Middleware communication optimization method and system based on industrial internet protocol

Abstract

The invention provides a middleware communication optimization method and system based on an industrial internet protocol, and relates to the technical field of communication optimization. The method comprises the following steps: acquiring historical data packet characteristic record data and historical data packet transmission record data of a plurality of data sources, and constructing a data characteristic matrix of each data source and a characteristic mapping matrix of 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 characteristic protocol optimization scheme of each data source; extracting historical communication quality record data from historical data packet transmission record data, and generating a communication quality characteristic matrix of each transmission protocol; and optimizing the data feature protocol optimization scheme of each data source to generate a target communication protocol optimization scheme of each data source, and performing communication optimization on the middleware based on the target communication protocol optimization scheme. According to the invention, the communication optimization of the industrial internet architecture middleware is realized.

Description

Technical Field

[0001] The present invention relates to the field of communication optimization technology, and in particular to a middleware communication optimization method and system based on an industrial Internet protocol. 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, equipment and control systems, requiring communication protocols to efficiently adapt to different network qualities and changing data loads. In the modern industrial Internet architecture, middleware plays a vital role as a bridge connecting different systems, devices and protocols. Middleware is responsible for processing data streams from various devices such as PLCs, sensors, smart devices, and control systems such as MES and ERP, ensuring the transmission and coordination of data at different levels and between different protocols.

[0003] In terms of communication optimization from the perspective of protocol adaptation, some statically configured protocol adaptation methods are difficult to respond to network fluctuations and data anomalies in real time because the protocol selection and data transmission strategies are usually set in advance and do not consider the dynamically changing network quality and data flow fluctuations. Dynamic protocol optimization solutions will be analyzed from the perspectives of network quality and data source characteristics, and there are complex interactions between these different perspectives. For example, the adaptability of the protocol may be affected by factors such as network delay and packet loss rate. Comprehensive analysis from different perspectives and in-depth consideration of the interaction of data characteristics from different perspectives can effectively evaluate and optimize the communication protocol comprehensively, generate middleware communication optimization strategies that can flexibly respond to the complex and changing network and data environment in the industrial Internet, and adapt to the needs 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 the industrial Internet protocol, which generates a middleware communication optimization strategy that can accurately adapt to the needs of different industrial application environments by analyzing the changes in the characteristics of data packets sent by the data source and the changes in network quality, and flexibly responds to the challenges brought by the complex and changeable network and data environment in the industrial Internet to the data communication of the middleware.

[0005] To achieve the above-mentioned object, the first aspect of the present invention provides a middleware communication optimization method based on the industrial Internet protocol, comprising: 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; The data feature protocol optimization scheme of each data source is optimized based on the communication quality characteristic matrix, a target communication protocol optimization scheme of each data source is generated, and communication optimization of the middleware is performed based on the target communication protocol optimization scheme.

[0006] Preferably, the period stable characteristic vector and period state resistance vector of each data source 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.

[0007] Preferably, the communication quality characteristic matrix of the transmission protocol further 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.

[0008] Preferably, the data feature protocol optimization solution for the data source 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.

[0009] Preferably, the data feature protocol optimization scheme of each data source is optimized based on the communication quality characteristic matrix to generate a 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.

[0010] A second aspect of the present invention provides a middleware communication optimization system based on an industrial Internet protocol, which is used to implement the above-mentioned middleware communication optimization method based on an industrial Internet protocol, including: 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.

[0011] Preferably, for the data feature optimization strategy generation module, a time period stable characteristic vector and a 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.

