Industrial control network data monitoring system based on cloud platform

Abstract

The invention discloses an industrial control network data monitoring system based on a cloud platform, and particularly relates to the technical field of industrial control network communication, comprising a data acquisition module for acquiring communication protocol data between an industrial control network and the cloud platform in real time to form a data set; the data analysis module analyzes the data set, calculates indexes such as network timeliness, throughput capacity and transmission efficiency, and generates a performance evaluation data set; the fuzzy reasoning module is used for deducing communication protocol performance based on the data set and dividing the communication protocol performance into high-class, medium-class and low-class; the feedback module sends corresponding signals according to protocol performance classification, and the adjusting module adjusts the size of a data monitoring window according to feedback signals; the method can be suitable for various industrial communication protocols, provides sufficient basis for whether the communication protocols need to be improved or not through a unified analysis mechanism, adapts to dynamic changes of an industrial network, achieves self-adaptive adjustment, and improves the stability and efficiency of the industrial control network.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial control network communication, and more specifically, to an industrial control network data monitoring system based on a cloud platform. Background Art

[0002] With the development of industrial Internet of Things and intelligent manufacturing, industrial control systems are gradually evolving from traditional closed and single-protocol local network architectures to cloud platform management, multi-protocol integration, and high-efficiency data transmission modes. In industrial control networks, common communication protocols include Modbus, PROFINET, EtherCAT, MQTT, etc. Different protocols have differences in aspects such as latency, throughput, and bandwidth utilization, making it difficult to manage and optimize them uniformly. Traditional monitoring methods mostly use fixed sampling or rule setting, which are difficult to adapt to the characteristic changes of different protocols, have weak data support, and cannot provide sufficient basis for whether communication protocols need to be improved. Moreover, industrial control systems need to accurately monitor the real-time performance, throughput capacity, and transmission efficiency of communication protocols to ensure the stable operation of production systems. Existing methods often rely on fixed thresholds or manual settings and cannot adaptively adjust the monitoring window, resulting in insufficient monitoring accuracy or waste of computing resources. Therefore, the present invention proposes an industrial control network data monitoring system based on a cloud platform to solve the above problems. Summary of the Invention

[0003] To achieve the above object, the present invention provides the following technical solutions:

[0004] An industrial control network data monitoring system based on a cloud platform includes a data acquisition module, a data analysis module, a fuzzy inference module, a feedback module, and an adjustment module;

[0005] The data acquisition module is used to collect data on the usage of communication protocols between the current industrial control network and the cloud platform in real time to obtain a communication protocol usage dataset;

[0006] The data analysis module is used to perform performance analysis of communication protocol usage based on the communication protocol usage dataset to obtain a performance evaluation data group and transmit it to the fuzzy inference module;

[0007] The fuzzy inference module is used to infer the performance of the currently used communication protocol based on the performance evaluation data group, and classify the performance of the currently used communication protocol into a high-performance protocol, a medium-performance protocol, and a low-performance protocol;

[0008] The feedback module is used to send a feedback signal of the corresponding type according to the classification type of the performance of the currently used communication protocol;

[0009] The adjustment module is used to adjust the length of the data monitoring window according to the type of the feedback signal.

[0010] In a preferred embodiment, performing communication protocol usage performance analysis to obtain a performance evaluation data set means:

[0011] Analyzing the data transmission delay situation of the communication protocol to obtain a network timeliness index;

[0012] Analyzing the data transmission efficiency of the communication protocol to obtain a network throughput capacity index;

[0013] Analyzing the data transmission overhead of the communication protocol to obtain a network transmission efficiency index;

[0014] Summarizing the network timeliness index, the network throughput capacity index, and the network transmission efficiency index to obtain a performance evaluation data set.

[0015] In a preferred embodiment, the acquisition logic of the network timeliness index is:

[0016] Within a fixed time window, collect the sending time and receiving time of each data packet, and then calculate the average value of the difference between the receiving time and the sending time to obtain the average transmission delay T avg ;

[0017] Then calculate the delay jitter value, and the calculation formula is:

[0018] T arrival，i+1 represents the (i + 1)-th receiving time, T send,i+1 represents the (i + 1)-th sending time, T arrival，i represents the i-th receiving time, T send，i represents the i-th sending time, N represents the total number of data packets, and J represents the delay jitter value;

[0019] Obtain the number of data packets whose difference between the receiving time and the sending time exceeds a preset standard duration, and divide it by the total number of data packets to obtain the timeout rate T timeout ;

[0020] The calculation formula of the network timeliness index is:

[0021] α, β, and γ are all preset non-zero proportional coefficients, and satisfy α + β + γ = 1; NTI represents the network timeliness index.

