An Industrial Internet Protocol-based Middleware Communication Optimization Method and System

Abstract

The present invention provides a middleware communication optimization method and system based on industrial Internet protocols, which relates to the technical field of communication optimization. The method includes: obtaining historical packet characteristic record data and historical packet transmission record data of multiple data sources, constructing a data characteristic matrix for each data source, and a feature mapping matrix for each transmission protocol; performing time period data fluctuation characteristic analysis and time period state resistance analysis on the data characteristic matrix to generate a data feature protocol optimization scheme for each data source; extracting historical communication quality record data from the historical packet transmission record data to generate a communication quality characteristic matrix for each transmission protocol; optimizing the data feature protocol optimization scheme for each data source to generate a target communication protocol optimization scheme for each data source, and performing communication optimization on the middleware based on the target communication protocol optimization scheme. The present invention realizes the communication optimization of the middleware in the industrial Internet architecture.

Description

Technical Field

[0001] The present invention relates to the technical field of communication optimization, and particularly to a middleware communication optimization method and system based on industrial Internet protocols. Background Art

[0002] In the industrial Internet environment, the stability, reliability, and real-time performance of data transmission are the core elements to ensure production efficiency and safety. With the improvement of industrial automation and intelligence, a large number of data sources send data in real time through sensors, devices, and control systems, requiring the communication protocol to efficiently adapt to different network qualities and continuously changing data loads. In a modern industrial Internet architecture, middleware, as a bridge connecting different systems, devices, and protocols, plays a crucial role. Middleware is responsible for processing data flows from various devices such as PLCs, sensors, intelligent devices, etc. and control systems such as MES, ERP, etc., ensuring the transmission and coordination of data between different levels and different protocols.

[0003] In terms of communication optimization from the perspective of protocol adaptation, some statically configured protocol adaptation methods are usually pre-set in terms of protocol selection and data transmission strategies, without considering dynamic network quality changes and data flow fluctuations, and it is difficult to respond to network fluctuations and data anomalies in actual operation. Dynamic protocol optimization schemes will analyze from perspectives such as network quality and data source characteristics, and there are complex interaction effects between these different perspectives. For example, the adaptability of the protocol may be affected by factors such as network latency and packet loss rate. Comprehensive analysis from different perspectives and in-depth consideration of the interaction of data characteristics under different perspectives can effectively evaluate and optimize the communication protocol comprehensively, generate middleware communication optimization strategies that can flexibly respond to the complex and changeable network and data environment in the industrial Internet, and adapt to the requirements of different industrial application environments. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a middleware communication optimization method and system based on industrial Internet protocols, which can generate middleware communication optimization strategies that can accurately adapt to the requirements of different industrial application environments by analyzing the characteristic changes of data packets sent by data sources and network quality changes, and flexibly respond to the challenges brought by the complex and changeable network and data environment in the industrial Internet to the data communication of middleware.

[0005] To achieve the above object, the first aspect of the present invention provides a middleware communication optimization method based on industrial Internet protocols, including:

[0006] Obtain the historical data packet characteristic record data and historical data packet transmission record data of multiple data sources, extract multiple groups of data packet characteristic parameters from each group of historical data packet characteristic record data, and construct a data characteristic matrix for each data source. Extract the transmission characteristic quality correlation data corresponding to various transmission protocols from each group of historical data packet transmission record data, including the transmission quality record data of data packets under different communication level data, and construct a feature mapping matrix for each transmission protocol regarding data characteristics and communication quality;

[0007] Determine the data reference state vector of each data characteristic matrix, perform time period data fluctuation characteristic analysis and time period state resistance analysis on the data characteristic matrix based on the data reference state vector, construct a time period stability characteristic vector and a time period state resistance vector for each data source, and generate a data characteristic protocol optimization plan for each data source;

[0008] Extract the historical communication quality record data used by the middleware for data communication from the historical data packet transmission record data, and analyze the historical communication quality record data based on the data reference state vector and the feature mapping matrix to generate a communication quality characteristic matrix for each transmission protocol;

[0009] Optimize the data characteristic protocol optimization plan of each data source based on the communication quality characteristic matrix to generate a target communication protocol optimization plan for each data source, and perform communication optimization on the middleware based on the target communication protocol optimization plan.

[0010] Preferably, for the time period stability characteristic vector and the time period state resistance vector of each data source, including:

[0011] Extract the local data state vector under each local time period from the data characteristic matrix of the data source, perform data fluctuation characteristic analysis on the local data state vector under each local time period through the data reference state vector respectively, obtain the data state difference parameters corresponding to the data source under multiple local time periods, and fuse the multiple data state difference parameters to construct the time period stability characteristic vector of the data source. The time period stability characteristic vector includes the global state difference parameters under each global reference time period;

[0012] Perform time period state resistance analysis on the data characteristic matrix through multiple data state difference parameters, including calculating the state change rate of each local time period based on the data state difference parameters and determining multiple local abnormal time periods, determining multiple state abnormal events of the data source according to the multiple local abnormal time periods, and using the following formula to calculate the local state resistance parameters of each state abnormal event:

[0013] ;

[0014] In the formula, The local state resistance parameter representing the state exception event, The maximum value of the data state difference parameters in multiple local exception periods of the state exception event, Indicates the State change rate of the nth local exception period in the state exception event, Indicates the total duration of the state exception event,

[0015] Using the local state resistance parameter of the state exception event as the local state resistance parameter for each local period under the state exception event, fusing the local state resistance parameters corresponding to multiple local periods included in the global reference period to generate the global state resistance parameter for each global reference period, and constructing a period state resistance vector containing the global state resistance parameters for each global reference period.

