Data acquisition method and device based on industrial internet identification analysis system

Abstract

The invention provides a data acquisition method and device based on an industrial internet identification analysis system, which are applied to the technical field of industrial internet, and are used for realizing data grading processing by acquiring data and priority parameters and distributing transmission queues according to the priority parameters; the method comprises the following steps: compressing and coding data, evaluating the quality of a wireless channel, selecting a transmission path, avoiding an interference channel, determining a retransmission strategy by combining a data priority and a network congestion state when transmission fails, receiving a data packet at an edge computing node according to the retransmission strategy, and uploading the data packet after local processing. The pressure of a core system is dispersed by utilizing edge computing power, and the overall data processing time delay can be reduced. Therefore, according to the method, through the steps of data priority distinguishing, data compression, channel quality evaluation, intelligent retransmission strategy, edge computing application and the like, the beneficial effect of anti-interference and low-delay data transmission in a wireless communication factory environment is achieved.

Description

Technical Field

[0001] This application relates to the technical field of industrial Internet, and particularly to a data collection method and device based on the industrial Internet identification and resolution system. Background Art

[0002] In large modern factories, a large number of sensors and industrial devices are deployed on the production line. These devices continuously generate a huge amount of data for real-time monitoring of the production status and optimization of the production process. To facilitate deployment and reduce wiring costs, wireless communication technologies are widely used in factories for data transmission. However, the electromagnetic interference in the industrial environment is complex, and the wireless network is vulnerable to interference. Coupled with the large number of devices, network congestion occurs frequently, seriously affecting the data transmission efficiency and real-time performance. For production links that require millisecond-level response, any delay in data collection may lead to a lag in production decisions and even cause production accidents.

[0003] Therefore, in a large factory environment that relies on wireless communication, in order to ensure real-time data collection required for production process optimization, how to ensure the high efficiency and real-time performance of industrial Internet identification and resolution system data collection in an industrial scenario with a complex and congested wireless network environment has become an urgent problem to be solved. Summary of the Invention

[0004] In view of the deficiencies of the above-mentioned prior art, the data collection method and device based on the industrial Internet identification and resolution system provided by this application are applied in the technical field of industrial Internet, and have the advantages of improving the data collection efficiency and real-time performance of the industrial Internet identification and resolution system in a complex wireless network environment.

[0005] In a first aspect, a data collection method based on the industrial Internet identification and resolution system is used to achieve anti-interference and low-latency data transmission in a factory environment using wireless communication. The method includes: S1: Obtain the data of industrial device sensors and the corresponding data priority parameters; S2: Allocate the data to transmission queues with different priorities according to the data priority parameters; S3: Perform compression encoding processing on the data in the transmission queue to generate a reduced data packet; S4: Evaluate the quality of the wireless channel and select the transmission path of the reduced data packet according to the wireless channel quality evaluation result; S5: When the transmission fails, determine a retransmission strategy based on the data priority parameters and the current network congestion status; S6: According to the retransmission strategy, receive the reduced data packet at the edge computing node, perform data parsing and feature extraction operations, and then upload it to the core parsing system.

[0006] A data acquisition method based on the industrial Internet identification and resolution system provided by this application lays a foundation for subsequent data processing by obtaining data and priority parameters, and differentiates the importance of data; realizes hierarchical data processing by allocating transmission queues according to priority parameters; reduces the size of data packets by compressing and encoding data, which is beneficial to reducing transmission delay and network load; ensures data transmission quality by evaluating the quality of the wireless channel and selecting a transmission path to avoid interference channels; determines a retransmission strategy by combining data priority and network congestion status when transmission fails, ensures that important data is retransmitted first, and considers the network condition to avoid exacerbating congestion; receives data packets at the edge computing node according to the retransmission strategy, performs local processing and then uploads them, uses the edge computing ability to disperse the pressure on the core system, and may reduce the overall data processing delay. Therefore, this method realizes the beneficial effect of anti-interference and low-delay data transmission in the wireless communication factory environment through steps such as data priority differentiation, data compression, channel quality evaluation, intelligent retransmission strategy, and edge computing application.

[0007] Further, step S5 includes: S51: Monitor the device operation status and production progress information of the production line in real time, and generate production status parameters; S52: Adjust the priority parameters of the data to be retransmitted according to the production status parameters; S53: Obtain the congestion degree parameter of the current wireless network; S54: Determine a retransmission strategy based on the adjusted data priority parameters and the congestion degree parameter; The retransmission strategy includes: when the data priority is higher than or equal to the first preset threshold and the network congestion degree is lower than the second preset threshold, perform retransmission immediately; when the data priority is lower than the first preset threshold or the network congestion degree is higher than the second preset threshold, delay retransmission.

[0008] A data collection method based on the industrial Internet identification and resolution system provided by the present application obtains production status parameters by monitoring the device operation status and production progress information of the production line in real time, so as to master the real-time dynamics of the production line; uses the obtained production status parameters to dynamically adjust the priority parameters of the data to be retransmitted, so that the data priority can reflect the actual needs of production; obtains the congestion degree parameter of the current wireless network to provide network environment information for formulating the retransmission strategy; comprehensively considers the adjusted data priority parameter and the network congestion degree parameter to determine the retransmission strategy. The specific retransmission strategy is set as follows: when the data priority is high and the network is not congested, retransmit immediately; when the data priority is low or the network is congested, delay retransmission. This strategy enables the retransmission mechanism to adaptively adjust according to the production status and network conditions, giving priority to ensuring the real-time transmission of important data and taking into account the reasonable utilization of network resources, thereby improving the efficiency and real-time performance of data collection in the industrial Internet identification and resolution system.

[0009] Further, step S52 includes: S521: Monitor the first change trend of the production status parameter and the second change trend of the current network congestion degree parameter; S522: Adjust the priority parameter of the reduced data packet to be retransmitted according to the first change trend and the second change trend; S523: Use the adjusted priority parameter to determine the retransmission strategy.

[0010] A data collection method based on the industrial Internet identification and resolution system provided by the present application can dynamically sense the changes in the industrial environment and network environment by monitoring the first change trend of the production status parameter and the second change trend of the network congestion degree parameter. Adjust the priority parameter of the data packet to be retransmitted according to the first change trend and the second change, so that the priority adjustment is no longer static or only based on the current state, but can predict future changes, thereby achieving more refined priority adjustment. Apply the adjusted priority parameter to determine the retransmission strategy to ensure that the retransmission strategy can fully consider the environmental change trend, and then improve the data collection efficiency and real-time performance in the dynamic industrial wireless network environment.