[0022] In a preferred embodiment, the acquisition logic of the network throughput capacity index is:

[0023] Within a fixed time window, sum up the data volume of each data packet to obtain the total data volume, then divide it by the duration corresponding to the fixed time window to obtain the actual throughput. Divide the number of lost data packets by the total number of data packets to obtain the packet loss rate;

[0024] The calculation formula for the network throughput capacity index is:

[0025] PLR is the packet loss rate, and T max is the preset maximum throughput, and T actual is the actual throughput. Both k and θ are preset non-zero control coefficients. Using the Sigmoid function, the throughput ratio is smoothly transitioned between low and high values. k controls the steepness of the curve, and θ controls the inflection point position. NTCI is the network throughput capacity index.

[0026] In a preferred embodiment, the acquisition logic of the network transmission efficiency index is as follows:

[0027] Within a fixed time window, obtain the available bandwidth values measured at each sampling time point and calculate the average value to obtain the average bandwidth. Divide the actual throughput by the average bandwidth to obtain the bandwidth utilization rate BU. Sum up the effective data volume of each data transmission to obtain the effective data load value FZ. Sum up the effective data and the additional overhead data volume of each data transmission to obtain the total transmitted data volume QB. The calculation formula for the network transmission efficiency index is: NTEI is the network transmission efficiency index.

[0028] In a preferred embodiment, the principle of using the fuzzy inference module is as follows:

[0029] Take the network timeliness index, network throughput capacity index, and network transmission efficiency index corresponding to the same fixed time window as input variables, and take the type to which the performance of the currently used communication protocol belongs as the output variable. Perform fuzzy processing on the input variables, convert the values of the input variables into fuzzy sets, perform fuzzy processing on the output variable, and convert the output variable into a fuzzy set. Formulate fuzzy rules to describe the adaptability of each type to which it belongs under different combinations of data types. Infer the type to which the performance of the currently used communication protocol belongs through the fuzzy rules for the fuzzy input variables.

[0030] In a preferred embodiment, the feedback module is used to send a feedback signal corresponding to the divided type of the performance of the currently used communication protocol, which means:

[0031] When the performance of the currently used communication protocol is classified as a high-performance protocol, send a type A1 feedback signal; when it is classified as a medium-performance protocol, send a type A2 feedback signal; when it is classified as a low-performance protocol, send a type B feedback signal.

[0032] In a preferred embodiment, the adjustment module is used to adjust the data monitoring window length according to the type of the feedback signal, which means that:

[0033] When the feedback signal type is A2, the adjustment formula is:

[0034] SJ2 is the adjusted data monitoring window length when the feedback signal type is A2;

[0035] When the feedback signal type is A1, the adjustment formula is:

[0036] SJ1 is the adjusted data monitoring window length when the feedback signal type is A1; Smin and Smax are respectively the preset adjustment range of the data monitoring window length, Smin < Smax, ε is the preset adjustment coefficient, and SC represents the current data monitoring window length;

[0037] When the feedback signal type is B, the data monitoring window length is restored to

[0038] Technical effects and advantages of the present invention:

[0039] There are various communication protocols in the industrial control network (such as Modbus, PROFINET, MQTT, etc.), and the data transmission performances of different protocols vary greatly. Traditional monitoring methods are difficult to be compatible with multi-protocol environments. Most traditional methods are optimized based on a single protocol and it is difficult to provide a general analysis solution for multiple protocols, resulting in insufficient monitoring accuracy. The present invention can, through the data acquisition module, monitor in real time the usage of communication protocols between the industrial control network and the cloud platform and form a data set of the usage of communication protocols, which is applicable to various industrial communication protocols. Through a unified analysis mechanism, calculate the network timeliness index (NTI), network throughput capacity index (NTCI), and network transmission efficiency index (NTEI), which can be adapted to different types of protocols, provide sufficient basis for whether the communication protocol needs to be improved, and can switch to other protocols for matching when the current protocol cannot be improved through the preset optimization rules, so that the communication quality remains in a good state.

[0040] Traditional methods often use fixed thresholds (such as fixed upper limits of time delay and lower limits of throughput) to judge the performance of protocols. However, the industrial network environment is complex, and the performance of communication protocols is affected by various factors (network congestion, bandwidth fluctuations, device status). The single-threshold judgment method is prone to misjudgment. The present invention adopts a fuzzy inference module. Through the fuzzy calculation of three indicators: network timeliness, throughput capacity, and transmission efficiency, the performance of the communication protocol is classified into high, medium, and low levels, realizing a more accurate protocol evaluation. Combining triangular membership functions or trapezoidal membership functions avoids the limitations of the traditional fixed-threshold method, enabling the system to more flexibly adapt to different industrial network environments. For example, if a certain protocol has a high throughput but large delay fluctuations, the traditional method may misjudge it as a high-performance protocol, while the fuzzy inference of the present invention can comprehensively evaluate multiple indicators to ensure a more reasonable classification result.

