task flow in the future for a period of time); the corresponding label is the real or simulated performance metric data (such as end-to-end delay, maximum jitter, packet loss rate, resource occupancy rate, etc.) generated under this policy configuration.

[0086] Subsequently, the input and label are used as training samples and fed into the TSN digital twin model for modeling training. The core of the model can select a hybrid model combining graph neural network (GNN) and attention mechanism: GNN is used to model the network topology structure and the multi-node scheduling coordination relationship, and the attention mechanism is used to dynamically focus on the local paths and task flows that have a significant impact on the key performance metrics under specific scheduling policies and traffic conditions. The model is trained in an end-to-end regression manner, and the goal is to minimize the error between the predicted performance metrics and the real metrics. A loss function such as weighted mean square error (WMSE) is used to emphasize the prediction accuracy of key metrics such as delay.

[0087] During the training process, diverse strategy configurations and multi-task load scenarios should be adopted for reinforcement, enabling the model to not only fit known strategy behaviors but also possess the generalization ability for undefined strategy combinations. In some scenarios, the model adaptability can be further extended by using simulation-enhanced datasets or transfer learning methods.

[0088] After training, the TSN digital twin model can accept the system time reference, task parameters, scheduling policy list, and load prediction results as inputs during actual deployment. Based on the learned relationship between scheduling behaviors and network characteristics internally, it can quickly execute simulation predictions under multiple strategies and output performance metric data under different strategy combinations, providing data support and optimization basis for scheduling decisions. The model also supports an incremental learning mechanism, periodically updating parameters by combining on-site feedback data to ensure high fitting degree and simulation accuracy at all times.

[0089] S106, determine the optimal scheduling strategy according to the performance metric data and the task parameters, and send the optimal scheduling strategy to the TSN switching device to complete the scheduling control configuration of various task flows in the network.

[0090] The optimal scheduling strategy can be a scheduling plan that can optimally meet the real-time, reliability, and resource utilization requirements of all task flows under the given network operating environment and task load conditions.

[0091] TSN switching devices can be Ethernet switching devices that support the TSN protocol family (IEEE802.1 series) and have deterministic forwarding capabilities. They are the key hardware infrastructure for ensuring the real-time and reliability of TSN networks.

[0092] After simulating multiple scheduling strategies, multi-dimensional performance metric data corresponding to each preset scheduling strategy can be obtained, such as: end-to-end delay, maximum jitter, bandwidth occupancy rate, scheduling success rate (i.e., whether it can be transmitted on time), number of scheduling conflicts, GCL configuration complexity, resource utilization rate, etc. for each task flow. Subsequently, the system matches and validates these performance metrics with the task parameters corresponding to various task flows (including: periodicity, maximum delay tolerance, bandwidth demand, jitter tolerance, and priority level). For example, if under a certain scheduling strategy, the delay of a periodic task flow exceeds its maximum delay tolerance or the jitter exceeds the tolerance, it is determined that this strategy does not meet the requirements and is removed from the candidate set.

[0093] Next, the system performs a comprehensive scoring process on the remaining scheduling policies. Specific methods can adopt multi-attribute decision-making models (such as TOPSIS, AHP, weighted scoring method, etc.). During this process, the system sets weight values for different performance indicators. For example, for high-real-time tasks, the weights of end-to-end delay and jitter are relatively high; for high-throughput tasks, the weight of bandwidth utilization is relatively high. The system calculates the normalized scores based on the indicator values of each scheduling policy and conducts weighted aggregation, finally generating a comprehensive score list of all scheduling policies, and selects the one with the highest score as the optimal scheduling policy.

[0094] Once the optimal scheduling policy is determined, the system structures its scheduling content, mainly including: scheduling type (such as TAS, CBS), GCL scheduling table (defining the time slot switch sequence of each port), priority mapping table (such as the mapping relationship between the PCP value of 802.1Q and tasks), port queuing rules and traffic shaping parameters (such as the token bucket parameters of CBS), and packages the above configurations into configuration instructions through the TSN control interface protocol (such as NETCONF, RESTCONF, OPCUA or a dedicated TSN SDN control protocol).

[0095] Subsequently, these instructions are sent by the system to each TSN switching device, and the device parses and loads the GCL table into the local scheduling control unit to configure the local queue priority and gating scheduling mechanism. Under the synchronous clock reference, the TSN switching device starts to execute the optimal scheduling policy to achieve precise timing forwarding control of all task flows in the network, ensuring that all critical service flows are delivered on time within the specified time window, meeting the high-reliability requirements of the industrial control network for deterministic communication.

[0096] In the embodiments of the present application, network node information of the industrial site is obtained, and based on the network node information, TSN topology deployment and configuration are performed to obtain an industrial network topology structure that supports the TSN protocol; the local system time information of each TSN network node in the industrial network topology structure is obtained, and precise synchronization processing is performed on the local system time information to obtain a unified system time reference for the entire network; historical communication data and service characteristic data of various task flows in the industrial site are obtained, and feature extraction is performed on the historical communication data and service characteristic data to obtain task parameters of various task flows; wherein, the task parameters include periodicity, maximum latency tolerance, bandwidth requirement, jitter tolerance, and priority level; historical load data of various task flows is obtained, the historical load data is time-aligned according to the system time reference to construct a unified time series, and the unified time series is input into a preset prediction model to obtain load prediction results of various task flows; the system time reference, task parameters, a preset scheduling policy list, and the load prediction results are input into a preset TSN digital twin network model, and simulation of multiple scheduling policies is performed to obtain performance index data under each scheduling policy; an optimal scheduling policy is determined according to the performance index data and the task parameters, and the optimal scheduling policy is sent to the TSN switching device to complete the scheduling control configuration of various task flows in the network. Through the above method for ensuring industrial real-time data transmission based on TSN, by integrating system time synchronization, task parameter modeling, load prediction, and TSN digital twin simulation, intelligent evaluation and automatic optimization and distribution of scheduling policies are realized, which not only significantly improves the real-time and deterministic communication capabilities of the industrial network, but also enhances the resource utilization efficiency and the adaptability of scheduling decisions, and ensures the reliable transmission of critical task flows on time.

