Industrial identification flow intelligent scheduling method and system

Abstract

The application relates to the technical field of industrial network identification flow scheduling, and discloses an intelligent industrial identification flow scheduling method and system.The method comprises the following steps: S1, receiving an identification analysis request sent by an industrial network terminal, and extracting a service urgency feature of an application layer and a transmission reliability feature of a transmission layer in the identification analysis request; S2, analyzing a dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generating a priority feature label corresponding to the request based on the dynamic coupling influence relationship and a preset balancing strategy; and S3, distributing the identification analysis request to a preset high-throughput processing path or a low-latency processing path based on the priority feature label.The application can solve the problems that the prior art is difficult to balance the conflict between service timeliness and transmission stability and difficult to meet the dynamic scheduling demand of diversified services in an industrial scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial network identification traffic scheduling, in particular to an industrial identification traffic intelligent scheduling method and system. BACKGROUND

[0002] In the field of industrial network identification traffic scheduling, the existing technology mostly adopts a fixed path allocation mechanism, which only allocates identification analysis requests to processing paths according to a single dimension, and cannot dynamically adapt to the complex correlation between business urgency and transmission reliability. When high-urgency business and high-reliability requirements appear at the same time, the fixed path is prone to cause the delay of urgent tasks due to the long time consumption of reliability verification, or the transmission error of tasks with strict reliability requirements due to the occupation of fast paths, and it is difficult to balance the conflict between business timeliness and transmission stability.

[0003] At the same time, the existing scheduling method lacks real-time feedback of path state and dynamic adjustment capability of resources. High-throughput paths are prone to cause processing delay due to queue accumulation, and low-latency paths may not be able to cope with sudden high-priority requests due to fixed bandwidth resources, and the ratio adjustment of terminal-side computing resources and transmission bandwidth lags behind the change of path state, resulting in low utilization of overall network resources, which is difficult to meet the dynamic scheduling needs of diversified business in industrial scenarios, especially when a large number of devices are concurrently communicating, the above defects will significantly reduce the response efficiency and reliability of identification analysis. SUMMARY

[0004] The present application provides an industrial identification traffic intelligent scheduling method and system, which mainly aims to solve the problems that the existing technology is difficult to balance the conflict between business timeliness and transmission stability, and is difficult to meet the dynamic scheduling needs of diversified business in industrial scenarios.

[0005] To achieve the above purpose, the present application provides an industrial identification traffic intelligent scheduling method, which comprises:

[0006] S1. receiving an identification analysis request sent by an industrial network terminal, and extracting the business urgency feature of the application layer and the transmission reliability feature of the transmission layer in the identification analysis request;

[0007] S2. analyzing the dynamic coupling influence relationship between the business urgency feature and the transmission reliability feature, and generating a priority feature label corresponding to the request based on the dynamic coupling influence relationship and a preset balancing strategy;

[0008] S3. distributing the identification analysis request to a preset high-throughput processing path or a low-latency processing path based on the priority feature label;

[0009] S4. real-time monitoring of the queue accumulation state of the high-throughput processing path and the response delay state of the low-latency processing path;

[0010] S5. dynamically adjusting a computing resource ratio of a processing unit and a bandwidth ratio of a transmission unit in the industrial network terminal based on a result of the real-time monitoring;

[0011] S6. resolving and feeding back the dynamically adjusted computing resource ratio and bandwidth ratio to a requesting terminal of the industrial network terminal.

[0012] Preferably, the receiving of the identification resolution request sent by the industrial network terminal comprises:

[0013] capturing a request transmission packet in network traffic sent by the industrial network terminal;

[0014] decapsulating a protocol of the request transmission packet to obtain an application layer load of the request transmission packet;

[0015] checking a request type of the application layer load to obtain an identification resolution request of the application layer load.

[0016] Preferably, the extracting of a service urgency feature of an application layer and a transmission reliability feature of a transmission layer in the identification resolution request comprises:

[0017] resolving a load field set and a header field set of the identification resolution request;

[0018] matching a request type of the load field set to obtain a service urgency feature value of the load field set;

[0019] separating a transmission reliability feature value of the header field set;

[0020] analyzing a transmission delay of the transmission reliability feature value, and taking the transmission delay as a transmission timeliness feature value;

[0021] taking a coupling feature of the transmission reliability feature value and the transmission timeliness feature value as a transmission reliability feature of the identification resolution request.

[0022] Preferably, the analyzing of a dynamic coupling influence relationship of the service urgency feature and the transmission reliability feature comprises:

[0023] taking a deviation value of the service urgency feature and the transmission reliability feature value as a basic conflict coefficient;

[0024] mapping the basic conflict coefficient to a static coupling strength value based on a coupling threshold group in the balancing strategy;

[0025] age-weighting the static coupling strength value and a dynamic correction factor in the balancing strategy to obtain a dynamic coupling influence relationship.

[0026] Preferably, the calculation formula of the dynamic correction factor is:

[0027]

[0028] wherein: is a dynamic correction factor, is a time-sensitive factor reference value corresponding to the balancing strategy, is a service urgency coefficient, is a service urgency characteristic value, is a transmission reliability characteristic value, is a basic conflict coefficient, is a reliability offset weight factor, is an ideal reliability threshold in the balancing strategy, is a reliability scaling factor.

[0029] Preferably, the assigning the identification resolution request to a preset high-throughput processing path or low-latency processing path based on the priority characteristic label comprises:

[0030] extracting a delay sensitivity index of the identification resolution request based on the priority characteristic label;

[0031] taking the delay sensitivity index as a path decision indication, and injecting the identification resolution request into a high-throughput processing path or low-latency processing path according to the path decision indication.

