Network comprehensive performance test system and method suitable for multi-protocol industrial Ethernet

Abstract

The invention belongs to the field of industrial Ethernet testing, and particularly relates to a network comprehensive performance testing system and method suitable for a multi-protocol industrial Ethernet. The system is mainly composed of a communication link monitoring unit, a signal excitation and acquisition front end and a network performance comprehensive evaluation module. Wherein the network performance comprehensive evaluation module comprises a network load analysis sub-module, a slave station synchronization precision analysis sub-module, a communication cycle precision analysis sub-module, a slave station transmission delay analysis sub-module and a full task execution cycle performance analysis sub-module; therefore, indexes such as network load, communication cycle precision, slave station transmission delay and synchronization precision, full task execution cycle performance and the like of industrial Ethernet control networks such as EtherCAT, Profinet and the like can be tested in parallel, and comprehensive evaluation of the performance of the industrial Ethernet network is realized through data fusion and correlation analysis. And guidance is provided for performance optimization of the industrial Ethernet system.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of industrial Ethernet test, and particularly relates to a network comprehensive performance test system and method suitable for multi-protocol industrial Ethernet. BACKGROUND

[0002] With the rapid development of artificial intelligence, Internet of Things, 5G and other technologies, industrial automation and intelligentization have become the general trend. Industrial Ethernet protocols such as EtherCAT and Profinet have been widely used in the field of industrial automation due to their high bandwidth, excellent real-time performance, low transmission delay and good network compatibility, which perfectly meet the requirements of real-time control systems such as logical control and motion control. With the in-depth application and scale expansion, it is imperative to comprehensively and accurately test and evaluate the performance of industrial Ethernet network.

[0003] In the prior art, the network performance test methods for industrial Ethernet mainly include using a network analyzer to capture Ethernet data frames, calling specific APIs of the master station protocol stack, and using an oscilloscope to measure the output signals of the master and slave stations. These test methods often have problems such as single test index, insufficient protocol depth analysis, poor multi-protocol applicability, etc. For users who need to fully understand the performance of a certain industrial Ethernet network, the solutions are not implementable. SUMMARY

[0004] In view of the problems existing in the prior art, the present application provides a network comprehensive performance test system and method suitable for multi-protocol industrial Ethernet. The system mainly comprises a communication link monitoring unit, a signal excitation and collection front end, and a network performance comprehensive evaluation module. It can perform parallel testing on multiple indexes of industrial Ethernet control network such as EtherCAT, Profinet, network load, communication cycle accuracy, slave station transmission delay and synchronization accuracy, and full task execution cycle performance. Through data fusion and correlation analysis, it realizes all-round evaluation of the performance of industrial Ethernet network, and provides guidance for performance optimization of industrial Ethernet system.

[0005] The technical scheme of the present application is as follows: A network comprehensive performance test system suitable for multi-protocol industrial Ethernet, comprising a communication link monitoring unit, a signal excitation and collection front end, and a network performance comprehensive evaluation module, wherein, The communication link monitoring unit is used for real-time capture and analysis of data communication in the multi-protocol industrial Ethernet network link; The signal excitation and collection front end is used for closed-loop interaction with the slave module of the multi-protocol industrial Ethernet at the electrical level; The network performance comprehensive evaluation module is used to perform fusion processing and deep correlation analysis on heterogeneous data from the communication link monitoring unit and the signal excitation and acquisition front end, so as to realize the comprehensive performance evaluation of multi-protocol industrial Ethernet.

[0006] Preferably, the communication link monitoring unit includes a network analyzer and a network packet analysis tool. The network analyzer is connected to the bus network, and the network packet analysis tool is deployed on a PC. The signal excitation and acquisition front end includes a signal generator and a multi-channel oscilloscope. The signal generator is connected to the input point of the slave module and is responsible for applying excitation signals to the slave. The multi-channel oscilloscope is connected to the output point of the slave module and is responsible for acquiring slave response signals. The network performance comprehensive evaluation module includes a network load analysis submodule, a slave synchronization accuracy analysis submodule, a communication cycle accuracy analysis submodule, a slave transmission delay analysis submodule, and a full task execution cycle performance analysis submodule.