[0097] On the basis of the above technical solution, optionally, after completing the scheduling control configuration of various task flows in the network, the method further includes:

[0098] Obtain the actual operation status data of various task flows, and compare and analyze the actual operation status data with the load prediction results;

[0099] If the deviation between the actual operation status data and the load prediction results exceeds a preset threshold, a preset feedback control mechanism is executed to dynamically update the optimal scheduling policy.

[0100] In this solution, the actual operation status data can be the operation data collected in real time from the TSN network during the actual operation of the task flow, which is used to reflect the real load situation of the current task flow communication, and can include but is not limited to real-time bandwidth usage, actual packet sending frequency and bytes, network transmission delay, link utilization rate, packet loss rate, and queue congestion information.

[0101] The preset threshold can be a judgment boundary pre-set by the system at the initial stage of deployment for judging the accuracy of load prediction and the reliability of scheduling effects, and can include a bandwidth prediction deviation threshold (such as ±10%), a latency error threshold (such as ±5 ms), a jitter change rate threshold, a lower limit of the task flow transmission success rate (such as 95%), and a congestion count threshold.

[0102] The preset feedback control mechanism can be a set of closed-loop dynamic optimization strategy systems pre-designed in an industrial TSN (Time-Sensitive Network) scheduling system to cope with significant deviations between the actual running state of the task flow and the prediction results, ensuring that the system can continuously adapt and optimize the scheduling effect during operation. When a significant deviation is detected between the actual running state data and the load prediction results, the system will trigger a response according to the preset threshold. First, it will adaptively adjust the prediction model, such as updating the prediction results by incremental learning, sliding window training, or fine-tuning the model parameters; then, it will re-execute the scheduling policy simulation in the TSN digital twin network model, re-evaluate the preset scheduling policy list based on the new prediction data, and select a new optimal scheduling policy; finally, it will send the updated optimal scheduling policy to the TSN switching device in real time, complete the scheduling configuration switch, and ensure the smooth transition of the scheduling switch through the time synchronization mechanism or the buffered scheduling window. At the same time, this mechanism also includes strategies to suppress frequent switching, such as setting a minimum adjustment interval time, a policy rollback mechanism, and policy version consistency management, to ensure that the system still operates stably and efficiently under dynamic changes.

[0103] The actual operating status data of various task flows can be obtained through a real-time acquisition mechanism deployed on TSN switching nodes and terminal devices. These data include key operating parameters such as the actual communication cycle (T_actual), average delay (D_avg_actual), maximum delay (D_max_actual), actual bandwidth occupancy (B_actual), packet loss rate (L_actual), and instantaneous jitter (J_actual) of each task flow. After these data are standardized in a form synchronized with the system time reference, they are compared item by item with the load prediction results output by the prediction model in the early stage. The load prediction results contain parameters corresponding one by one to the actual values, such as the prediction cycle (T_pred), predicted bandwidth requirement (B_pred), predicted delay (D_pred), and predicted jitter range (J_pred). The deviation judgment in the comparison and analysis process is carried out through predefined threshold values. For example, the allowed error range is set as: |T_actual - T_pred| ≤ ε_T (such as a 5% cycle error), |B_actual - B_pred| ≤ ε_B (such as a ±10% bandwidth error), |D_max_actual - D_pred| ≤ ε_D (such as 3 ms), |J_actual - J_pred| ≤ ε_J (such as 0.5 ms). Only when any deviation exceeds these threshold limits does the system trigger the feedback control mechanism. The feedback control mechanism consists of two parts: one is to update the load prediction model using an incremental learning strategy. At this time, the system introduces the actual operating data as new samples and retrains or fine-tunes the prediction model parameters through a short-term sliding window (such as the data in the recent 5 minutes). Common algorithms include time series prediction networks such as LSTM and GRU. The other is to re-enter the updated predicted values, the current system time reference, and task parameters (such as priority P_i, maximum delay tolerance D_i, bandwidth requirement B_i, and periodicity T_i) into the TSN digital twin network model, and perform a full simulation under the existing scheduling strategy list (such as cyclic scheduling, preemptive scheduling, gated scheduling, bandwidth-aware scheduling, etc.) to evaluate the performance index data of each strategy under the new load conditions, such as scheduling success rate S_i, average delay D_i_avg, packet loss rate L_i, and link utilization rate U_i. Finally, based on the task parameters (especially priority and delay requirements) and the performance index data, a weighted ranking is performed to select the new optimal scheduling strategy and send it to the TSN switching device through the centralized controller.

[0104] In this solution, by predicting parameters such as the bandwidth, cycle, and delay of future task flows, potential congestion or conflicts can be identified in advance, improving the system's forward-looking scheduling ability.

[0105] On the basis of the above technical solution, optionally, after completing the scheduling control configuration of various task flows in the network, the method further includes:

[0106] Determine the critical task flows of various task flows according to the task parameters, obtain the data packets of the critical task flows, perform frame replication and path distribution on the data packets of the critical task flows, and obtain the data packets for multi-path redundant transmission;

[0107] Based on the data packets for multi-path redundant transmission, perform comparison of data frame sequence numbers and redundant data elimination to obtain fault-tolerant data frames, and send the fault-tolerant data frames to the control system for task instruction execution.

[0108] In this solution, the critical task flow can refer to the task flow with extremely high requirements for real-time performance, reliability, and security in the industrial TSN network. Such tasks usually carry core production control instructions, such as PLC control instructions, industrial robot motion instructions, safety control commands, etc. Their characteristic parameters (such as maximum delay tolerance, jitter tolerance, priority level) are highly sensitive in the task parameters. Once transmission errors or delays occur, it may lead to system failures or safety risks.

[0109] The data packets of the critical task flow can refer to the frame-level or packet-level data units generated by the critical task flow during the communication process, carrying specific service data and control information. These data packets have specific periodicity, bandwidth requirements, and priority identifiers, and need to be reliably delivered to the target device within a specified time window.

[0110] The data packets for multi-path redundant transmission can refer to the set of copies that are frame-replicated from the data packets of the critical task flow and transmitted in parallel through multiple different link paths in the TSN network. This is the key means to implement "frame replication and elimination" in the TSN fault-tolerant mechanism, ensuring that the data transmitted by the secondary path can still be delivered on time in case of main path failures or interference.