[0016] Preferably, for the communication quality characteristic matrix of the transport protocol, it further includes:

[0017] Extracting the local communication state vector for each local period from the historical communication quality record data, using the data reference state vector as the data reference state vector for each local period, analyzing the transmission quality of the target transport protocol in multiple local periods through the feature mapping matrix, analyzing the reference transmission quality parameters of the target transport protocol in each local period based on the preset transmission reference standard of the data source, and constructing the communication quality characteristic matrix of the data source for the target transport protocol according to the reference transmission quality parameters corresponding to multiple local periods.

[0018] Preferably, for the data characteristic protocol optimization scheme of the data source, it further includes:

[0019] Calculating the global data exception characteristic parameter for each global reference period through the period stability characteristic vector and the period state resistance vector of the data source, and calculating the local exception characteristic parameter for each local period according to the data state difference parameter and the local state resistance parameter of each local period included in each global reference period;

[0020] Extracting the transport protocol adopted in each local period from the historical data packet transmission record data of the data source, constructing the abnormal reference range for each transport protocol in each global reference period, determining the reference abnormal characteristic parameter for each transport protocol in each global reference period according to multiple abnormal reference ranges, and determining the protocol candidate list for each global reference period according to the global data exception characteristic parameter of the global reference period, to obtain the data characteristic protocol optimization scheme of the data source including the protocol candidate list for each global reference period.

[0021] Preferably, optimize the data feature protocol optimization scheme for each data source based on the communication quality characteristic matrix to generate the target communication protocol optimization scheme for each data source, including:

[0022] Extract the communication quality reference feature vectors of each transmission protocol in the global reference period from the communication quality characteristic matrix of the transmission protocol, extract the historical transmission protocols actually adopted in each local period according to the historical data packet transmission record data of the data source, and determine the real-time matching parameters of each transmission protocol in each local period according to the historical transmission protocols actually adopted in each local period, and construct the real-time matching feature vectors of each transmission protocol;

[0023] Perform protocol timing adaptation analysis on the protocol candidate list in the global reference period according to the reference communication quality feature vector and the real-time matching feature vector, and calculate the protocol timing adaptation parameters of each transmission protocol in the global reference period by using the following formula:

[0024] ;

[0025] In the formula, represents the protocol timing adaptation parameter, , respectively represent the local adaptation parameters of the transmission protocol in the th and th local periods, is the number of local periods, represents the timing optimization factor, where the local adaptation parameter of the local period is calculated according to the reference transmission quality parameter and the real-time matching parameter of the transmission protocol in the local period;

[0026] After calculating the protocol timing adaptation parameters of each transmission protocol in the global reference period, optimize the protocol candidate list in the global reference period based on the protocol timing adaptation parameters to generate a protocol adaptation correction list, and determine the target communication protocol optimization scheme of the data source according to the protocol adaptation correction list.

[0027] The second aspect of the present invention provides a middleware communication optimization system based on the industrial Internet protocol for implementing the above-mentioned middleware communication optimization method based on the industrial Internet protocol, including:

[0028] A data acquisition module for obtaining the historical data packet characteristic record data and historical data packet transmission record data of multiple data sources;

[0029] A data characteristic analysis module for extracting multiple groups of data packet characteristic parameters from each group of historical data packet characteristic record data and constructing the data characteristic matrix of each data source;

[0030] A transmission characteristic quality correlation analysis module, which is used to extract transmission characteristic quality correlation data corresponding to multiple transmission protocols from each set of historical data packet transmission record data, including communication quality record data of data packets under different communication level data, and construct a characteristic mapping matrix of each transmission protocol regarding data characteristics and communication quality;

[0031] A data feature optimization strategy generation module, which is used to determine the data reference state vector of each data feature matrix, perform time period data fluctuation characteristic analysis and time period state resistance analysis on the data feature matrix based on the data reference state vector, construct a time period stability characteristic vector and a time period state resistance vector for each data source, and generate a data feature protocol optimization plan for each data source;

[0032] A communication quality analysis module, which is used to extract historical communication quality record data of middleware for data communication from historical data packet transmission record data, and analyze the historical communication quality record data based on the data reference state vector and the characteristic mapping matrix to generate a communication quality characteristic matrix for each transmission protocol;

[0033] A data network fusion optimization module, which is used to optimize the data feature protocol optimization plan of each data source based on the communication quality characteristic matrix, generate a target communication protocol optimization plan for each data source, and perform communication optimization on the middleware based on the target communication protocol optimization plan.

[0034] Preferably, for the data feature optimization strategy generation module, constructing a time period stability characteristic vector and a time period state resistance vector for each data source includes:

[0035] Extracting a local data state vector under each local time period from the data feature matrix of the data source, performing data fluctuation characteristic analysis on the local data state vector under each local time period through the data reference state vector respectively, obtaining data state difference parameters corresponding to the data source under multiple local time periods respectively, fusing the multiple data state difference parameters to construct a time period stability characteristic vector of the data source, and the time period stability characteristic vector includes global state difference parameters under each global reference time period;

[0036] Performing time period state resistance analysis on the data feature matrix through multiple data state difference parameters, including calculating the state change rate of each local time period based on the data state difference parameters and determining multiple local abnormal time periods, determining multiple state abnormal events of the data source according to the multiple local abnormal time periods, and calculating the local state resistance parameter of each state abnormal event by using the following formula:

[0037] ;

[0038] In the formula, The local state resistance parameter indicating a state anomaly event, The maximum value of the data state difference parameters in multiple local anomaly time periods of the state anomaly event, Indicating the State change rate of the nth local anomaly time period in the state anomaly event, Indicating the total duration of the state anomaly event, Indicating the number of local anomaly time periods in the state anomaly event;

[0039] Take the local state resistance parameter of the state anomaly event as the local state resistance parameter of each local time period under the state anomaly event, fuse the local state resistance parameters corresponding to multiple local time periods included in the global reference time period to generate the global state resistance parameter of each global reference time period, and construct a time period state resistance vector containing the global state resistance parameter of each global reference time period.