[0007] Preferably, the network analyzer accesses the bus network losslessly via a physical TAP, captures all data frames exchanged between the master and slave stations in real time, and adds a high-precision hardware timestamp to each data frame. The network packet analysis tool filters periodic data frames and extracts process data from them.

[0008] Preferably, the signal generator generates a standard digital or analog excitation signal and applies it to the input point of the slave module, and the multi-channel oscilloscope acquires the response signal at the output point of the slave module.

[0009] Preferably, the network load analysis submodule performs statistics on data frames, identifies the proportion of various frames and calculates bandwidth utilization, automatically identifies atypical communication patterns based on a deep learning-based traffic anomaly monitoring model, predicts load trends, and diagnoses frame structure anomalies.

[0010] Preferably, the slave station synchronization accuracy analysis submodule evaluates the synchronization performance by measuring the response time difference of multiple slave station modules to output signal commands, uses a clustering algorithm to group and identify the synchronization performance of the slave station modules, and identifies abnormal synchronization nodes through a collaborative filtering method.

[0011] Preferably, the communication cycle accuracy analysis submodule uses Kalman filtering and multi-cycle dynamic time warping algorithm based on the timestamp of the periodic data frame to analyze whether there are periodic jitter, long-term drift trends and sudden abnormal modes in the communication cycle, and finally obtains the communication cycle accuracy.

[0012] Preferably, the slave station transmission delay analysis submodule calculates the cumulative transmission delay of a single station or multiple stations by capturing the time difference between each slave station module before and after in the data frame path, and separates deterministic delays and identifies long-term trends based on signal decomposition and regression analysis methods, thereby achieving interpretable analysis of the delay.

[0013] Preferably, the full task execution cycle performance analysis submodule takes the frequency domain response curve as the core and integrates parameter identification algorithms. It uses machine learning models to establish a dynamic mathematical model of the system, analyzes the impact of different system parameters on the dynamic characteristics of multi-protocol industrial Ethernet, comprehensively evaluates the dynamic response performance of multi-protocol industrial Ethernet under real task conditions, and provides guidance for the optimization of master station controller parameters.

[0014] A network comprehensive performance testing method based on the aforementioned network comprehensive performance testing system includes the following steps: Step S1: Build an industrial Ethernet control network. The master station controller writes a PLC program to send control and data acquisition tasks to each slave station module. The master and slave stations exchange data through periodic communication. Step S2: Connect the network analyzer in the communication link monitoring unit to the master-slave station or between slave stations to capture all data frames in the communication link in real time. The network packet analysis tool processes the captured data frames, filters out periodic data frames, and extracts process data.

[0015] Step S3: The signal generator in the signal excitation and acquisition front end outputs a sinusoidal sweep frequency signal to each slave station analog input module, and the multi-channel oscilloscope acquires the first response signal output by each slave station analog output module; Step S4: In response to the synchronization accuracy test requirements of the slave modules, the master controller controls the digital output modules of each slave module to output step signals, and a multi-channel oscilloscope acquires the second response signals output by each slave module. Step S5: The communication link monitoring unit and the signal excitation and acquisition front end send the obtained data frames and response signals to the network performance comprehensive evaluation module. Each performance analysis sub-module performs data processing and analysis to obtain the test results of each performance index. Step S6: The network performance comprehensive evaluation module performs data fusion and correlation analysis on the test results of each sub-module, reveals their inherent coupling correlation, comprehensively evaluates network performance, and outputs a comprehensive performance evaluation report of multi-protocol industrial Ethernet.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention constructs a multi-protocol, multi-dimensional, non-intrusive network comprehensive performance testing system, which is compatible with various industrial Ethernet protocols (based on the IEEE 802.3 standard), such as EtherCAT, Profinet, and Ethernet / IP. It supports multi-dimensional parallel testing of key performance indicators such as communication real-time performance, synchronization accuracy, and dynamic performance. By integrating correlation analysis and constructing a performance correlation reasoning model based on knowledge graphs, it reveals the inherent coupling relationship of various performance indicators, comprehensively evaluates the overall network performance, and automatically generates a comprehensive evaluation report, which can provide guidance for the performance optimization of industrial Ethernet systems.