[0111] The fault-tolerant data frame can refer to the final valid frame generated by the receiving end (such as a TSN switch or controller) after identifying the sequence numbers of the multi-path redundant data packet copies, removing duplicate frames, and reorganizing them in order. The fault-tolerant data frame is a reliable data frame that has been processed by the redundancy mechanism to remove the impact of faults and can be directly used for the execution and feedback of industrial tasks.

[0112] The control system can refer to the system unit that interfaces with the TSN network and is responsible for actual industrial equipment control and task scheduling, such as PLC (Programmable Logic Controller), DCS (Distributed Control System), motion controller, industrial server, etc. The fault-tolerant data frames are finally delivered to the control system to drive the equipment to complete precise control operations.

[0113] In an industrial TSN network, to ensure the real-time performance and reliability of critical control instructions, it is first necessary to classify and evaluate the priority of all task flows based on task parameters (including periodicity, maximum latency tolerance, bandwidth requirement, jitter tolerance, priority level). By constructing a priority discrimination model (such as the AHP method based on weighted multi-index decision-making or the threshold rule method), task flows with low latency tolerance, high jitter requirements, short data periods, and high priority levels can be identified, and these are defined as critical task flows. Then, the data packets corresponding to these critical task flows are captured in real time in the industrial communication system and handed over to the Frame Replication and Elimination for Reliability (FRER) mechanism for processing. Specifically, the critical data packets are frame-replicated at the source node or the ingress TSN switching device, and according to a predefined path distribution strategy (such as based on minimum path conflict, maximum link bandwidth redundancy, or historical reliability metrics), two or more independent transmission paths are dynamically allocated for each replicated frame and sent in parallel through different paths in the TSN network, thereby generating data packets for multi-path redundant transmission. Frame aggregation and comparison operations are performed based on the data frame sequence numbers embedded in each frame of the data packets for multi-path redundant transmission. Each received data copy is placed in a buffer indexed by the sequence number. The system monitors in real time whether there are data frames on multiple paths with the same sequence number. Once multiple copies under the same sequence number are identified, the system will initiate the redundancy elimination logic: adopting the "first arrival first" or "check result first" strategy, selecting a copy with correct content and minimum latency as the valid frame, and immediately discarding the remaining copies. At the same time, the system can choose to perform content consistency verification on these copies to further enhance the error detection ability. The finally obtained unique valid frame is marked as a fault-tolerant data frame.

[0114] These fault-tolerant data frames are then sent to an industrial control system (such as PLC, DCS, motion controller, etc.) to execute the relevant task instruction control logic, thereby ensuring the accurate, safe, and highly real-time execution of critical control instructions.

[0115] In this solution, by embedding data frame sequence numbers in the data packets for multi-path redundant transmission, the system can accurately identify and eliminate duplicate frames at the receiving end, retaining only the first valid copy, and improving the reliability and fault tolerance of critical task flows in the industrial network.

[0116] Based on the above technical solution, optionally, after completing the scheduling control configuration of various task flows in the network, the method further includes:

[0117] Obtain a collaborative control instruction, and extract the periodic scheduling requirements, time-sensitive level, and resource requirement information of various task flows according to the collaborative control instruction;

[0118] Map the cycle scheduling requirements to corresponding time windows, map the time-sensitive levels to corresponding time slot resource configurations, and map the resource demand information to network transmission path configurations;

[0119] Generate a synchronous scheduling policy based on the time window, time slot resource configuration, and network transmission path configuration, and schedule and execute various task flows according to the synchronous scheduling policy.

[0120] In this solution, the collaborative control instruction can refer to a control command issued by an upper control system (such as an industrial control master station, a dispatching control center, etc.) for coordinating the synchronous execution of multiple task flows. These instructions usually include the dependency relationships between tasks, scheduling priorities, synchronization requirements, etc., aiming to ensure that multiple tasks or devices operate in coordination according to the established time and resource strategies.

[0121] The cycle scheduling requirement can refer to the repeated execution frequency and periodicity of various task flows in time. For example, a task is triggered every 5 ms or every 10 ms. It reflects the periodic scheduling characteristics of the task and is the basis for formulating the time window.

[0122] The time-sensitive level can be an indicator to measure the tolerance of a task to time delay, transmission jitter, and scheduling accuracy, usually divided into high, medium, and low levels. For example, motion control tasks belong to the high time-sensitive level, and ordinary data acquisition tasks may be at the low level. This indicator determines the urgency and priority of scheduling.

[0123] The resource demand information can include parameters such as the network bandwidth, routing path, buffer capacity, and computing resources required by the task flow during transmission. For example, a high-definition video stream task may require a continuous bandwidth of ≥10 Mbps, while a control instruction transmission may only require a few hundred bytes.

[0124] The time window can refer to the transmission time period divided under the unified time benchmark of the whole network to meet the cycle scheduling requirements. Each periodic task is assigned to one or more exclusive time windows to ensure timely triggering and avoid conflicts. For example, the time window set for a 5-ms periodic task may be a 1-ms segment every 5 ms.

[0125] The time slot resource configuration can be the result of finely dividing and allocating the time resources within the scheduling cycle according to the time-sensitive level. Tasks with a high sensitive level will be assigned time slots that are more forward and have less jitter. For example, in a scheduling cycle, the first few microseconds may be dedicated to time-sensitive tasks.

[0126] Network transmission path configuration can select and preset the optimal transmission path for the task flow according to the resource demand information, combined with the network topology of TSN (Time-Sensitive Network), which may include the main path and backup path, bandwidth reservation, path conflict avoidance, etc.

[0127] The synchronous scheduling strategy can be a unified scheduling execution plan generated for all task flows by combining the time window, time slot configuration, and transmission path. This strategy exists in the form of a scheduling table or scheduling rules, controlling when, where, and at what bandwidth a task is transmitted, whether to enable the redundancy mechanism, etc., on a specific time and path, so as to ensure that each task is executed efficiently and orderly according to the collaboration requirements.