[0041] Traditional monitoring systems often require manual adjustment of data acquisition parameters, such as adjusting the size of the monitoring window, resulting in slow response speed and low automation. When an abnormality occurs in the communication protocol (such as sudden network jitter and increased packet loss rate), the system cannot adjust the monitoring strategy in time, easily leading to data loss or inaccurate monitoring. The present invention adopts a feedback module, which automatically sends different types of feedback signals according to the performance level of the communication protocol: High-performance protocol (Class A1) → Narrow the monitoring window to reduce the system's computational burden. Medium-performance protocol (Class A2) → Widen the monitoring window to improve data monitoring accuracy. Low-performance protocol (Class B) → Restore the monitoring window to the default value to ensure sufficient data sampling for protocol optimization. The feedback mechanism realizes automatic optimization without manual intervention, ensuring that the system can adapt to the dynamic changes of the industrial network in real time, improving stability and response speed.

[0042] Traditional systems generally use a fixed time window for data monitoring. If the window is too small, it will lead to insufficient data sampling. If the window is too large, it will increase the computational overhead and affect real-time performance. The adjustment module of the present invention can adaptively adjust the length of the data monitoring window according to the type of feedback signal. By dynamically adjusting the window size, the monitoring system can reduce unnecessary consumption of computing resources while ensuring data accuracy, improving the acquisition efficiency. Description of the Drawings

[0043] For the convenience of those skilled in the art to understand, the present invention will be further described below in conjunction with the drawings;

[0044] Figure 1 It is the schematic diagram of the industrial control network data monitoring system based on the cloud platform in the present invention. Detailed Embodiments

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0046] Embodiment 1:

[0047] With the rapid development of industrial automation and industrial Internet, industrial control networks have become important infrastructures in key industries such as manufacturing, energy, and transportation. However, traditional industrial control networks face various challenges in data monitoring, transmission optimization, and intelligent regulation: Industrial control networks usually involve multiple communication protocols (such as Modbus, OPCUA, EtherCAT, etc.), and different protocols have differences in transmission delay, throughput capacity, and data overhead. Traditional local monitoring methods are difficult to effectively coordinate multi-protocol environments, resulting in great difficulties in network performance monitoring and optimization.

[0048] In industrial production, the performance of communication protocols directly affects production efficiency, device collaboration, and the stability of remote control. Traditional monitoring systems usually rely on fixed rule settings and cannot be intelligently adjusted according to the dynamic changes of network status. Modern industrial control systems are evolving towards a "cloud + edge + end" architecture. By using the powerful data processing capabilities of cloud computing, intelligent analysis, optimization, and prediction of network data can be achieved. However, existing industrial control systems generally lack a cloud monitoring and intelligent regulation mechanism for the usage of communication protocols.

[0049] The present invention aims to solve the following key problems:

[0050] Intelligent monitoring and evaluation of communication protocols are required: Due to the performance differences of different communication protocols, the system needs to collect the protocol usage in real time and calculate the network timeliness index (NTI), network throughput capacity index (NTCI), and network transmission efficiency index (NTEI) to comprehensively evaluate the transmission performance of the protocol.

[0051] Adaptive adjustment of data monitoring strategies is required: The monitoring windows of existing industrial networks are usually fixed and difficult to adapt to the dynamically changing network environment. Through a fuzzy inference module, the performance of communication protocols can be classified as high, medium, and low, and accordingly, the length of the data monitoring window can be adjusted to improve monitoring accuracy and network adaptability.

[0052] Optimization of network transmission based on a feedback mechanism is required: According to the actual performance of the communication protocol, the system needs to automatically generate feedback signals (A1, A2, B types) to guide the adjustment strategy of the monitoring system to adapt to the performance characteristics of different protocols.

[0053] Improving the Stability and Efficiency of Industrial Control Networks: Through in-depth analysis and optimization of the performance of communication protocols, this system can reduce data transmission latency, improve throughput capacity, and reduce protocol overhead, thereby enhancing the data transmission efficiency of the entire industrial control network.

[0054] The present invention proposes an industrial control network data monitoring system based on a cloud platform, including a data acquisition module, a data analysis module, a fuzzy inference module, a feedback module, and an adjustment module;

[0055] The data acquisition module is used to collect data on the usage of communication protocols between the current industrial control network and the cloud platform in real time, obtaining a dataset of communication protocol usage; the performance of the communication protocol of the industrial control network changes over time, and this module ensures the acquisition of the latest network data. The collected data is used as input for the data analysis module to evaluate the performance of the communication protocol. Communication protocols include Modbus, OPCUA, MQTT, EtherCAT, etc.