[0032] Preferably, the real-time monitoring of the queue accumulation state of the high-throughput processing path and the response delay state of the low-latency processing path comprises:

[0033] obtaining a transmission occupation time sequence set of the high-throughput processing path and the low-latency processing path;

[0034] calculating a relative deviation of a preset reference time sequence set and the transmission occupation time sequence set, and synchronously separating a transmission link congestion index and an end-to-end delay drift amplitude in the relative deviation;

[0035] taking the transmission link congestion index as the queue accumulation state of the high-throughput processing path;

[0036] taking the end-to-end delay drift amplitude as the response delay state of the low-latency processing path.

[0037] Preferably, the dynamically adjusting the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal based on the result of the real-time monitoring comprises:

[0038] The queue accumulation state is subjected to accumulation depth index calculation to obtain a queue accumulation depth index;

[0039] The calculation resource ratio of the processing unit and the bandwidth ratio of the transmission unit in the industrial network terminal are dynamically adjusted based on the queue accumulation depth index and the response delay state.

[0040] Preferably, the calculation resource ratio and the bandwidth ratio after dynamic adjustment are parsed and fed back to the requesting terminal of the industrial network terminal, including

[0041] The calculation resource ratio and the bandwidth ratio are packaged into a feedback instruction package;

[0042] The feedback instruction package is subjected to network priority transmission to obtain a resource receiving confirmation signal of the industrial network terminal;

[0043] The resource receiving confirmation signal is parsed, and the parsed result is fed back to the requesting terminal.

[0044] An industrial identification flow intelligent scheduling system, the system comprising:

[0045] A parsing module is configured to receive an identification parsing request sent by an industrial network terminal, and extract a service urgency feature of an application layer and a transmission reliability feature of a transmission layer in the identification parsing request;

[0046] A priority module is configured to analyze a dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generate a priority feature label corresponding to a request based on the dynamic coupling influence relationship and a preset balancing strategy;

[0047] A distribution module is configured to distribute the identification parsing request to a preset high-throughput processing path or a low-latency processing path based on the priority feature label;

[0048] A monitoring module is configured to monitor a queue accumulation state of the high-throughput processing path and a response delay state of the low-latency processing path in real time;

[0049] An adjustment module is configured to dynamically adjust a calculation resource ratio of a processing unit and a bandwidth ratio of a transmission unit in the industrial network terminal based on a result of the real-time monitoring;

[0050] A feedback module is configured to parse and feed back the calculation resource ratio and the bandwidth ratio after dynamic adjustment to a requesting terminal of the industrial network terminal.

[0051] Compared with the prior art, the present application has the following beneficial effects:

[0052] 1. By extracting the application layer service urgency characteristics and the transmission layer transmission reliability characteristics in the identification analysis request, the priority label is generated by combining the dynamic coupling influence relationship of the two, which realizes the precision of request allocation. Based on the priority mechanism of multi-dimensional feature coupling, the resource mismatch caused by single-dimensional scheduling can be avoided, which not only ensures that high-urgency services are preferentially passed through low-delay paths for rapid response, but also ensures that high-reliability demand services are stably transmitted in high-throughput paths, effectively balancing the conflict between service timeliness and transmission stability.

[0053] 2. By real-time monitoring of the queue accumulation and response delay state of the path, the computing resource ratio of the terminal processing unit and the bandwidth ratio of the transmission unit are dynamically adjusted, and the adjustment result is fed back to the terminal, forming a closed-loop dynamic optimization mechanism, which can adapt to the path state change in real time, avoid the processing delay of high-throughput paths caused by queue accumulation, and the problem that low-delay paths cannot cope with sudden requests due to fixed bandwidth, significantly improving the utilization rate of network resources.

[0054] 3. Dynamic feedback adjustment ensures that the terminal resource allocation matches the actual scheduling demand in real time, which can effectively improve the overall response efficiency and reliability of identification analysis in large-scale device concurrent communication scenarios, and meet the dynamic scheduling needs of diversified services in industrial scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0055] Figure 1 A flowchart of an intelligent industrial identification traffic scheduling method according to an embodiment of the present application is provided.

[0056] Figure 2 A functional module diagram of an intelligent industrial identification traffic scheduling system according to an embodiment of the present application is provided. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments belong to part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0058] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0059] Depending on context, the word "if" or "if" can be interpreted to mean "when" or "while" or "in response to determining" or "in response to detecting," depending on context. Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon being determined," or "in response to determining" or "upon detecting [the stated condition or event]," or "in response to detecting [the stated condition or event]," depending on context.

[0060] The embodiment of the present application provides a kind of industrial identification flow intelligent scheduling method.The execution subject of the kind of industrial identification flow intelligent scheduling method includes but is not limited to at least one of the electronic equipment that can be configured to execute the method provided by the embodiment of the present application, such as server, terminal etc.In other words, the kind of industrial identification flow intelligent scheduling method can be executed by software or hardware installed in terminal equipment or server equipment.The server includes but is not limited to: single server, server cluster, cloud server or cloud server cluster etc.The server can be independent server, can also be cloud server that provides cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, content distribution network and basic cloud computing services such as big data and artificial intelligence platform.

[0061] As Figure 1 As shown in the figure, it is a flowchart of the industrial identification flow intelligent scheduling method of the present application, in the embodiment, the method includes:

[0062] S1. receive the identification analysis request sent by industrial network terminal, extract the service urgency feature of application layer and the transmission reliability feature of transmission layer in the identification analysis request.

[0063] In the embodiment, the identification analysis request sent by the industrial network terminal includes:

[0064] Capture the request transmission package in the network flow sent by the industrial network terminal;

[0065] Unseal the protocol of the request transmission package, obtain the application layer load of the request transmission package;

[0066] Verify the request type of the application layer load, obtain the identification analysis request of the application layer load.

[0067] Specifically, industrial network terminal refers to various devices in factory, such as numerical control machine tool, robot, sensor etc., and they are the source of sending identification analysis request.In physical environment, these terminals are distributed in different positions of workshop, and access to industrial network through wired or wireless mode.

[0068] The identification resolution request is a request for querying the corresponding information of a device or data identification in an industrial network, such as a request for querying the manufacturer, model parameters, and the like of a certain sensor.

