Network slice configuration method based on flow analysis
By setting data monitoring points in the industrial Internet of Things and building data prediction models, and configuring network slices based on timeliness and link performance, massive data management problems are solved, real-time monitoring and control of data in the industrial Internet of Things are realized, and the System stability and reliability.
Patent Information
- Application Number
- CN202410962075.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-19
AI Technical Summary
The existing technology is difficult to effectively manage massive data in the industrial Internet of Things, resulting in unreasonable division of network resources, resulting in degradation of system performance, increased service delay, and it is difficult to ensure real-time transmission and monitoring of key data.
By setting data monitoring points in the industrial Internet of Things, building a data prediction model, and performing network slice configurations based on the timeliness of data monitoring indicators and communication link performance configuration parameters to ensure that high-timed data is divided into high-performance links. And adapt to different data needs through dynamic priority adjustments, realizing differentiated processing and resource optimization.
Real-time monitoring and control of industrial production data, improve system stability and reliability, ensure timely transmission of key data, avoid data congestion and delay, flexibly manage bandwidth resources, and improve network link efficiency and performance .
Smart Images

Figure CN118984276A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of traffic analysis technology, and specifically to a network slicing configuration method based on traffic analysis. Background Art
[0002] The prior art CN115514646A "A network slice configuration method and related device based on traffic analysis" obtains the traffic data of the current server within the preset time to obtain the current traffic data, which includes the duration, behavior and traffic size; inputs the current traffic data into the preset GRU model for load prediction to obtain the current load prediction result, and the preset GRU model includes an autoencoder and a GRU network; performs network slice configuration on the VNF network service according to the current load prediction result, the preset related load prediction result and the load threshold. This application can solve the technical problems that the prior art lacks pertinence and is prone to frequent migration of VNFs, resulting in more network resources being occupied, reducing system performance and increasing service latency.
[0003] The prior art CN112737808A "A method and device for configuring a network slice instance" sends network configuration request information to a network data analysis module; receives network configuration information sent by the network data analysis module; determines the load status of the network slice instance according to the network configuration information, and determines the target network slice instance to be accessed by the access terminal according to the load status. The network slice instance configuration side actively obtains network configuration information from the network data analysis module, and configures the network slice instance to be accessed by the access terminal according to the load status of the network slice instance in the network configuration information, thereby avoiding the access terminal accessing a network slice instance with a relatively large load, making the load of the network slice instance more uniform; at the same time, since the load status of multiple network slice instances can be obtained, the network slice instance configuration side directly performs the configuration when configuring the network slice instance, thereby reducing the load of the network slice instance configuration system.
[0004] Faced with the massive industrial data generated in the modern industrial Internet of Things, how to use network slicing technology in the industrial Internet of Things to divide the entire Internet of Things infrastructure and network resources to meet the network requirements in different industrial application scenarios, how to configure network slices in communication links, and perform priority management according to the timeliness of different data monitoring indicators, effectively manage massive data, and ensure the real-time transmission and monitoring of key data, thereby improving the efficiency, safety and reliability of the industrial production process, are issues that we urgently need to solve. Summary of the invention
[0005] In order to solve the above technical problems, the object of the present invention is to provide a network slice configuration method based on traffic analysis, comprising the following steps: Step S1: Setting data monitoring points for each industrial production equipment in the target industrial Internet of Things to obtain production data information generated by the industrial production equipment; Step S2: construct a data prediction model, and obtain the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data prediction model; Step S3: According to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, complete the network slicing configuration operation of each data monitoring indicator in the communication link; Step S4: Compare the real-time production data information in the current collection cycle with the data volume prediction value of each data monitoring indicator, and generate traffic anomaly alarm information or perform ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator based on the comparison result.
[0006] Furthermore, a data monitoring point is set for each industrial production equipment in the target industrial Internet of Things, and the process of obtaining the production data information generated by the industrial production equipment includes: Several industrial production equipment in the target industrial Internet of Things are connected to the monitoring center through a communication link, and data monitoring points are set for each industrial production equipment in the target industrial Internet of Things. The data monitoring points are used to collect production data information generated by the industrial production equipment to which they belong, and mark the collection cycle. The production data information includes several data monitoring indicators corresponding to each industrial production equipment and the data volume corresponding to each data monitoring indicator.
[0007] Furthermore, the process of obtaining the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data prediction model includes: Construct a data prediction model based on deep learning, obtain historical production data information generated by each industrial production equipment in several historical collection cycles, use the historical production data information generated by each industrial production equipment in several historical collection cycles as a training set and a test set, input the training set into the data prediction model for training until the loss function training is stable, and save the model parameters, test the data prediction model through the test set until it meets the preset requirements, and output the data prediction model; The production data prediction information of each industrial production equipment in the current collection cycle is obtained according to the data prediction model, and the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle is obtained according to the production data prediction information.