[0056] The data analysis module is used to analyze the performance of communication protocol usage based on the dataset of communication protocol usage, obtaining a performance evaluation data group and delivering it to the fuzzy inference module; by analyzing the dataset of communication protocol usage, calculate the key performance indicators of the communication protocol (Network Timeliness Index NTI, Network Throughput Capacity Index NTCI, Network Transmission Efficiency Index NTEI), form a performance evaluation data group, and deliver it to the fuzzy inference module. The performance differences of different protocols are obvious, and this module can provide objective indicators to ensure data-based optimization decisions. Problems such as high latency, low throughput, and excessive overhead can be discovered in advance through data analysis, reducing the impact of communication failures. The calculated performance evaluation data group is the basis for the fuzzy inference module to classify the protocol performance.

[0057] The fuzzy inference module is used to infer the performance of the currently used communication protocol based on the performance evaluation data group, classifying the performance of the currently used communication protocol into high-performance protocols, medium-performance protocols, and low-performance protocols; based on the performance evaluation data group (NTI, NTCI, NTEI), classify the currently used communication protocol into high-performance protocols, medium-performance protocols, and low-performance protocols. Use fuzzy logic (FuzzyLogic) for inference to make the classification more flexible and adaptable to complex network environments. The industrial network environment is complex, and the performance of communication protocols is not absolutely fixed. Fuzzy inference can handle non-linearity and uncertainty, improving adaptability. The classification result directly affects the signal output of the feedback module, providing a basis for subsequent optimization operations. For example, when classified as a low-performance protocol, communication can be improved through preset rules, including but not limited to protocol switching and other methods.

[0058] The feedback module is used to send feedback signals of corresponding types according to the classification types of the performance of the currently used communication protocol; the feedback signals determine whether it is necessary to adjust the length of the data monitoring window to improve the system response speed. It avoids manual intervention, realizes intelligent network monitoring and adjustment, and improves communication efficiency. When the performance of the communication protocol fluctuates, the feedback mechanism can respond quickly and optimize the monitoring strategy.

[0059] The adjustment module is used to adjust the length of the data monitoring window according to the type of the feedback signal. When the performance of the communication protocol is poor, it automatically increases the monitoring intensity to capture the cause of the problem. When the performance of the communication protocol is stable, it reduces the consumption of monitoring resources and improves the system operation efficiency. In different network environments, the length of the monitoring window can be dynamically adjusted to make the monitoring system more intelligent.

[0060] The following is an example of improving communication through preset rules mentioned in the present invention:

[0061] Select efficient protocols: MQTT based on the TCP / IP protocol: suitable for industrial Internet of Things, supports QoS levels, and can reduce data loss and retransmission. PROFINET, EtherCAT: optimized for industrial control systems, with lower network latency, suitable for high-real-time scenarios.

[0062] Optimize the data transmission method: Batch data transmission: Avoid uploading a small amount of data each time, but use batch uploading to reduce the number of requests. Edge computing preprocessing: Preprocess, compress, and aggregate data on the local gateway or edge device to reduce the amount of data uploaded.

[0063] Protocol conversion optimization: Use a protocol gateway to convert inefficient protocols into efficient protocols, reducing the time cost of data parsing and conversion. For example, ModbusRTU can be used on field devices, but converted to MQTT at the gateway for cloud transmission to improve transmission efficiency.

[0064] Studying the upload delay and cloud computing response delay caused by low-quality communication protocols has important value in the fields of industrial control, Internet of Things, intelligent manufacturing, etc. In industrial automation and remote monitoring, device status data needs to be synchronized to the cloud in real time. If the communication protocol causes data lag, it may lead to misjudgment and affect production efficiency. For example, in applications such as oil pipeline monitoring and smart grid, if the sensor data upload is delayed, it may affect accident warning and real-time scheduling.

[0065] Studying communication protocol optimization can reduce problems such as data loss and duplicate transmission, ensure the integrity and consistency of industrial data, and improve the accuracy of fault diagnosis. Inefficient protocols may lead to chaotic data formats and excessive redundant data, increasing the costs of cloud storage and computing. Studying more efficient communication protocols and data compression methods helps reduce the burden on cloud computing and lower enterprise operating costs.

[0066] Performing a performance analysis of the communication protocol to obtain a performance evaluation data set means:

[0067] By analyzing the data transmission delay of the communication protocol, obtain the network timeliness index;

[0068] By analyzing the data transmission efficiency of the communication protocol, obtain the network throughput capacity index;

[0069] By analyzing the data transmission overhead of the communication protocol, obtain the network transmission efficiency index;

[0070] Summarize the network timeliness index, network throughput capacity index, and network transmission efficiency index to obtain the performance evaluation data set. Quantitatively analyze the usage of the communication protocol to evaluate its transmission delay, throughput capacity, and transmission overhead in the industrial control network, providing data support for optimizing network monitoring and regulation strategies. Network timeliness index (NTI): Evaluate the delay and stability of data transmission to ensure network real-time performance. Network throughput capacity index (NTCI): Evaluate the data transmission efficiency to ensure that the throughput capacity of the protocol meets the requirements of industrial applications. Network transmission efficiency index (NTEI): Evaluate the additional overhead of the protocol (such as protocol headers, control data, etc.) and measure the bandwidth utilization and transmission efficiency.