[0040] The present invention has the following beneficial effects:

[0041] By analyzing the historical record data of the middleware, the present invention analyzes the data characteristics of the data source from the perspective of data source characteristics, constructs a data characteristic matrix, conducts time period data fluctuation characteristic analysis and time period state resistance analysis on the data source, combines the data fluctuation level and abnormal state resistance level of the data source to generate a data characteristic protocol optimization scheme, and determines the adaptability of different communication protocols under the data source characteristics; at the same time, analyzes the network state of the middleware for data transmission from the network quality, constructs a characteristic mapping matrix representing the association between data characteristics and communication quality, combines the changes in network quality, conducts transmission adaptation analysis on different communication protocols, and finally fuses the analysis results of network quality and data source characteristics to dynamically optimize the protocol selection strategy, ensuring that the protocol selection can not only adapt to the changes in network conditions, but also meet the transmission requirements of the data source, and has wide adaptability in actual industrial Internet applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic flow chart of a middleware communication optimization method based on industrial Internet protocol in an implementation process of the present invention.

[0043] Figure 2 It is a schematic structural diagram of a middleware communication optimization system based on industrial Internet protocol in an implementation process of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0044] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0045] Figure 1 It shows a schematic flow chart of a middleware communication optimization method based on industrial Internet protocol in an implementation process of the present invention.

[0046] Please refer to Figure 1 , a middleware communication optimization method based on industrial Internet protocol provided by the present invention can be applied to the middleware layer in the industrial Internet architecture, and is used to realize communication optimization of the middleware in the industrial Internet from the perspective of protocol adaptation. The method specifically includes the following steps:

[0047] Step S1: Obtain the historical packet characteristic record data and historical packet transmission record data of multiple data sources, extract multiple groups of packet characteristic parameters, and construct a data characteristic matrix for each data source. Extract the transmission characteristic quality correlation data corresponding to various transmission protocols from each group of historical packet transmission record data, and construct a characteristic mapping matrix of each transmission protocol regarding data characteristics and communication quality.

[0048] In the modern industrial Internet architecture, the middleware, as a bridge connecting different systems, devices, and protocols, plays a crucial role. Its main functions include data transmission and protocol management, etc., and it is responsible for transmitting data between devices, applications, and networks to ensure that data is not lost during the communication process.

[0049] In this embodiment, for different data sources such as sensor networks, production lines, or other production control systems, the corresponding historical packet characteristic record data and historical packet transmission record data contain some characteristics shown by the packets received by the middleware from the data source during the transmission process, such as packet size and quantity, packet loss situation, sending frequency, packet type, etc., which can provide important basis for the selection of communication protocols, and the relevant transmission information record data after these data are transmitted by the middleware to other network layers, such as the network conditions of transmission, the transmission protocols used, and the transmission quality performance of the packets after being transmitted, such as throughput, delay, packet loss rate, transmission success rate, etc.

[0050] Regarding the historical data packet characteristic record data of different data sources, considering that there may be differences in the data transmission requirements of different data sources, analyze the historical data packet characteristic record data of each data source, and extract some data packet characteristic parameters of the data packets received by the data source at different times within a certain period of history, such as size, quantity, type, etc., which are some key information affecting the selection of the transmission protocol. Specifically, taking a historical week as an example, each day can be divided into multiple local time periods, such as each hour as a local time period, and a set of data packet characteristic parameters corresponding to the data packets received in each local time period are extracted, so as to construct a data characteristic matrix of the data source within a certain period of history regarding the data packet characteristic information received in different time periods.

[0051] Regarding the historical data packet transmission record data of different data sources, for the transmission-related record data contained therein, extract the transmission characteristic quality correlation data corresponding to each transmission protocol. Specifically, in the industrial Internet, the network and data environment are complex and changeable. For different data sources and different network environments, the middleware needs to select an appropriate communication protocol for transmission according to the actual situation. For example, in some scenarios where low-latency transmission requirements are needed, communication protocols with relatively low latency such as QUIC or UDP can be selected, while in some scenarios where higher reliability, that is, the packet loss rate cannot be too high, communication protocols with higher reliability such as TCP can be selected. During the actual transmission process, the transmission quality is affected by multiple factors such as the characteristics of the data packets themselves and the network quality. Extract the transmission characteristic quality correlation data corresponding to multiple transmission protocols from each set of historical data packet transmission record data, that is, the transmission quality record data of specific data packets under different communication level data, such as for a certain type and size of data packet, in a network environment with specific bandwidth, latency, jitter, etc., the transmission quality information such as latency, packet loss rate, throughput, etc. recorded after the data packet finally uses a certain communication protocol for data transmission. Through these data, a characteristic mapping matrix of each transmission protocol regarding data characteristics and communication quality can be constructed, including information such as network status, data packet characteristics, and final transmission effects. Through the matrix, the transmission quality under different data characteristics and network conditions can be estimated, such as the successful transmission rate when using a certain protocol for data packet transmission at a specific data packet size and sending frequency, or the adaptability performance under different packet loss rates and latencies.

[0052] Step S2: Determine the data reference state vector of each data characteristic matrix, perform time period data fluctuation characteristic analysis and time period state resistance analysis on the data characteristic matrix based on the data reference state vector, construct the time period stability characteristic vector and time period state resistance vector of each data source, and generate the data characteristic protocol optimization plan for each data source.