[0008] Furthermore, according to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, the process of completing the network slice configuration operation of each data monitoring indicator in the communication link includes: Acquire all communication links between each industrial production equipment and the monitoring center, acquire performance configuration parameters of each communication link, the performance configuration parameters including link bandwidth and average data transmission rate, sort each communication link in positive order according to the average data transmission rate of each communication link, and generate a communication link rate queue; Preset the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment, sort each data monitoring indicator in reverse order according to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment, and generate a data monitoring indicator timeliness queue; Construct an indicator traversal pointer and a link traversal pointer, wherein the indicator traversal pointer is used to traverse and retrieve the data monitoring indicator timeliness queue in the order from the first position of the queue to the last position of the queue, and the link traversal pointer is used to traverse and retrieve the communication link rate queue in the order from the first position of the queue to the last position of the queue; First, the indicator traversal pointer performs a traversal search on the data monitoring indicator timeliness queue. When the indicator traversal pointer retrieves the data monitoring indicator, it generates an indicator configuration instruction for the data monitoring indicator, suspends the data monitoring indicator traversal search, and sends the indicator configuration instruction to the link traversal pointer. When the link traversal pointer receives the indicator configuration instruction, the link traversal pointer performs a communication link traversal search on the communication link rate queue. When the link traversal pointer retrieves the communication link, the link bandwidth of the communication link is configured for a network slice according to the indicator configuration instruction. After completing the network slicing configuration operation of the data monitoring indicator corresponding to the indicator configuration instruction, if the remaining link bandwidth of the communication link is zero, the communication link is removed from the communication link rate queue, and then the communication link traversal search is suspended, and an indicator configuration completion instruction is sent to the indicator traversal pointer. After the indicator traversal pointer receives the indicator configuration completion instruction, the data monitoring indicator is removed from the data monitoring indicator timeliness queue, and the data monitoring indicator traversal search is resumed, and so on, until the network slicing configuration operation of all data monitoring indicators in the data monitoring indicator timeliness queue is completed.
[0009] Further, when the link traversal pointer retrieves a communication link, the process of performing a network slicing configuration operation on the link bandwidth of the communication link according to the indicator configuration instruction includes: According to the indicator configuration instruction, the data monitoring indicator that needs to perform the network slice configuration operation is obtained, the data volume prediction value of the data monitoring indicator in the current collection cycle is obtained, and the link bandwidth occupied by the data volume prediction value is obtained, and at the same time, the link bandwidth of the communication link retrieved by the link traversal pointer is obtained, and the link bandwidth of the communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is compared with the remaining link bandwidth of the communication link; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is less than or equal to the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and a link bandwidth consistent with the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is allocated from the remaining link bandwidth of the communication link to the network slice, thereby completing the network slice configuration operation of the data monitoring indicator; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is greater than the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and the remaining link bandwidth of the communication link is allocated to the network slice of the data monitoring indicator, and the remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle and the link bandwidth in the network slice allocated to the data monitoring indicator, and the link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue, and when the link traversal pointer retrieves another communication link, the other communication link is marked as a diversion communication link of the data monitoring indicator, and the link bandwidth of the diversion communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link; If the remaining link bandwidth to be allocated is less than or equal to the remaining link bandwidth of the offload communication link, a network slice of the data monitoring indicator in the offload communication link is constructed, and a link bandwidth consistent with the remaining link bandwidth to be allocated from the remaining link bandwidth of the offload communication link is allocated to the network slice of the data monitoring indicator, thereby completing the network slice configuration operation of the data monitoring indicator in the communication link and the offload communication link; If the remaining link bandwidth to be allocated is greater than the remaining link bandwidth of the diversion communication link, a network slice of the data monitoring indicator in the diversion communication link is constructed, and the remaining link bandwidth of the diversion communication link is allocated to the network slice. The remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the predicted value of the data volume of the data monitoring indicator in the current collection cycle and the link bandwidth allocated to the data monitoring indicator in the communication link and the network slice in the diversion communication link. The link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue. When the link traversal pointer retrieves other communication links, the other retrieved communication links are also marked as diversion communication links of the data monitoring indicator. Subsequently, the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link, and so on, until the network slice configuration operation of the data monitoring indicator in the communication link and several diversion communication links is completed.