[0071] The acquisition logic of the network timeliness index is:

[0072] Within a fixed time window, collect the sending time and receiving time of each data packet, and then calculate the average value of the difference between the receiving time and the sending time to obtain the average transmission delay T avg , which reasonably reflects the overall delay level of network transmission and is applicable to different industrial communication protocols (TCP / IP, PROFINET);

[0073] Then calculate the delay jitter value, and the calculation formula is:

[0074] T arrival，i+1 represents the (i + 1)-th receiving time, T send，i+1 represents the (i + 1)-th sending time, T arrival，i represents the i-th receiving time, T send，iDenote the \(i\)-th transmission time as \(t_i\), \(N\) represents the total number of data packets, and \(J\) represents the jitter value of the delay; calculating the jitter using the delay difference between adjacent data packets can capture the instantaneous fluctuations of the network and is applicable to industrial control systems with requirements for low latency and high stability. Further, if the network environment has high noise, the root mean square (RMS) jitter can be considered to replace the absolute value average to avoid the influence of a single extreme value:

[0075] Obtain the number of data packets whose difference between the received time and the transmitted time exceeds the preset standard duration, and divide it by the total number of data packets to get the timeout rate \(T\). timeout ; Calculating the timeout rate of data packets can intuitively reflect the network stability and packet loss situation, and is applicable to scenarios sensitive to timeouts such as industrial automation and remote control.

[0076] The calculation formula for the network timeliness index is:

[0077] \(\alpha\), \(\beta\), and \(\gamma\) are all preset non-zero proportional coefficients to adapt to different application scenarios (such as industrial Internet of Things vs real-time video transmission), and satisfy \(\alpha+\beta+\gamma = 1\); \(NTI\) represents the network timeliness index. Calculating \(NTI\) using an inverse proportional relationship ensures that the numerical range is in \((0,1]\), which can adapt to different network states. The larger the network timeliness index (\(NTI\)), the smaller the data transmission delay, the more stable the delay jitter, and the lower the timeout rate of the current communication protocol, that is, the better the timeliness of the network and the higher the real-time performance of data transmission. In an industrial control network, a larger \(NTI\) indicates that the communication protocol can respond to data requests faster, reduce transmission delays, and improve the stability and reliability of the system, and is applicable to scenarios with high requirements for high real-time and high stability, such as industrial automation control, remote monitoring, and intelligent manufacturing. If \(NTI\) is too small, it means that there are large delays or instabilities in communication, which may affect production efficiency and device synchronization.

[0078] The acquisition logic for the network throughput capacity index is as follows:

[0079] Within a fixed time window, sum the data volume of each data packet to obtain the total data volume, then divide it by the duration corresponding to the fixed time window to get the actual throughput, and divide the number of lost data packets by the total number of data packets to get the packet loss rate;

[0080] The calculation formula for the network throughput capacity index is:

[0081] \(PLR\) is the packet loss rate, \(T\) max is the preset maximum throughput, \(T\) actual is the actual throughput, \(k\) and \(\theta\) are both preset non-zero control coefficients, and \(NTCI\) is the network throughput capacity index. Using the throughput ratio As a core measurement indicator, this ratio directly reflects whether the actual throughput of the current communication protocol is close to the maximum theoretical value. If the ratio is close to 1, it means that the protocol is close to the maximum throughput and the data transmission efficiency is high. The Sigmoid function is used to smooth the impact of the throughput ratio on the index. Traditional linear calculation may cause NTCI to drop too steeply when the throughput is low, and NTCI to approach 1 too quickly when the throughput is high. The Sigmoid function throughput ratio makes the throughput ratio transition smoothly between low and high values. k controls the steepness of the curve, and θ controls the position of the inflection point to improve the stability and adaptability of the calculation. The hyperbolic tangent function 1-tanh (PLR) is introduced to correct the impact of the packet loss rate on the throughput. Too high a packet loss rate (PLR) will cause data retransmission and reduce the actual throughput. Directly using the packet loss rate may cause excessive impact in extreme cases. Therefore, tanh (PLR) is used for nonlinear mapping, so that the impact of the packet loss rate on NTCI is small at low values ​​and significantly increased at high values. This ensures that when the packet loss rate is close to 0, NTCI will not be affected too much, and when the packet loss rate is higher, NTCI will drop more significantly, reflecting the negative effects of packet loss.