[0128] Collaborative control instructions can be obtained through the master control scheduling system, edge devices, or control servers. These instructions contain the scheduling and communication requirements of multiple tasks. The system uses protocol parsing and structured semantic analysis techniques to extract the periodic scheduling requirements, time sensitivity levels, and resource demand information of each type of task flow from the collaborative control instructions. Specifically, the periodic scheduling requirement refers to the triggering frequency of the task flow (such as executing once every 5 ms), which is used to describe the periodic scheduling characteristics of the task; the time sensitivity level is used to measure the tolerance of the task to network transmission delay, usually divided by priority levels (such as TSN priorities 0 - 7); the resource demand information refers to the occupation demand of the task for network resources, including the required bandwidth (such as 50 Mbps), the required transmission path (such as via switching nodes A, B, C), and the transmission reliability level. After extracting the above parameters, the system first maps the periodic scheduling requirement to the corresponding time window through a unified network clock mechanism. For example, a task with a period of 10 ms is assigned to the 0 - 1 ms time period of each scheduling cycle for execution. Subsequently, using the time sensitivity - priority mapping table, the time sensitivity level is converted into specific time slot resource configurations, that is, tasks of different levels are scheduled to different time slice resources, and a scheduling protection mechanism (such as GuardBand isolation area) is configured for tasks with high time sensitivity. At the same time, combined with the resource demand information, network topology awareness and constrained path search techniques are used to determine the network transmission path configuration that meets the task requirements, and control strategies such as path identification, traffic shaping, and bandwidth reservation are set. Based on the above time window, time slot resource configuration, and network transmission path configuration, the system automatically generates a synchronous scheduling strategy for the whole network task synchronization scheduling. This strategy contains key scheduling parameters such as the start time, duration, assigned channel, data frame transmission time slot, and transmission path of the task scheduling. Finally, this synchronous scheduling strategy is sent to the relevant switching devices and terminal nodes in the TSN network to ensure that various task flows achieve time synchronization, high-priority guarantee, and resource conflict avoidance in the multi-source heterogeneous industrial network.

[0129] In this solution, by parsing the collaborative control instructions into specific time windows, time slot resources, and network path configurations, a synchronous scheduling strategy is generated, achieving precise scheduling of task flows and resource collaboration in the industrial network.

[0130] Based on the above technical solution, optionally, after completing the scheduling control configuration for various task flows in the network, the method further includes:

[0131] Obtain the operation status log data during the operation of the industrial network and the data statistical information reported by network devices, perform multi-dimensional fusion processing on the operation status log data and the data statistical information, and obtain a fused network status evaluation data set;

[0132] Use statistical analysis methods and rule discrimination algorithms to process the network status evaluation data set and obtain the results of network operation performance indicators.

[0133] In this solution, the industrial network can refer to a communication network used in an industrial automation system, connecting devices such as sensors, actuators, controllers, PLCs, and industrial hosts to achieve data exchange and control information transmission. It supports communication protocols with high real-time performance, determinism, and reliability.

[0134] The operation status log data can refer to the behavior data automatically recorded by industrial network devices (such as switches, routers, terminal nodes, etc.) during operation, usually including timestamp information such as task flow transmission records, packet loss records, device exception logs, interface up / down times, forwarding delays, and queuing delays.

[0135] Network devices can refer to the hardware involved in data forwarding, routing, management, or terminal communication in the industrial network, including TSN switches, industrial gateways, terminal nodes (such as PLCs or embedded devices), etc. These devices report performance statistical information periodically or event-driven.

[0136] The data statistical information can be the operation index data collected and reported by network devices in real-time or periodically, including throughput, packet forwarding rate, link utilization rate, delay, packet loss rate, clock synchronization error, resource occupancy, etc., reflecting the current network performance status.

[0137] The network status evaluation data set can be a structured data set formed by fusing and organizing the operation status log data and the data statistical information in the time dimension, device dimension, task flow dimension, etc., used to reflect the comprehensive operation status of the entire industrial network currently or during a certain period.

[0138] Statistical analysis methods can include, for example, time series analysis, mean-variance calculation, correlation analysis, moving window average, anomaly detection, etc., used to quantify data trends and fluctuations.

[0139] The rule discrimination algorithm can perform conditional judgment and classification discrimination on the data set based on predefined business rules or an expert system rule base (such as "delays exceeding X ms are regarded as anomalies") to assist in identifying faults or potential risks.

[0140] The results of network operation performance indicators can be the result output after statistical analysis and rule discrimination processing, usually including network delay indicators, packet loss rate, clock synchronization accuracy, bandwidth utilization rate, task flow transmission stability level, device health score, etc., which can be used for performance evaluation, operation and maintenance decision-making, or dynamic scheduling optimization.

[0141] The operating status log data during the operation of the industrial network and the data statistical information reported by network devices can be obtained in real time through a distributed acquisition system. Among them, the operating status log data can be continuously collected through log proxy programs (such as acquisition modules based on Fluentd or Logstash) deployed at key network nodes (such as switches, controllers, and terminal devices), including the transmission timestamp of data frames, path hop counts, task flow execution results, retransmission records, exception event codes, etc.; while the data statistical information is periodically reported through the SNMP interface, Netconf interface, or proprietary API of industrial network devices, and the content covers real-time statistical metrics such as port bandwidth utilization, queue length, forwarding delay, synchronization error, link load, retransmission count, and packet loss ratio. The processing module on the edge computing node or the central server side performs multi-dimensional fusion processing on the above two types of data, specifically including: aligning timestamps based on a unified time reference (such as PTP clock), associating and matching task flow identifiers, constructing a fusion matrix in the node dimension, link dimension, and service type dimension, and using data cleaning, normalization, and feature mapping to generate a unified structured fused network status evaluation data set. This data set is centered around the task flow, associating the performance data of each node on this flow to form a complete network status map. A combined statistical analysis method and rule discrimination algorithm are used to perform intelligent processing on the fused data set. In terms of statistical analysis, techniques such as mean / variance calculation under a sliding time window, trend line fitting, clustering anomaly detection (such as DBSCAN or LOF), and periodic fluctuation modeling are used to analyze the stability and volatility of key indicators; in terms of rule discrimination, through preset network operation rule templates (such as "packet loss rate > 1% is abnormal", "synchronization error exceeding ±100ns requires a warning"), each task flow and node status are classified and judged and graded. The output network operation performance index results include task flow stability level, node health score, link bottleneck identification result, service delay evaluation value, abnormal alarm event list, etc., providing a decision-making basis for network dynamic scheduling, fault tolerance switching, and fault warning. Throughout the process, a time series database (such as InfluxDB), an index calculation engine (such as Prometheus), and a data visualization tool (such as Grafana) are relied on for real-time presentation and response.