[0012] The present invention has the following beneficial effects: The present invention analyzes the historical record data of the middleware, analyzes the data characteristics of the data source from the perspective of data source characteristics and constructs a data characteristic matrix, and performs time period data fluctuation characteristic analysis and time period state resistance analysis on the data source. In combination with the data fluctuation level and abnormal state resistance level of the data source, a data characteristic protocol optimization scheme is generated to determine the adaptability of different communication protocols under the data source characteristics; at the same time, from the network quality analysis middleware, the network state used for data transmission is constructed to obtain a feature mapping matrix representing the association between data characteristics and communication quality, and in combination with the changes in network quality, transmission adaptation analysis is performed on different communication protocols. Finally, the network quality and data source characteristic analysis results are integrated to dynamically optimize the protocol selection strategy to ensure that the protocol selection can not only adapt to 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

[0013] Figure 1 The present invention is a flowchart of a middleware communication optimization method based on the Industrial Internet Protocol during an implementation process of the present invention.

[0014] Figure 2 This is a structural schematic diagram of a middleware communication optimization system based on the Industrial Internet Protocol during an implementation of the present invention. DETAILED DESCRIPTION

[0015] In order to make those skilled in the art better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below 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.

[0016] Figure 1 A flow chart of a middleware communication optimization method based on the Industrial Internet Protocol in an implementation process of the present invention is shown.

[0017] See also Figure 1The present invention provides a middleware communication optimization method based on the industrial Internet protocol, which can be applied to the middleware layer in the industrial Internet architecture to optimize the communication of the middleware in the industrial Internet from the perspective of protocol adaptation. The method specifically includes the following steps: Step S1, obtain 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 and construct a data characteristic matrix for each data source, extract the transmission characteristic quality correlation data corresponding to multiple transmission protocols from each group of historical data packet transmission record data, and construct a feature mapping matrix about data characteristics and communication quality for each transmission protocol.

[0018] In the modern industrial Internet architecture, middleware plays a vital role as a bridge connecting different systems, devices and protocols. Its main functions include data transmission and protocol management. It is responsible for transmitting data between devices, applications and networks to ensure that data is not lost during the communication process.

[0019] In this embodiment, for different data sources such as sensor networks, production lines or other production control systems, the corresponding historical data packet characteristic record data and historical data packet transmission record data include some characteristics of the data packets received by the middleware from the data source during the transmission process, such as data packet size and quantity, packet loss, sending frequency, data packet type and other information, which can provide an important basis for the selection of communication protocols, as well as transmission information record data related to these data after being transmitted to other network layers by the middleware, such as the transmission network conditions, the transmission protocol used, and the transmission quality performance of the data packet after being transmitted, such as throughput, delay, packet loss rate, transmission success rate and other information.

[0020] For the historical data packet characteristic record data of different data sources, considering that different data sources may have different data transmission requirements, the historical data packet characteristic record data of each data source is analyzed to extract some data packet characteristic parameters of the data packets received by the data source at different times in a historical period, such as size, quantity, type and other key information that affects 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 packet characteristic information received by the data source at different time periods in a historical period.

[0021] For the historical data packet transmission record data from different data sources, and for the transmission-related record data contained therein, the transmission characteristic quality-related data corresponding to each transmission protocol is extracted. Specifically, the network and data environment in the industrial Internet are complex and changeable. For different data sources and different network environments, the middleware needs to select the appropriate communication protocol for transmission according to the actual situation. For example, some scenarios require low-latency transmission requirements, then you can choose QUIC or UDP and other communication protocols with relatively low latency, while some scenarios may require higher reliability, that is, the packet loss rate cannot be too high, and you can choose TCP and other communication protocols with higher reliability. In the actual transmission process, the transmission quality will be affected by multiple factors such as the characteristics of the data packet itself and the network quality. From each set of historical data packet transmission record data, the transmission characteristic quality associated data corresponding to multiple transmission protocols are extracted, that is, the transmission quality record data of specific data packets under different communication level data. For example, for a certain type and size of data packets, under a specific bandwidth, delay, jitter and other network environment, the final data packet uses a certain communication protocol for data transmission. The recorded transmission quality information such as delay, packet loss rate, throughput, etc. can be constructed through these data. The feature mapping matrix of data characteristics and communication quality of each transmission protocol can be constructed, including information about network status, data packet characteristics and the final transmission effect. The matrix can be used to estimate the transmission quality under different data characteristics and network conditions, such as the successful transmission rate under a specific data packet size and sending frequency when using a certain protocol for data packet transmission, or the adaptability performance under different packet loss rates and delays.