[0017] 2. This invention proposes a network comprehensive performance testing system suitable for multi-protocol industrial Ethernet. It innovatively introduces frequency domain response analysis methods used in linear control systems into industrial Ethernet control systems, introduces the concept of full-task execution cycle performance, and employs a "frequency sweep signal excitation-response acquisition" approach to evaluate the dynamic response characteristics, system stability, and other comprehensive performance indicators of the entire bus network system. By establishing a system dynamic model through system identification and machine learning algorithms, the system's full-task cycle performance is analyzed, comprehensively improving the accuracy and depth of the evaluation. Automatic parameter tuning is achieved based on reinforcement learning methods, providing a basis for optimizing the master station controller parameters.

[0018] 3. The network comprehensive performance testing system for multi-protocol industrial Ethernet proposed in this invention integrates multiple advanced algorithms to achieve intelligent analysis. These include: introducing a deep learning model for self-diagnosis and prediction of traffic anomalies; employing Kalman filtering and DTW algorithms to improve the accuracy of periodic measurements; combining EMD and regression analysis to achieve interpretable analysis of transmission delay; and applying clustering algorithms to evaluate slave station synchronization accuracy and identify abnormal nodes. By introducing these intelligent algorithms, the system further identifies abnormal network patterns and performance change trends, serving network performance prediction, fault diagnosis, and comprehensive performance optimization.

[0019] 4. The network comprehensive performance testing system for multi-protocol industrial Ethernet proposed in this invention achieves zero-interference access to the original communication link through hardware TAP at the physical layer, and performs closed-loop signal interaction with the system under test through the electrical interface at the application layer. It does not require parsing or modifying the Ethernet protocol stack, nor does it require embedding any special test code in the main station. It has strong versatility and platform independence, and reduces the operational requirements for technical personnel. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the embodiments will be briefly introduced below. The features and advantages of the present invention can be more clearly understood by referring to the accompanying drawings. The accompanying drawings are schematic and should not be construed as limiting the present invention in any way. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the network comprehensive performance testing system for multi-protocol industrial Ethernet proposed in this invention.

[0022] Figure 2 This is a flowchart of the network comprehensive performance testing method proposed in this invention.

[0023] Figure 3 This is the network comprehensive performance evaluation report template of the present invention.

[0024] Figure 4 This is the network comprehensive performance evaluation report obtained in Example 1. Detailed Implementation

[0025] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments of the present invention and the features thereof can be combined with each other.

[0026] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0027] The architecture of the network comprehensive performance testing system for multi-protocol industrial Ethernet of the present invention is as follows: Figure 1 As shown, it consists of three parts: a communication link monitoring unit, a signal excitation and acquisition front-end, and a network performance comprehensive evaluation module.

[0028] The communication link monitoring unit, consisting of a network analyzer and a network packet analysis tool, is responsible for real-time capture and analysis of data communication in the network link. The network analyzer connects to the bus network losslessly via a physical TAP, capturing all data frames exchanged between the master and slave stations in real time and adding a high-precision hardware timestamp to each frame. The network packet analysis tool processes the captured data frames, filters periodic frames, and extracts process data, providing a structured data foundation for subsequent network performance testing and in-depth analysis.

[0029] The signal excitation and acquisition front-end, consisting of a signal generator and a multi-channel oscilloscope, enables closed-loop interaction with the slave modules of the network under test (DUT) at the electrical level. It executes a highly reliable excitation-response test process. The high-precision signal generator produces adjustable frequency and amplitude digital or analog excitation signals (such as step signals and sinusoidal sweep signals), which are applied to the input points of the slave modules in the DUT to simulate the state changes of real switches or sensors. The multi-channel oscilloscope synchronously acquires the response signals from the output points of the slave modules in the network, providing data support for the back-end network performance comprehensive evaluation module to assess the network's dynamic performance and frequency response characteristics.

