Data-real fusion test time-varying network determination delay transmission method

Abstract

The invention discloses a data-real fusion test time-varying network determination delay transmission method, a hardware part of the method is realized by adopting an Artix7 chip of the Xilinx company, a software part of the method is realized by adopting a digital test system developed by Unity software, and the method specifically comprises the following steps of: constructing a multi-dimensional test environment, specifically, constructing a full-physical simulation field, setting a multi-dimensional test target, and constructing a full-physical simulation field; configuring dynamic test data; a dynamic congestion network is constructed, a digital real communication network composed of physical communication nodes constructed based on an FPGA and digital twin nodes with a data communication function is included, meanwhile, corresponding functions of the nodes are configured, and the state of the dynamic nodes is set; and constructing a multi-agent model, specifically comprising a multi-agent network model, a multi-agent deep reinforcement learning algorithm model and a traffic scheduling mechanism. According to the invention, stable delay communication of a digital-real communication network can be realized, and efficient operation of a digital-real fusion test can be effectively supported.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital-real fusion, and in particular relates to a method for determining delayed transmission in a time-varying network through digital-real fusion testing. Background Art

[0002] Equipment research and development is a comprehensive project involving multiple disciplines and fields. From design, manufacturing to testing and deployment, each link requires precise calculations and rigorous testing to ensure the performance and safety of the equipment. With the widespread application of digital and intelligent technologies, the combination of digital testing and physical testing has become an important trend. Based on semi-physical simulation, digital-physical fusion testing further emphasizes the integration of physical testing and digital testing, fully considers the influence of environmental factors, and constructs a comprehensive physical field test environment and multi-dimensional precise scene-object model. During the testing process, attention is paid to the fusion analysis of physical test data and digital test data to improve the accuracy and credibility of the test.

[0003] Digital-physical fusion testing requires data communication between the digital domain and the physical domain, and the digital communication nodes and physical communication nodes are interconnected to form a digital-physical communication network. The premise for implementing digital-physical fusion testing is that the digital domain and the physical domain must maintain a high degree of consistency, which requires stable information transmission with low latency between different nodes, that is, delayed deterministic communication. However, in the actual digital-physical fusion testing process, due to the large amount of test tasks and limited communication resources, there are congested nodes in the digital-physical communication network, and as the test objectives change, the congested nodes will change. Communication node congestion means that the amount of data transmitted at this time is greater than the bandwidth that the communication protocol or communication equipment can bear, so data queuing or even data loss will occur, which will lead to inconsistency or even loss of multi-source test data.

[0004] The current solutions to this problem can be roughly divided into two categories. One is to classify the data to be transmitted according to priority, and then allocate communication resources to ensure the transmission of high-priority data first, and then transmit low-priority data after the high-priority data transmission is completed. This type of method can achieve delayed deterministic transmission of high-priority data, but in the context of digital-real fusion testing, almost all test data have strong consistency and real-time requirements. The second method is to plan another transmission path for the data to be transmitted, and allocate the congested data flow to other relay nodes for transmission to the destination. This scheduling method achieves load balancing of each node in the communication network to a certain extent, but when scheduling traffic for congested nodes, it may affect the normal function of the scheduled nodes, and different data are transmitted from different paths, and the delay to the terminal is also different, which cannot effectively meet the requirements of multi-source data transmission consistency. Summary of the invention

[0005] The technical problem to be solved by the present invention is: to provide a method for determining the delay transmission of a time-varying network in a digital-real fusion test, which can realize the delay deterministic transmission of a digital-real communication network in a digital-real fusion test.