[0022] Step S2, determine the data baseline 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 baseline state vector, construct the time period stability characteristic vector and time period state resistance vector of each data source, and generate a data feature protocol optimization plan for each data source.

[0023] In this embodiment, the data baseline state vector is obtained by modeling the standard transmission conditions of each data source, and is used to reflect the baseline level of the data packets sent by the data source to the middleware under relatively ideal conditions, such as the average value of the data packet size, data type, data transmission frequency, standard value of the packet loss rate, etc., and data can be recorded according to the historical data packet characteristics of the data source, and the data packet characteristics of different time periods are counted, and the data packet characteristics received at the normal level without abnormal conditions are used as the benchmark. In actual industrial scenarios, when production anomalies occur, such as a production line failure, the Internet of Things system will use a higher frequency to collect data, and return to the original collection frequency after returning to normal. Then, for this period of time, 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 increase abnormally.

[0024] Using the data baseline state vector as a reference under normal production conditions, data fluctuation characteristics analysis and period state resistance analysis are performed for each period in the data characteristic matrix. The purpose of data fluctuation characteristics analysis is to identify changes in the characteristic state of data packets received from a data source during the fluctuation period, such as identifying the magnitude of change in the amount of data in each period, that is, whether there is an abnormal increase or decrease in data, and quantifying the impact of such fluctuations on protocol adaptation. By comparing the deviation of the data characteristics of each period from the baseline state vector, the fluctuation index of each period is obtained, which reflects the degree of data flow fluctuation within the period, and the period stability characteristic vector of the data source is constructed.

[0025] The purpose of period state resistance analysis is to identify the recovery capability of the data source within the period, that is, the ability of the system to recover to normal levels after data anomalies such as increased data volume and increased latency occur. By analyzing the highest data volume level within the period and the time required to recover to normal levels, the data source's resistance to data state anomalies is evaluated. Finally, by analyzing the recovery level shown by historical data, the recovery capability of the data source in different periods is evaluated, and the period state resistance vector of the data source is constructed. Based on the period stability characteristic vector and period state resistance vector of each data source, the stability of the amount of data transmitted from the data source to the middleware and the recovery level in the face of abnormal conditions are comprehensively analyzed, and the matching between multiple communication protocols in different periods is evaluated. Finally, the data feature protocol optimization scheme of each data source based on the adaptation status between different communication protocols is generated from the perspective of data source characteristics.

[0026] Step S3: extract the historical communication quality record data used by the middleware for data communication from the 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.

[0027] In this embodiment, the historical communication quality record data includes the network status for data communication transmission in different time periods, reflecting the specific parameters of the communication network in different time periods during the actual communication process. And the data packet state level represented by the data reference state vector is used as the reference data, combined with the feature mapping matrix, to analyze whether the data packet transmission requirements can be met under different network conditions if the data packet represented by the data reference state vector needs to be transmitted, so as to obtain the transmission quality that can be achieved by using a certain protocol for transmission under the condition that the transmission network conditions of each time period are known and the data packet does not have abnormal fluctuations under the ideal condition, and finally construct the communication quality characteristic matrix of each transmission protocol, that is, the communication quality level theoretically shown by using a certain protocol for data transmission under the actual network conditions and specific transmitted data is quantified.

[0028] In this embodiment, the process of analyzing the historical communication quality record data, taking any communication protocol as an example and recording it as the target communication protocol, includes extracting the local communication state vector in each local time period from the historical communication quality record data, that is, the network state condition of the local time period, and taking the data reference state vector as the data reference state vector in each local time period, that is, the data packet characteristics, combining the feature mapping matrix determined in the above steps, and estimating the transmission effect under the specific network state and data packet characteristics according to the feature mapping matrix of the target communication protocol, and analyzing the reference transmission quality parameters of the target transmission protocol in each local time period 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 actual conditions, 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, which is qualified, otherwise it is recorded as 0 to indicate that the transmission quality is unqualified. After determining the reference transmission quality parameter 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. The historical communication quality record data is analyzed to generate the communication quality characteristic matrix of each transmission protocol.