[0053] In this embodiment, the data reference state vector is obtained by modeling the standard transmission situation of each data source, and is used to reflect the reference level of the data packets sent by the data source to the middleware under relatively ideal conditions. For example, information such as the average value of the data packet size, data type, data sending frequency, and standard value of the packet loss rate can be recorded according to the historical data packet characteristics of the data source, and the data packet characteristics in different time periods can be statistically analyzed. The data packet characteristics received at the normal level without abnormalities are used as the reference. In an actual industrial scenario, when a production anomaly occurs, such as a failure in a certain production line, the Internet of Things system will collect data at a higher frequency. After returning to normal, it will return to the original collection frequency. Then, during this period, the amount of data from the production system will be higher than the performance under normal production levels, and the amount of data sent to the middleware will show an abnormal increase.

[0054] Taking the data reference state vector as a reference under the normal production state, analyze the data fluctuation characteristics and time period state resistance of each time period in the data characteristic matrix. The purpose of the data fluctuation characteristics analysis is to identify the change situation of the data packet characteristic state received from a certain data source during the fluctuation time period. For example, identify the change range of the data volume in each time period, that is, whether there is an abnormal increase or decrease in the data, and quantify the impact of this fluctuation on protocol adaptation. By comparing the deviation between the data characteristics of each time period and the reference state vector, the fluctuation index of each time period is obtained, which reflects the degree of data flow fluctuation within the time period, and the time period stability characteristic vector of the data source is constructed.

[0055] The purpose of the time period state resistance analysis is to identify the recovery ability of the data source within this time period, that is, the ability of the system to return to the normal level after data anomalies such as an increase in data volume and an increase in delay occur. By analyzing information such as the highest data volume level within the time period and the duration required to return to the normal level, evaluate the resistance of the data source to data state anomalies. Finally, evaluate the recovery ability of the data source in different time periods by analyzing the recovery level shown by historical data, and construct the time period state resistance vector of the data source. Based on the time period stability characteristic vector and the time period state resistance vector of each data source, comprehensively analyze the stability of the data volume transmitted by the data source to the middleware and the recovery level in the face of abnormal states, evaluate the matching situation with multiple communication protocols in different time periods, and finally generate a data characteristic protocol optimization plan for the adaptation state between each data source and different communication protocols from the perspective of data source characteristics.

[0056] Step S3: Extract the historical communication quality record data used by the middleware for data communication from the historical data packet transmission record data, and analyze the historical communication quality record data based on the data reference state vector and the feature mapping matrix to generate the communication quality characteristic matrix of each transmission protocol.

[0057] In this embodiment, the historical communication quality record data includes the network status for data communication transmission at different time periods, reflecting the specific parameters of the communication network at different time periods during the actual communication process. Taking the packet status level represented by the data reference state vector as the reference data, and combining with the feature mapping matrix, it is analyzed whether the transmission requirements of the packets represented by the data reference state vector can be met under different network conditions, so as to obtain the transmission quality that can be achieved by using a certain protocol for transmission under the condition that the ideal packets do not show abnormal fluctuations when the transmission network conditions of each time period are known. Finally, the communication quality characteristic matrix of each transmission protocol, that is, the quantization table, is constructed to show the communication quality level theoretically demonstrated by using a certain protocol for data transmission under the actual network conditions and specific data to be transmitted.

[0058] In this embodiment, the process of analyzing the historical communication quality record data takes any one communication protocol as an example and is denoted as the target communication protocol, including extracting the local communication state vectors at each local time period from the historical communication quality record data, that is, representing the network state conditions of the local time period, and taking the data reference state vector as the data reference state vector at each local time period, that is, the packet characteristics. Combining with the feature mapping matrix determined in the previous step, according to the feature mapping matrix of the target communication protocol, the transmission effect under specific network states and packet characteristics is estimated, and the reference transmission quality parameters of the target transmission protocol at each local time period are analyzed according to the preset transmission reference standard of the data source. Among them, the preset transmission reference standard analysis can be a standard set for the data transmission quality of different data sources according to the actual situation, such as limiting specific information such as delay and packet loss rate. If the estimated transmission quality meets the preset conditions, the reference transmission quality parameter can be recorded as 1, indicating qualified, otherwise it is recorded as 0 indicating unqualified transmission quality. After determining the reference transmission quality parameters of each local time period in this way, the communication quality characteristic matrix of the target transmission protocol under the data source can be constructed according to the reference transmission quality parameters corresponding to multiple local time periods respectively. The analysis of the historical communication quality record data is realized to generate the communication quality characteristic matrix of each transmission protocol.

[0059] Step S4: Optimize the data characteristic protocol optimization scheme of each data source based on the communication quality characteristic matrix to generate the target communication protocol optimization scheme of each data source, and perform communication optimization on the middleware based on the target communication protocol optimization scheme.

[0060] In this embodiment, the data feature protocol optimization solution specifically analyzes the communication protocols adapted by data sources at different time periods from the perspective of packet feature changes. The communication quality characteristic matrix further considers the impact of actual network conditions on data communication. By organically integrating the analysis results from perspectives such as network quality and data source characteristics, a target communication protocol optimization solution for each data source is finally generated. Based on this solution, dynamic communication optimization of the middleware can be carried out to achieve flexible and intelligent protocol selection and efficient data transmission. Optimizing in this way enables the middleware to adapt to dynamically changing network conditions and data traffic requirements, ensures the best communication protocol selection in different industrial application scenarios, avoids resource waste at the same time, and improves network performance.

[0061] In an alternative implementation process, for step S2 above, in order to be able to select an appropriate transmission protocol, considering the characteristic changes of the data source, a comprehensive analysis of the time period data characteristics and time period state resistance of each data source is carried out, and a time period stability characteristic vector and a time period state resistance vector of each data source are constructed. This process includes the following content:

[0062] Extract the local data state vector under each local time period from the data characteristic matrix of the data source, and perform data fluctuation characteristic analysis on the local data state vector under each local time period through the data reference state vector to obtain the data state difference parameters corresponding to the data source under multiple local time periods.