[0030] The network performance comprehensive evaluation module is responsible for fusing and deeply analyzing heterogeneous data from the first two parts (communication link monitoring unit and signal excitation and acquisition front end) to achieve a comprehensive evaluation of the performance of multi-protocol industrial Ethernet networks. This module consists of sub-modules for network load analysis, slave synchronization accuracy analysis, communication cycle accuracy analysis, slave transmission delay analysis, and full task execution cycle performance analysis. It can test basic indicators such as network bandwidth utilization, clock synchronization accuracy, communication cycle, and slave transmission delay, as well as dynamic performance. Through data fusion and correlation analysis, it constructs a performance correlation reasoning model based on a knowledge graph, revealing the inherent coupling relationships between various indicators and performing quantitative analysis. For example, increased network load leads to decreased communication cycle accuracy; communication cycle accuracy and slave transmission delay affect slave synchronization accuracy; communication cycle accuracy and slave synchronization accuracy directly determine the full task execution cycle performance of the bus; and improving the full task execution cycle performance (such as shortening the communication cycle and transmitting more process data) leads to increased network load. Through graph query and graph reasoning algorithms, isolated performance indicators are transformed into an interconnected knowledge network, thereby achieving a deep evaluation of the performance of the industrial Ethernet system. By introducing intelligent algorithms to identify abnormal patterns and performance change trends in the network, guidance is provided for fault diagnosis and comprehensive performance optimization of industrial Ethernet networks.

[0031] The network performance comprehensive evaluation module is the core part of the network comprehensive performance testing system of this invention. It integrates multiple sub-functional modules such as network load analysis, slave synchronization accuracy analysis, communication cycle accuracy analysis and slave transmission delay analysis, and full task execution cycle performance analysis, so as to realize a multi-dimensional and in-depth comprehensive evaluation of the performance of industrial Ethernet networks.

[0032] The network load analysis submodule performs statistical analysis on data frames, identifies the proportion of different frame types (such as process data frames, clock synchronization frames, and link detection frames), and calculates bandwidth utilization. It introduces a deep learning-based traffic anomaly monitoring model to automatically identify atypical communication patterns, predict load trends, and possesses self-diagnostic capabilities for frame structure anomalies. Its bandwidth utilization calculation formula is as follows:

[0033] Where N is the total number of data frames, For network load rate, R represents the number of bytes occupied by all Ethernet data frames, and R represents the physical bandwidth of the industrial Ethernet (100Mbps).

[0034] The communication cycle accuracy analysis submodule uses the timestamp of the periodic process data frame and employs Kalman filtering and multi-cycle dynamic time warping (DTW) algorithm to smooth the interval between adjacent periodic data frames and extract jitter features. It analyzes whether there is periodic jitter, long-term drift trend and sudden abnormal mode in the communication cycle, thereby improving the measurement accuracy and reliability of the communication cycle.

[0035] The slave transmission delay analysis submodule captures the time difference before and after each slave station along the data frame path, calculates the cumulative transmission delay of a single station or multiple stations, isolates jitter components by combining signal decomposition methods such as Empirical Mode Decomposition (EMD), and applies regression analysis to model the correlation between delay and potential factors such as load, temperature, and humidity, separates deterministic delays and identifies long-term trends, and achieves interpretable delay analysis.

[0036] The slave synchronization accuracy analysis submodule evaluates synchronization performance by measuring the response time difference of multiple slave modules to output signal commands. It uses clustering algorithms (such as K-means) to group and identify slave synchronization performance, identifies abnormal synchronization nodes through collaborative filtering, and finally uses time series analysis to predict the degradation trend of slave synchronization performance and identify systematic synchronization accuracy decay.

[0037] The full-task execution cycle performance analysis submodule is used to comprehensively evaluate the dynamic response and control performance of industrial Ethernet under real-world task conditions. It introduces frequency domain response analysis methods used in linear control systems into industrial Ethernet control systems, integrating system identification and adaptive signal processing techniques. First, based on the imported "excitation-response" signals from the signal excitation and acquisition front-end, and after windowing the signals to suppress spectral leakage, the amplitude ratio (attenuation) and phase difference (hysteresis) of the signals at each frequency point are obtained through Fast Fourier Transform (FFT). Based on this, a frequency domain response curve is plotted, visually displaying the system's frequency response characteristics and determining the system bandwidth. To further enhance the evaluation depth and model generalization ability, the full-task execution cycle performance analysis submodule integrates a system identification algorithm. Utilizing parametric models such as ARX (Autoregressive Exogenous Input) or nonparametric models such as neural networks, it automatically identifies and establishes the system's dynamic mathematical model based on input-output data pairs. Furthermore, adaptive filters and phase compensation strategies are introduced to analyze the impact of different communication cycles and communication cycle accuracy on the dynamic characteristics of the system. The communication cycle and network load are balanced, and reinforcement learning algorithms are used for automatic parameter tuning to achieve adaptive dynamic optimization of parameters, providing a basis for the optimization of the main station controller parameters.