[0011] Further, step S522 includes: S5221: Calculate the Pearson correlation coefficient ρ(t) between the first change trend ΔS(t) and the second change trend ΔC(t), where the calculation formula of ρ(t) is: ρ(t) = Cov(ΔS(t), ΔC(t)) / (σ(ΔS(t)) * σ(ΔC(t))), where Cov(ΔS(t), ΔC(t)) is the covariance of ΔS(t) and ΔC(t), σ(ΔS(t)) is the standard deviation of ΔS(t), and σ(ΔC(t)) is the standard deviation of ΔC(t); S5222: Determine the adjustment factor M(ρ(t)) of the priority parameter according to the Pearson correlation coefficient ρ(t); S5223: Calculate the adjustment amplitude ΔP(t) of the priority parameter according to the adjustment factor M(ρ(t)) of the priority parameter, where the calculation formula of ΔP(t) is: ΔP(t) = α_base * M(ρ(t)), where α_base is the basic adjustment amplitude of the priority parameter; S5224: Use the adjustment amplitude ΔP(t) of the priority parameter to adjust the priority parameter of the reduced data packet to be retransmitted.

[0012] A data collection method based on the industrial Internet identification and resolution system provided by this application. This method is based on the Pearson correlation coefficient, associates the change trend of production status parameters with the change trend of network congestion degree parameters, and quantifies the method of adjusting the priority parameter of data packets. This method makes the adjustment of the priority parameter more refined and adaptive, and can improve the efficiency and real-time performance of data collection in the industrial Internet identification and resolution system in industrial scenarios with complex network environments and easy congestion.

[0013] Further, in step S6, the step of receiving the reduced data packet at the edge computing node according to the retransmission strategy includes: S61: When executing the retransmission strategy, monitor the network congestion status of multiple edge computing nodes; S62: Compare the network congestion status of the multiple edge computing nodes; S63: Select the edge computing node with the lowest network congestion status; S64: Receive the reduced data packet at the edge computing node with the lowest network congestion status.

[0014] Further, in step S6, the steps of uploading to the core resolution system after performing data parsing and feature extraction operations include: S65: Parse the reduced data packet and extract multiple data features; S66: Evaluate the relevance between the data features and the preset production optimization target to obtain a relevance parameter; S67: Filter the data features based on the relevance parameter, and retain the data features with a relevance higher than the third preset threshold; S68: Upload the retained data features to the core resolution system.

[0015] Further, step S66 includes: S661: At time t, obtain multiple said data features, said preset production optimization goal, and production environment parameters; S662: Determine the dynamic weight coefficient of said preset production optimization goal under said production environment parameters; S663: For each said data feature and said preset production optimization goal, calculate the environment perception relevance under said production environment parameters; S663: Calculate said relevance parameter according to said dynamic weight coefficient and said environment perception relevance.

[0016] Further, step S3 includes: S31: Identify the type of data in the data queue to be transmitted; S32: According to the data type, select a matching compression and encoding method from multiple preset compression and encoding methods; S33: Use the selected compression and encoding method to perform compression and encoding processing on the data in the data queue to be transmitted, and generate a streamlined data packet.

[0017] Further, step S4 includes: S41: Periodically monitor the channel quality parameters of multiple wireless channels; S42: Based on the monitored channel quality parameters, calculate the channel quality sliding average value and channel stability parameter of each wireless channel; S43: According to said channel quality sliding average value and said channel stability parameter, determine the channel selection priority of each wireless channel; S44: Select the wireless channel with the highest channel selection priority as the transmission path of said streamlined data packet.

[0018] In a second aspect, a data acquisition device based on the industrial Internet identification and resolution system is applied to the steps of any one of the above methods. The device includes: An acquisition module, configured to acquire data of industrial device sensors and corresponding data priority parameters; An allocation module, configured to allocate said data to transmission queues with different priorities according to said data priority parameters; An encoding module, configured to perform compression and encoding processing on said data in the transmission queue to generate a streamlined data packet; An evaluation module, configured to evaluate the quality of wireless channels, and select the transmission path of said streamlined data packet according to the wireless channel quality evaluation result; A policy module, configured to determine a retransmission policy based on said data priority parameters and the current network congestion state in case of transmission failure; A retransmission module, configured to receive the streamlined data packet at an edge computing node according to the retransmission policy, and upload it to the core parsing system after performing data parsing and feature extraction operations.

[0019] Advantageous effects: The data acquisition method and device based on the industrial Internet identification resolution system proposed in this application lay a foundation for subsequent data processing by obtaining data and priority parameters, and distinguish the importance of data; by allocating transmission queues according to priority parameters, hierarchical data processing is achieved; by compressing and encoding data, the size of data packets is reduced, which helps to reduce transmission delay and network load; by evaluating the quality of the wireless channel and selecting a transmission path, interference channels are avoided to ensure the quality of data transmission; by determining the retransmission policy in combination with data priority and network congestion status when transmission fails, important data is preferentially retransmitted, and the network condition is considered to avoid aggravating congestion; by receiving data packets at the edge computing node according to the retransmission policy, performing local processing and then uploading, the edge computing ability is utilized to disperse the pressure on the core system and may reduce the overall data processing delay. Therefore, through steps such as data priority differentiation, data compression, channel quality evaluation, intelligent retransmission policy, and edge computing application, this method achieves the beneficial effect of anti-interference and low-latency data transmission in a wireless communication factory environment. Description of the Drawings

[0020] Figure 1 It is a flowchart of a data acquisition method based on the industrial Internet identification resolution system proposed in this application.

[0021] Figure 2 It is a structural diagram of a data acquisition device based on the industrial Internet identification resolution system proposed in this application.

[0022] Label description: 201, acquisition module; 202, allocation module; 203, encoding module; 204, evaluation module; 205, policy module; 206, retransmission module. Detailed Embodiments

[0023] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and marked in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0024] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first", "second", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.

[0025] In the prior art, there is a lack of a solution to ensure the high efficiency and real-time nature of data collection in the industrial Internet of Things identification and resolution system in an industrial scenario where the wireless network environment is complex and prone to congestion. Therefore, to solve this problem, the present application provides a data collection method and device based on the industrial Internet of Things identification and resolution system, specifically: Please refer to Figure 1 In a first aspect, the present application proposes a data collection method based on the industrial Internet of Things identification and resolution system, which is used to achieve anti-interference and low-latency data transmission in a factory environment using wireless communication. The method includes: S1: Obtain the data of industrial equipment sensors and the corresponding data priority parameters; S2: Allocate the data to transmission queues with different priorities according to the data priority parameters; S3: Perform compression encoding processing on the data in the transmission queue to generate a streamlined data packet; S4: Evaluate the quality of the wireless channel and select a transmission path for the streamlined data packet according to the evaluation result of the wireless channel quality; S5: When the transmission fails, determine a retransmission strategy based on the data priority parameters and the current network congestion status; S6: According to the retransmission strategy, receive the streamlined data packet at the edge computing node, perform data parsing and feature extraction operations, and then upload it to the core parsing system.