[0063] In this embodiment, the local time period is specifically a pre-determined data segmentation parameter. For example, every half hour or one hour is taken as a local time period. For the data within the past two weeks, multiple local time periods can be obtained. The data characteristic matrix of the data source contains multiple characteristic parameters of the data packets received by the middleware at different time periods, and the local data state vectors containing these characteristic parameters can be extracted for each local time period. Taking the reference state vector as a reference, the differences between the data characteristics at different local time periods and the normal level are measured. Specifically, data fluctuation characteristic analysis is performed on the local data state vectors of each local time period by using the data reference state vector respectively. For example, the Euclidean distance between the data reference state vector and the local data state vector is calculated as the data state difference parameter for measuring the difference in the characteristics of the two data packets. In an industrial scenario, the abnormal state rules of data packets are relatively unified. For example, it may be due to an increase in the amount of data caused by a production failure. By quantifying the difference between the actual level and the reference level, the degree of data abnormality can be characterized to a certain extent. Considering the periodicity of production activities, the possibility of abnormal states may be relatively high at different times of the day due to the particularity of the production process. Therefore, the data state difference parameters for multiple local time periods are fused. Specifically, multiple global reference time periods are determined. For example, from 8:00 to 9:00 in the morning every day is taken as a global reference time period. During this time period, the state of production equipment needs to be detected, and a high-frequency sampling strategy may be adopted for data collection to determine the states of different equipment and ensure the normal operation of the equipment. Each global reference time period contains multiple local time periods. The average method can be used to calculate the mean value of the data state difference parameters corresponding to the multiple local time periods included in the global reference time period, so as to obtain the global state difference parameter for each global reference time period. Finally, a time period stability characteristic vector of the data source including the global state difference parameters for each global reference time period is constructed to reflect characteristics such as the stability of the data sent by the data source at different time periods.

[0064] Through multiple data state difference parameters, time period state resistance analysis is performed on the data characteristic matrix, and the local state resistance parameters of each local time period are calculated. Finally, a time period state resistance vector of the data source is constructed based on the local state resistance parameters of multiple local time periods.

[0065] In this embodiment, for the calculation process of the local state resistance parameter in a local time period, first calculate the state change rate of each local time period based on the data state difference parameter. Specifically, the difference between the local state resistance parameters of the current local time period and the previous local time period can be calculated. Based on the duration of the local time period, the state change rate representing the change level of the data state per unit time is obtained. At the same time, the local time periods with the data state difference parameter greater than the preset state difference threshold are recorded as local abnormal time periods, that is, the time periods with significant differences in state compared to the normal level are screened out and marked. Multiple state abnormal events can be determined according to the continuity of the local abnormal time periods. A state abnormal event may involve one or more local abnormal time periods, that is, the data state is abnormal within a certain period of time.

[0066] After determining multiple state abnormal events, use the following formula to calculate the local state resistance parameter of each state abnormal event:

[0067] ;

[0068] In the formula, represents the local state resistance parameter of the state abnormal event, represents the maximum value of the data state difference parameter in multiple local abnormal time periods of the state abnormal event, which is used to describe the severity of the state abnormal event, represents the th state change rate of the local abnormal time period in the state abnormal event, which represents the data state change speed of the data source in the state abnormal event, represents the total duration of the state abnormal event, represents the number of local abnormal time periods in the state abnormal event.

[0069] The above method considers the data characteristics of each time period. Specifically, by combining the abnormal amplitude and the recovery level, it can well simulate the dynamic process of the data source recovering after experiencing a state abnormality. The amplitude of the abnormal state and the resistance to the abnormal state will both affect the data communication transmission. Different scenarios need to select appropriate transmission protocols according to actual needs. Through the above method, the stability of the data source sending data and the resistance to abnormal events can be evaluated, so as to provide a reference for the middleware to select an appropriate transmission protocol.

[0070] Through the above method, the local state resistance parameters of each state exception event can be calculated. Then, the local state resistance parameters of the state exception event are used as the local state resistance parameters of each local time period under the state exception event, and the local state resistance parameters corresponding to multiple local time periods included in the global reference time period are fused. For example, the mean value of multiple local state resistance parameters is taken to generate the global state resistance parameters of each global reference time period. Finally, a time period state resistance vector including the global state resistance parameters of each global reference time period is constructed.

[0071] In an alternative implementation process, for the above step S2, for the data feature protocol optimization scheme of the data source, the following content is used for generation:

[0072] The global data exception characteristic parameters of each global reference time period are calculated through the time period stability characteristic vector and the time period state resistance vector of the data source. For multiple local time periods included in each global reference time period, the local exception characteristic parameters of each local time period are calculated according to the data state difference parameters and the local state resistance parameters of the local time period.

[0073] In this embodiment, the time period stability characteristic vector and the time period state resistance vector of the data source are fused. Specifically, the product between the global state difference parameter and the global state resistance parameter is calculated to obtain the global data exception characteristic parameters representing the data fluctuation amplitude and recovery ability of each global reference time period. For each local time period, in a similar manner, the product of the data state difference parameter and the local state resistance parameter of the local time period is calculated as the local exception characteristic parameter of the local time period.

[0074] Then, the transmission protocol adopted in each local time period is extracted from the historical data packet transmission record data of the data source, so as to obtain the local exception characteristic parameters of each transmission protocol in different local time periods. The multiple local exception characteristic parameters belonging to the same global reference time period are summarized to construct the exception reference range of each transmission protocol in each global reference time period. Finally, the reference exception characteristic parameters of each transmission protocol in each global reference time period are determined according to multiple exception reference ranges. For example, the central value of the exception reference range is taken as the corresponding reference exception characteristic parameter.