[0038] Based on this, the present invention also proposes a method for testing the overall performance of a network, such as... Figure 2 As shown, it includes the following steps: Step 1: Build an industrial Ethernet control network. Connect the master controller and each slave station via network cables to establish a linear hardware topology network, and complete the network configuration using configuration software. The master controller writes a PLC program to send control and data acquisition tasks to the slave stations. The master and slave stations exchange data through periodic communication.

[0039] Step 2: Connect the monitoring port of the network analyzer in the communication link monitoring unit to the master / slave station or between slave stations, and set its monitoring protocol type to enable it to capture data frames in the network in real time. Deploy a network packet analysis tool on the PC to perform preliminary processing on the captured data frames, filter out periodic data frames, and extract the process data from the data frames.

[0040] Step 3: At the signal excitation and acquisition front end, connect the signal output channel of the signal generator to the analog input module of the network slave station, and connect the signal acquisition channel of the multi-channel oscilloscope to the analog output module of the slave station to establish a signal channel for input excitation and output response. Set the signal generator to output a sinusoidal sweep frequency signal to each slave station's analog input module, and the oscilloscope acquires the response signal output by each slave station's analog output module.

[0041] Step 4: Specifically, for the requirements of slave station synchronization accuracy testing, the master station controller writes a control program to control the digital output modules of each slave station module to synchronously output step signals. The signal acquisition channel of the multi-channel oscilloscope acquires the output signals of each slave station module and sends them to the network performance comprehensive evaluation module for evaluation.

[0042] Step 5: The communication link monitoring unit and the signal excitation and acquisition front end send the acquired signals to the network performance comprehensive evaluation module. Each submodule processes and analyzes the data to obtain the test results of each performance evaluation index. Specifically, the network load analysis submodule calculates bandwidth utilization; the communication cycle accuracy analysis submodule and the slave transmission delay analysis submodule calculate communication cycle accuracy and slave transmission delay using data frame timestamps; the slave synchronization accuracy analysis submodule calculates slave synchronization accuracy using the delay time of different slave output digital or analog signals; and the full task execution cycle performance analysis submodule analyzes the system's dynamic performance by plotting slave frequency domain response curves.

[0043] Step 6: The network performance comprehensive evaluation module performs data fusion and correlation analysis on the test results of each submodule, establishes a performance correlation reasoning model based on a knowledge graph, and reveals the inherent correlation of each performance indicator through graph query and graph reasoning algorithms, and performs quantitative analysis to comprehensively evaluate the overall performance of the tested network. Intelligent algorithms are introduced into each submodule to identify abnormal patterns and performance change trends in the network, providing guidance for fault diagnosis and comprehensive performance optimization of industrial Ethernet networks. Finally, a comprehensive performance evaluation report for multi-protocol industrial Ethernet is output, with the following format: Figure 3 As shown.

[0044] Example 1 This embodiment constructs a typical EtherCAT control network covering logic control and I / O interaction. A Beckhoff CX2043 embedded controller is used as the master station, and three sets of I / O slave stations are configured, each consisting of an EK1100 coupler, an EL1252 / EL2252 digital input module, and an EL3702 / EL4732 analog input module. The master and slave stations, as well as the slave stations themselves, are connected via RJ45 network cables to establish a linear hardware topology network. TwinCAT3 software is used to complete the network configuration.

[0045] The PLC program is written using ST language in TwinCAT3 software. The program logic is that the EtherCAT master station collects the digital and analog signals from the IO slave input modules in real time and controls the output modules to output the collected signals in real time.