[0026] Among them, in step S1, the data of industrial equipment sensors is obtained in real time through various sensors deployed on the production line, and the data priority parameters can be preset according to the data type, data importance, or key nature of the production link. For example, data related to production safety can be set as high priority, equipment operation status data can be set as medium priority, and environmental monitoring data can be set as low priority.

[0027] In step S2, the data of industrial equipment sensors are assigned to different transmission queues based on the set priority parameters. For example, three priority queues, namely high, medium, and low, can be established. Data with high priority enters the high-priority queue, and data with low priority enters the low-priority queue to ensure that high-priority data is processed first. Specifically, the priority parameters can be scored for different data. If the score is higher than or equal to 85, the corresponding data enters the high-priority queue. If the score is lower than 50, it enters the low-priority queue. If the score is between 50 and 85, it enters the medium-priority queue.

[0028] Among them, even for data packets in the same queue, such as production equipment data packets and production product data packets both in the high-priority queue, they need to be transmitted sequentially according to the priority parameters of the data packets. For example, if the priority parameter score of the production equipment data packet is 86 and that of the production product data packet is 87, then in the high-priority queue, the production product data packet is transmitted first.

[0029] In step S3, for different types of data, appropriate compression coding methods are selected. For example, for numerical data such as temperature and humidity, differential coding or predictive coding can be used for compression; for non-numerical data such as images and videos, compression algorithms such as JPEG and MPEG can be used. The purpose of compression coding processing is to reduce the size of data packets, reduce the transmission burden on the wireless channel, and reduce the transmission delay.

[0030] In step S4, the wireless channel quality assessment can be performed periodically, and the assessment parameters include signal strength, signal-to-noise ratio, bit error rate, etc. The channel quality assessment results are used to guide the selection of the data packet transmission path. Selecting a path with high channel quality for data transmission can reduce the probability of data transmission failure and improve the reliability of data transmission.

[0031] In step S5, the determination of the retransmission strategy comprehensively considers the data priority and the current network congestion status. When the data priority is high and the network congestion level is low, retransmission is immediately executed to ensure that important data is delivered in a timely manner. When the data priority is low or the network congestion level is high, retransmission is delayed to avoid exacerbating network congestion. The network congestion status can be obtained by monitoring parameters such as network packet loss rate and delay.

[0032] In step S6, after receiving the streamlined data packet, the edge computing node first performs data parsing to restore the original data. Then, according to the preset feature extraction algorithm, valuable data features are extracted from the original data, such as equipment operation status features, production process parameter features, etc. Finally, the extracted data features are uploaded to the core parsing system for subsequent data analysis, decision support, and other applications. The deployment of edge computing nodes can sink data processing tasks to the network edge, reduce the computing pressure on the core parsing system, and reduce the data transmission delay.

[0033] Further, step S5 includes: S51: Monitor the device operation status and production progress information of the production line in real time, and generate production status parameters; S52: Adjust the priority parameter of the data to be retransmitted according to the production status parameters; S53: Obtain the congestion degree parameter of the current wireless network; S54: Determine the retransmission strategy based on the adjusted data priority parameter and congestion degree parameter; The retransmission strategy includes: when the data priority is higher than the first preset threshold and the network congestion degree is lower than the second preset threshold, perform retransmission immediately; when the data priority is lower than the first preset threshold or the network congestion degree is higher than the second preset threshold, delay the retransmission.

[0034] Among them, in step S51, the production status parameters may include parameters such as device operation speed, production rhythm, and product qualification rate that can reflect the operation status of the production line.

[0035] In step S52, the adjustment of the priority parameter of the data to be retransmitted can be specifically that when the production status parameters indicate that the production line is operating well, for example, the device is operating stably and the production progress is normal, the priority of the data to be retransmitted can be appropriately reduced; on the contrary, when the production status parameters indicate that the production line is operating abnormally, for example, the device fails or the production progress lags behind, the priority of the data to be retransmitted can be increased to ensure that the data related to the abnormal production status can be preferentially retransmitted.

[0036] In step S53, the network congestion degree parameter can be obtained by monitoring indicators such as the packet loss rate, delay, and throughput of the wireless network.

[0037] In step S54, the first preset threshold and the second preset threshold are preset values, which can be adjusted according to the actual application scenario and requirements. For example, the first preset threshold can be set to 80, and the second preset threshold can be set to 20. The specific execution method of the retransmission strategy can be that when the condition for immediate retransmission is met, the system immediately starts the retransmission mechanism and resends the reduced data packet; when the condition for delayed retransmission is met, the system puts the data packet to be retransmitted into the delayed retransmission queue and retransmits it when the network condition improves or the data priority increases.

[0038] In some specific embodiments, the production status parameters further include the production beat and the product qualification rate. The production beat data and the product qualification rate data are collected in real time by sensors deployed on the production line, and these data are used as production status parameters. The priority parameter of the data to be retransmitted is adjusted according to the change rate of the production beat and the level of the product qualification rate. For example, when the production beat slows down and the product qualification rate decreases, it indicates that the production line may be abnormal. At this time, the priority of the data to be retransmitted is increased to ensure that the relevant data can be preferentially retransmitted for timely analysis and solution of production problems. On the contrary, when the production beat is stable and the product qualification rate remains at a high level, the retransmission priority of non-critical data can be appropriately reduced.

[0039] Further, step S52 includes: S521: Monitoring the first change trend of the production status parameters and the second change trend of the current network congestion degree parameter; S522: Adjusting the priority parameter of the reduced data packet to be retransmitted according to the first change trend and the second change trend; S523: Using the adjusted priority parameter to determine the retransmission strategy.

[0040] Among them, in step S521, the first change trend is obtained by analyzing the changes of the production status parameters in a previous period of time. For example, it can be the average growth rate or decline rate of the monitored production status parameters in the past 5 sampling cycles. The second change trend is obtained by analyzing the changes of the network congestion degree parameters in a previous period of time. For example, it can be the average growth rate or decline rate of the monitored network congestion degree parameters in the past 5 sampling cycles.

[0041] In step S522, the way to adjust the priority parameter can be that when the first change trend indicates the deterioration of the production status and the second change trend indicates the intensification of network congestion, the adjustment range of the priority parameter is increased. On the contrary, when the first change trend indicates the improvement of the production status and the second change trend indicates the alleviation of network congestion, the adjustment range of the priority parameter is decreased. Therefore, according to the adjustment range of the priority parameter, the priority parameter can be accurately adjusted.

[0042] In step S523, the adjusted priority parameter will be used to determine the subsequent retransmission strategy to ensure that the retransmission strategy can adapt to the dynamically changing industrial environment.