[0142] In this solution, by obtaining the operating status log data and data statistical information and performing fusion analysis, the operating status of the industrial network can be comprehensively, real-time, and accurately grasped, and potential faults and performance bottlenecks can be discovered in a timely manner.

[0143] Figure 2 It is a schematic flowchart of a method for ensuring industrial real-time data transmission based on TSN provided by an embodiment of the present disclosure. The method may include the following steps:

[0144] S201, obtaining network node information of an industrial site, performing TSN topology deployment and configuration according to the network node information, and obtaining an industrial network topology structure supporting the TSN protocol.

[0145] S202, obtaining the local system time information of each TSN network node in the industrial network topology structure, accurately synchronizing the local system time information, and obtaining a unified system time reference for the entire network.

[0146] S203, obtaining historical communication data and service characteristic data of various task flows in the industrial site, performing feature extraction on the historical communication data and service characteristic data, and obtaining task parameters of various task flows; wherein the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level.

[0147] S204, obtaining historical load data of various task flows, time-aligning the historical load data according to the system time reference, constructing a unified time series, inputting the unified time series into a preset prediction model, and obtaining load prediction results of various task flows.

[0148] S205, input the system time benchmark, task parameters, preset scheduling strategy list and load forecast results into the preset TSN digital twin network model, perform simulation of multiple scheduling strategies, and obtain performance indicator data under each scheduling strategy.

[0149] S206, determining an optimal scheduling strategy according to the performance indicator data and the task parameters, sending the optimal scheduling strategy to the TSN switching device, and completing the scheduling control configuration of various task flows in the network.

[0150] S207, obtaining the load prediction results, scheduling resource allocation parameters and network bandwidth configuration data of each type of task flow within the preset evaluation time window, and calculating the predicted sending load of each type of task flow in each scheduling cycle within the preset evaluation time window based on the load prediction results and scheduling resource allocation parameters.

[0151] The preset evaluation time window can be the time period set by the system when performing load forecasting and bandwidth allocation evaluation. For example, it can be set to the next 5 minutes or 30 scheduling cycles. The system will evaluate the load changes, resource allocation and available bandwidth of the task flow within this time period to support more accurate scheduling decisions.

[0152] Scheduling resource allocation parameters may refer to resource scheduling ratio parameters pre-set by the system for various task flows, usually including the number of allocated time slots, priority level, maximum allowable sending load, sending cycle, etc. These parameters are output by the scheduling strategy generator or TSN digital twin model to guide the transmission behavior of task flows in the network.

[0153] The network bandwidth configuration data can be the bandwidth resource configuration of each link or communication path in the current industrial network. For example, the total bandwidth, reserved bandwidth, remaining available bandwidth, and the bandwidth ratio reserved for different priority flows of each link. It reflects the resource carrying capacity of the network within the scheduling period.

[0154] The scheduling period can be the time unit for the scheduling system to schedule the task flow. For example, scheduling is performed every 1 ms, that is, the scheduling period is 1 ms. The entire evaluation time window is divided into several scheduling periods for fine-grained resource prediction and allocation.

[0155] The predicted transmission load can refer to the amount of data (usually expressed in bytes / bit) that the task flow is expected to send calculated based on historical load data and prediction models (such as LSTM, time series regression, etc.) in each scheduling period. This is an important indicator for judging whether there is an overload risk.

[0156] The load prediction results, scheduling resource allocation parameters, and current network bandwidth configuration data of various task flows within the preset evaluation time window can be obtained from the scheduling control center or the upper-layer scheduling system. Based on the historical load data and current business trends of each task flow, the system outputs the predicted transmission load in each scheduling period within the evaluation time window through time series prediction models (such as LSTM neural network, moving window weighted average method, etc.), that is, the amount of data expected to be transmitted per period. Specifically, the historical load data of each task flow in previous scheduling periods is collected, and combined with the current business growth trend or fluctuation characteristics to form a continuous load change sequence. These data are processed using time series prediction models: if the long short-term memory neural network model is used, the system uses the historical load of multiple consecutive periods as input to train the model to identify the time dependence relationship and trend characteristics therein, and predict the load changes in subsequent scheduling periods; if the moving window weighted average method is used, based on the load values of a fixed number of historical periods, different weights are assigned to different periods and weighted averaged to obtain the predicted transmission load for each subsequent scheduling period. Finally, the system outputs the predicted load results of each type of task flow in each scheduling period within the entire evaluation time window for reference in the scheduling and resource allocation processes. Subsequently, according to the scheduling resource allocation parameters such as the period length, priority level, and bandwidth ratio allocated to each type of task flow in the scheduling resource allocation, the predicted load is mapped to the available scheduling resources to calculate the target transmission volume of the task flow in each scheduling period, ensuring that it matches the bandwidth and period resources based on the predicted value and achieving refined resource management driven by prediction.

[0157] S208. Obtain the scheduling parameters for each scheduling period, and calculate the upper limit of the allocable bandwidth load of each type of task flow in each scheduling period within the preset evaluation time window according to the network bandwidth configuration data and the scheduling parameters.

[0158] The scheduling parameters can refer to the scheduling control parameters set by the system for the task flow in each scheduling period, including available transmission time slots, maximum data volume that can be transmitted, task flow scheduling priority, path selection strategy, etc. It guides the specific scheduling execution actions in each period.

[0159] The upper limit of the allocable bandwidth load can refer to the maximum allowable bandwidth load limit calculated by the system for each task flow in each scheduling period after comprehensively considering factors such as the current link bandwidth, task flow priority, and allocated resources. Its function is to prevent link overload and ensure the transmission reliability of high-priority flows.