[0046] Deploy a communication link monitoring unit, connecting the TAP A and TAP B listening ports of the HIS NANL-B500G-RE network analyzer in series to the master / slave station or between slave stations, respectively. Set its listening protocol type to EtherCAT to enable real-time capture of data frames in the linear topology network (data frame timestamp resolution 1 nanosecond). Deploy the WireShark network packet analysis tool on the PC, filter the data frames captured by the network analyzer, obtain the EtherCAT periodic data frames, and analyze the service commands (such as LRD, LWR, LRW, etc.) within them.

[0047] The signal excitation and acquisition front-end is deployed by connecting the output channel of the GED 33600A signal generator to the analog input module of the IO slave station, and the signal input channel of the GED DSOX2024A four-channel oscilloscope to the analog output module of the IO slave station, thus establishing a signal channel for input excitation and output response. The signal generator is configured to output a sine wave signal with adjustable frequency and amplitude (frequency 1-1000Hz, amplitude 5V-10V), and the multi-channel oscilloscope captures the response signal from the output module.

[0048] A comprehensive network performance evaluation module is deployed on the PC. Based on the data collected by the communication link monitoring unit and the signal excitation and acquisition front end, a comprehensive evaluation of network performance is carried out. This module is a custom host computer that integrates sub-modules such as network load, communication cycle accuracy, slave transmission delay, slave synchronization accuracy and full task execution cycle performance analysis. The network load analysis submodule assesses network load by calculating bandwidth utilization and enables traffic anomaly monitoring and frame structure anomaly self-diagnosis. The communication cycle accuracy analysis submodule calculates the timestamps of periodic process data frames and introduces filtering and normalization algorithms to evaluate communication cycle accuracy. The slave station transmission delay analysis submodule calculates single-station or multi-station transmission delays using data frame timestamps and uses signal decomposition and regression analysis methods to achieve interpretable delay analysis. The slave station synchronization accuracy analysis submodule assesses synchronization performance by measuring the response time difference of multiple slave stations to output command signals and uses clustering algorithms and collaborative filtering methods to identify synchronization anomaly nodes. The full task execution cycle performance analysis submodule uses the frequency domain response curve as the core and integrates parameter identification algorithms to establish a dynamic mathematical model of the system and analyzes the impact of different communication cycles and communication cycle accuracies on the dynamic characteristics of the system, comprehensively evaluating the dynamic response performance of the network system under real task conditions.

[0049] The network performance comprehensive evaluation module performs data fusion and correlation analysis on the test results of each submodule, establishes a performance correlation reasoning model based on knowledge graphs, and reveals the inherent correlation of each performance indicator through graph query and graph reasoning algorithms, and performs quantitative analysis to comprehensively evaluate the overall performance of the EtherCAT network. Intelligent algorithms are introduced into each submodule to identify abnormal patterns and performance change trends in the network, providing guidance for EtherCAT network fault diagnosis and comprehensive performance optimization. Finally, a comprehensive performance evaluation report of this EtherCAT control network system is output. An example evaluation report is shown below. Figure 4 As shown.

[0050] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0051] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0052] In this invention, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. The term "multiple" refers to two or more unless otherwise expressly defined.

[0053] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A network integrated performance test system suitable for multi-protocol industrial Ethernet, characterized in that, The application relates to a communication link monitoring unit, a signal excitation and acquisition front end and a network performance comprehensive evaluation module, wherein, The communication link monitoring unit is used for real-time acquisition and analysis of data communication in a multi-protocol industrial Ethernet network link; The signal excitation and acquisition front end is used for closed-loop interaction with a slave module of the multi-protocol industrial Ethernet at an electrical level; The network performance comprehensive evaluation module is used for fusion processing and deep correlation analysis of heterogeneous data from the communication link monitoring unit and the signal excitation and acquisition front end, so as to realize comprehensive performance evaluation of the multi-protocol industrial Ethernet.

2. The network comprehensive performance test system of claim 1, wherein, The communication link monitoring unit comprises a network analyzer and a network data packet analysis tool, the network analyzer is connected to a bus network, and the network data packet analysis tool is arranged on a PC; the signal excitation and acquisition front end comprises a signal generator and a multi-channel oscilloscope, the signal generator is connected to an input point of the slave module and is responsible for applying an excitation signal to the slave module, and the multi-channel oscilloscope is connected to an output point of the slave module and is responsible for collecting a response signal of the slave module; The network performance comprehensive evaluation module comprises a network load analysis submodule, a slave synchronization precision analysis submodule, a communication cycle precision analysis submodule, a slave transmission delay analysis submodule and a full-task execution cycle performance analysis submodule.