[0043] Specifically, during the data retransmission process of the data collection method in the industrial Internet identification and resolution system, in order to more finely adjust the priority of data retransmission, first, in step S521, the system will monitor the production status parameters and network congestion degree parameters in real time. The first change trend of the production status parameters and the second change trend of the network congestion degree parameters will be calculated. Then, in step S522, according to the two change trends obtained in step S521, the adjustment of the priority parameter will be executed. For example, if the production status parameter shows that the production efficiency is rapidly decreasing (the first change trend), and at the same time the network congestion degree is also rapidly increasing (the second change trend), then the priority parameter of the data packet to be retransmitted will be greatly increased. For example, the data packet to be retransmitted rises from the medium-priority queue to the high-priority queue to ensure the priority retransmission of important production data. On the contrary, if the production status is stable or improving, and the network congestion is alleviated, the adjustment range of the priority parameter will be reduced, and the data packet corresponding to the adjusted priority parameter will still remain in the original priority queue. Finally, in step S523, the adjusted priority parameter will be directly used to determine the specific retransmission strategy. For example, whether to retransmit immediately or delay retransmission, and when the network resources are limited, which data packets should be retransmitted first. In this way, the formulation of the retransmission strategy can be more flexible and intelligent, so as to improve the data collection efficiency and real-time performance in the dynamically changing industrial wireless network environment. Thus, the data collection method can better adapt to the complex and dynamically changing environment of the industrial site, ensure the reliability and real-time performance of data transmission, and reduce the impact of data delay on the production process.

[0044] In some specific embodiments, assume that the production status parameter is the production line speed and the network congestion degree parameter is the data packet loss rate. In step S521, the system monitors that the production line speed has decreased by 5 on average in the past 10 seconds, and the data packet loss rate has increased by 2 on average. In step S522, based on these two change trends, the system determines that the production condition tends to deteriorate and the network environment tends to be congested. Therefore, the priority parameter of the data packet to be retransmitted is greatly increased. For example, 10 points are added to the original priority parameter. In step S523, the adjusted priority parameter is used to evaluate the retransmission strategy. If the adjusted priority exceeds the first preset threshold, the system will immediately start the retransmission mechanism and retransmit the data packet first to ensure that key data can be delivered in time in case of production anomalies and network congestion, and to ensure the stable operation of the production process.

[0045] Furthermore, step S522 includes: S5221: Calculate the Pearson correlation coefficient ρ(t) of the first change trend ΔS(t) and the second change trend ΔC(t), where the calculation formula of ρ(t) is: ρ(t) = Cov(ΔS(t),ΔC(t)) / (σ(ΔS(t))*σ(ΔC(t))), where Cov(ΔS(t), ΔC(t)) is the covariance of ΔS(t) and ΔC(t), σ(ΔS(t)) is the standard deviation of ΔS(t)), and σ(ΔC(t)) is the standard deviation of ΔC(t); S5222: Determine the adjustment factor M(ρ(t)) of the priority parameter according to the Pearson correlation coefficient ρ(t); S5223: Calculate the adjustment amplitude ΔP(t) of the priority parameter according to the adjustment factor M(ρ(t)) of the priority parameter. Among them, the calculation formula of ΔP(t) is: ΔP(t) = α_base * M(ρ(t)), where α_base is the basic adjustment amplitude of the priority parameter; S5224: Use the adjustment amplitude ΔP(t) of the priority parameter to adjust the priority parameter of the reduced data packet to be retransmitted.

[0046] Among them, step S521 is to monitor the first change trend ΔS(t) of the production status parameter and the second change trend ΔC(t) of the current network congestion degree parameter. Specifically, the first change trend can be obtained by calculating the difference between parameters at consecutive time points. For example, ΔS(t) can be expressed as S(t) - S(t - 1), where S(t) is the production status parameter at time t. Similarly, the second change trend ΔC(t) can be expressed as C(t) - C(t - 1), where C(t) is the network congestion degree parameter at time t. The monitoring process can be executed periodically to track the dynamic changes of parameters in real time.

[0047] Step S522 is to adjust the priority parameter of the reduced data packet to be retransmitted according to the first change trend and the second change trend. Step S522 further includes steps S5221 to S5224 to provide a specific method for adjusting the priority parameter.

[0048] Step S5221 is to calculate the Pearson correlation coefficient ρ(t) of the first change trend ΔS(t) and the second change trend ΔC(t). The Pearson correlation coefficient is used to quantify the degree of linear correlation between two variables. In the formula, Cov(ΔS(t), ΔC(t)) represents the covariance of ΔS(t) and ΔC(t), measuring the degree of synchronous change of two variables; σ(ΔS(t)) and σ(ΔC(t)) respectively represent the standard deviations of ΔS(t) and ΔC(t), measuring their respective degrees of dispersion. The value range of the Pearson correlation coefficient ρ(t) is [-1, 1]. The closer the value is to the extreme value 1, the stronger the positive correlation; the closer the value is to the extreme value -1, the stronger the negative correlation; and the value close to 0 indicates a weak linear correlation.

[0049] Step S5222 is to determine the adjustment factor M(ρ(t)) of the priority parameter according to the Pearson correlation coefficient ρ(t). The adjustment factor M(ρ(t)) is a function of the Pearson correlation coefficient ρ(t) and is used to establish the mapping relationship between the correlation coefficient and the priority adjustment amplitude. The specific functional form of M(ρ(t)) can be designed according to the actual application scenario and requirements. For example, when the positive correlation is strong, a larger adjustment factor can be set, and vice versa.

[0050] Specifically, the calculation rule of M(ρ(t)) is as follows: When ρ(t) > ρ(threshold_ positive), it indicates that the positive correlation is strong, and the adjustment factor is: M(ρ(t)) = 1 / (1 + k(positive correlation) * ρ(t)) When ρ(t) < ρ(threshold_ negative), it indicates that the negative correlation or non - correlation is strong, and the adjustment factor is: M(ρ(t)) = 1 + k(negative correlation) * |ρ(t)| When ρ(threshold_ negative) ≤ ρ(t) ≤ ρ(threshold_ positive), it indicates that the correlation is not strong, and the adjustment factor is: M(ρ(t)) = 1.

[0051] Among them, ρ(threshold_ positive) is the threshold for judging strong positive correlation, ρ(threshold_ negative) is the threshold for judging negative correlation or non - correlation, k(positive correlation) is the adjustment amplitude attenuation coefficient for positive correlation, and k(negative correlation) is the adjustment amplitude enhancement coefficient for negative correlation or non - correlation. k(positive correlation) and k(negative correlation) can be manually set according to historical experience.

[0052] Step S5223 is to calculate the priority parameter adjustment amplitude ΔP(t) according to the adjustment factor M(ρ(t)) of the priority parameter. The calculation formula is ΔP(t) = α_base * M(ρ(t)), where α_base is the basic priority parameter adjustment amplitude, which is a preset constant and is used to control the basic step size of the priority adjustment. By correcting the basic priority parameter adjustment amplitude through the adjustment factor M(ρ(t)), the adaptive adjustment of the priority parameter based on the correlation can be realized.