[0010] Furthermore, the process of comparing the real-time production data information in the current collection cycle with the data volume prediction value of each data monitoring indicator, and generating the traffic abnormality alarm information or the network slice ultra-short-term dynamic adjustment operation of the data monitoring indicator according to the comparison result includes: Obtain the real-time production data information generated by each industrial production equipment at the current moment in the current collection cycle collected by each data monitoring point, obtain the real-time data volume of each data monitoring indicator generated by each industrial production equipment according to the real-time production data information generated by each industrial production equipment, and compare the real-time data volume of each data monitoring indicator at the current moment in the current collection cycle with the data volume prediction value of each data monitoring indicator; If the real-time data volume of a data monitoring indicator is greater than the corresponding data volume prediction value, a preset error threshold is used to obtain the data volume difference between the real-time data volume of the data monitoring indicator and the data volume prediction value, and the data volume difference is compared with the error threshold. If the data volume difference is greater than the error threshold, flow abnormality alarm information of the industrial production equipment to which the data monitoring indicator belongs is generated and fed back to the monitoring center; If the data volume difference is less than or equal to the error threshold, an ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator is performed.
[0011] Furthermore, the process of performing ultra-short-term dynamic adjustment of network slices of data monitoring indicators includes: Obtain the time requirement of each data monitoring indicator, preset the network slice calling time of each data monitoring indicator, the initial value of the network slice calling time is zero, generate the dynamic priority of each data monitoring indicator according to the time requirement of each data monitoring indicator and the network slice calling time, sort each data monitoring indicator in positive order according to the dynamic priority of each data monitoring indicator, and build a dynamic queue to be adjusted, and the dynamic queue to be adjusted updates the order of each data monitoring indicator in the dynamic queue to be adjusted in real time according to the dynamic priority of each data monitoring indicator; Obtain the data volume difference between the data monitoring indicator and the data volume prediction value, obtain the data monitoring indicator at the top of the dynamic queue to be adjusted at the current moment, obtain the network slice of the data monitoring indicator in the communication link and the link bandwidth allocated to the network slice, and determine whether the link bandwidth is greater than or equal to the data volume difference. If less than, filter out the data monitoring indicator of the number of dynamic queues to be adjusted, and so on, until the data monitoring indicator whose link bandwidth in the network slice allocated to the data monitoring indicator in the dynamic queue to be adjusted is greater than or equal to the data volume difference is filtered out, and the data monitoring indicator is marked as the data monitoring indicator to be called; The network slice of the data monitoring indicator to be called in the communication link is called to the data monitoring indicator, and the time when the network slice of the data monitoring indicator to be called in the communication link is called to the network slice of the data monitoring indicator is recorded.
[0012] Furthermore, when the ranking of each data monitoring indicator in the dynamic queue to be adjusted is updated, the network slice ultra-short-term dynamic adjustment operation of the data monitoring indicator is performed again.
[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. According to the time requirements of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, complete the network slicing configuration operation of each data monitoring indicator in the communication link. By dividing the data monitoring indicators with high timeliness into the communication links with high performance configuration, it is ensured that these data can be transmitted with the lowest delay, so as to realize real-time monitoring and control of the production process, which helps to detect abnormal situations in time and take measures to ensure the stability and safety of production. Dividing the data monitoring indicators with low timeliness requirements into the communication links with low performance configuration can manage bandwidth resources more flexibly, give priority to ensuring the bandwidth required for real-time monitoring and control, and will not be affected by other data congestion. Through network slicing configuration, different data monitoring indicators can be treated differently to avoid data congestion or delay affecting the transmission of key data, significantly improve the stability and reliability of the system, and ensure the normal operation of industrial production equipment.
[0014] 2. Generate the dynamic priority of each data monitoring indicator according to the time required by each data monitoring indicator and the time when the network slice is called. When the time when the network slice of each data monitoring indicator is called is zero, the longer the time required by each data monitoring indicator is, the higher the dynamic priority is. That is, the network slice of the data monitoring indicator with the lower time requirement is easier to be called. When the network slice of the data monitoring indicator is called, as the time when the network slice of the data monitoring indicator is called increases, the dynamic priority of the data monitoring indicator gradually decreases. When the dynamic priority of the data monitoring indicator is lower than that of other queues in the dynamic queue to be adjusted, the dynamic queue to be adjusted updates the order of the data monitoring indicators in the queue. The called network slice is converted to call the new network slice of the data monitoring indicator with higher dynamic priority, which avoids the original network slice being called all the time, resulting in the data monitoring indicator to which the original network slice belongs not transmitting data for a long time. It ensures that even if the network slice of the data monitoring indicator with low timeliness requirement is called by the data monitoring indicator with high timeliness requirement, when the data monitoring indicator with low timeliness requirement waits for data transmission long enough, its dynamic priority can be reduced to a very low level, thereby obtaining the opportunity for the data monitoring indicator with low timeliness requirement to transmit data. By dynamically adjusting the configuration strategy of the network slice to adapt to the actual needs of different data monitoring indicators, the efficiency and performance of network link data transmission are improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 This is a schematic diagram of a network slice configuration method based on traffic analysis in an embodiment of the present application. DETAILED DESCRIPTION
[0016] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.
