Semi-physical simulation experiment online verification system and method

Abstract

The invention discloses a semi-physical simulation experiment online verification system and method, and belongs to the technical field of digital simulation. The system comprises a Monte Carlo equipment experiment platform used for configuring experiment parameters, providing digital simulation key sample data and generating random interference parameters; the semi-physical simulation integrated management system is used for carrying out scheduling management on the semi-physical simulation system; performing a semi-physical simulation experiment on the semi-physical simulation system by using the digital simulation key sample data according to the experiment plan; random interference parameters are loaded in the experiment operation process; and the digital simulation system is used for realizing visualization of the experiment process. Seamless integration of an experimental equipment key parameter module, a pneumatic analysis module and an equipment simulation model in the environment is achieved, a Monte Carlo equipment test platform is developed and designed to complete equipment Monte Carlo experiment rapid design, and then a systematic large sample simulation test is completed.

Description

Technical Field

[0001] The invention belongs to the technical field of digital simulation, and in particular relates to an online verification system and method for a semi-physical simulation experiment. Background Art

[0002] Semi-physical simulation is subject to strong constraints from simulation hardware and digital models, which leads to complex organization and insufficient flexibility of simulation experiments. To address this problem, China mainly improves the scheduling and management efficiency of semi-physical simulation models through intelligent transformation of simulation framework software and universal modeling technology of simulation models. From the mid-to-late 1990s to the present, driven by the development of various new aircraft, a group of high-level and large-scale semi-physical simulation laboratories have carried out equipment application simulation experiments based on semi-physical simulation technology, breaking through a large number of key technologies, and advanced simulation technologies such as distributed interactive simulation and virtual reality have been widely used. Machine learning provides analytical means for data analysis technology, mainly studying the theory and algorithm of how to extract patterns and models from data, that is, learning algorithms, which are the inevitable product of the development of artificial intelligence to a certain stage. Many methods in data mining are derived from machine learning. Traditional machine learning algorithms such as linear discriminant analysis, logistic regression, decision trees, random forests, neural networks, support vector machines, etc. have been derived into deep learning, reinforcement learning, deep reinforcement learning and other algorithms.

[0003] Prior art Chinese patent application CN202410186768.6 discloses a virtual-reality interaction system based on Unreal Engine and semi-physical simulation, including an Unreal Engine and a semi-physical simulation system; the semi-physical simulation system includes a controller, a communication middleware and a semi-physical simulation model; the Unreal Engine exchanges data with the semi-physical simulation model through the communication middleware; the Unreal Engine issues simulation control instructions through the controller, which are sent to the semi-physical simulation model via the communication middleware; the semi-physical simulation model sends the model data generated by the simulation process through the communication middleware.

[0004] The existing technology has the following shortcomings: 1. The simulation verification parameters of key parameters of existing pure digital models of equipment are usually set fixed values, without considering the actual situation of the application scenario, and there is a problem of low environmental and model accuracy. 2. During the construction process, traditional semi-physical simulation models are restricted by the connection to the physical object, and usually have a rigid structure and a single function, making it impossible to carry out large-scale multi-functional simulation tests. 3. Large-sample pure digital experimental simulation experiments have a large number of experimental factors, multiple dimensions, and complex relationships. Accurate qualitative evaluation of key experimental samples requires proofing field equipment verification experiments, and the evaluation results are obtained over a long period of time and with great difficulty.

[0005] In addition, in order to obtain the actual data of equipment experiments under random interference factors, there are two commonly used test methods: equipment actual test experiments and computer simulation equipment test experiments. Obtaining data through actual field tests of equipment requires a large number of experiments, which greatly increases the cost of experimental equipment development and prolongs the equipment design cycle. With the continuous development of computer technology, simulation technology has gradually become an indispensable means for the development and management of modern equipment systems. Using computer simulation technology to simulate the equipment developed can quickly obtain the statistical results of the equipment test parameters required by designers at a very low cost, providing methods and basis for equipment research and development design and performance verification. In the Monte Carlo equipment experiment work, designers are required to write corresponding simulation analysis modules and data interfaces between modules, and develop corresponding equipment simulation model programs to conduct a large number of Monte Carlo verification tests. The design task is arduous and prone to errors. In addition, due to the problems that different simulation modules or software tools often have poor interface compatibility and difficulty in data interaction, the scalability and reusability of models and codes are greatly limited, further increasing the workload of the early development of Monte Carlo simulation tests of experimental equipment. Summary of the invention

[0006] The purpose of the present invention is to provide an online verification system and method for semi-physical simulation experiments, partially solve or alleviate the above-mentioned deficiencies in the prior art, develop and design a Monte Carlo equipment test platform to complete the rapid design of equipment Monte Carlo experiments and then complete systematic large-sample simulation experiments.