[0082] The larger the network throughput index NTCI is, the higher the data transmission efficiency and throughput of the current communication protocol is, that is, the protocol can transmit data closer to its maximum throughput, with a lower packet loss rate and more stable data transmission. In industrial control networks, a higher NTCI indicates that the communication protocol is suitable for high-bandwidth demand scenarios such as large data volume transmission, streaming media, and remote monitoring, and can ensure faster data flow and less transmission loss. If the NTCI is too small, it means that the throughput is limited, and there may be problems such as low bandwidth utilization, high packet loss rate, or poor transmission efficiency of the protocol itself, which will affect the stable operation of industrial equipment.

[0083] The logic for obtaining the network transmission efficiency index is as follows:

[0084] In a fixed time window, the available bandwidth value measured at each sampling time point is obtained and the average value is calculated to obtain the average bandwidth. The actual throughput is divided by the average bandwidth to obtain the bandwidth utilization BU. The effective data volume of each data transmission is summed to obtain the effective data load value FZ. The effective data and the additional overhead data volume of each data transmission are summed together to obtain the total transmission data volume QB. The network transmission efficiency index calculation formula is: NTEI is the Network Transmission Efficiency Index. The Bandwidth Utilization (BU) is introduced to measure the usage of network resources. This metric can reflect the utilization rate of network resources by the current communication protocol. A higher BU indicates that the protocol effectively uses the network bandwidth and avoids waste. The logarithmic function is used to process the proportion of payload data, FZ (value of valid data payload): representing the truly useful amount of information in the transmitted data. QB (total amount of transmitted data, including protocol overhead): includes valid data and protocol additional overheads, such as control messages, retransmitted data, etc. Through logarithmic transformation, the limitations of linear calculation are avoided, making the change of NTEI smoother and preventing extreme impacts when FZ and QB are too large or too small. The relative scaling of FZ / QB is performed to ensure that NTEI can reasonably reflect the transmission efficiency of the protocol under different data scales. Combining BU and the data payload ratio improves the rationality of the calculation. The higher the proportion of the valid data payload value FZ and the smaller the protocol additional overhead QB, the closer NTEI is to 1, indicating a higher transmission efficiency of the communication protocol. The larger the Network Transmission Efficiency Index NTEI, the higher the data transmission efficiency of the current communication protocol, that is: the bandwidth utilization is higher, and the communication protocol makes full use of the available bandwidth resources. The protocol overhead is smaller, and the proportion of the protocol's control data, handshake data, and retransmitted data is low, and more bandwidth is utilized by the valid data payload (FZ). The data transmission efficiency is high, suitable for application scenarios with low latency and high throughput requirements, such as industrial automation control, 5G communication, intelligent manufacturing, etc. If NTEI is too small, it means: there may be bandwidth waste, and the actual throughput is much lower than the available bandwidth (low BU). It may be due to a relatively high protocol additional overhead, resulting in a low proportion of valid data transmission and affecting the communication efficiency. There may be problems such as frequent retransmissions, too large protocol headers, or flow control mechanism restrictions, reducing the transmission efficiency.

[0085] The principle of using the fuzzy inference module is as follows: The network timeliness index, network throughput capacity index, and network transmission efficiency index corresponding to the same fixed time window are used as input variables together, and the type to which the performance of the currently used communication protocol belongs is used as the output variable. The input variables are fuzzified, converting the values of the input variables into fuzzy sets, the output variable is fuzzified, converting the output variable into a fuzzy set, and fuzzy rules are formulated to describe the fitness of each type to which it belongs under different combinations of data types. The fuzzified input variables are inferred through the fuzzy rules to obtain the type to which the performance of the currently used communication protocol belongs.

[0086] The fuzzy inference module is used to intelligently classify the performance of the communication protocol. Based on the network timeliness index, network throughput capacity index, and network transmission efficiency index, it infers the performance level of the currently used communication protocol and classifies it into high-performance, medium-performance, and low-performance. This module mainly includes the following steps:

[0087] First, convert the network timeliness, network throughput capacity, and network transmission efficiency calculated within a fixed time window into fuzzy variables. Since these metrics are continuous numerical values, directly using them in practical applications may lead to unclear boundary conditions, so fuzzy processing is required.

[0088] Fuzzy method: Set up fuzzy sets. For example, network timeliness can be divided into: high, medium, low; network throughput capacity can be divided into: high, medium, low; network transmission efficiency can be divided into: high, medium, low; Use membership functions to represent the matching degree between the input variables and each fuzzy set. Trigonometric functions can be used for membership degree calculation. Similar to the input variables, the performance levels (high, medium, low) of communication protocols also need to be fuzzified: High performance is applicable to communication protocols with low latency, high throughput, and high transmission efficiency. Medium performance is applicable to general industrial applications, allowing a certain amount of delay or lower throughput. Low performance is applicable to communication protocols with network instability and low transmission efficiency. The membership function can be defined using triangular or trapezoidal functions similar to the input variables for different performance levels.