[0053] Step S5224 is set to use the priority parameter adjustment amplitude ΔP(t) to adjust the priority parameter of the reduced data packet to be retransmitted. Specifically, the calculated priority parameter adjustment amplitude ΔP(t) can be added to the current priority parameter of the data packet to be retransmitted to obtain the adjusted priority parameter. Among them, the larger the priority parameter adjustment amplitude ΔP(t), the higher the value of the adjusted priority parameter, so that the corresponding data packet to be retransmitted enters the high-priority queue from the low-priority queue; the smaller the priority parameter adjustment amplitude ΔP(t), the smaller the value of the adjusted priority parameter, and the priority queue of its corresponding data packet to be retransmitted remains unchanged, but the transmission order may be improved in the original priority queue. The adjusted priority parameter will be used for subsequent retransmission policy determination, thus affecting the retransmission timing and resource allocation of the data packet.

[0054] In some specific embodiments, the adjustment factor M(ρ(t)) is set to a piecewise function. For example, when the Pearson correlation coefficient ρ(t) is greater than 0.5, M(ρ(t)) takes the value of 2; when ρ(t) is between 0.1 and 0.5, M(ρ(t)) takes the value of 1; when ρ(t) is less than 0.1, M(ρ(t)) takes the value of 0.5. The basic priority parameter adjustment amplitude α_base is set to 5.

[0055] Suppose at a certain moment t, the change trend ΔS(t) of the monitored production state parameter is positive, indicating that the production state tends to be busy; at the same time, the change trend ΔC(t) of the monitored network congestion degree parameter is also positive, indicating that the network congestion degree is increasing. The calculated Pearson correlation coefficient ρ(t) is 0.6, which is greater than 0.5. According to the setting of step S5222, the adjustment factor M(ρ(t)) takes the value of 2.

[0056] Then, in step S5223, according to the formula ΔP(t) = α_base * M(ρ(t)), the priority parameter adjustment amplitude ΔP(t) = 5 * 2 = 10 is calculated.

[0057] Finally, in step S5224, the priority parameter adjustment amplitude 10 is used to adjust the priority parameter of the reduced data packet to be retransmitted. If the original priority parameter of the data packet to be retransmitted is 30, the adjusted priority parameter is 30 + 10 = 40. The adjusted priority parameter will be used for subsequent retransmission policy decision-making, making it more likely for this data packet to be retransmitted preferentially, thereby reducing the data transmission delay and ensuring the real-time nature of data collection.

[0058] Through the above specific embodiments, it shows specifically how to use the Pearson correlation coefficient and the adjustment factor to quantify and adjust the data packet priority parameters, realizes the adaptive adjustment of the priority parameters, and improves the efficiency and real-time performance of data collection in the industrial wireless network environment.

[0059] In step S5, according to the retransmission policy, the steps for the edge computing node to receive the refined data packet include: S51: When executing the retransmission policy, monitor the network congestion status of multiple edge computing nodes; S52: Compare the network congestion status of multiple edge computing nodes; S53: Select the edge computing node with the network congestion status; S54: Receive the refined data packet at the edge computing node with the network congestion status.

[0060] Among them, in step S51, when the system determines that the data packet retransmission policy needs to be executed, start monitoring the network congestion status of multiple edge computing nodes in the factory environment. The monitoring of the network congestion status can be achieved in various ways. For example, probe packets can be periodically sent to each edge computing node, and the network congestion degree can be evaluated according to parameters such as the response time and packet loss rate of the probe packets.

[0061] In step S52, compare and analyze the network congestion status parameters of multiple edge computing nodes monitored in step S51, with the aim of finding the edge computing node with relatively light network load.

[0062] In step S53, based on the comparison result of step S52, select the edge computing node with the network congestion status, that is, select the edge computing node with the lowest current network load or the most unobstructed network to provide a better network environment for the retransmission of data packets.

[0063] In step S54, at the edge computing node with the network congestion status selected in step S53, execute the operation of receiving the refined data packet to ensure that the retransmitted data packet can be received through a relatively idle network path, thereby reducing the risk of retransmission failure and improving the reliability of data transmission.

[0064] Specifically, during the data collection process of the industrial Internet identification and resolution system, when a data packet transmission fails for the first time and a retransmission operation needs to be performed, the system first monitors the network congestion situation of multiple edge computing nodes deployed in the factory environment in real time. The monitoring process can adopt the method of sending ICMP Echo requests to calculate the round-trip time RTT of the data packet. The higher the RTT value, the more congested the network is. Then, the RTT values of each edge computing node collected are sorted or compared, and the edge computing node with the lowest RTT value, that is, the node with the lowest network congestion degree, is selected. Finally, the retransmitted data packet is routed to the selected edge computing node for reception. Through the above steps, nodes with network congestion can be avoided, and edge computing nodes with good network conditions can be selected for data retransmission, thereby improving the retransmission success rate and ensuring the real-time and reliability of data collection.

[0065] In some specific embodiments, it is assumed that there are three edge computing nodes deployed in the factory environment, which are respectively marked as node A, node B, and node C. When data needs to be retransmitted, the system first sends network probe packets to node A, node B, and node C respectively, and monitors the network response time of each node. The monitoring results show that the average response time of node A is 10 ms, the average response time of node B is 5 ms, and the average response time of node C is 20 ms. By comparison, the response time of node B is the shortest, indicating that the network congestion state of node B is the best, and it can be set as the edge computing node. Therefore, the system selects node B as the receiving node for the retransmitted data packet and sends the retransmitted data packet to node B. Thus, the data packet is retransmitted through node B with relatively unobstructed network, reducing the possibility of retransmission failure caused by network congestion and improving the data collection efficiency.

[0066] Further, in step S6, the steps of uploading to the core parsing system after performing data parsing and feature extraction operations include: S65: Parse the streamlined data packet and extract multiple data features; S66: Evaluate the relevance between the data features and the preset production optimization target to obtain a relevance parameter; S67: Screen the data features based on the relevance parameter and retain the data features with a relevance higher than the third preset threshold; S68: Upload the retained data features to the core parsing system.

[0067] Among them, in step S65, after the streamlined data packet is received, it is parsed through a pre-set data parsing protocol. The data parsing protocol can include, but is not limited to, industrial general protocols such as Modbus, OPC-UA, or MQTT. According to the protocol specifications, multiple data features that can reflect the operating state of industrial equipment are extracted from the data packet. The data features can be, for example, temperature data, pressure data, vibration frequency data, or current data.