3. The network synthetic performance testing system of claim 2, wherein, The network analyzer is connected to the bus network in a lossless manner through a physical TAP, can capture all data frames interacted between a master and a slave in real time, and adds a high-precision hardware time stamp to each data frame; the network data packet analysis tool screens periodic data frames and extracts process data therefrom.

4. The network synthetic performance testing system of claim 2, wherein, The signal generator generates a standard digital or analog excitation signal and applies the excitation signal to the input point of the slave module; the multi-channel oscilloscope collects the response signal of the output point of the slave module.

5. The network synthetic performance testing system of claim 2, wherein, The network load analysis submodule performs statistics on the data frames, identifies the proportion of various frames and calculates the bandwidth utilization rate; a traffic anomaly monitoring model based on deep learning automatically identifies atypical communication modes, predicts load trends and diagnoses frame structure anomalies.

6. The network synthetic performance testing system of claim 2, wherein, The slave synchronization precision analysis submodule evaluates synchronization performance by measuring the response time difference of multiple slave modules to an output signal instruction, adopts a clustering algorithm to group and identify the synchronization performance of the slave modules, and identifies abnormal synchronization nodes through a collaborative filtering method.

7. The network synthetic performance testing system of claim 2, wherein, The communication cycle precision analysis submodule adopts Kalman filtering and a multi-cycle dynamic time warping algorithm to analyze whether there is periodic jitter, long-term drift trend and burst anomaly mode in the communication cycle based on the time stamp of the periodic data frame, and finally obtains the communication cycle precision.

8. The network synthetic performance testing system of claim 2, wherein, The slave transmission delay analysis submodule calculates the cumulative transmission delay of a single station or multiple stations by capturing the time difference before and after each slave module in the data frame path, separates the deterministic delay and identifies the long-term trend based on signal decomposition and regression analysis, and realizes interpretable analysis of the delay.

9. The network synthetic performance testing system of claim 2, wherein, The full task execution cycle performance analysis submodule takes the frequency domain response curve as the core and integrates the parameter identification algorithm, uses the machine learning model to establish the system dynamic mathematical model, analyzes the influence of different system parameters on the dynamic characteristics of the multi-protocol industrial Ethernet, comprehensively evaluates the dynamic response performance of the multi-protocol industrial Ethernet under the real task condition, and provides guidance for the parameter optimization of the master station controller.

10. A network comprehensive performance testing method based on the network comprehensive performance testing system of any one of claims 2-9, characterized in that, It comprises the following steps: Step S1, an industrial Ethernet control network is built, a master station controller writes a PLC program to issue control and data acquisition tasks to each slave module, and data interaction is performed between the master station and the slave stations through periodic communication; Step S2, a network analyzer in the communication link monitoring unit is connected between the master station and the slave stations or between each slave station, all data frames in the communication link are captured in real time, the captured data frames are processed by a network data packet analysis tool, periodic data frames are screened out and process data is extracted; Step S3, a signal generator in the signal excitation and acquisition front end outputs a sine sweep signal to the analog input of each slave module, and a multi-channel oscilloscope collects the first response signal output by the analog output module of each slave station; Step S4, for the synchronization accuracy test requirement of the slave module, the master station controller controls the digital output module of each slave module to output a step signal, and a multi-channel oscilloscope collects the second response signal output by each slave module; Step S5, the communication link monitoring unit and the signal excitation and acquisition front end send the obtained data frames and response signals to the network performance comprehensive evaluation module, each performance analysis submodule performs data processing and analysis, and the test results of each performance index are obtained; Step S6, the network performance comprehensive evaluation module performs data fusion and correlation analysis on the test results of each submodule, reveals the internal coupling and correlation, comprehensively evaluates the network performance, and outputs a comprehensive performance evaluation report of the multi-protocol industrial Ethernet.

Images