[0006] The present invention solves the technical problem by adopting the following technical solution: a method for determining delayed transmission of a time-varying network by digital-real fusion testing, comprising:

[0007] Step 1: Build a full physical simulation environment, including:

[0008] a. Based on the actual operation scenario of the equipment to be tested, a full physical simulation field environment is constructed using a digital simulation system;

[0009] b. Carry out systematic configuration for various test targets of the equipment to be tested, including: configuring the mechanism model of the test function to support the needs of different test targets; clarifying the test scenarios, including environmental conditions, operating procedures and potential interference factors; formulating detailed test requirements, covering performance indicators, durability testing, and reliability evaluation; establishing the evaluation criteria for test results; c. After completing the configuration of the full physical simulation field environment and various test targets, construct dynamic test data, including: generating corresponding test data for different test targets to ensure coverage of all possible working conditions and extreme conditions; then generating different volumes of test data according to the requirements of test granularity and precision; dynamically adjusting test parameters through real-time data monitoring and feedback mechanism;

[0010] Step 2: Build a dynamic congestion network, including:

[0011] a. Build a digital-physical communication network, which consists of digital nodes, physical nodes, and corresponding communication protocols and clock synchronization protocols;

[0012] b. Node function configuration: According to the design specifications and actual operation of the equipment to be tested, the functional characteristics of each node are analyzed one by one; through digital simulation technology, the functional modules of each node are rebuilt to ensure that they can accurately reflect the actual operation behavior in the simulation environment; for each functional node, the communication resources consumed when performing a specific task, such as bandwidth requirements, are analyzed, and then the node evaluation model is established to analyze the resource occupancy rate and the execution of the node function;

[0013] c. Congestion node setting: dynamically generate congestion nodes in the digital communication network constructed above according to the test content and objectives;

[0014] Step 3: Build a multi-agent model, including:

[0015] For the digital nodes and physical nodes in the digital-real communication network, the corresponding Agent model is constructed. The function of the Agent model is to ensure the normal operation of the node test function and participate in the decision-making of the network traffic scheduling plan; multiple Agent models are interconnected, and then a multi-agent deep reinforcement learning model is constructed to train the digital-real communication network to realize the data traffic scheduling function of congested nodes without affecting the operation of node functions.

[0016] The present invention is implemented by using the Artix7 chip and Unity software of Xilinx Company.

[0017] The beneficial effects of the present invention compared with the prior art are:

[0018] (1) By constructing a multi-dimensional digital-real communication network in the digital-real fusion test, a comprehensive digital test scenario is constructed by combining multi-target test scenarios and full physical field models, so that the results of the digital test have a high degree of credibility.

[0019] (2) A traffic scheduling method based on multi-agent reinforcement learning is proposed. When faced with network congestion, traffic scheduling can be performed and delayed deterministic transmission can be achieved without affecting the normal operation of nodes.

[0020] (3) A traffic scheduling method is proposed when the network faces dynamic congested nodes, which improves the universality and reliability of digital communication networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is an architecture diagram of a method for determining delayed transmission in a time-varying network by using digital-real fusion testing according to the present invention. DETAILED DESCRIPTION

[0022] The present invention is described in further detail below in conjunction with the accompanying drawings.

[0023] The present invention relates to a method for determining delayed transmission of a time-varying network in a digital-real fusion test, which is implemented by using Xilinx's Artix7 FPGA chip and Unity software, and includes multi-dimensional test environment construction, dynamic congestion network construction, and multi-agent model construction. Aiming at the deterministic delayed communication of a digital-real communication network in a digital-real fusion test of complex equipment in various fields, the present invention can realize the determined delayed transmission of a digital-real communication network under a dynamic congestion node.

[0024] Figure 1 This is a diagram of the architecture of a method for determining delayed transmission in a time-varying network by using digital-real fusion testing in the present invention. Figure 1 Describe the transmission method. The transmission method includes:

[0025] Step 1: Build a full physical simulation field environment. The specific contents are as follows:

[0026] a. Based on the actual operation scenario of the equipment to be tested, a full physical simulation field environment is constructed using a digital simulation system. This process includes configuring relevant mechanism models to reproduce the physical test scenario as realistically as possible. For example, for aircraft testing, an aerodynamic model can be used to simulate the impact of airflow during flight. By accurately simulating various physical factors in the environment, including temperature, pressure, humidity, etc., and by using high-precision sensor data and historical operation data, the model parameters are adjusted to make the simulation results more reliable and effective.