[0075] After determining the reference anomaly characteristic parameters of each transmission protocol under each global reference period, for any global reference period, the adaptability between it and multiple transmission protocols can be determined according to the global data anomaly characteristic parameters of the global reference period. Specifically, it can be to judge which transmission protocol's reference anomaly characteristic parameters are the closest to the global data anomaly characteristic parameters of the global reference period, and then this transmission protocol is recorded as the best transmission protocol for the current global reference period. Since the historical data contains different state anomaly events, by analyzing the protocols actually used in a large amount of time, the quantitative analysis of the anomaly degree is realized, so as to determine which transmission protocol is widely used in different state anomaly events under any global reference period. After determining the best data transmission protocol, other transmission protocols with performance higher than this protocol are also screened, and finally a protocol candidate list for each global reference period is constructed, and a data characteristic protocol optimization scheme for the data source including the protocol candidate list for each global reference period is obtained. It is used to indicate the strategy for optimizing middleware communication from the perspective of data source characteristics.

[0076] In an optional implementation process, for step S4, optimize the data characteristic protocol optimization scheme of each data source based on the communication quality characteristic matrix to generate the target communication protocol optimization scheme of each data source, specifically including:

[0077] Extract the communication quality reference feature vector of each transmission protocol in the global reference period from the communication quality characteristic matrix of the transmission protocol, extract the historical transmission protocol actually used in each local period according to the historical data packet transmission record data of the data source, and determine the real-time matching parameter of each transmission protocol in each local period according to the historical transmission protocol actually used in each local period, and construct the real-time matching feature vector of each transmission protocol.

[0078] In this embodiment, the communication quality reference feature vector of the global reference period represents the transmission quality qualification information of a specific transmission protocol for transmitting a data packet used as a benchmark under specific network conditions, and contains reference transmission quality parameters corresponding to multiple local periods respectively. At the same time, further analyze the historical data packet transmission record data of the data source to determine which protocol is specifically used to transmit the data packet of this data source in each local period, so as to determine the real-time matching parameters of different transmission protocols in each local period. For example, if protocol A is specifically used in a certain local period, the real-time matching parameter of protocol A can be recorded as 1, otherwise it is recorded as 0. In this way, the real-time matching feature vector of each transmission protocol can be constructed to indicate the actual matching situation of each transmission protocol in different local periods.

[0079] The reference communication quality feature vector and the real-time matching feature vector respectively quantify the adaptability of the transmission protocol in different local time periods from the perspectives of the ideal data packet characteristic state and the best protocol actually selected. Then, based on the reference communication quality feature vector and the real-time matching feature vector, protocol timing adaptability analysis is performed on the protocol candidate list in the global reference time period, and the following formula is used to calculate the protocol timing adaptability parameter of each transmission protocol in the global reference time period:

[0080] ;

[0081] In the formula, represents the protocol timing adaptability parameter, 、 respectively represent the local adaptability parameters of the transmission protocol in the th and th local time periods, is the number of local time periods, represents the timing optimization factor, which is used to control the priorities of different time periods. Among them, the local adaptability parameter of the local time period is calculated according to the reference transmission quality parameter and the real-time matching parameter of the transmission protocol in the local time period. For example, the sum of the reference transmission quality parameter and the real-time matching parameter is taken as the local adaptability parameter of the local time period.

[0082] The above formula takes into account the transmission quality that can be achieved by using a specific transmission protocol when the data packet state of the data source is stable under the historical transmission network conditions of the transmission protocol, and further considers the specific best transmission protocol in different time periods under the condition of abnormal data fluctuations. At the same time, it considers the evolutionary trend of the protocol adaptation situation, that is, evaluates the adaptation ability of the protocol to fluctuate with the network state in the long term, and realizes the self-adaptive evolution of the protocol adaptation state.

[0083] After calculating the protocol timing adaptability parameter of each transmission protocol in the global reference time period, the protocol candidate list in the global reference time period is optimized based on the protocol timing adaptability parameter to generate a protocol adaptation correction list, and the target communication protocol optimization scheme of the data source is determined according to the protocol adaptation correction list. In this way, the communication strategy for different time periods can be preset in advance based on historical data, without the need to monitor the actual network conditions and data characteristics in real time and analyze them frequently at high frequency. At the same time, it reduces the resource waste caused by protocol switching, and realizes efficient and stable communication protocol optimization. Through the target communication protocol optimization scheme for middleware communication optimization, not only can the accuracy of protocol selection be improved, but also it can better cope with network fluctuations and data load changes in the industrial Internet environment, and has wide adaptability in actual industrial Internet applications, especially suitable for those scenarios that need to flexibly respond to network quality fluctuations and efficient data transmission.

[0084] Figure 2 It shows a schematic structural diagram of a middleware communication optimization system based on industrial Internet protocol in an implementation process of the present invention.