[0068] In step S66, the evaluation of the relevance parameter is achieved by calculating the statistical correlation between the data features and the preset production optimization goal. The statistical correlation calculation method can adopt the Pearson correlation coefficient, mutual information, or canonical correlation analysis, etc. The preset production optimization goal is determined according to the actual production requirements of the factory. For example, it can be to improve the product yield, reduce energy consumption, or reduce the equipment failure rate.

[0069] In step S67, a third preset threshold is preset to define the level of relevance between the data features and the production optimization goal. The third preset threshold can be adjusted according to the actual application scenario and the strictness of data screening. When the relevance parameter of the data feature is higher than the third preset threshold, the data feature is determined to be highly relevant to the production optimization goal and is retained; otherwise, it is filtered out.

[0070] In step S68, the data features retained after screening are uploaded to the core analysis system through a wireless network or a wired network for subsequent data analysis, decision support, and production optimization.

[0071] Specifically, after the edge computing node receives the refined data packet, it first performs protocol analysis on the data packet. For example, if the data packet is encapsulated using the Modbus protocol, then according to the frame format of the Modbus protocol, information such as the function code, register address, and data field are parsed. The sensor measurement values contained in the data field are the data features to be extracted. Assume that the extracted data features include three features: equipment temperature, spindle speed, and motor current. Subsequently, for each data feature, its relevance to the preset production optimization goal is evaluated. The preset production optimization goal is assumed to be reducing the product defective rate, and the evaluation method uses the Pearson correlation coefficient. The Pearson correlation coefficients between the equipment temperature, spindle speed, and motor current and the historical data of the product defective rate are calculated respectively, obtaining three relevance parameters. Assume that the relevance parameter of the equipment temperature is 0.85, the relevance parameter of the spindle speed is 0.62, and the relevance parameter of the motor current is 0.23. Then, a third preset threshold of 0.7 is set, and the calculated relevance parameters are compared with the third preset threshold. The relevance parameter of the equipment temperature, 0.85, is higher than 0.7, so the equipment temperature data feature is retained. The relevance parameters of the spindle speed and the motor current are both lower than 0.7, so these two data features are filtered out. Finally, only the retained equipment temperature data feature is uploaded to the core analysis system.

[0072] Furthermore, step S66 includes: S661: At time t, obtain multiple data features, the preset production optimization goal, and production environment parameters; S662: Determine the dynamic weight coefficient of the preset production optimization goal under the production environment parameters; S663: Calculate the environmental perception relevance under the production environment parameters for each data feature and the preset production optimization goal; S664: Calculate the relevance parameter based on the dynamic weight coefficient and the environmental perception relevance.

[0073] Among them, in step S661, the data feature can be the parsed device operation status data, product quality data, process parameter data, etc. The preset production optimization goal can be to improve production efficiency, reduce energy consumption, or increase the product yield rate, etc. The production environment parameters can be external environmental factors such as temperature, humidity, noise, or vibration that affect the production process.

[0074] In step S662, the determination of the dynamic weight coefficient can be achieved through a preset rule or model. For example, when the environmental temperature is too high, the weight coefficient for reducing energy consumption can be increased, while the weight coefficient for improving production efficiency can be decreased.

[0075] In step S663, the calculation of the environmental perception relevance can adopt the modified Pearson correlation coefficient, mutual information, or other correlation measurement methods that can reflect the environmental impact. In step S664, the calculation of the relevance parameter can be the product of the dynamic weight coefficient and the environmental perception relevance, or a weighted sum, etc.

[0076] Specifically, at time t, n data features are obtained as D_i(t), and m preset production optimization goals are obtained as G_j(t), where n is the total number of data features, i is the serial number of the data feature, i = 1, 2, 3...n; m is the total number of preset production optimization goals, and j is the serial number of the preset production optimization goal, j = 1, 2, 3...m.

[0077] The production environment parameter is E_z(t), and the production environment parameter includes at least one of the current production stage, device operation status, environmental temperature, and network congestion status. z is the serial number of the production environment parameter. The dynamic weight coefficient of the preset production optimization goal under the production environment parameter is W_j(E_z(t)), where W_j(E_z(t)) is dynamically adjusted based on E_z(t) according to a preset rule or a machine learning model, and the value range of the dynamic weight coefficient is [0, 1] or [0, 100].

[0078] The calculation function of the environmental perception relevance is Corr(D_i(t), G_j(t), E_z(t)); The calculation formula for calculating the relevance parameter based on the dynamic weight coefficient and the environmental perception relevance is: $R_i(t)=\sum_{j = 1}^{m}(W_j(E_z(t)) * Corr(D_i(t), G_j(t), E_z(t)))$, where $R_i(t)$ is the correlation parameter.

[0079] Further, step S3 includes: S31: Identify the data types in the queue to be transmitted; S32: According to the data types, select the matching compression and encoding method from a variety of preset compression and encoding methods; S33: Use the selected compression and encoding method to perform compression and encoding processing on the data in the queue to be transmitted, generating a streamlined data packet.

[0080] In step S31, the data type identification can be implemented based on the type identifiers preset in the data packet header information, and the identifiers are associated with the sensor type or data content. For example, the data packet of the temperature sensor can be marked as "temperature", the data packet of the pressure sensor can be marked as "pressure", and the data packet of the image can be marked as "image".

[0081] In step S32, the preset library of various compression and encoding methods can include lossless compression algorithms such as DEFLATE and LZ4, as well as lossy compression algorithms such as JPEG and MPEG. For numerical data such as temperature or pressure, DEFLATE can be selected to ensure data accuracy; for image data, JPEG can be selected to obtain a higher compression ratio. The matching process can use the look-up table method to establish a mapping table between the data type and the compression algorithm, and quickly find and select the corresponding compression algorithm according to the identified data type.

[0082] In step S33, the selected compression and encoding method is used for the data in the queue to be transmitted, generating a data packet with a smaller size. Thus, on the premise of ensuring the integrity of the data information or meeting the accuracy requirements, the effective compression of the data volume is achieved, reducing the network burden of subsequent wireless data transmission.

[0083] Further, step S4 includes: S41: Periodically monitor the channel quality parameters of multiple wireless channels; S42: Based on the monitored channel quality parameters, calculate the channel quality sliding average value and the channel stability parameter of each wireless channel; S43: According to the channel quality sliding average value and the channel stability parameter, determine the channel selection priority of each wireless channel; S44: Select the wireless channel with the highest channel selection priority as the transmission path for the streamlined data packet.

[0084] Among them, in step S41, the periodic monitoring can be implemented as follows: the data acquisition device polls multiple wireless channels at fixed time intervals. Each time it polls, the data acquisition device receives the signals of each wireless channel and measures the channel quality parameters. The channel quality parameters can include at least one of parameters such as Received Signal Strength Indicator (RSSI), Signal-to-Noise Ratio (SNR), and Bit Error Rate (BER).