[0027] b. Systematically configure the various test functions of the equipment to be tested. This process is specifically manifested as: configuring the mechanism model of the test function to support the needs of different test objectives. For example, for performance testing, a dynamic model can be configured, while for durability testing, a material fatigue model may be required; clarifying the test scenario, including environmental conditions (such as temperature range, humidity level), operating procedures (such as specific steps for starting, running, and stopping) and potential interference factors (such as electromagnetic interference, vibration, etc.), to ensure that all factors that may affect the test results are taken into account; formulating detailed test requirements, covering performance indicators, durability testing, and reliability evaluation; establishing evaluation criteria for test results, clarifying the performance indicators of the equipment to be tested under different working conditions, such as response time, output power, stability, etc., to ensure that the test results are comparable and repeatable, providing a solid foundation for subsequent analysis.

[0028] c. After completing the configuration of the full physical simulation field environment and multiple test targets, dynamic test data is constructed. This process includes: generating corresponding test data for different test targets to ensure coverage of all possible working conditions and extreme conditions, such as high temperature, strong magnetic field interference and other environments for aerospace equipment; then generating different volumes of test data according to the requirements of test granularity and precision, so as to facilitate in-depth analysis and comparison; dynamically adjusting test parameters through real-time data monitoring and feedback mechanism to improve the accuracy and comprehensiveness of the test.

[0029] Step 2: Build a dynamic congestion network. The specific process is as follows:

[0030] a. Build a digital-real communication network. Build a digital-real communication network in the full physical field simulation environment described in step 1 above, which can also be called a digital-real fusion test synchronization network, consisting of digital nodes, physical nodes, and corresponding communication protocols and clock synchronization protocols. The physical node is a communication unit based on FPGA technology that supports TCP / UDP protocol for data transmission; the digital node is a virtual communication unit created in Unity software that has communication capabilities and simulates the communication resource status and functions of the physical node. The physical node and the digital node are interconnected to form a complete digital-real communication network.

[0031] b. Node function configuration. According to the design specifications and actual operation of the equipment to be tested, the functional characteristics of each node are analyzed one by one. Through digital simulation technology, the functional modules of each node are rebuilt to ensure that they can accurately reflect the actual operation behavior in the simulation environment. For each node, the communication resources consumed by it when performing a specific task, such as bandwidth requirements, are analyzed, and then the node evaluation model is established to analyze the resource occupancy rate and the execution of node functions.

[0032] c. Congested node setting. In the digital communication network constructed above, congested nodes are dynamically generated according to the test content and objectives. Congested nodes will be set to transmit a large amount of test data, and the data of other nodes will face data congestion when passing through this node, resulting in large communication delays and packet loss rates.

[0033] Step 3: Build a multi-agent model. The details are as follows:

[0034] The multi-agent model is a distributed artificial intelligence model composed of multiple agents. Each agent has the ability to make autonomous decisions and interact, and can collaborate, compete or negotiate with each other to complete common or individual tasks. For digital nodes and physical nodes in the digital-real communication network, the corresponding agent model is constructed. The function of the agent model is to ensure the normal operation of the node test function and monitor the operation status of the node in real time. For example, the agent model can track each link of data collection, processing and transmission, and identify potential problems in time. At the same time, the agent model can participate in the decision-making of network traffic scheduling: the agent model analyzes the current network status, reasonably allocates communication resources, and optimizes the data transmission path. Multiple agent models are connected to each other, and the communication mechanism between agent models is designed to enable them to share information, such as the load status of the node, data transmission delay, etc. Through collaboration, the agent model can make more effective scheduling decisions together. Subsequently, a multi-agent deep reinforcement learning model is constructed. The input of the model is the state of the full physical field simulation environment, the topological structure of the digital-real communication network, the traffic data state of the digital-real communication network, the test objectives, etc. Then, the full physical field simulation environment is used as the training environment for reinforcement learning. By dynamically setting the data traffic, the digital-real communication network is trained to realize the data traffic scheduling function of congested nodes without affecting the functional operation of the nodes. The trained multi-agent deep reinforcement learning model is used as the output. During the actual digital-real fusion test, the trained multi-agent deep reinforcement learning model is used to perform traffic scheduling for dynamic congested nodes, thereby reducing data transmission delay and packet loss rate, and achieving determined communication delay.