[0069] The request transmission packet is a network data packet carrying the identification resolution request, like a package carrying the request information, which is transmitted in the transmission medium of the industrial network, such as a network cable or a wireless signal.

[0070] The application layer load is the specific data content carried by the application layer in the request transmission packet, and is the core information of the identification resolution request, which is equivalent to the contents of the package.

[0071] The service urgency feature (application layer) is a feature that reflects the service urgency of the identification resolution request. For example, when a production line suddenly breaks down, the request for querying the relevant information of the faulty device will have a very high service urgency feature value.

[0072] The transmission reliability feature (transmission layer) is a feature that reflects the reliability of the identification resolution request in the transmission process, including the coupling feature of the transmission reliability feature value and the transmission timeliness feature value. The transmission reliability feature value can be understood as the probability of no data loss or error in the transmission process.

[0073] The transmission timeliness feature value, i.e., the transmission delay, is the time interval from the issuance of the request to the start of processing. In industrial scenarios with extremely high requirements for data accuracy, such as the transmission of control instructions for precision machining equipment, the transmission reliability feature is particularly important.

[0074] In detail, capturing the request transmission packet is to accurately capture the transmission data packet containing the identification resolution request from the network traffic sent by the industrial network terminal. This action is like installing a special monitoring probe in the complex transmission pipeline of the factory to real-time capture all flowing information packets, ensuring that no identification resolution request packet is missed.

[0075] The unsealing request transmission packet protocol is a protocol unsealing operation on the captured request transmission packet. It can be understood as unpacking the outer packaging of the information packet, parsing layer by layer according to the network protocol specification, and finally obtaining the application layer load inside the transmission packet. For example, in an industrial Ethernet environment, the protocol headers of each layer of the application layer data are removed according to the TCP / IP protocol, and the core application layer data is obtained.

[0076] The application layer load request type is verified, and it is determined whether it is an identification resolution request. This is similar to checking whether the contents of the information packet are of a specific type required, and only after confirming that it is an identification resolution request, subsequent processing is performed.

[0077] In the embodiment, the extracting the service urgency feature of the application layer and the transmission reliability feature of the transmission layer in the identification resolution request comprises:

[0078] Resolving the payload field set and the header field set of the identification resolution request;

[0079] Matching the request type of the payload field set to obtain the service urgency feature value of the payload field set;

[0080] Separating the transmission reliability feature value of the header field set;

[0081] Analyzing the transmission delay of the transmission reliability feature value, and taking the transmission delay as a transmission timeliness feature value;

[0082] Taking the coupling feature of the transmission reliability feature value and the transmission timeliness feature value as the transmission reliability feature of the identification resolution request.

[0083] Specifically, the identification resolution request is a network request sent by a device in an industrial scene, such as a factory, to query the parameters, fault information, etc. corresponding to a component identifier, which is an object to be processed by traffic scheduling. For example, a numerical control machine needs to resolve a tool identifier to obtain tool life data.

[0084] The payload field set is the actual service data content carried in the request, such as the specific business information about tool use conditions and current life query requirements in the tool identifier resolution request described above. It is the core business content of the request, which is equivalent to the goods in the package.

[0085] The header field set is the information in the network protocol header, including the source address, destination address, protocol type, transmission control parameters such as the sequence number and acknowledgement number in TCP, which is responsible for guiding the data transmission path and controlling the transmission process.

[0086] In detail, when a device sends an identification resolution request in an industrial network transmission link, such as an industrial Ethernet or wireless network in a workshop, a network receiving node such as an edge gateway or an industrial router disassembles the received request packet by following the industrial applicable network protocol, such as EtherNet / IP, through a protocol resolution tool. First, identify the protocol header structure, extract the source IP, which may be the network address of a device such as 192.168.1.100 representing a machine tool in the workshop, the destination IP, the identification resolution server address, etc. header field; then further into the data packet, take out the payload content carrying business query, instruction, etc. in it, complete the resolution of the two field sets, and prepare for subsequent judgment of request priority and scheduling strategy.

[0087] Specifically, the request type is a classification of the industrial identification resolution request, such as fault diagnosis, device emergency alarm, urgent identification resolution to find fault reasons, regular parameter query, regular device operation parameter query, program update, and identification verification before pushing a new control program to the device.

[0088] The business emergency degree characteristic value is a value for measuring the request business emergency degree, such as the fault diagnosis request which may be assigned a value of 10 (urgent) and the regular query which may be assigned a value of 3 (general). The higher the value, the more urgent the processing is required, which is related to whether the production can be quickly recovered and loss can be avoided.

[0089] In detail, a request type-emergency degree mapping library is established in advance in the industrial system. After the load field set is resolved, the business appeal key information is extracted, such as the fault code X in the load which needs to be diagnosed. These information is used to match the mapping library.

[0090] If the fault diagnosis type is matched, the corresponding high emergency degree characteristic value such as 10 is assigned according to the library setting. If it is a regular parameter query, a low characteristic value such as 3 is assigned, which is convenient for subsequent arrangement of processing order.

[0091] Specifically, the transmission reliability characteristic value is an ability index for ensuring accurate and lossless data transmission. The header field includes the TCP protocol confirmation mechanism parameters, error check codes such as CRC check related fields, and can reflect the requirements of the transmission process for reliable data delivery. For example, some industrial control instructions require 100% accurate transmission, and the corresponding reliability characteristic value is high. While some non-critical state reporting has a lower reliability requirement.

[0092] In detail, the network protocol analysis module is used to filter out the fields related to transmission reliability from the header field set, such as the ACK confirmation bit of the TCP header and the reliable transmission control information implied in the window size field. These fields are extracted from the complex header information and quantified into characteristic values. The degree of requirement for transmission reliability of this package is determined, which is convenient for subsequent arrangement of transmission resources.