[0089] According to the basic principles of fuzzy logic, establish a rule base to map different combinations of input variables to output variables. For example: If the network timeliness is high, the network throughput capacity is high, and the network transmission efficiency is high, then the communication protocol belongs to high performance. If the network timeliness is medium, the network throughput capacity is medium, and the network transmission efficiency is high, then the communication protocol belongs to medium performance. If the network timeliness is low, the network throughput capacity is low, and the network transmission efficiency is low, then the communication protocol belongs to low performance. According to the fuzzified values of the input variables and the rule base, use a fuzzy inference algorithm to calculate the membership of the output variable. Common methods include: Maximum membership degree method: Select the classification with the highest membership degree in the output variable as the final result. Weighted average method: Calculate the comprehensive weights of each classification to obtain a more accurate performance classification result.

[0090] The result of fuzzy inference is still a fuzzy set, and defuzzification (anti-fuzzification) is required to convert it into specific high, medium, and low performance classifications for communication protocols. The weighted average method can be used to calculate the final output value, and the obtained value can be used to determine the performance category of the communication protocol. Example:

[0091] Suppose the network timeliness of a certain communication protocol is 0.8 (high), the throughput capacity is 0.6 (medium), and the transmission efficiency is 0.7 (high). Through fuzzy rule inference, this protocol may be classified as high performance and trigger a type A1 feedback signal. In this way, the fuzzy inference module can intelligently adapt to different network conditions, making the performance classification of communication protocols more accurate and flexible.

[0092] The feedback module is used to send feedback signals of corresponding types according to the classification types of the performance of the currently used communication protocol, which means that when the performance of the currently used communication protocol is classified as a high-performance protocol, it sends A1-class feedback signals; when it is classified as a medium-performance protocol, it sends A2-class feedback signals; when it is classified as a low-performance protocol, it sends B-class feedback signals. When the communication protocol performance is good (high-performance protocol): feedback signal of A1 class → appropriately reduce the monitoring window, reduce the consumption of monitoring resources, and improve the system operation efficiency. When the communication protocol performance is average (medium-performance protocol): feedback signal of A2 class → appropriately expand the monitoring window, improve the monitoring accuracy to ensure data stability. When the communication protocol performance is poor (low-performance protocol): feedback signal of B class → restore the monitoring window length to the default value to ensure that the system can conduct sufficient monitoring to identify problems with the optimized communication protocol. The common optimization methods have been exemplified in the previous content.

[0093] The adjustment module is used to adjust the length of the data monitoring window according to the type of the feedback signal, which means that:

[0094] When the type of the feedback signal is A2 class, the adjustment formula is:

[0095] SJ2 is the adjusted data monitoring window length when the feedback signal type is A2 class;

[0096] When the type of the feedback signal is A1 class, the adjustment formula is:

[0097] SJ1 is the adjusted data monitoring window length when the feedback signal type is A1 class; Smin and Smax are the preset adjustment ranges of the data monitoring window length respectively, Smin < Smax, ε is the preset adjustment coefficient, and SC represents the current data monitoring window length;

[0098] When the type of the feedback signal is B class, the data monitoring window length is restored to

[0099] Feedback signal of A2 class: The monitoring window expands, which is applicable to the situation where the communication protocol performance is average, indicating that there may be certain fluctuations or load pressure in the network. After the monitoring window expands, more data samples can be obtained, improving the perception ability of the protocol state to more precisely adjust the network optimization strategy.

[0100] Feedback signal of A1 class: The monitoring window shrinks, which is applicable to the situation where the communication protocol performance is good, indicating that the protocol transmission is stable, with small errors and an ideal network environment. After the monitoring window shrinks, the overhead of monitoring calculations can be reduced, the consumption of system resources can be lowered, and the monitoring efficiency can be improved simultaneously.

[0101] Feedback signal type B: The monitoring window is restored to the default value, which is applicable to the situation where the communication protocol performance is poor, indicating unstable network quality and low protocol performance. At this time, improvement and optimization processing are required, and the monitoring window is restored to the default value to ensure normal monitoring of the improved and optimized communication protocol.

[0102] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0103] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution is prior or posterior. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0104] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0105] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0106] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. The industrial control network data monitoring system based on cloud platform is characterized by: It includes data acquisition module, data analysis module, fuzzy reasoning module, feedback module and adjustment module; The data acquisition module is used to collect the communication protocol usage data between the current industrial control network and the cloud platform in real time to obtain the communication protocol usage data set; The data analysis module is used to analyze the performance of the communication protocol according to the communication protocol usage data set, obtain the performance evaluation data set and transmit it to the fuzzy reasoning module; The fuzzy reasoning module is used to infer the performance of the currently used communication protocol according to the performance evaluation data group, and divide the performance of the currently used communication protocol into high performance protocol, medium performance protocol and low performance protocol; The feedback module is used to send a feedback signal of a corresponding type according to the classification type of the communication protocol performance currently used; The adjustment module is used to adjust the length of the data monitoring window according to the type of the feedback signal.