[0029] Step S4: Optimize the data feature protocol optimization scheme for each data source based on the communication quality characteristic matrix, 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.

[0030] In this embodiment, the data feature protocol optimization scheme specifically analyzes the communication protocol adapted by the data source in different time periods from the perspective of changes in data packet characteristics, and the communication quality characteristic matrix further considers the impact of actual network conditions on data communication, and organically integrates the analysis results from the perspectives of network quality and data source characteristics, and finally generates a target communication protocol optimization scheme for each data source. Based on this scheme, dynamic communication optimization of the middleware can achieve flexible and intelligent protocol selection and efficient data transmission. Optimization in this way can enable the middleware to adapt to dynamically changing network conditions and data flow requirements, ensure the best communication protocol selection in different industrial application scenarios, and avoid resource waste and improve network performance.

[0031] In an optional implementation process, for the above step S2, in order to select a suitable transmission protocol, taking into account the characteristic changes of the data source, the period data characteristics and period state resistance of each data source are comprehensively analyzed to construct a period stability characteristic vector and a period state resistance vector of each data source. The process includes the following contents: The local data state vector in each local time period is extracted from the data characteristic matrix of the data source, and the data fluctuation characteristics of the local data state vector in each local time period are analyzed through the data reference state vector to obtain the data state difference parameters corresponding to the data source in multiple local time periods.

[0032] In this embodiment, the local time period is specifically a predetermined data segmentation parameter, such as every half hour or one hour as a local time period, and for the data within two weeks of history, multiple local time periods can be divided, and the data characteristic matrix of the data source contains multiple characteristic parameters of the data packets received by the middleware in different time periods, and the local data state vector containing these characteristic parameters in each local time period can be extracted. With the reference state vector as a reference, the difference between the data characteristics in different local time periods and the normal level is measured, and the data fluctuation characteristics of the local data state vector in each local time period are analyzed respectively by the data reference state vector, for example, the Euclidean distance between the data reference state vector and the local data state vector is calculated as a data state difference parameter to measure the difference in data packet characteristics between the two. In industrial scenarios, the abnormal state rules of data packets are relatively uniform, for example, the amount of data may increase due to production failures. By quantifying the difference between the actual level and the reference level, the degree of data abnormality can be characterized to a certain extent. At the same time, considering the cyclical nature of production activities, the possibility of abnormal conditions occurring at different times of the day may be higher due to the particularity of the production process. Therefore, the data state difference parameters under multiple local time periods are fused, specifically to determine multiple global reference time periods, such as 8:00 to 9:00 a.m. every day as a global reference time period, during which the status of production equipment needs to be detected, and a high-frequency sampling strategy may be used for data collection to determine the status of different equipment to ensure the normal operation of the equipment. Each global reference time period contains multiple local time periods, and the average method can be used to calculate the mean of the data state difference parameters corresponding to the multiple local time periods contained in the global reference time period, and obtain the global state difference parameters under each global reference time period. Finally, the time period stability characteristic vector of the data source including the global state difference parameters under each global reference time period is constructed to reflect the stability of data sent by the data source in different time periods and other characteristics.

[0033] The data feature matrix is ​​subjected to period state resistance analysis through multiple data state difference parameters, and the local state resistance parameters of each local period are calculated. Finally, the period state resistance vector of the data source is constructed according to the local state resistance parameters of multiple local periods.

[0034] In this embodiment, for the calculation process of the local state resistance parameter of the local time period, the state change rate of each local time period is first calculated based on the data state difference parameter. Specifically, the difference between the local state resistance parameter of the current local time period and the previous local time period can be calculated. Based on the length of the local time period, the state change rate that characterizes the level of change of the data state in unit time is obtained. At the same time, the local time period with a data state difference parameter greater than a preset state difference threshold is recorded as a local abnormal time period, that is, those time periods with a more significant difference in state compared to the normal level are screened out and marked. Multiple state abnormal events can be determined based on the continuity of the local abnormal time period. The state abnormal event may involve one or more local abnormal time periods, that is, the data state is abnormal for a period of time.