[0093] Specifically, the transmission delay is the time spent in network transmission from the sending end such as sensors and controllers in the factory to the receiving end, including link transmission time and device processing and forwarding time. The industrial scene is sensitive to delay, such as identification resolution of motion control instructions, which will cause uncoordinated device action if the delay is large. While the uploading of large data volume logs has a higher tolerance to delay.

[0094] The transmission timeliness characteristic value is an index for measuring the time sensitivity of transmission using the delay value. The smaller the delay, the stronger the urgency reflected by the timeliness characteristic value. The algorithm can be used for conversion, such as setting a delay threshold and deducting points in proportion if the threshold is exceeded.

[0095] In detail, with the network timestamp mechanism, the device and the server can synchronize high-precision clocks in the industrial network, record the time when the identification resolution request is sent and the time when the processing starts, calculate the difference between the two to obtain the transmission delay. Then this delay value is used as a transmission timeliness characteristic value for subsequent judgment of transmission efficiency and scheduling priority.

[0096] Specifically, the coupling feature: the two dimensions of transmission reliability and timeliness are fused into a comprehensive feature through algorithms such as weighted calculation and correlation model establishment, which can more comprehensively reflect the request for transmission quality. For example, high reliability and low latency requests, such as emergency shutdown instructions for identification resolution, have high coupling feature values and should be prioritized and accurately scheduled network resources; while low reliability requirements and slightly larger latency are acceptable, such as device regular state statistics, which have low coupling feature values and can be flexibly scheduled.

[0097] In detail, with the pre-designed coupling algorithm, the weight requirements of reliability and timeliness in the industrial scene are combined, such as production control requests with a reliability weight of 0.6 and a timeliness weight of 0.4. The transmission reliability feature value and the transmission timeliness feature value are substituted into the calculation. For example, the reliability feature value is 0.8 and the timeliness feature value is 0.9. The coupling feature value is calculated as 0.8x0.6+0.9x0.4=0.84.

[0098] This coupling feature is used as a comprehensive evaluation basis for the transmission reliability of the request to guide subsequent traffic scheduling, such as allocating low-latency and high-reliability transmission channels to requests with high coupling feature values, and allowing requests with low coupling feature values to use shared links or peak-shifting scheduling to ensure that different importance levels of identification resolution requests in industrial production can adapt to network resources and ensure smooth production processes.

[0099] S2. Analyze the dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generate a priority feature label for the corresponding request based on the dynamic coupling influence relationship combined with a pre-set balancing strategy.

[0100] In this embodiment, the analysis of the dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature comprises:

[0101] The deviation value of the service urgency feature and the transmission reliability feature value is taken as a basic conflict coefficient;

[0102] The basic conflict coefficient is mapped to a static coupling strength value based on the coupling threshold group in the balancing strategy;

[0103] The static coupling strength value and the dynamic correction factor in the balancing strategy are time-weighted to obtain a dynamic coupling influence relationship.

[0104] Specifically, the business urgency feature is a value representing the urgency of the business level of the industrial identifier resolution request, such as assigning a value of 9 to the urgency of a fault diagnosis request and a value of 3 to the urgency of a regular data query.

[0105] The transmission reliability feature value is a value reflecting the request for data accuracy and losslessness during transmission, such as assigning a value of 8 to the transmission reliability requirement of a control instruction and a value of 4 to the transmission reliability requirement of a log upload.

[0106] The deviation value is the difference between the two feature values (absolute value processing can be done), such as urgency 9 and reliability 4, deviation value |9-4|=5; urgency 3 and reliability 8, deviation value |3-8|=5.

[0107] The basic conflict coefficient is used to measure the mismatch between business demand and transmission demand, and a large deviation indicates a large difference between the two demands, which requires subsequent scheduling to be coordinated.

[0108] In detail, in the traffic scheduling control module of the industrial network, such as the scheduling program of the workshop edge server and the industrial gateway, the business urgency feature value and the transmission reliability feature value of the identifier resolution request are first read. Then, through mathematical operation, the deviation value of the two feature values is calculated, and this deviation value is directly taken as the basic conflict coefficient, which is recorded for subsequent judgment of the internal conflict degree during request scheduling.

[0109] Specifically, the balance strategy is a set of pre-configured scheduling rules in the industrial scene, including how to balance the contradiction between business urgency and transmission reliability.

[0110] The coupling threshold group is a set of threshold values that divide the basic conflict coefficient interval, which is used to classify the conflict degree and guide the subsequent mapping rules.

[0111] The static coupling strength value is a value representing the static conflict strength obtained by mapping the basic conflict coefficient to the coupling threshold group, such as a low conflict corresponding to a strength value of 2, a medium conflict corresponding to a strength value of 5, and a high conflict corresponding to a strength value of 8. It is a standardized and rule-matching value.

[0112] In detail, in the traffic scheduling system, the pre-configured balance strategy is retrieved from the strategy database of the industrial server, and the coupling threshold group is found according to the factory production demand, historical fault review, etc. Then, the basic conflict coefficient calculated in the first step is matched with the threshold interval, such as a basic conflict coefficient of 6 falling into the medium conflict interval of [4-7].

[0113] According to the mapping rules corresponding to the threshold group, the basic conflict coefficient is converted into a static coupling strength value. The pre-configured level division table (threshold group) is used to classify the difference conflict into low, medium, and high intensity (static coupling strength value), which facilitates subsequent processing by level.

[0114] Specifically, the dynamic correction factor is a coefficient dynamically adjusted according to the real-time state of the industrial scene. When the production line is busy, the correction factor of the emergency request is increased, so that the scheduling is more inclined to prioritize protection. When the network is idle, the correction factor is more moderate.

[0115] The time-weighted is combined with the time dimension, and the static coupling strength value and the dynamic correction factor are multiplied to obtain a result more suitable for the real-time scene.

[0116] The dynamic coupling influence relationship is the result of the dynamic mutual influence degree of the business urgency and the transmission reliability in the current industrial environment, which is used to accurately generate the request priority.