[0085] Please refer to Figure 2 , based on the same concept of the above-mentioned middleware communication optimization method based on industrial Internet protocol, the present invention also provides a middleware communication optimization system based on industrial Internet protocol, specifically including:

[0086] A data acquisition module, configured to obtain historical packet characteristic record data and historical packet transmission record data of multiple data sources;

[0087] A data characteristic analysis module, configured to extract multiple groups of packet characteristic parameters from each group of historical packet characteristic record data and construct a data characteristic matrix for each data source;

[0088] A transmission characteristic quality correlation analysis module, configured to extract transmission characteristic quality correlation data corresponding to multiple transmission protocols from each group of historical packet transmission record data, including transmission quality record data of packets under different communication level data, and construct a characteristic mapping matrix of each transmission protocol regarding data characteristics and communication quality;

[0089] A data feature optimization strategy generation module, configured to determine a data reference state vector of each data characteristic matrix, perform time period data fluctuation characteristic analysis and time period state resistance analysis on the data characteristic matrix based on the data reference state vector, construct a time period stability characteristic vector and a time period state resistance vector for each data source, and generate a data feature protocol optimization scheme for each data source;

[0090] Among them, constructing a time period stability characteristic vector and a time period state resistance vector for each data source includes:

[0091] Extracting a local data state vector in each local time period from the data characteristic matrix of the data source, performing data fluctuation characteristic analysis on the local data state vector in each local time period respectively through the data reference state vector, obtaining data state difference parameters corresponding to the data source in multiple local time periods respectively, and fusing multiple data state difference parameters to construct a time period stability characteristic vector of the data source, and the time period stability characteristic vector includes global state difference parameters in each global reference time period;

[0092] Performing time period state resistance analysis on the data feature matrix through multiple data state difference parameters, including calculating a state change rate of each local time period based on the data state difference parameters and determining multiple local abnormal time periods, determining multiple state abnormal events of the data source according to the multiple local abnormal time periods, and calculating local state resistance parameters of each state abnormal event by using the following formula:

[0093] ;

[0094] In the formula, represents the local state resistance parameter of the state anomaly event, represents the maximum value of the data state difference parameters in multiple local anomaly periods of the state anomaly event, represents the th state change rate of the local anomaly period in the state anomaly event, represents the total duration of the state anomaly event, represents the number of local anomaly periods in the state anomaly event;

[0095] Taking the local state resistance parameter of the state anomaly event as the local state resistance parameter of each local period under the state anomaly event, fusing the local state resistance parameters corresponding to multiple local periods included in the global reference period to generate the global state resistance parameter of each global reference period, and constructing a period state resistance vector containing the global state resistance parameter of each global reference period.

[0096] A communication quality analysis module, configured to extract the historical communication quality record data used by the middleware for data communication from the historical data packet transmission record data, and analyze the historical communication quality record data based on the data reference state vector and the feature mapping matrix to generate a communication quality characteristic matrix for each transmission protocol;

[0097] A data network fusion and optimization module, configured to optimize the data feature protocol optimization scheme of each data source based on the communication quality characteristic matrix to generate a target communication protocol optimization scheme for each data source, and perform communication optimization on the middleware based on the target communication protocol optimization scheme.

[0098] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The parts not described in detail in this specification belong to the prior art well-known to those skilled in the art.

Claims

1. A middleware communication optimization method based on industrial Internet protocol, characterized in that: include: Acquire historical data packet characteristic record data and historical data packet transmission record data from multiple data sources, extract multiple groups of data packet characteristic parameters from each group of historical data packet characteristic record data and construct a data characteristic matrix for each data source, extract transmission characteristic quality associated data corresponding to multiple transmission protocols from each group of historical data packet transmission record data, including transmission quality record data of data packets under different communication level data, and construct a feature mapping matrix about data characteristics and communication quality for each transmission protocol; Determine the data baseline state vector of each data feature matrix, perform period data fluctuation characteristic analysis and period state resistance analysis on the data feature matrix based on the data baseline state vector, construct the period stability characteristic vector and period state resistance vector of each data source, and generate the data feature protocol optimization plan for each data source; Extracting historical communication quality record data used by the middleware for data communication from historical data packet transmission record data, analyzing the historical communication quality record data based on the data reference state vector and the feature mapping matrix, and generating a communication quality characteristic matrix for each transmission protocol; Optimize the data feature protocol optimization scheme for each data source based on the communication quality characteristic matrix, generate the target communication protocol optimization scheme for each data source, and optimize the communication of the middleware based on the target communication protocol optimization scheme; For each data source, the period stable characteristic vector and period state resistance vector include: Extract the local data state vector in each local time period from the data characteristic matrix of the data source, perform data fluctuation characteristic analysis on the local data state vector in each local time period through the data reference state vector, obtain the data state difference parameters corresponding to the data source in multiple local time periods, fuse the multiple data state difference parameters to obtain the time period stable characteristic vector of the data source, the time period stable characteristic vector includes the global state difference parameters in each global reference time period; The data feature matrix is ​​subjected to period state resistance analysis through multiple data state difference parameters, including calculating the state change rate of each local period based on the data state difference parameters and determining multiple local abnormal periods, determining multiple state abnormal events of the data source according to the multiple local abnormal periods, and calculating the local state resistance parameter of each state abnormal event using the following formula: ； In the formula, represents the local state resistance parameter to state abnormality events, The maximum value of the data state difference parameter in multiple local abnormal time periods representing the state abnormality event, Indicates the first The state change rate of a local abnormal period, Indicates the total duration of abnormal status events. Indicates the number of local abnormal periods in the state abnormal event; The local state resistance parameter of the state abnormality event is used as the local state resistance parameter of each local time period under the state abnormality event. The local state resistance parameters corresponding to multiple local time periods contained in the global reference time period are fused to generate the global state resistance parameter under each global reference time period, and a time period state resistance vector containing the global state resistance parameter under each global reference time period is constructed.

2. According to a method for optimizing middleware communication based on industrial Internet protocol according to claim 1, it is characterized in that: The communication quality characteristic matrix for the transmission protocol also includes: The local communication state vector in each local time period is extracted from the historical communication quality record data, and the data baseline state vector is used as the data reference state vector in each local time period. The transmission quality of the target transmission protocol in multiple local time periods is analyzed through the feature mapping matrix. The reference transmission quality parameters of the target transmission protocol in each local time period are analyzed based on the preset transmission reference standard of the data source. According to the reference transmission quality parameters corresponding to multiple local time periods, the communication quality characteristic matrix of the target transmission protocol under the data source is constructed.