[0035] After determining multiple state abnormality events, the local state resistance parameter of each state abnormality event is calculated using the following formula: ; In the formula, represents the local state resistance parameter to state abnormality events, Indicates the maximum value of the data state difference parameters in multiple local abnormal time periods of the state abnormal event, which is used to describe the severity of the state abnormal event. Indicates the first The state change rate of a local abnormal period indicates the speed at which the data state of the data source changes during the state abnormality event. Indicates the total duration of abnormal status events. Indicates the number of local abnormal periods in the status abnormal event.

[0036] The above method takes into account the data characteristics of each time period. Combined with the abnormal amplitude and recovery level, it can well simulate the dynamic process of data source recovery after experiencing abnormal state. The amplitude of the abnormal state and the resistance to the abnormal state will affect the communication transmission of data. Different scenarios need to select appropriate transmission protocols according to actual needs. The above method can be used to evaluate the stability of data sent by the data source and its resistance to abnormal events, thereby providing a reference for the middleware to select appropriate transmission protocols.

[0037] The local state resistance parameters of each state abnormality event can be calculated in the above manner, and then the local state resistance parameters of the state abnormality event are used as the local state resistance parameters of each local time period under the state abnormality event, and the local state resistance parameters corresponding to multiple local time periods contained in the global reference time period are fused, for example, the average of multiple local state resistance parameters is taken to generate the global state resistance parameters under each global reference time period, and finally a time period state resistance vector containing the global state resistance parameters under each global reference time period is constructed.

[0038] In an optional implementation process, for the above step S2, the data feature protocol optimization scheme for the data source is generated using the following content: 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 based on the data state difference parameters and the local state resistance parameters of the local period.

[0039] In this embodiment, the period stability characteristic vector and the 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 anomaly characteristic parameter characterizing the data fluctuation amplitude and recovery ability of each global reference period. For each local period, a similar method is used to calculate the product of the data state difference parameter and the local state resistance parameter of the local period as the local anomaly characteristic parameter under the local period.

[0040] Then, the transmission protocol used in each local period is extracted from the historical data packet transmission record data of the data source, so as to obtain the local abnormal characteristic parameters of each transmission protocol in different local periods, and the multiple local abnormal characteristic parameters belonging to the same global reference period are summarized to construct the abnormal reference range of each transmission protocol in each global reference period. Finally, the reference abnormal characteristic parameters of each transmission protocol in each global reference period are determined according to the multiple abnormal reference ranges, for example, the central value of the abnormal reference range is taken as the corresponding reference abnormal characteristic parameter.

[0041] After determining the reference abnormal characteristic parameters of each transmission protocol in each global reference period, for any global reference period, the compatibility between it and multiple transmission protocols can be determined according to the global data abnormal characteristic parameters of the global reference period. Specifically, it can be determined that the global data abnormal characteristic parameters of the global reference period are closest to the reference abnormal characteristic parameters of which transmission protocol, and then the transmission protocol is recorded as the best transmission protocol for the current global reference period. Because the historical data contains different state abnormal events, by analyzing the protocols actually used in a large amount of time, a quantitative analysis of the degree of abnormality is achieved, so as to determine which transmission protocol is widely used in different state abnormal events in any global reference period. After determining the best data transmission protocol, other transmission protocols with higher performance than the protocol are also screened, and finally a candidate list of protocols for each global reference period is constructed, and a data feature protocol optimization scheme for the data source containing the candidate list of protocols 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.