[0117] In detail, the scheduling system first obtains the real-time state data of the current industrial environment, and calculates the dynamic correction factor according to the data. Then, according to the time-weighted algorithm specified in the balance strategy, the static coupling strength value and the dynamic correction factor are substituted into the calculation, and the result is the dynamic coupling influence relationship, which re-adjusts the judgment of the conflict, so that the arrangement of the task is more suitable for the actual production rhythm, and the urgent and important identification analysis request is prioritized and reliably transmitted, and the production is not delayed due to network congestion and unreasonable resource allocation.

[0118] In the embodiment, the calculation formula of the dynamic correction factor is:

[0119]

[0120] Among them: is the dynamic correction factor, is the time-sensitive factor reference value corresponding to the balance strategy, is the business urgency coefficient, is the characteristic value of the business urgency, is the transmission reliability characteristic value, is the basic conflict coefficient, is the reliability offset weight factor, is the ideal reliability threshold in the balance strategy, is the reliability scaling factor.

[0121] Specifically, is the dynamic correction factor, which dynamically reflects the mutual influence of the business urgency and the transmission reliability demand in the industrial scene, and adjusts the priority of the identification analysis request scheduling. For example, a high dynamic correction factor indicates that the request needs to be scheduled and prioritized.

[0122] is the time-sensitive factor reference value corresponding to the balance strategy, which reflects the overall requirement of the industrial scene for time sensitivity. The faster the production rhythm and the more real-time the factory, the higher the value.

[0123] The business urgency coefficient is used to amplify / reduce the impact of business urgency on dynamic correction, highlighting the weight of business urgency in scheduling.

[0124] The aforementioned service urgency feature value is used to quantify the urgency of the identifier resolution request. The higher the value, the more urgent the request. Identifier resolution requests initiated when a device experiences a sudden shutdown failure are more urgent. The value is set to 9, which periodically sends requests to upload device status. Set the value to 3.

[0125] The transmission reliability characteristic value is used to quantify the degree of reliability required by the request for data accuracy, no packet loss, and low latency during the transmission process. The higher the value, the more stringent the requirements.

[0126] The basic conflict coefficient reflects the conflict between the urgency of the service and the requirements for transmission reliability. The larger the difference, the higher the conflict coefficient. The larger the value, the more difficult it is to match the demands of the two parties.

[0127] This is a reliability offset weighting factor used to adjust the weight of the impact of transmission reliability requirements on dynamic correction, emphasizing the assurance of transmission reliability. Let's assume it's large.

[0128] The ideal reliability threshold in the balancing strategy is the preset ideal value for transmission reliability in the balancing strategy. If the transmission reliability is close to this value, it means that the transmission reliability meets the factory requirements.

[0129] This is a reliability scaling factor used to scale the transmission reliability characteristic value. ,let and The comparison is more reasonable and better suited to the factory's precise requirements for reliability.

[0130] Furthermore, focus on Higher than In industrial settings, situations like sudden equipment malfunction alarms (…) (High), but the transmission link was originally intended for regular data ( (For applications with low demand), where the urgency of the business outweighs transmission reliability, a conflict arises where the business forces a transmission upgrade; if ≤ For example, in routine status queries, high transmission reliability is required. If the result is 0, this part of the impact is zero and will not trigger a business-driven correction.

[0131] , the conflict is amplified by the service pressure over the transmission. The larger the conflict is amplified, the more prominent the contradiction between the service urgency and the transmission status is.

[0132] More specifically, The conflict impact is compressed by an exponential function. When the conflict is very small, close to 1, the 1-result is close to 0, indicating that the service demand for transmission is weak; when the conflict is very large, close to 0, the 1-result is close to 1, the service demand for transmission upgrade is the largest, and it is strongly required to improve the transmission guarantee.

[0133] Further, The service urgency coefficient weights the part of the impact. In the factory, if you want to highlight the scheduling priority of the fault class high-urgency service, set large, so that the conflict impact dominated by this type of service is more prominent, and the scheduling system is promoted to respond preferentially.

[0134] In detail, The transmission reliability characteristic value is adjusted by a reliability scaling factor . That is, the actual value is mapped to a fine interval that the factory is concerned about, and the comparison with the ideal threshold is more in line with the actual demand.

[0135] S3. Distribute the identity resolution request to a preset high-throughput processing path or low-latency processing path based on the priority feature label.

[0136] In this embodiment, the distributing the identity resolution request to a preset high-throughput processing path or low-latency processing path based on the priority feature label comprises:

[0137] Extracting a delay sensitivity index of the identity resolution request based on the priority feature label;

[0138] Taking the delay sensitivity index as a path decision indication, and injecting the identity resolution request into a high-throughput processing path or a low-latency processing path according to the path decision indication.

[0139] Specifically, the delay sensitivity index quantifies a value of how sensitive the identity resolution request is to transmission delay.

[0140] ​In detail, in the traffic scheduling system of the industrial network, a priority characteristic label of an identification resolution request is first invoked. Then, according to a preset label-delay sensitivity mapping rule, a delay sensitivity index of the request is extracted, which is set by a factory operation and maintenance personnel according to production requirements, such as that a high emergency and high reliability label corresponds to a delay sensitivity index of 0.7-1, a normal label corresponds to a delay sensitivity index of 0.3-0.6, and the like.

[0141] Specifically, the path decision indication is to use the delay sensitivity index as a scheduling instruction to tell the system whether the request is suitable for a high-throughput path or a low-latency path.

[0142] The high-throughput processing path is a transmission channel in the industrial network that is specially optimized for large data transmission and can carry high load, such as a link connecting a large data storage server in a factory, which has high bandwidth and can transmit a large amount of identification resolution request data at a time, but the delay may be slightly high.

[0143] The low-latency processing path is a transmission channel that is optimized for low transmission delay and fast response. The delay is extremely low, but the bandwidth may not be so large, and it is suitable for fast transmission of a small amount of critical requests.