[0160] The execution plan of multiple scheduling cycles can be preset through the scheduling control strategy configuration mechanism in the control architecture of the industrial network. In actual operation, the scheduling configuration information in the network is collected in real time through the network control node (such as a centralized master controller or an edge scheduling node) to obtain the key scheduling parameters of each scheduling cycle. These scheduling parameters specifically include the scheduling priority of various task flows (which can be determined by the task level division rules or based on the task delay sensitivity calculation), the transmission time slice or time slot assigned to each task flow (based on the task real-time and bandwidth requirements, such as time slot rotation, the earliest deadline priority and other strategies), and the start and end time of each task flow allowed to be transmitted (based on the task release time and the latest completion time setting). After obtaining the scheduling parameters, the system enters the next stage, that is, using the network topology information and real-time link information to obtain the network bandwidth configuration data of the current network. This step is completed through the network status perception mechanism, using the following technologies: real-time link bandwidth monitoring algorithm: including end-to-end link capacity measurement based on active detection (such as sending specific detection packets and analyzing the return time), link load statistical analysis method: such as window sliding statistics of the average and peak traffic in the current time, used to estimate the available bandwidth, link congestion assessment model: combined with historical bandwidth utilization and delay trends, to determine whether the link is congested. Subsequently, the system divides the entire preset evaluation time window into multiple continuous scheduling cycles, and for each cycle, combines the scheduling parameters obtained in the previous step with the network bandwidth configuration data to calculate the allocatable bandwidth load upper limit of each task flow in the cycle. This calculation step uses the following methods: path analysis and mapping: determine the transmission path of each type of task flow (i.e., the sequence of network node links that the task passes through from the source node to the target node), link bandwidth cross-constraint analysis: traverse each link on the path, and take the minimum available bandwidth on the path as the bandwidth bottleneck, task bandwidth demand evaluation: combine the task data size and transmission cycle length in the scheduling parameters to calculate the theoretical bandwidth demand of the task, bandwidth load matching and limit correction: compare the theoretical bandwidth demand with the path bottleneck bandwidth. If the bottleneck bandwidth is insufficient, the adjusted allocable bandwidth load upper limit is calculated through a weighted allocation algorithm (such as proportional allocation based on task priority weighting). Finally, the system outputs the allocable bandwidth load upper limit for each task flow in each scheduling cycle.

[0161] In this embodiment, the load demand and network bandwidth resource matching of the task flow in each future scheduling cycle can be evaluated in advance, and the resource allocation strategy can be dynamically adjusted to avoid scheduling conflicts and task delays caused by load bursts or insufficient bandwidth, thereby improving network transmission efficiency and system operation stability.

[0162] Based on the above technical solution, optionally, after calculating the allocatable bandwidth load upper limit of each task flow in each scheduling period within the preset evaluation time window, the method further includes:

[0163] If there is a task flow whose predicted transmission load exceeds the upper limit of the allocable bandwidth load, the corresponding task flow is regarded as the target task flow;

[0164] Generate a delay warning message according to the predicted transmission load of the target task flow, the upper limit of the allocable bandwidth load, and the scheduling resource allocation parameters.

[0165] In this solution, the target task flow may refer to a task flow whose predicted transmission load exceeds the upper limit of the allocable bandwidth load in a certain scheduling period. Due to insufficient bandwidth resources, such task flows may face risks such as increased transmission delay or data congestion, so they are identified as objects that need to be monitored and processed key.

[0166] The delay warning message may be a warning result generated after analyzing the predicted transmission load of the target task flow, the upper limit of the allocable bandwidth load, and the scheduling resource allocation parameters. This information is used to prompt the system that there may be risks of insufficient scheduling resources or network congestion, and helps to adjust resources in advance or optimize the scheduling strategy to reduce the probability of task execution failure due to delay.

[0167] After detecting that the predicted transmission load of a certain task flow exceeds the upper limit of the corresponding allocable bandwidth load, the system takes it as the target task flow and further conducts a detailed analysis of the potential delay risk of this task flow. The system extracts the relevant parameters of the target task flow, including the predicted transmission load value within the current scheduling period, the corresponding upper limit value of the allocable bandwidth load, the scheduling resource allocation parameters of the task (such as task priority, task periodicity, tolerance delay threshold, etc.), and the time-sensitive level of the task itself. Then, the system performs multi-dimensional calculations according to the set evaluation model. The first step is to calculate the bandwidth gap ratio of the task, that is, dividing the difference between the predicted transmission load and the upper limit of the allocable bandwidth load by the upper limit value to quantify the shortage degree of the current bandwidth resources. The second step is to evaluate the congestion impact index of the task, which can be weighted and analyzed by comprehensively considering the traffic change trend in historical data and the competition situation of other task flows for shared resources in the current period. The third step is to combine the time-sensitive level to judge the tolerance degree of the task to scheduling delay. For example, high-level control tasks often require millisecond-level response, and scheduling lag will directly affect the system stability. The system inputs the evaluation indicators of the above dimensions into a rule judgment engine (or a rule-based scoring model) to quantitatively classify the delay risk level. For example, it can be set to three levels: "low risk", "medium risk", and "high risk", and generate a delay warning message for high-risk task flows. The warning message includes but is not limited to key contents such as task flow identifier, calculated bandwidth gap percentage, delay risk level, corresponding scheduling period number, predicted worst transmission delay range, etc., to ensure that subsequent resource scheduling systems or operation and maintenance personnel can perform dynamic response measures such as scheduling optimization, task readjustment, or resource expansion according to this information, so as to effectively avoid the problem of critical task delay caused by network congestion.

[0168] The training process of the set evaluation model is as follows: The set evaluation model is mainly trained through supervised learning, and its training data comes from the operation records of a large number of historical task flows, including the transmission load, available bandwidth, task priority, scheduling resource allocation situation, time-sensitive level, actual transmission delay results, etc. of each task flow in historical scheduling periods. During the training process, first, these multi-dimensional feature data are cleaned, normalized, and feature engineering is performed to extract key input variables; then, whether the historical task flow has a serious delay (such as exceeding the set delay threshold) is used as the target label to construct a classification model or regression model for the delay risk prediction problem. Commonly used models include decision trees, random forests, gradient boosting trees, support vector machines, or neural networks, etc. The hyperparameters are tuned through cross-validation, and finally an evaluation model that can predict the delay risk level of the task flow according to real-time input is obtained. After this model is deployed, it can run in real time within each scheduling period to predict the delay risk level of the detected target task flow.