[0042] In an optional implementation process, for step S4, the data feature protocol optimization scheme of each data source is optimized based on the communication quality characteristic matrix to generate a target communication protocol optimization scheme for each data source, specifically including: The communication quality reference feature vector of each transmission protocol in the global reference time period is extracted from the communication quality characteristic matrix of the transmission protocol, and the historical transmission protocol actually adopted in each local time period is extracted according to the historical data packet transmission record data of the data source. According to the historical transmission protocol actually adopted in each local time period, the real-time matching parameters of each transmission protocol in each local time period are determined to construct the real-time matching feature vector of each transmission protocol.

[0043] In this embodiment, the communication quality reference feature vector of the global reference period represents the transmission quality qualification information of the data packet transmitted as a benchmark under a specific transmission protocol and specific network conditions, and includes reference transmission quality parameters corresponding to multiple local periods. At the same time, the historical data packet transmission record data of the data source is further analyzed to determine which protocol is specifically used to transmit the data packet of the 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 a local period specifically uses protocol A, the real-time matching parameter of protocol A can be recorded as 1, otherwise it is recorded as 0. In this way, a 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.

[0044] The reference communication quality feature vector and the real-time matching feature vector quantify the adaptability of the transmission protocol in different local time periods from the perspective of the ideal data packet characteristic state and the perspective of the best protocol actually selected, respectively. Then, the protocol timing adaptation analysis is performed on the protocol candidate list of the global reference time period according to the reference communication quality feature vector and the real-time matching feature vector. The protocol timing adaptation parameters of each transmission protocol in the global reference time 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, It represents a timing optimization factor, which is used to control the priorities of different time periods. The local adaptation parameters of the local time period are calculated based on the reference transmission quality parameters and real-time matching parameters of the transmission protocol in the local time period. For example, the sum of the reference transmission quality parameters and the real-time matching parameters is taken as the local adaptation parameters of the local time period.

[0045] The above formula takes into account the transmission quality that can be achieved by using a specific transmission protocol under historical transmission network conditions when the data packet status of the data source is stable. It further considers the specific optimal transmission protocol at different time periods when the data fluctuates abnormally. It also considers the evolutionary trend of the protocol adaptation situation, that is, it evaluates the adaptability of the protocol to fluctuations in the network status in the long term, and realizes the adaptive evolution of the protocol adaptation status.

[0046] After calculating the protocol timing adaptation parameters of each transmission protocol in the global reference period, the protocol candidate list of the global reference 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. In this way, the communication strategies for different time periods can be preset in advance based on historical data, without the need for real-time monitoring and high-frequency analysis of actual network conditions and data characteristics, while reducing the waste of resources caused by protocol switching, and achieving efficient and stable communication protocol optimization. Optimizing the communication of the middleware through the target communication protocol optimization scheme can not only improve the accuracy of protocol selection, but also better cope with network fluctuations and data load changes in the industrial Internet environment. It has a wide range of adaptability in actual industrial Internet applications, and is particularly suitable for scenarios that require flexible response to network quality fluctuations and efficient data transmission.

[0047] Figure 2 Shown is a structural diagram of a middleware communication optimization system based on the Industrial Internet Protocol during an implementation of the present invention.

[0048] See also Figure 2 Based on the same concept of the above-mentioned middleware communication optimization method based on the industrial Internet protocol, the present invention also provides a middleware communication optimization system based on the industrial Internet protocol, which specifically includes: 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 transmission 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; Among them, the time period stability characteristic 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.

[0049] 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.

[0050] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. Parts not described in detail in this specification belong to the prior art 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; The data feature protocol optimization scheme of each data source is optimized based on the communication quality characteristic matrix, a target communication protocol optimization scheme of each data source is generated, and communication optimization of the middleware is performed based on the target communication protocol optimization scheme.

2. According to a method for optimizing middleware communication based on industrial Internet protocol according to claim 1, it is characterized in that: 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.

3. The middleware communication optimization method based on the industrial Internet protocol according to claim 2 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.

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 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.

5. The middleware communication optimization method based on the industrial Internet protocol according to claim 4 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.

6. 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 5, 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.

7. The middleware communication optimization system based on the industrial Internet protocol according to claim 6 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