[0085] In step S42, the moving average of the channel quality is obtained by averaging the channel quality parameters monitored in the recent period of time. For example, the moving window average method can be adopted, and the window size can be adjusted according to the actual application scenario and network environment. The channel stability parameter can be characterized by the standard deviation of the channel quality parameters. The smaller the standard deviation, the smaller the fluctuation of the channel quality and the better the channel stability.

[0086] In step S43, the channel selection priority can be determined as the weighted sum of the moving average of the channel quality and the channel stability parameter. The weight coefficients can be set according to actual needs. For example, the weight of the moving average of the channel quality can be set higher than the weight of the channel stability parameter. Compare the final scores of the channel selection priorities of each channel to ensure that the preferentially selected channel has better average quality.

[0087] In step S44, the wireless channel with the highest channel selection priority is determined by comparing the numerical values of the channel selection priorities of each wireless channel. The wireless channel corresponding to the largest channel selection weight value is selected as the transmission path for the reduced data packet.

[0088] In some specific embodiments, it is assumed that there are three wireless channels in the industrial wireless network environment, which are respectively labeled as Channel 1, Channel 2, and Channel 3. The data acquisition device periodically monitors the Received Signal Strength Indicator (RSSI) values of these three channels every 1 second. Calculate the moving average of the RSSI of each channel in the past 10 seconds as the moving average of the channel quality, and calculate the standard deviation of the RSSI of each channel in the past 10 seconds as the channel stability parameter. The channel selection priority is determined as the moving average of the channel quality minus the channel stability parameter. The formula is expressed as: channel selection priority = K1 * moving average of the channel quality + K2 * channel stability parameter, where K1 and K2 are weight coefficients. Compare the channel selection priorities of Channel 1, Channel 2, and Channel 3, and select the channel with the largest channel selection priority value as the transmission path for the reduced data packet. For example, if the channel selection priority of Channel 1 is the highest, then select Channel 1 as the transmission path for the reduced data packet. Through the above implementation, more stable and reliable wireless channel selection can be achieved, improving the efficiency and stability of data acquisition in the industrial Internet identification resolution system.

[0089] The present application further provides a data acquisition device for the industrial Internet identification and resolution system, which is applied to any of the above method steps. The device includes: An acquisition module 201, configured to acquire data of industrial equipment sensors and corresponding data priority parameters; An allocation module 202, configured to allocate data to transmission queues with different priorities according to the data priority parameters; An encoding module 203, configured to perform compression encoding processing on the data in the transmission queue to generate a streamlined data packet; An evaluation module 204, configured to evaluate the quality of the wireless channel and select a transmission path for the streamlined data packet according to the wireless channel quality evaluation result; A policy module 205, configured to determine a retransmission policy based on the data priority parameters and the current network congestion state when transmission fails; A retransmission module 206, configured to receive the streamlined data packet at the edge computing node according to the retransmission policy, perform data parsing and feature extraction operations, and then upload the data to the core parsing system.

[0090] Among them, the acquisition module 201 is configured to collect raw data from the sensors of industrial equipment and determine the importance level of each data unit. The data priority parameters can be determined by the sensor type, data generation frequency, or criticality in the production process.

[0091] The allocation module 202 receives the data and priority information from the acquisition module and places the data into different transmission queues according to the preset priority rules. For example, the high-priority queue can be used for data with high real-time requirements, and the low-priority queue can be used for data that tolerates delays.

[0092] The encoding module 203 applies a suitable compression algorithm to different priority queues, aiming to reduce the data volume and speed up the transmission speed. The selection of the compression algorithm can be based on the data type and priority.

[0093] The evaluation module 204 continuously monitors the quality of the wireless communication channel. The channel quality parameters can include signal strength, bit error rate, or signal-to-noise ratio. The evaluation result is used to select the best transmission path.

[0094] The policy module 205 is activated when the data packet transmission fails. It considers the priority of the data packet and the congestion degree of the current network. The network congestion degree can be reflected by the data packet loss rate or delay time. The retransmission policy determines when and how to retransmit the failed data packet.

[0095] The retransmission module 206 executes the retransmission policy formulated by the policy module. When receiving data packets at the edge computing node, the edge computing node performs preliminary data processing tasks, including data parsing and feature extraction. The extracted data features are finally sent to the core parsing system for more advanced analysis and applications.

[0096] Specifically, when the data acquisition device of the industrial Internet identification and resolution system is working, first, the acquisition module 201 collects production process data from various sensors in the factory workshop, such as temperature sensors, pressure sensors, displacement sensors, etc., and assigns a priority parameter to each data packet. The priority parameter reflects the importance of the data.

[0097] After receiving these data with priorities, the allocation module 202 allocates them to different transmission queues. For example, the emergency stop signal is allocated to the highest priority queue, and the environmental monitoring data is allocated to a lower priority queue.

[0098] The encoding module 203 performs compression processing on the data in different queues. For example, for high-priority control instructions, lossless compression is used to ensure data integrity, and for low-priority monitoring data, lossy compression is used to reduce the data volume.

[0099] The evaluation module 204 monitors the factory wireless network environment in real time and evaluates the quality of different channels. For example, by measuring the signal strength and interference level of each channel, the channel quality is evaluated. When data needs to be transmitted, the evaluation module 204 will select the channel with the best quality. If a data packet is lost during transmission, the policy module 205 will formulate a retransmission policy according to the priority of the data packet and the current network congestion situation. For example, high-priority data is retransmitted immediately, and low-priority data is retransmitted when the network is idle.

[0100] The retransmission module 206 sends the data packet to the edge computing node with the lowest network congestion degree according to the formulated retransmission policy. After receiving the data packet, the edge computing node first performs decompression and data parsing, then extracts key production features, such as equipment operating status, product quality parameters, etc., and finally uploads these feature data to the core parsing system for production process monitoring, fault diagnosis, and optimization decision-making.

[0101] In some specific embodiments, the data acquisition device of the industrial Internet identification and resolution system can be deployed on the production line of an intelligent factory. The acquisition module 201 is connected to the PLC controller and various sensors on the production line to collect device operation data and environmental data in real time. The allocation module 202 distributes the data to three priority queues according to preset rules, such as according to the data type or source. The high-priority queue is used to process emergency alarm data, the medium-priority queue is used to process key production parameter data, and the low-priority queue is used to process non-critical monitoring data. The encoding module 203 adopts the LZ4 fast compression algorithm for the high-priority queue and the Zstd algorithm for the medium- and low-priority queues to balance the compression ratio and compression speed. The evaluation module 204 periodically scans the wireless channels in the 2.4 GHz and 5 GHz frequency bands and calculates the channel quality scores of each channel. The scores comprehensively consider the signal strength, signal-to-noise ratio, and channel occupancy rate. When the policy module 205 detects packet loss, it checks the priority of the packet. If the priority is higher than the preset threshold and the current network average packet loss rate is lower than 5, it immediately starts retransmission; otherwise, it delays retransmission. When performing retransmission, the retransmission module 206 monitors the load conditions of the three surrounding edge computing nodes, selects the edge computing node with the lowest CPU load as the data receiving node. After receiving the data, the edge computing node uses the decompression algorithm corresponding to the encoding module to decompress the data, then parses the data packet, extracts features such as device rotation speed, temperature, and current, and uploads these feature data to the core resolution system of the factory through the MQTT protocol.