[0017] like Figure 1 As shown, a network slice configuration method based on traffic analysis includes the following steps: Step S1: Setting data monitoring points for each industrial production equipment in the target industrial Internet of Things to obtain production data information generated by the industrial production equipment; Step S2: construct a data prediction model, and obtain the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data prediction model; Step S3: According to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, complete the network slicing configuration operation of each data monitoring indicator in the communication link; Step S4: Compare the real-time production data information in the current collection cycle with the data volume prediction value of each data monitoring indicator, and generate traffic anomaly alarm information or perform ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator based on the comparison result.
[0018] It should be further explained that, in the specific implementation process, data monitoring points are set for each industrial production equipment in the target industrial Internet of Things, and the process of obtaining production data information generated by the industrial production equipment includes: Several industrial production equipment in the target industrial Internet of Things are connected to the monitoring center through a communication link, and data monitoring points are set for each industrial production equipment in the target industrial Internet of Things. The data monitoring points are used to collect production data information generated by the industrial production equipment to which they belong, and mark the collection cycle. The production data information includes several data monitoring indicators corresponding to each industrial production equipment and the data volume corresponding to each data monitoring indicator.
[0019] It should be further explained that, in the present invention, industrial production equipment in the industrial Internet of Things refers to various devices and machines connected and managed through the Internet of Things technology in an industrial production environment. These devices are usually used to monitor, control, optimize and automate industrial production processes to improve production efficiency, reduce costs, and improve product quality. Industrial production equipment may include various types of mechanical equipment, sensors, actuators, controllers, etc., and their functions and uses are different; The data monitoring indicators of industrial production equipment may vary depending on the specific equipment type, production process and application scenario. In the present invention, the data monitoring indicators of industrial production equipment include but are not limited to: temperature, pressure, humidity, current and voltage, production rate, environmental parameters, etc.
[0020] It should be further explained that, in the specific implementation process, the process of constructing a data prediction model and obtaining the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data prediction model includes: Construct a data prediction model based on deep learning, obtain historical production data information generated by each industrial production equipment in several historical collection cycles, use the historical production data information generated by each industrial production equipment in several historical collection cycles as a training set and a test set, input the training set into the data prediction model for training until the loss function training is stable, and save the model parameters, test the data prediction model through the test set until it meets the preset requirements, and output the data prediction model; The production data prediction information of each industrial production equipment in the current collection cycle is obtained according to the data prediction model, and the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle is obtained according to the production data prediction information.
[0021] It should be further explained that, in the specific implementation process, according to the timeliness requirement time of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, the process of completing the network slicing configuration operation of each data monitoring indicator in the communication link includes: obtaining all communication links between each industrial production equipment and the monitoring center, obtaining the performance configuration parameters of each communication link, the performance configuration parameters including the link bandwidth and the average data transmission rate, sorting each communication link in positive order according to the average data transmission rate of each communication link, generating a communication link rate queue, the faster the average data transmission rate of the communication link, the higher the ranking of the communication link in the communication link rate queue; Preset the time requirement of each data monitoring indicator corresponding to each industrial production equipment, sort each data monitoring indicator in reverse order according to the time requirement of each data monitoring indicator corresponding to each industrial production equipment, and generate a data monitoring indicator time requirement queue. The shorter the time requirement of the data monitoring indicator, the higher the ranking of the data monitoring indicator in the data monitoring indicator time requirement queue; Construct an indicator traversal pointer and a link traversal pointer, wherein the indicator traversal pointer is used to traverse and retrieve the data monitoring indicator timeliness queue in the order from the first position of the queue to the last position of the queue, and the link traversal pointer is used to traverse and retrieve the communication link rate queue in the order from the first position of the queue to the last position of the queue; First, the indicator traversal pointer performs a traversal search on the data monitoring indicator timeliness queue. When the indicator traversal pointer retrieves the data monitoring indicator, it generates an indicator configuration instruction for the data monitoring indicator, suspends the data monitoring indicator traversal search, and sends the indicator configuration instruction to the link traversal pointer. When the link traversal pointer receives the indicator configuration instruction, the link traversal pointer performs a communication link traversal search on the communication link rate queue. When the link traversal pointer retrieves the communication link, the link bandwidth of the communication link is configured for a network slice according to the indicator configuration instruction. After completing the network slicing configuration operation of the data monitoring indicator corresponding to the indicator configuration instruction, if the remaining link bandwidth of the communication link is zero, the communication link is removed from the communication link rate queue, and then the communication link traversal search is suspended, and an indicator configuration completion instruction is sent to the indicator traversal pointer. After the indicator traversal pointer receives the indicator configuration completion instruction, the data monitoring indicator is removed from the data monitoring indicator timeliness queue, and the data monitoring indicator traversal search is resumed, and so on, until the network slicing configuration operation of all data monitoring indicators in the data monitoring indicator timeliness queue is completed.