[0144] In detail, after the scheduling system obtains the delay sensitivity index, it uses the index as a command signal. If the index is high, it is determined to be a delay-sensitive request, and the scheduling system triggers the low-latency path selection logic to inject the request into the low-latency processing path.

[0145] In the physical network, the request is made to go through the pre-planned low-latency link through the configured network strategy.

[0146] If the index is low, it is determined to be a throughput-sensitive request, and the scheduling system triggers the high-throughput path selection logic to inject the request into the high-throughput processing path.

[0147] Physically, it guides the request to go through a link with sufficient bandwidth and can transmit a large amount of data in parallel, so that the request can be efficiently transmitted with other similar requests, and the high-bandwidth advantage is used to quickly complete data interaction.

[0148] S4. Real-time monitoring of the queue accumulation state of the high-throughput processing path and the response delay state of the low-latency processing path.

[0149] In the embodiment, the real-time monitoring of the queue accumulation state of the high-throughput processing path and the response delay state of the low-latency processing path comprises:

[0150] Obtaining a transmission occupation time sequence set of the high-throughput processing path and the low-latency processing path;

[0151] calculating a relative deviation of a preset reference timing set and the transmission occupancy timing set, and separating out a transmission link congestion index and an end-to-end delay drift amplitude in the relative deviation;

[0152] taking the transmission link congestion index as a queue accumulation state of the high-throughput processing path;

[0153] taking the end-to-end delay drift amplitude as a response delay state of the low-latency processing path.

[0154] Specifically, the reference timing set is a path ideal state timing data preset by the factory. For example, the bandwidth occupancy rate of the high-throughput path should be stabilized at 30% to 50% under normal load, and the corresponding timing set is [t1: 40%, t2: 35%…]; the ideal forwarding delay of the low-latency path should be ≤1 millisecond, and the timing set records the stable value of the ideal delay value over time.

[0155] The relative deviation is the difference between the transmission occupancy timing set and the reference timing set, which reflects the deviation degree of the actual state and the ideal state of the path. For example, the actual bandwidth occupancy of the high-throughput path is 60%, and the reference is 40%, with a deviation of +20%.

[0156] The transmission link congestion index is a special index from the relative deviation, which reflects whether the high-throughput path is congested. The part of the deviation with high bandwidth occupancy and serious data packet queue accumulation is quantified into the congestion index, and the higher the value represents the more congested.

[0157] The end-to-end delay drift amplitude is a special index from the relative deviation, which reflects the delay fluctuation of the low-latency path. The part of the deviation with unstable and high-low forwarding delay is quantified into the drift amplitude, and the larger the value represents the more unstable the delay.

[0158] In detail, in the relative deviation calculation scheduling system, the data of the transmission occupancy timing set and the reference timing set at the corresponding time points are subtracted to obtain the deviation at each time point, and then the deviations are arranged in time sequence to form a deviation timing set. For example, the actual bandwidth of the high-throughput path at t1 is 60%, the reference is 40%, the deviation is +20%; the actual bandwidth at t2 is 55%, the reference is 35%, and the deviation is +20%.

[0159] Separating the congestion index and the drift amplitude is to use an algorithm, such as filtering and feature extraction, to separate the two problems from the deviation timing set:

[0160] For the high-throughput path, the deviation part with bandwidth continuously exceeding the reference and queue length increasing is selected, and the transmission link congestion index is calculated through integration, weighting, etc. It can be understood as the period when the traffic load exceeds the load and the queue length becomes longer.

[0161] For the low latency path, the deviation part with large delay fluctuation and out of the stable range is screened out, and is also quantified into the end-to-end delay drift amplitude by algorithm. It can be understood that the delivery time is fast and slow, and exceeds the normal fluctuation.

[0162] Specifically, the queue accumulation state is the congestion degree of data packet queuing waiting for transmission in the high throughput processing path, which is quantified by the transmission link congestion index. The higher the index, the more serious the queue accumulation.

[0163] The response delay state is the delay fluctuation degree of data packet from initiation to reception in the low latency processing path, which is quantified by the end-to-end delay drift amplitude. The larger the amplitude, the more unstable the response delay.

[0164] In detail, the scheduling system directly assigns the transmission link congestion index to the queue accumulation state of the high throughput processing path, and assigns the end-to-end delay drift amplitude to the response delay state of the low latency processing path. Then, the state data is stored in the path state database for subsequent scheduling decision.

[0165] S5. Based on the results of real-time monitoring, dynamically adjust the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal.

[0166] In this embodiment, the dynamic adjustment of the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal based on the results of real-time monitoring comprises:

[0167] The queue accumulation state is calculated by the accumulation depth index to obtain the queue accumulation depth index;

[0168] Based on the queue accumulation depth index and the response delay state, the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal are dynamically adjusted.

[0169] Specifically, the queue accumulation state is obtained in the previous step, and the state of data packet queuing congestion in the high throughput processing path is embodied by the transmission link congestion index and the like.

[0170] The accumulation depth index is a congestion severity value obtained by further quantifying the queue accumulation state. The higher the value, the deeper the queue accumulation, and the busier the processing unit and the transmission unit.

[0171] In detail, in the resource scheduling module of the industrial network terminal, the queue accumulation state data is retrieved. Then, by using a preset algorithm such as integral algorithm and exponential weighting algorithm, the factors such as the duration of queue accumulation and the congestion degree are combined to calculate the accumulation depth index.

[0172] For example, if the high-throughput path has congestion index ≥ 0.6 for 5 consecutive seconds, the algorithm assigns a high value (e.g., 0.8) to the queue depth index; if it is occasionally congested, the index is low (e.g., 0.3).

[0173] Specifically, the computing resource ratio is the proportion of resources allocated by the processing unit (such as CPU, GPU) in the industrial network terminal to different services. For example, 60% of CPU resources are allocated to high emergency requests, and 40% of CPU resources are allocated to regular requests.

[0174] The bandwidth ratio is the proportion of bandwidth allocated by the transmission unit, such as the network card and wireless module, in the industrial network terminal to different paths.