[0035] In summary, the present invention discloses a method for determining delayed transmission in a time-varying network for digital-real fusion testing, including multi-dimensional test environment construction, dynamic congestion network construction, and multi-agent model construction, which can realize the digital-real communication network under dynamic congestion conditions, and perform traffic scheduling on congested nodes without interfering with the execution of normal test functions of the nodes. The present invention can improve the consistency and reliability of the digital-real communication network under complex dynamic environments, and can effectively support the results of digital-real fusion testing.

[0036] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in this field.

[0037] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for determining delayed transmission in a time-varying network by using digital-real fusion testing, characterized in that: The following steps are involved: Step 1: Build a full physical simulation environment, including: a. Based on the actual operation scenario of the equipment to be tested, use the digital simulation system to build a full physical simulation field environment; b. Systematically configure the various test targets of the equipment to be tested, including: configuring the mechanism model of the test function to support the needs of different test targets; clarifying the test scenario, including environmental conditions, operating procedures and potential interference factors; formulating detailed test requirements, covering performance indicators, durability testing, and reliability evaluation; establishing the evaluation criteria for test results; c. After completing the configuration of the full physical simulation field environment and various test targets, construct dynamic test data, including: generating corresponding test data for different test targets to ensure coverage of all possible working conditions and extreme conditions; then generating different volumes of test data according to the requirements of test granularity and precision; dynamically adjusting test parameters through real-time data monitoring and feedback mechanism; Step 2: Build a dynamic congestion network, including: a. Build a digital-physical communication network, which consists of digital nodes, physical nodes, and corresponding communication protocols and clock synchronization protocols; b. Node function configuration: According to the design specifications and actual operation of the equipment to be tested, the functional characteristics of each node are analyzed one by one; through digital simulation technology, the functional modules of each node are rebuilt to ensure that they can accurately reflect the actual operation behavior in the simulation environment; for each functional node, the communication resources consumed when performing a specific task, such as bandwidth requirements, are analyzed, and then the node evaluation model is established to analyze the resource occupancy rate and the execution of the node function; c. Congestion node setting: dynamically generate congestion nodes in the digital communication network constructed above according to the test content and objectives; Step 3: Build a multi-agent model, including: For the digital nodes and physical nodes in the digital-real communication network, the corresponding Agent model is constructed. The function of the Agent model is to ensure the normal operation of the node test function and participate in the decision-making of the network traffic scheduling plan; multiple Agent models are interconnected, and then a multi-agent deep reinforcement learning model is constructed to train the digital-real communication network to realize the data traffic scheduling function of congested nodes without affecting the operation of node functions.

2. The method for determining delayed transmission in a time-varying network by digital-real fusion testing as claimed in claim 1, characterized in that: Step 1 includes: configuring relevant mechanism models to reproduce the physical test scenario; simulating various physical factors in the environment, including temperature, pressure, and humidity.

3. The method for determining delayed transmission in a time-varying network by digital-real fusion testing as claimed in claim 1, characterized in that: The physical node is a communication unit based on FPGA technology, which supports TCP / UDP protocol for data transmission; the digital node is a virtual communication unit created in Unity software, which has communication capabilities and simulates the communication resource status and functions of the physical node. The physical node and the digital node are interconnected to form a complete digital-real communication network.

4. The method for determining delayed transmission in a time-varying network by digital-real fusion testing according to claim 1, characterized in that: The method for determining delayed transmission in a time-varying network by using digital-real fusion testing is implemented using Xilinx's Artix7 chip and Unity software.

Images