[0022] By dividing the data monitoring indicators with high timeliness (such as temperature, pressure, etc.) into high-performance communication links, it is ensured that these data can be transmitted with the lowest delay, so as to achieve real-time monitoring and control of the production process, which helps to detect abnormal situations in time and take measures to ensure the stability and safety of production; By assigning data monitoring indicators that do not require high timeliness (such as energy consumption, production rate, etc.) to low-performance communication links, bandwidth resources can be managed more flexibly, giving priority to bandwidth required for real-time monitoring and control without being affected by other data congestion; It should be further explained that, in a specific implementation process, when the link traversal pointer retrieves a communication link, the process of performing a network slicing configuration operation on the link bandwidth of the communication link according to the indicator configuration instruction includes: According to the indicator configuration instruction, the data monitoring indicator that needs to perform the network slice configuration operation is obtained, the data volume prediction value of the data monitoring indicator in the current collection cycle is obtained, and the link bandwidth occupied by the data volume prediction value is obtained, and at the same time, the link bandwidth of the communication link retrieved by the link traversal pointer is obtained, and the link bandwidth of the communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is compared with the remaining link bandwidth of the communication link; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is less than or equal to the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and a link bandwidth consistent with the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is allocated from the remaining link bandwidth of the communication link to the network slice, thereby completing the network slice configuration operation of the data monitoring indicator; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is greater than the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and the remaining link bandwidth of the communication link is allocated to the network slice of the data monitoring indicator, and the remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle and the link bandwidth in the network slice allocated to the data monitoring indicator, and the link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue, and when the link traversal pointer retrieves another communication link, the other communication link is marked as a diversion communication link of the data monitoring indicator, and the link bandwidth of the diversion communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link; If the remaining link bandwidth to be allocated is less than or equal to the remaining link bandwidth of the offload communication link, a network slice of the data monitoring indicator in the offload communication link is constructed, and a link bandwidth consistent with the remaining link bandwidth to be allocated from the remaining link bandwidth of the offload communication link is allocated to the network slice of the data monitoring indicator, thereby completing the network slice configuration operation of the data monitoring indicator in the communication link and the offload communication link; If the remaining link bandwidth to be allocated is greater than the remaining link bandwidth of the diversion communication link, a network slice of the data monitoring indicator in the diversion communication link is constructed, and the remaining link bandwidth of the diversion communication link is allocated to the network slice. The remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the predicted value of the data volume of the data monitoring indicator in the current collection cycle and the link bandwidth allocated to the data monitoring indicator in the communication link and the network slice in the diversion communication link. The link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue. When the link traversal pointer retrieves other communication links, the other retrieved communication links are also marked as diversion communication links of the data monitoring indicator. Subsequently, the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link, and so on, until the network slice configuration operation of the data monitoring indicator in the communication link and several diversion communication links is completed.
[0023] Through network slicing configuration, differentiated processing of different data monitoring indicators can be achieved to avoid data congestion or delays affecting the transmission of key data. This helps to improve the stability and reliability of the system and ensure the normal operation of industrial production equipment.
[0024] It should be further explained that, in the specific implementation process, the real-time production data information in the current collection cycle is compared with the data volume prediction value of each data monitoring indicator, and the process of generating traffic abnormality alarm information or network slice ultra-short-term dynamic adjustment operation of the data monitoring indicator according to the comparison result includes: Obtain the real-time production data information generated by each industrial production equipment at the current moment in the current collection cycle collected by each data monitoring point, obtain the real-time data volume of each data monitoring indicator generated by each industrial production equipment according to the real-time production data information generated by each industrial production equipment, and compare the real-time data volume of each data monitoring indicator at the current moment in the current collection cycle with the data volume prediction value of each data monitoring indicator; If the real-time data volume of a data monitoring indicator is greater than the corresponding data volume prediction value, a preset error threshold is used to obtain the data volume difference between the real-time data volume of the data monitoring indicator and the data volume prediction value, and the data volume difference is compared with the error threshold. If the data volume difference is greater than the error threshold, flow abnormality alarm information of the industrial production equipment to which the data monitoring indicator belongs is generated and fed back to the monitoring center; If the data volume difference is less than or equal to the error threshold, an ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator is performed.