3. The middleware communication optimization method based on the industrial Internet protocol according to claim 2 is characterized in that: The data feature protocol optimization solution for data sources also includes: The global data anomaly characteristic parameters of each global reference period are calculated by using the period stability characteristic vector and the period state resistance vector of the data source. For the multiple local periods contained in each global reference period, the local anomaly characteristic parameters of each local period are calculated according to the data state difference parameters and the local state resistance parameters of the local period. The transmission protocol adopted in each local time period is extracted from the historical data packet transmission record data of the data source, and the abnormal reference range of each transmission protocol in each global reference time period is constructed. The reference abnormal characteristic parameters of each transmission protocol in each global reference time period are determined according to multiple abnormal reference ranges. The protocol candidate list of each global reference time period is determined according to the global data abnormal characteristic parameters of the global reference time period, and the data feature protocol optimization scheme of the data source including the protocol candidate list for each global reference time period is obtained.

4. The middleware communication optimization method based on the industrial Internet protocol according to claim 3 is characterized in that: The data feature protocol optimization scheme for each data source is optimized based on the communication quality characteristic matrix to generate the target communication protocol optimization scheme for each data source, including: Extract the communication quality reference feature vector of each transmission protocol in the global reference period from the communication quality characteristic matrix of the transmission protocol, extract the historical transmission protocol actually used in each local period according to the historical data packet transmission record data of the data source, determine the real-time matching parameters of each transmission protocol in each local period according to the historical transmission protocol actually used in each local period, and construct a real-time matching feature vector for each transmission protocol; According to the reference communication quality feature vector and the real-time matching feature vector, the protocol timing adaptation analysis is performed on the protocol candidate list of the global reference period, and the protocol timing adaptation parameters of each transmission protocol in the global reference period are calculated using the following formula: ； In the formula, Indicates protocol timing adaptation parameters, , Respectively expressed in and The local adaptation parameters of the transmission protocol in a local time period, is the number of local time periods, represents a timing optimization factor, wherein the local adaptation parameter of the local time period is calculated according to the reference transmission quality parameter and the real-time matching parameter of the transmission protocol in the local time period; After calculating the protocol timing adaptation parameters of each transmission protocol in the global reference time period, the protocol candidate list of the global reference time period is optimized based on the protocol timing adaptation parameters to generate a protocol adaptation correction list, and the target communication protocol optimization scheme of the data source is determined according to the protocol adaptation correction list.

5. A middleware communication optimization system based on industrial Internet protocol, characterized in that: The system is used to implement a middleware communication optimization method based on an industrial Internet protocol as described in any one of claims 1 to 4 above, comprising: A data acquisition module, used to obtain historical data packet characteristic record data and historical data packet transmission record data from multiple data sources; A data characteristic analysis module is used to extract multiple groups of data packet characteristic parameters from each group of historical data packet characteristic record data and construct a data characteristic matrix for each data source; The transmission characteristic quality correlation analysis module is used to extract the transmission characteristic quality correlation data corresponding to multiple transmission protocols from each group of historical data packet transmission record data, including the communication quality record data of the data packet under different communication level data, and construct a feature mapping matrix about data characteristics and communication quality for each transmission protocol; The data feature optimization strategy generation module is used to determine the data baseline state vector of each data feature matrix, perform period data fluctuation characteristic analysis and period state resistance analysis on the data feature matrix based on the data baseline state vector, construct the period stability characteristic vector and period state resistance vector of each data source, and generate the data feature protocol optimization plan for each data source; A communication quality analysis module is used to extract historical communication quality record data used by the middleware for data communication from historical data packet transmission record data, analyze the historical communication quality record data based on the data reference state vector and the feature mapping matrix, and generate a communication quality characteristic matrix for each transmission protocol; The data network fusion optimization module is used to optimize the data feature protocol optimization plan of each data source based on the communication quality characteristic matrix, generate the target communication protocol optimization plan for each data source, and optimize the communication of the middleware based on the target communication protocol optimization plan.

6. The middleware communication optimization system based on the industrial Internet protocol according to claim 5 is characterized in that: For the data feature optimization strategy generation module, the time period stable feature vector and time period state resistance vector of each data source are constructed, including: Extract the local data state vector in each local time period from the data characteristic matrix of the data source, perform data fluctuation characteristic analysis on the local data state vector in each local time period through the data reference state vector, obtain the data state difference parameters corresponding to the data source in multiple local time periods, fuse the multiple data state difference parameters to obtain the time period stable characteristic vector of the data source, the time period stable characteristic vector includes the global state difference parameters in each global reference time period; The data feature matrix is ​​subjected to period state resistance analysis through multiple data state difference parameters, including calculating the state change rate of each local period based on the data state difference parameters and determining multiple local abnormal periods, determining multiple state abnormal events of the data source according to the multiple local abnormal periods, and calculating the local state resistance parameter of each state abnormal event using the following formula: ； In the formula, represents the local state resistance parameter to state abnormality events, The maximum value of the data state difference parameter in multiple local abnormal time periods representing the state abnormality event, Indicates the first The state change rate of a local abnormal period, Indicates the total duration of abnormal status events. Indicates the number of local abnormal periods in the state abnormal event; The local state resistance parameter of the state abnormality event is used as the local state resistance parameter of each local time period under the state abnormality event. The local state resistance parameters corresponding to multiple local time periods contained in the global reference time period are fused to generate the global state resistance parameter under each global reference time period, and a time period state resistance vector containing the global state resistance parameter under each global reference time period is constructed.

Images