[0102] The above are only the embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A data collection method based on an industrial Internet identification resolution system, used to achieve anti-interference, low-latency data transmission in a factory environment using wireless communication, characterized in that: The method includes: S1: Obtain the data of industrial equipment sensors and the corresponding data priority parameters; S2: Allocating the data to transmission queues of different priorities according to the data priority parameter; S3: compress and encode the data in the transmission queue to generate a simplified data packet; S4: Evaluate the quality of the wireless channel, and select a transmission path for the simplified data packet according to the wireless channel quality evaluation result; S5: When transmission fails, determining a retransmission strategy based on the data priority parameter and the current network congestion status; S6: According to the retransmission strategy, the streamlined data packet is received at the edge computing node, and data parsing and feature extraction operations are performed and then uploaded to the core parsing system.

2. According to claim 1, a data collection method based on an industrial Internet identification resolution system is characterized in that: Step S5 includes: S51: monitor the equipment operation status and production progress information of the production line in real time and generate production status parameters; S52: adjusting the priority parameter of the data to be retransmitted according to the production status parameter; S53: Obtaining a congestion level parameter of the current wireless network; S54: Determine a retransmission strategy based on the adjusted data priority parameter and the congestion level parameter; The retransmission strategy includes: when the data priority is higher than or equal to a first preset threshold and the network congestion level is lower than a second preset threshold, retransmission is performed immediately; when the data priority is lower than the first preset threshold, or the network congestion level is higher than the second preset threshold, retransmission is performed with delay.

3. According to claim 2, a data collection method based on an industrial Internet identification resolution system is characterized in that: Step S52 includes: S521: monitoring a first change trend of a production status parameter and a second change trend of a current wireless network congestion degree parameter; S522: Adjusting the priority parameter of the simplified data packet to be retransmitted according to the first change trend and the second change trend; S523: Use the adjusted priority parameter to determine the retransmission strategy.

4. According to claim 3, a data collection method based on an industrial Internet identification resolution system is characterized in that: Step S522 includes: S5221: Calculate the Pearson correlation coefficient ρ(t) between the first change trend ΔS(t) and the second change trend ΔC(t), wherein the calculation formula of ρ(t) is: ρ(t) = Cov(ΔS(t), ΔC(t)) / (σ(ΔS(t)) * σ(ΔC(t))), where Cov(ΔS(t), ΔC(t)) is the covariance of ΔS(t) and ΔC(t), σ(ΔS(t)) is the standard deviation of ΔS(t), σ(ΔC(t)) is the standard deviation of ΔC(t); S5222: Determine an adjustment factor M(ρ(t)) of the priority parameter according to the Pearson correlation coefficient ρ(t); S5223: Calculate the priority parameter adjustment amplitude ΔP(t) according to the priority parameter adjustment factor M(ρ(t)), wherein the calculation formula of ΔP(t) is: ΔP(t) = α_base * M(ρ(t)), wherein α_base is the basic priority parameter adjustment amplitude; S5224: Use the priority parameter adjustment amplitude ΔP(t) to adjust the priority parameter of the simplified data packet to be retransmitted.

5. The data collection method based on the industrial Internet identification resolution system according to claim 1 is characterized in that: In step S6, the step of receiving the simplified data packet at the edge computing node according to the retransmission strategy includes: S61: When executing the retransmission strategy, monitor the network congestion status of multiple edge computing nodes; S62: Compare the network congestion states of the multiple edge computing nodes; S63: Select an edge computing node with the lowest network congestion status; S64: Receive the simplified data packet at the edge computing node with the lowest network congestion status.

6. A data collection method based on the industrial Internet identification resolution system according to claim 5, characterized in that: In step S6, the steps of performing data analysis and feature extraction operations and uploading to the core analysis system include: S65: parsing the simplified data packet to extract multiple data features; S66: Evaluate the correlation between the data feature and the preset production optimization target to obtain a correlation parameter; S67: Filtering data features based on the correlation parameter, and retaining data features with correlation higher than a third preset threshold; S68: Upload the retained data features to the core analysis system.

7. A data collection method based on the industrial Internet identification resolution system according to claim 6, characterized in that: Step S66 includes: S661: At time t, obtaining a plurality of the data features, the preset production optimization goals, and production environment parameters; S662: Determine a dynamic weight coefficient of the preset production optimization target under the production environment parameters; S663: Calculating the environmental perception relevance under the production environment parameters for each of the data features and the preset production optimization target; S664: Calculate the relevance parameter according to the dynamic weight coefficient and the environmental perception relevance.

8. The data collection method based on the industrial Internet identification resolution system according to claim 1 is characterized in that: Step S3 includes: S31: Identify the type of data in the queue to be transmitted; S32: selecting a matching compression encoding method from a plurality of preset compression encoding methods according to the data type; S33: Using the selected compression encoding method to perform compression encoding processing on the data in the transmission queue to generate a simplified data packet.

9. The data collection method based on the industrial Internet identification resolution system according to claim 1 is characterized in that: Step S4 includes: S41: periodically monitoring channel quality parameters of multiple wireless channels; S42: Calculate the channel quality sliding average and channel stability parameter of each wireless channel based on the monitored channel quality parameters; S43: determining a channel selection priority of each wireless channel according to the channel quality sliding average value and the channel stability parameter; S44: Select a channel to select a wireless channel with the highest priority as a transmission path for the simplified data packet.

10. A data collection device based on an industrial Internet identification resolution system, applied to the steps of the method described in any one of claims 1 to 9 above, characterized in that: The device includes: An acquisition module, used to acquire data from industrial equipment sensors and corresponding data priority parameters; an allocation module, configured to allocate the data to transmission queues of different priorities according to the data priority parameter; An encoding module, used for performing compression encoding processing on the data in the transmission queue to generate a simplified data packet; An evaluation module, used for evaluating the quality of a wireless channel and selecting a transmission path for the simplified data packet according to the evaluation result of the wireless channel quality; A policy module, used to determine a retransmission strategy based on the data priority parameter and the current network congestion status when a transmission fails; The retransmission module is used to receive the simplified data packet at the edge computing node according to the retransmission strategy, perform data parsing and feature extraction operations, and then upload it to the core parsing system.

Images