[0025] It should be further explained that, in the specific implementation process, the process of performing the ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator includes: Obtain the time requirement of each data monitoring indicator, preset the network slice calling time of each data monitoring indicator, the initial value of the network slice calling time is zero, generate the dynamic priority of each data monitoring indicator according to the time requirement of each data monitoring indicator and the network slice calling time, sort each data monitoring indicator in positive order according to the dynamic priority of each data monitoring indicator, and build a dynamic queue to be adjusted, and the dynamic queue to be adjusted updates the order of each data monitoring indicator in the dynamic queue to be adjusted in real time according to the dynamic priority of each data monitoring indicator; Obtain the data volume difference between the data monitoring indicator and the data volume prediction value, obtain the data monitoring indicator at the top of the dynamic queue to be adjusted at the current moment, obtain the network slice of the data monitoring indicator in the communication link and the link bandwidth allocated to the network slice, and determine whether the link bandwidth is greater than or equal to the data volume difference. If less than, filter out the data monitoring indicator of the number of dynamic queues to be adjusted, and so on, until the data monitoring indicator whose link bandwidth in the network slice allocated to the data monitoring indicator in the dynamic queue to be adjusted is greater than or equal to the data volume difference is filtered out, and the data monitoring indicator is marked as the data monitoring indicator to be called; The network slice of the data monitoring indicator to be called in the communication link is called to the data monitoring indicator, and the time when the network slice of the data monitoring indicator to be called in the communication link is called to the network slice of the data monitoring indicator is recorded.
[0026] It should be further explained that, in the specific implementation process, the formula for generating the dynamic priority of each data monitoring indicator according to the timeliness requirement time of each data monitoring indicator and the time when the network slice is called is:
[0027] in, Indicates the dynamic priority of the i-th category data monitoring indicator, Indicates the timeliness requirement of the i-th type of data monitoring indicator, Indicates the time when the network slice of the i-th type of data monitoring indicator is called, Represents a non-zero bias coefficient; when the network slices of the data monitoring indicators are called at zero time, the longer the timeliness requirement of each data monitoring indicator, the higher the dynamic priority, that is, the network slice of the data monitoring indicator with the lower timeliness requirement is easier to be called; when the network slice of the data monitoring indicator is called, as the calling time of the network slice of the data monitoring indicator increases, the dynamic priority of the data monitoring indicator gradually decreases. When the dynamic priority of the data monitoring indicator is lower than that of other queues in the dynamic waiting queue, the dynamic waiting queue updates the order of the data monitoring indicators in the queue, and the original called network slice is converted to call the new dynamic priority The network slices with higher-level data monitoring indicators avoid the original network slices being called all the time, resulting in the data monitoring indicators to which the original network slices belonged not transmitting data for a long time. This ensures that even if the network slices with data monitoring indicators with low timeliness requirements are called by data monitoring indicators with high timeliness requirements, when the data monitoring indicators with low timeliness requirements wait for data transmission long enough, their dynamic priority can be reduced to a very low level, thereby obtaining the opportunity for data monitoring indicators with low timeliness requirements to transmit data. By dynamically adjusting the configuration strategy of network slices to adapt to the actual needs of different data monitoring indicators, the efficiency and performance of network link data transmission are improved.
[0028] It should be further explained that, in the specific implementation process, when the ranking of each data monitoring indicator in the dynamic queue to be adjusted is updated, the network slicing ultra-short-term dynamic adjustment operation of the data monitoring indicator is performed again.
[0029] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A network slice configuration method based on traffic analysis, characterized in that: The following steps are involved: Step S1: Setting data monitoring points for each industrial production equipment in the target industrial Internet of Things to obtain production data information generated by the industrial production equipment; Step S2: construct a data prediction model, and obtain the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data prediction model; Step S3: According to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, complete the network slicing configuration operation of each data monitoring indicator in the communication link; Step S4: Compare the real-time production data information in the current collection cycle with the data volume prediction value of each data monitoring indicator, and generate traffic anomaly alarm information or perform ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator based on the comparison result.
2. According to the method for configuring a network slice based on traffic analysis in claim 1, it is characterized in that: The process of setting data monitoring points for each industrial production equipment in the target industrial Internet of Things and obtaining the production data information generated by the industrial production equipment includes: Several industrial production equipment in the target industrial Internet of Things are connected to the monitoring center through a communication link, and data monitoring points are set for each industrial production equipment in the target industrial Internet of Things. The data monitoring points are used to collect production data information generated by the industrial production equipment to which they belong, and mark the collection cycle. The production data information includes several data monitoring indicators corresponding to each industrial production equipment and the data volume corresponding to each data monitoring indicator.