[0175] In detail, the decision logic generation is the resource scheduling module of the terminal, which takes the queue depth index and response delay state as input, runs the pre-set resource allocation strategy, and the factory operation and maintenance personnel sets according to the production demand, such as when the queue depth index is high and the response delay is large, the low latency path is prioritized.

[0176] Adjusting the computing resource is that if the queue is deep, it means that there are a large number of requests accumulated in the high-throughput path, and the processing unit needs more resources to process these requests, so the scheduling module increases the computing resource ratio of the processing unit to the high-throughput path request; if the response delay is large, it means that the request of the low latency path is delay sensitive, so the resource ratio of the processing unit to the low latency path request is increased.

[0177] Adjusting the transmission bandwidth is that when the queue is deep, the transmission unit allocates more bandwidth to the high-throughput path to make the accumulated requests faster; when the response delay is large, more bandwidth is allocated to the low latency path to reduce the delay fluctuation.

[0178] S6. Analyze and feedback the dynamically adjusted computing resource ratio and bandwidth ratio to the request terminal of the industrial network terminal.

[0179] In this embodiment, the analysis and feedback of the dynamically adjusted computing resource ratio and bandwidth ratio to the request terminal of the industrial network terminal comprises

[0180] The computing resource ratio and the bandwidth ratio are packaged into a feedback instruction package;

[0181] The feedback instruction package is transmitted with network priority to obtain a resource reception confirmation signal of the industrial network terminal;

[0182] Analyze the resource reception confirmation signal and feed back the analysis result to the request terminal.

[0183] Specifically, the computing resource ratio is the proportion of resources allocated by the processing unit (such as CPU, GPU) in the industrial network terminal to different services. For example, 60% of CPU resources are allocated to high emergency requests, and 40% of CPU resources are allocated to regular requests.

[0184] After the bandwidth allocation is adjusted, the transmission unit allocates a certain proportion of bandwidth to different paths, such as allocating 40% bandwidth to low-latency paths.

[0185] Feedback instruction packets are network data packets that package computing resource allocation and bandwidth allocation. They are like instruction envelopes, containing information that tells the requesting terminal how to adapt to the new resources.

[0186] In detail, in the feedback module of the industrial network terminal, the computing resource allocation and bandwidth allocation data are first converted into a format that the network can transmit. Then, according to the preset instruction packet structure and the industrial protocol format defined by the factory, such as a data packet structure conforming to the EtherNet / IP protocol, these two types of allocation data are loaded to form a feedback instruction packet.

[0187] Specifically, network priority transmission marks feedback instruction packets with high priority, allowing the industrial network to transmit them first, avoiding being overtaken by other low-priority data, and ensuring timely delivery of instructions.

[0188] In detail, the priority labeling is a transmission module of the industrial network terminal that assigns a high-priority label to the feedback instruction packet.

[0189] The receiving confirmation is a request for the terminal to immediately generate a resource receiving confirmation signal after receiving the feedback instruction packet, and send it back to the industrial network terminal according to a preset protocol format, such as including the receiving time, terminal ID, and confirmation code.

[0190] Specifically, parsing involves converting data such as the receipt time and acknowledgment code in the resource reception confirmation signal into information that humans / systems can understand, such as terminal A receiving the instruction at 10:00:01 with a correct acknowledgment code.

[0191] The parsing result feedback involves sending the parsed simple confirmation information back to the requesting terminal, forming a secondary confirmation.

[0192] In detail, the parsing confirmation signal is the parsing module of the industrial network terminal. After receiving the resource reception confirmation signal, it splits the data according to the protocol format and extracts key information.

[0193] The feedback result is generated based on the parsed information, resulting in a simple feedback result.

[0194] The feedback to the requesting terminal is to repackage the feedback result into a small data packet and send it back to the requesting terminal.

[0195] like Figure 2 The diagram shown is a functional block diagram of an industrial identification traffic intelligent scheduling system provided in an embodiment of the present invention.

[0196] The industrial identification flow intelligent scheduling system 100 can be installed in an electronic device. According to the functions implemented, the industrial identification flow intelligent scheduling system 100 can include an analysis module 101, a priority module 102, an allocation module 103, a monitoring module 104, an adjustment module 105, and a feedback module 106. The modules in the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, and are stored in the memory of the electronic device.

[0197] In the present embodiment, the functions of each module / unit are as follows:

[0198] The analysis module 101 is used to receive an identification analysis request sent by an industrial network terminal, and extract the service urgency feature of the application layer and the transmission reliability feature of the transmission layer in the identification analysis request;

[0199] The priority module 102 is used to analyze the dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generate a priority feature label corresponding to the request based on the dynamic coupling influence relationship and a preset balancing strategy;

[0200] The allocation module 103 is used to allocate the identification analysis request to a preset high-throughput processing path or a low-latency processing path based on the priority feature label;

[0201] The monitoring module 104 is used to monitor the queue accumulation state of the high-throughput processing path and the response delay state of the low-latency processing path in real time;

[0202] The adjustment module 105 is used to dynamically adjust the computing resource ratio of the processing unit and the bandwidth ratio of the transmission unit in the industrial network terminal based on the results of the real-time monitoring;

[0203] The feedback module 106 is used to analyze and feed back the dynamically adjusted computing resource ratio and bandwidth ratio to the requesting terminal of the industrial network terminal.

[0204] In several embodiments provided in the present application, it should be understood that the disclosed method and system can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there can be another division way when actually implemented.

[0205] The modules described as separate components may or may not be physically separate, and the components displayed as modules may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0206] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0207] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0208] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence is to use digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, obtain knowledge and use knowledge to obtain the best results.