2. The cloud platform-based industrial control network data monitoring system according to claim 1 is characterized in that: The performance analysis of the communication protocol is performed to obtain the performance evaluation data set, which refers to: By analyzing the data transmission delay of the communication protocol, the network timeliness index is obtained; By analyzing the data transmission efficiency of the communication protocol, the network throughput index is obtained; By analyzing the data transmission overhead of the communication protocol, the network transmission efficiency index is obtained; The network timeliness index, network throughput index, and network transmission efficiency index are summarized to obtain a performance evaluation data set.

3. The cloud platform-based industrial control network data monitoring system according to claim 2 is characterized in that: The logic for obtaining the network timeliness index is: In a fixed time window, the sending time and receiving time of each data packet are collected, and then the average difference between the receiving time and the sending time is calculated to obtain the average transmission delay T avg ; Then calculate the delay jitter value, the calculation formula is: Indicates the i+1th receiving time, T send，i+1 Indicates the i+1th sending time, T arrival，i represents the i-th receiving time, T send，i represents the i-th sending time, N represents the total number of data packets, and J represents the delay jitter value; Obtain the number of packets whose difference between the receiving time and the sending time exceeds the preset standard time length, and divide it by the total number of packets to obtain the timeout rate T timeout ; The network timeliness index calculation formula is: α, β, and γ are all preset non-zero proportional coefficients, and satisfy α+β+γ=1; NTI represents the network timeliness index.

4. The cloud platform-based industrial control network data monitoring system according to claim 3 is characterized in that: The logic for obtaining the network throughput index is as follows: In a fixed time window, the data volume of each data packet is summed to obtain the total data volume, and then divided by the duration corresponding to the fixed time window to obtain the actual throughput. The number of lost data packets is divided by the total number of data packets to obtain the packet loss rate. The network throughput index calculation formula is: PLR is the packet loss rate, T max is the preset maximum throughput, T actual is the actual throughput, k and θ are both preset non-zero control coefficients, and NTCI is the network throughput index.

5. The cloud platform-based industrial control network data monitoring system according to claim 4 is characterized in that: The logic for obtaining the network transmission efficiency index is as follows: In a fixed time window, the available bandwidth value measured at each sampling time point is obtained and the average value is calculated to obtain the average bandwidth. The actual throughput is divided by the average bandwidth to obtain the bandwidth utilization BU. The effective data volume of each data transmission is summed to obtain the effective data load value FZ. The effective data and the additional overhead data volume of each data transmission are summed together to obtain the total transmission data volume QB. The network transmission efficiency index calculation formula is: NTEI is the Network Transmission Efficiency Index.

6. The cloud platform-based industrial control network data monitoring system according to claim 5, characterized in that: The usage principle of the fuzzy reasoning module is: The network timeliness index, network throughput index, and network transmission efficiency index corresponding to the same fixed time window are taken as input variables, and the type of the currently used communication protocol performance is taken as the output variable. The input variables are fuzzified and the values ​​of the input variables are converted into fuzzy sets. The output variables are fuzzified and the output variables are converted into fuzzy sets. Fuzzy rules are formulated to describe the adaptability of each type under different data type combinations. The fuzzified input variables are inferred through fuzzy rules to obtain the type of the currently used communication protocol performance.

7. The cloud platform-based industrial control network data monitoring system according to claim 6, characterized in that: The feedback module is used to send out feedback signals of corresponding types according to the classification type of the performance of the currently used communication protocol: When the performance of the currently used communication protocol is classified as a high-performance protocol, a Class A1 feedback signal is issued; when it is classified as a medium-performance protocol, a Class A2 feedback signal is issued; when it is classified as a low-performance protocol, a Class B feedback signal is issued.

8. The cloud platform-based industrial control network data monitoring system according to claim 7, characterized in that: The adjustment module is used to adjust the data monitoring window length according to the type of feedback signal, which refers to: When the feedback signal type is A2, the adjustment formula is: SJ2 is the adjusted data monitoring window length when the feedback signal type is A2; When the feedback signal type is A1, the adjustment formula is: When the feedback signal type of SJ1 is type A1, it is the adjusted data monitoring window length; Smin and Smax are respectively the preset adjustment range of the data monitoring window length, Smin < Smax, ε is the preset adjustment coefficient, and SC represents the current data monitoring window length; When the feedback signal type is Class B, the data monitoring window length is restored to

Images