3. According to the method for configuring a network slice based on traffic analysis in claim 2, it is characterized in that: The process of obtaining the data volume forecast value of each data monitoring indicator of each industrial production equipment in the current collection cycle according to the data forecast model includes: Construct a data prediction model based on deep learning, obtain historical production data information generated by each industrial production equipment in several historical collection cycles, use the historical production data information generated by each industrial production equipment in several historical collection cycles as a training set and a test set, input the training set into the data prediction model for training until the loss function training is stable, and save the model parameters, test the data prediction model through the test set until it meets the preset requirements, and output the data prediction model; The production data prediction information of each industrial production equipment in the current collection cycle is obtained according to the data prediction model, and the data volume prediction value of each data monitoring indicator of each industrial production equipment in the current collection cycle is obtained according to the production data prediction information.
4. According to the method for configuring network slices based on traffic analysis in claim 3, it is characterized in that: According to the timeliness requirements of each data monitoring indicator corresponding to each industrial production equipment and the performance configuration parameters of each communication link between each industrial production equipment and the monitoring center, the process of completing the network slice configuration operation of each data monitoring indicator in the communication link includes: Acquire all communication links between each industrial production equipment and the monitoring center, acquire performance configuration parameters of each communication link, the performance configuration parameters including link bandwidth and average data transmission rate, sort each communication link in positive order according to the average data transmission rate of each communication link, and generate a communication link rate queue; Preset the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment, sort each data monitoring indicator in reverse order according to the timeliness requirement of each data monitoring indicator corresponding to each industrial production equipment, and generate a data monitoring indicator timeliness queue; Construct an indicator traversal pointer and a link traversal pointer, wherein the indicator traversal pointer is used to traverse and retrieve the data monitoring indicator timeliness queue in the order from the first position of the queue to the last position of the queue, and the link traversal pointer is used to traverse and retrieve the communication link rate queue in the order from the first position of the queue to the last position of the queue; First, the indicator traversal pointer performs a traversal search on the data monitoring indicator timeliness queue. When the indicator traversal pointer retrieves the data monitoring indicator, it generates an indicator configuration instruction for the data monitoring indicator, suspends the data monitoring indicator traversal search, and sends the indicator configuration instruction to the link traversal pointer. When the link traversal pointer receives the indicator configuration instruction, the link traversal pointer performs a communication link traversal search on the communication link rate queue. When the link traversal pointer retrieves the communication link, the link bandwidth of the communication link is configured for a network slice according to the indicator configuration instruction. After completing the network slicing configuration operation of the data monitoring indicator corresponding to the indicator configuration instruction, if the remaining link bandwidth of the communication link is zero, the communication link is removed from the communication link rate queue, and then the communication link traversal search is suspended, and an indicator configuration completion instruction is sent to the indicator traversal pointer. After the indicator traversal pointer receives the indicator configuration completion instruction, the data monitoring indicator is removed from the data monitoring indicator timeliness queue, and the data monitoring indicator traversal search is resumed, and so on, until the network slicing configuration operation of all data monitoring indicators in the data monitoring indicator timeliness queue is completed.