[0209] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. An intelligent scheduling method for industrial traffic flow identification, characterized in that, The method includes: S1. Receive the identifier resolution request sent by the industrial network terminal, and extract the service urgency feature of the application layer and the transmission reliability feature of the transmission layer from the identifier resolution request; S2. Analyze the dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generate priority feature labels for corresponding requests based on the dynamic coupling influence relationship and a preset balancing strategy, including: using the deviation value between the service urgency feature and the transmission reliability feature as the basic conflict coefficient; The basic conflict coefficients are mapped to static coupling strength values ​​based on the coupling threshold group in the balancing strategy. By applying a time-dependent weighted average to the static coupling strength value and the dynamic correction factor in the balancing strategy, the dynamic coupling influence relationship is obtained. The formula for calculating the dynamic correction factor includes: ； in: As a dynamic correction factor, The benchmark value of the time-sensitivity factor corresponding to the aforementioned balancing strategy. This is the business urgency coefficient. The urgency feature value of the service. The transmission reliability characteristic value is... Basic conflict coefficient, As a reliability offset weighting factor, This represents the ideal reliability threshold in the balancing strategy. This is a reliability scaling factor; S3. Based on the priority feature tag, allocate the identifier resolution request to a preset high-throughput processing path or low-latency processing path; S4. Monitor the queue backlog status of the high-throughput processing path and the response delay status of the low-latency processing path in real time; S5. Based on the results of the real-time monitoring, dynamically adjust the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal; S6. The dynamically adjusted computing resource allocation and bandwidth allocation are parsed and fed back to the requesting terminal of the industrial network terminal.

2. The industrial identification traffic intelligent scheduling method as described in claim 1, characterized in that, The receipt of the identifier resolution request sent by the industrial network terminal includes: Capture request packets in the network traffic sent by the industrial network terminal; Deseal the protocol of the request transport packet to obtain the application layer payload of the request transport packet; Verify the request type of the application layer payload to obtain the identifier resolution request of the application layer payload.

3. The industrial identification traffic intelligent scheduling method as described in claim 1, characterized in that, The extraction of the application layer's service urgency feature and the transport layer's transmission reliability feature from the identifier resolution request includes: Parse the payload field set and header field set of the identifier parsing request; Match the request type of the payload field set to obtain the business urgency feature value of the payload field set; Separate the transmission reliability feature value of the header field set; Analyze the transmission delay of the transmission reliability characteristic value, and use the transmission delay as the transmission timeliness characteristic value; The coupling feature of the transmission reliability feature value and the transmission timeliness feature value is used as the transmission reliability feature of the identifier resolution request.

4. The industrial identification traffic intelligent scheduling method as described in claim 1, characterized in that, The step of allocating the identifier resolution request to a preset high-throughput processing path or low-latency processing path based on the priority feature tag includes: The latency sensitivity index of the identifier parsing request is extracted based on the priority feature label; The latency sensitivity index is used as a path decision indicator, and the identifier resolution request is injected into a high-throughput processing path or a low-latency processing path according to the path decision indicator.

5. The industrial identification traffic intelligent scheduling method as described in claim 1, characterized in that, The real-time monitoring of the queue backlog status of the high-throughput processing path and the response latency status of the low-latency processing path includes: Obtain the transmission occupancy time set of the high-throughput processing path and the low-latency processing path; Calculate the relative deviation between the preset reference timing set and the transmission occupancy timing set, and simultaneously separate the transmission link congestion index and end-to-end delay drift amplitude in the relative deviation. The transmission link congestion index is used as the queue backlog state of the high-throughput processing path; The end-to-end delay drift magnitude is used as the response delay state of the low-latency processing path.

6. The industrial identification traffic intelligent scheduling method as described in claim 1, characterized in that, The dynamic adjustment of the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal based on the results of the real-time monitoring includes: The queue stacking state is calculated using a stacking depth index to obtain the queue stacking depth index. The computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal are dynamically adjusted based on the queue backlog depth index and the response latency status.

7. The industrial identification traffic intelligent scheduling method as described in claim 6, characterized in that, The requesting terminal that parses and feeds back the dynamically adjusted computing resource allocation and bandwidth allocation to the industrial network terminal includes... The computing resource allocation and the bandwidth allocation are encapsulated into a feedback instruction package; The feedback instruction packet is transmitted with network priority to obtain the resource reception confirmation signal of the industrial network terminal; The resource receipt confirmation signal is parsed, and the parsing result is fed back to the requesting terminal.

8. An intelligent scheduling system for industrial traffic flow identification, characterized in that, The system includes: The parsing module is used to receive identifier resolution requests sent by industrial network terminals and extract the application layer service urgency characteristics and the transport layer transmission reliability characteristics from the identifier resolution requests. Priority module: used to analyze the dynamic coupling influence relationship between the service urgency feature and the transmission reliability feature, and generate priority feature labels for corresponding requests based on the dynamic coupling influence relationship and a preset balancing strategy, including: using the deviation value between the service urgency feature and the transmission reliability feature value as the basic conflict coefficient; The basic conflict coefficients are mapped to static coupling strength values ​​based on the coupling threshold group in the balancing strategy. By applying a time-dependent weighted average to the static coupling strength value and the dynamic correction factor in the balancing strategy, the dynamic coupling influence relationship is obtained. The formula for calculating the dynamic correction factor includes: ； in: As a dynamic correction factor, The benchmark value of the time-sensitivity factor corresponding to the aforementioned balancing strategy. This is the business urgency coefficient. The urgency feature value of the service. The transmission reliability characteristic value is... Basic conflict coefficient, As a reliability offset weighting factor, This represents the ideal reliability threshold in the balancing strategy. This is a reliability scaling factor; Allocation module: used to allocate the identifier resolution request to a preset high-throughput processing path or low-latency processing path based on the priority feature label; Monitoring module: used to monitor the queue backlog status of the high-throughput processing path and the response latency status of the low-latency processing path in real time; Adjustment module: used to dynamically adjust the computing resource allocation of the processing unit and the bandwidth allocation of the transmission unit in the industrial network terminal based on the results of the real-time monitoring; Feedback module: Used to parse the dynamically adjusted computing resource allocation and bandwidth allocation and feed them back to the request terminal of the industrial network terminal.

Images