[0169] The construction and training method of the rule judgment engine is as follows: The rule judgment engine is constructed based on the expert knowledge base and historical operation rules, and is trained by using the methods of rule extraction and weight learning. First, domain experts set a series of judgment rules in combination with network scheduling experience and business characteristics, such as "if the bandwidth gap ratio exceeds 30% and the task priority is high, it is marked as high risk"; these rules are input into the system as the initial rule set. Subsequently, the system conducts offline simulation playback on a large amount of historical data, and adjusts the triggering conditions and weights of each rule by comparing the rule output results with the actual delay situation to improve the accuracy and generalization ability. In addition, fuzzy logic and rule scoring mechanism can be combined to comprehensively score the task flow triggered by multiple rules simultaneously, further quantifying the delay risk level. Finally, the rule judgment engine can quickly judge the risk level according to the current task parameters and output structured delay warning information to support the dynamic decision-making of the scheduling system.

[0170] In this solution, by combining historical operation data and real-time scheduling parameters, and using the evaluation model or rule judgment engine to identify and warn the delay risk of the task flow in advance, it is possible to locate potential bottlenecks before the problem occurs, timely adjust the resource scheduling strategy, and avoid serious transmission delays of tasks due to insufficient bandwidth.

[0171] Based on the above technical solution, optionally, after calculating the upper limit of the allocable bandwidth load of each scheduling cycle of various task flows within the preset evaluation time window, the method further includes:

[0172] If there is no task flow with a predicted transmission load exceeding the upper limit of the allocable bandwidth load, every time the preset evaluation interval is reached, the preset evaluation time window is updated, and the load prediction results, scheduling resource allocation parameters, and network bandwidth configuration data of various task flows within the preset evaluation time window are re-obtained. According to the load prediction results and scheduling resource allocation parameters, the predicted transmission load of various task flows in each scheduling cycle within the preset evaluation time window is recalculated;

[0173] The scheduling parameters of each scheduling cycle are re-obtained, and according to the network bandwidth configuration data and scheduling parameters, the upper limit of the allocable bandwidth load of various task flows in each scheduling cycle within the preset evaluation time window is recalculated.

[0174] In this solution, the preset evaluation interval can be a fixed time period set by the system for periodic prediction and evaluation, used to trigger a new round of task flow prediction and network resource evaluation process. This evaluation interval can be configured according to the dynamics of the network and the real-time requirements of tasks. For example, it can be set to 5 seconds, 30 seconds, or 1 minute, which not only ensures the real-time response of the system but also takes into account the consumption of computing resources.

[0175] The system is internally equipped with a high-precision timer or a task scheduling mechanism. When the timer reaches a complete evaluation interval, the system enters a new prediction-evaluation process. First, the system slides forward the set time length based on the current moment to update the preset evaluation time window. For example, it shifts the start and end times of the original evaluation window forward by the length of an evaluation interval, thus forming a new time window coverage. Then, it loops again to execute obtaining the load prediction results, scheduling resource allocation parameters, and network bandwidth configuration data of various task flows within the preset evaluation time window. According to the load prediction results and scheduling resource allocation parameters, it recalculates the predicted transmission load of various task flows in each scheduling cycle within the preset evaluation time window; it obtains the scheduling parameters of each scheduling cycle again, and according to the network bandwidth configuration data and scheduling parameters, it recalculates the upper limit of the allocable bandwidth load of various task flows in each scheduling cycle within the preset evaluation time window.

[0176] In this solution, by continuously updating the evaluation time window within each preset evaluation interval and dynamically recalculating the predicted transmission load and the upper limit of the allocable bandwidth of the task flows, the system can achieve real-time monitoring of the network status and forward-looking scheduling optimization, so as to discover potential load change trends in advance.

[0177] Figure 3 It is a schematic block diagram of an industrial real-time data transmission guarantee system based on TSN provided by an embodiment of the present disclosure. It is characterized in that the system includes:

[0178] The TSN network topology deployment module 301 is used to obtain the network node information of the industrial field, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure that supports the TSN protocol;

[0179] The system time synchronization module 302 is used to obtain the local system time information of each TSN network node in the industrial network topology, perform precise synchronization processing on the local system time information, and obtain a unified system time reference for the whole network;

[0180] The task parameter extraction module 303 is used to obtain the historical communication data and service characteristic data of various task flows in the industrial field, perform feature extraction on the historical communication data and service characteristic data, and obtain the task parameters of various task flows; among them, the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level;

[0181] The load prediction module 304 is used to obtain the historical load data of various task flows, align the historical load data according to the system time reference, construct a unified time series, and input the unified time series into a preset prediction model to obtain the load prediction results of various task flows;

[0182] The TSN simulation module 305 is configured to input a system time reference, task parameters, a preset scheduling policy list, and a load prediction result into a preset TSN digital twin network model, perform simulations of multiple scheduling policies, and obtain performance metric data under each scheduling policy;

[0183] The scheduling optimization module 306 is configured to determine an optimal scheduling policy according to the performance metric data and the task parameters, and send the optimal scheduling policy to the TSN switching device to complete the scheduling control configuration of various task flows in the network.

[0184] Figure 4 FIG. shows a schematic block diagram of an electronic device 400 that can be used to implement the embodiments of the present disclosure, including a processor 401, a memory 402, a program or instruction stored on the memory 402 and executable on the processor 401. When the program or instruction is executed by the processor 401, it implements each process of the above-described embodiment of the method for ensuring industrial real-time data transmission based on TSN and can achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0185] It should be noted that the electronic device in the embodiments of the present application includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0186] It should be noted that in this document, the terms "include", "comprise" or any other variation thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or system including a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or system. Without further limitation, an element defined by the phrase "including a..." does not exclude the presence of additional identical elements in the process, method, article or system including the element. In addition, it should be pointed out that the scope of the methods and systems in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in a reverse order according to the functions involved. For example, the methods described may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0187] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described example methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation. Based on such an understanding, the technical solution of the present application, in essence or the part that contributes to the prior art, can be embodied in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions for causing a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0188] The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present application, those of ordinary skill in the art can also make many forms without departing from the purpose of the present application and the scope protected by the claims, and all of them fall within the protection scope of the present application.

[0189] The above is only the preferred embodiment of the present application and the technical principles applied. The present application is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments and substitutions that can be made by those skilled in the art will not depart from the protection scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments. Without departing from the concept of the present application, it can also include more other equivalent embodiments, and the scope of the present application is determined by the scope of the claims.