5. According to the method for configuring network slices based on traffic analysis in claim 4, it is characterized in that: When the link traversal pointer retrieves a communication link, the process of performing a network slice configuration operation on the link bandwidth of the communication link according to the indicator configuration instruction includes: According to the indicator configuration instruction, the data monitoring indicator that needs to perform the network slice configuration operation is obtained, the data volume prediction value of the data monitoring indicator in the current collection cycle is obtained, and the link bandwidth occupied by the data volume prediction value is obtained, and at the same time, the link bandwidth of the communication link retrieved by the link traversal pointer is obtained, and the link bandwidth of the communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is compared with the remaining link bandwidth of the communication link; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is less than or equal to the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and a link bandwidth consistent with the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is allocated from the remaining link bandwidth of the communication link to the network slice, thereby completing the network slice configuration operation of the data monitoring indicator; If the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle is greater than the remaining link bandwidth of the communication link, a network slice of the data monitoring indicator in the communication link is constructed, and the remaining link bandwidth of the communication link is allocated to the network slice of the data monitoring indicator, and the remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the data volume prediction value of the data monitoring indicator in the current collection cycle and the link bandwidth in the network slice allocated to the data monitoring indicator, and the link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue, and when the link traversal pointer retrieves another communication link, the other communication link is marked as a diversion communication link of the data monitoring indicator, and the link bandwidth of the diversion communication link that has been allocated to the network slice and the remaining link bandwidth that has not been allocated to the network slice are obtained, and the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link; If the remaining link bandwidth to be allocated is less than or equal to the remaining link bandwidth of the offload communication link, a network slice of the data monitoring indicator in the offload communication link is constructed, and a link bandwidth consistent with the remaining link bandwidth to be allocated from the remaining link bandwidth of the offload communication link is allocated to the network slice of the data monitoring indicator, thereby completing the network slice configuration operation of the data monitoring indicator in the communication link and the offload communication link; If the remaining link bandwidth to be allocated is greater than the remaining link bandwidth of the diversion communication link, a network slice of the data monitoring indicator in the diversion communication link is constructed, and the remaining link bandwidth of the diversion communication link is allocated to the network slice. The remaining link bandwidth to be allocated of the data monitoring indicator is obtained according to the link bandwidth occupied by the predicted value of the data volume of the data monitoring indicator in the current collection cycle and the link bandwidth allocated to the data monitoring indicator in the communication link and the network slice in the diversion communication link. The link traversal pointer continues to perform communication link traversal retrieval on the communication link rate queue. When the link traversal pointer retrieves other communication links, the other retrieved communication links are also marked as diversion communication links of the data monitoring indicator. Subsequently, the remaining link bandwidth to be allocated of the data monitoring indicator is compared with the remaining link bandwidth of the diversion communication link, and so on, until the network slice configuration operation of the data monitoring indicator in the communication link and several diversion communication links is completed.
6. A network slice configuration method based on traffic analysis according to claim 5, characterized in that: The process of comparing the real-time production data information in the current collection cycle with the data volume prediction value of each data monitoring indicator, and generating traffic anomaly alarm information or network slice ultra-short-term dynamic adjustment operation of the data monitoring indicator according to the comparison result includes: Obtain the real-time production data information generated by each industrial production equipment at the current moment in the current collection cycle collected by each data monitoring point, obtain the real-time data volume of each data monitoring indicator generated by each industrial production equipment according to the real-time production data information generated by each industrial production equipment, and compare the real-time data volume of each data monitoring indicator at the current moment in the current collection cycle with the data volume prediction value of each data monitoring indicator; If the real-time data volume of a data monitoring indicator is greater than the corresponding data volume prediction value, a preset error threshold is used to obtain the data volume difference between the real-time data volume of the data monitoring indicator and the data volume prediction value, and the data volume difference is compared with the error threshold. If the data volume difference is greater than the error threshold, flow abnormality alarm information of the industrial production equipment to which the data monitoring indicator belongs is generated and fed back to the monitoring center; If the data volume difference is less than or equal to the error threshold, an ultra-short-term dynamic adjustment operation of the network slice of the data monitoring indicator is performed.
7. A network slice configuration method based on traffic analysis according to claim 6, characterized in that: The process of performing ultra-short-term dynamic adjustment of network slices for data monitoring indicators includes: Obtain the time requirement of each data monitoring indicator, preset the network slice calling time of each data monitoring indicator, the initial value of the network slice calling time is zero, generate the dynamic priority of each data monitoring indicator according to the time requirement of each data monitoring indicator and the network slice calling time, sort each data monitoring indicator in positive order according to the dynamic priority of each data monitoring indicator, and build a dynamic queue to be adjusted, and the dynamic queue to be adjusted updates the order of each data monitoring indicator in the dynamic queue to be adjusted in real time according to the dynamic priority of each data monitoring indicator; Obtain the data volume difference between the data monitoring indicator and the data volume prediction value, obtain the data monitoring indicator at the top of the dynamic queue to be adjusted at the current moment, obtain the network slice of the data monitoring indicator in the communication link and the link bandwidth allocated to the network slice, and determine whether the link bandwidth is greater than or equal to the data volume difference. If less than, filter out the data monitoring indicator of the number of dynamic queues to be adjusted, and so on, until the data monitoring indicator whose link bandwidth in the network slice allocated to the data monitoring indicator in the dynamic queue to be adjusted is greater than or equal to the data volume difference is filtered out, and the data monitoring indicator is marked as the data monitoring indicator to be called; The network slice of the data monitoring indicator to be called in the communication link is called to the data monitoring indicator, and the time when the network slice of the data monitoring indicator to be called in the communication link is called to the network slice of the data monitoring indicator is recorded.
8. The method for configuring a network slice based on traffic analysis according to claim 7, characterized in that: When the ranking of each data monitoring indicator in the dynamic queue to be adjusted is updated, the network slice ultra-short-term dynamic adjustment operation of the data monitoring indicator is performed again.