Claims

1. An industrial real-time data transmission guarantee method based on TSN, characterized in that, The method comprises: Acquire network node information of the industrial site, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure supporting the TSN protocol; Obtain the local system time information of each TSN network node in the industrial network topology, accurately synchronize the local system time information, and obtain a unified system time benchmark for the entire network; Acquire historical communication data and service characteristic data of various task flows in the industrial site, perform feature extraction on the historical communication data and service characteristic data, and obtain task parameters of various task flows; wherein the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level; Obtain historical load data of various task flows, perform time alignment on the historical load data according to the system time reference, construct a unified time series, input the unified time series into a preset prediction model, and obtain load prediction results of various task flows; Input the system time benchmark, task parameters, preset scheduling strategy list and load forecast results into the preset TSN digital twin network model, perform simulations of multiple scheduling strategies, and obtain performance indicator data under each scheduling strategy; The optimal scheduling strategy is determined according to the performance indicator data and the task parameters, and the optimal scheduling strategy is sent to the TSN switching device to complete the scheduling control configuration of various task flows in the network.

2. The method according to claim 1, wherein in, After completing the scheduling control configuration of various task flows in the network, the method further includes: Obtaining actual running status data of various task flows, and comparing and analyzing the actual running status data with load forecast results; If the deviation between the actual operating status data and the load forecast result exceeds the preset threshold, the preset feedback control mechanism is executed to dynamically update the optimal scheduling strategy.

3. The method according to claim 1, characterized in that, in, After completing the scheduling control configuration of various task flows in the network, the method further includes: Determine the key task flow of each task flow according to the task parameters, obtain the data packet of the key task flow, perform frame replication and path distribution on the data packet of the key task flow, and obtain the data packet of multi-path redundant transmission; Based on the data packets transmitted redundantly over multiple paths, comparison of data frame sequence numbers and elimination of redundant data are performed to obtain fault-tolerant data frames, which are then sent to a control system for task instruction execution.

4. The method according to claim 1, wherein in, After completing the scheduling control configuration of various task flows in the network, the method further includes: Acquire collaborative control instructions, and extract periodic scheduling requirements, time sensitivity levels, and resource demand information of various task flows according to the collaborative control instructions; Mapping the periodic scheduling requirements to corresponding time windows, mapping the time sensitivity levels to corresponding time slot resource configurations, and mapping the resource demand information to network transmission path configurations; A synchronous scheduling strategy is generated according to the time window, time slot resource configuration and network transmission path configuration, and various task flows are scheduled and executed according to the synchronous scheduling strategy.

5. The method according to claim 1, characterized in that, in, After completing the scheduling control configuration of various task flows in the network, the method further includes: Obtaining the operation status log data during the operation of the industrial network and the data statistical information reported by the network equipment, performing multi-dimensional fusion processing on the operation status log data and the data statistical information to obtain a fused network status evaluation data set; The network status evaluation data set is processed using a statistical analysis method and a rule discrimination algorithm to obtain network operation performance indicator results.

6. The method according to claim 1, wherein in, After completing the scheduling control configuration of various task flows in the network, the method further includes: Obtain load prediction results, scheduling resource allocation parameters, and network bandwidth configuration data for various task flows within a preset evaluation time window, and calculate the predicted sending load of various task flows in each scheduling cycle within the preset evaluation time window based on the load prediction results and scheduling resource allocation parameters; Obtain the scheduling parameters of each scheduling period, and calculate the allocatable bandwidth load upper limit of each scheduling period of each task flow within a preset evaluation time window according to the network bandwidth configuration data and the scheduling parameters.

7. The method according to claim 6, wherein in, After calculating the allocatable bandwidth load upper limit of each task flow in each scheduling cycle within the preset evaluation time window, the method further includes: If there is a task flow whose predicted sending load exceeds the upper limit of the allocatable bandwidth load, the corresponding task flow will be used as the target task flow; Generate delay warning information based on the predicted sending load of the target task flow, the upper limit of the allocatable bandwidth load, and the scheduling resource allocation parameters.

8. The method according to claim 6, wherein in, After calculating the allocatable bandwidth load upper limit of each task flow in each scheduling cycle within the preset evaluation time window, the method further includes: If there is no task flow whose predicted sending load exceeds the upper limit of the allocatable bandwidth load, the preset evaluation time window is updated every time the preset evaluation interval is reached, and the load prediction results, scheduling resource allocation parameters and network bandwidth configuration data of various task flows within the preset evaluation time window are re-obtained, and the predicted sending load of various task flows in each scheduling cycle within the preset evaluation time window is recalculated according to the load prediction results and scheduling resource allocation parameters; The scheduling parameters of each scheduling period are re-acquired, and the allocatable bandwidth load upper limit of each scheduling period of each type of task flow in a preset evaluation time window is recalculated according to the network bandwidth configuration data and the scheduling parameters.

9. An industrial real-time data transmission guarantee system based on TSN, for performing the method according to any one of claims 1-8, characterized in that, The system comprises: A TSN network topology deployment module is used to obtain network node information of the industrial site, perform TSN topology deployment and configuration according to the network node information, and obtain an industrial network topology structure that supports the TSN protocol; The system time synchronization module is used to obtain the local system time information of each TSN network node in the industrial network topology structure, accurately synchronize the local system time information, and obtain a unified system time benchmark for the entire network; A task parameter extraction module is used to obtain historical communication data and service characteristic data of various task flows in the industrial site, perform feature extraction on the historical communication data and service characteristic data, and obtain task parameters of various task flows; wherein the task parameters include periodicity, maximum delay tolerance, bandwidth requirement, jitter tolerance, and priority level; A load prediction module, configured to obtain historical load data of various task flows, perform time alignment on the historical load data according to the system time reference, construct a unified time series, and input the unified time series into a preset prediction model to obtain load prediction results of various task flows; A TSN simulation module, configured to input the system time reference, task parameters, a preset scheduling policy list, and the load prediction results into a preset TSN digital twin network model, perform simulation of multiple scheduling policies, and obtain performance index data under each scheduling policy; A scheduling optimization module, configured to determine an optimal scheduling policy according to the performance index data and the task parameters, and send the optimal scheduling policy to a TSN switching device to complete the scheduling control configuration of various task flows in the network.

10. An electronic device, characterized in that, It includes a processor, a memory, and a program or instruction stored on the memory and executable on the processor. When the program or instruction is executed by the processor, the steps of the method for ensuring industrial real-time data transmission based on TSN according to any one of claims 1-8 are implemented.

Images