[0007] In order to solve the above-mentioned technical problems, the present invention specifically adopts the following technical solutions: A semi-physical simulation experiment online verification system comprises: a Monte Carlo equipment experiment platform, which is used to configure experiment parameters, provide digital simulation key sample data and generate random interference parameters; statistically analyze the experimental equipment operation parameters and equipment evaluation results, so as to obtain the statistical characteristic values ​​of the key parameters of the experimental equipment; a semi-physical simulation integrated management system, which is used to schedule and manage the semi-physical simulation system; the semi-physical simulation system is used to establish a semi-physical simulation model of the experimental equipment according to the motion characteristics and motion laws of the experimental equipment; perform experiment planning according to the experiment parameters; perform semi-physical simulation experiments on the semi-physical simulation system using digital simulation key sample data according to the experiment planning; and load the random interference parameters issued by the Monte Carlo equipment experiment platform during the experiment operation, so as to obtain the experimental equipment operation parameters and equipment evaluation results after disturbance; a digital simulation system is used to perform three-dimensional construction of the experimental equipment and the experimental scene according to the semi-physical simulation model and the data in the experiment process, so as to realize the visualization of the experimental process.

[0008] As an improvement, the Monte Carlo equipment experiment platform includes: an experimental parameter configuration module, which is used to configure the number of experimental tasks, set the number of threads to be started, and the experimental data to be configured for each task experiment; the experimental data includes the injection type and injection time of random interference; an experimental data generation module, which is used to extract digital simulation key sample data from large sample experiments for semi-physical simulation experiments, and generate random interference parameters.

[0009] As an improvement, the semi-physical simulation system is a distributed system, including several semi-physical simulation nodes; the three-dimensional model of the experimental equipment constructed by the digital simulation system corresponds one-to-one to the semi-physical simulation node, and during the experiment, the digital simulation system and the semi-physical simulation system are synchronized in time.

[0010] As an improvement, data is transmitted between the semi-physical simulation nodes via optical fiber reflection memory.

[0011] As an improvement, the semi-physical simulation integrated management system includes: an experiment planning module, which is used to plan experiments according to experimental parameters, and send experimental data and digital simulation key sample data to semi-physical simulation nodes for experiments; a node scheduling module, which is used to schedule corresponding semi-physical simulation nodes to participate in experiments according to the number of threads set by the Monte Carlo equipment experiment platform.

[0012] As an improvement, the digital simulation system includes: a virtual engine for constructing a three-dimensional geometric model of the experimental equipment and creating a visual experimental scene, thereby realizing a visual reconstruction of the equipment state; and a physical engine for simulating a physical experimental scene and providing physical interaction.

[0013] The present invention also provides an online verification method for a semi-physical simulation experiment, comprising: According to the motion characteristics and motion laws of the experimental equipment, a semi-physical simulation model of the experimental equipment is established; Extract key sample data of digital simulation from large sample experiments; Determine the random interference factors and their distribution patterns during the operation of the experimental equipment, and generate random interference parameters based on the random interference factors and their distribution patterns; Configure experimental parameters, plan experiments according to the experimental parameters, conduct semi-physical simulation experiments on the semi-physical simulation system using key digital simulation sample data according to the experimental plan, and load random interference parameters during the experiment; The experimental equipment and experimental scene are constructed in three dimensions according to the semi-physical simulation model and the data in the experimental process to realize the visualization of the experimental process; Obtain the experimental equipment operating parameters and equipment evaluation results after disturbance; Statistics are collected on the experimental equipment operating parameters and equipment evaluation results to obtain the statistical characteristic values ​​of the key parameters of the experimental equipment.

[0014] As an improvement, a semi-physical simulation model of the experimental equipment is established using a semi-physical simulation system, and a semi-physical simulation experiment is carried out based on the key sample data of the digital simulation; based on time synchronization with the semi-physical simulation system, the digital simulation system performs three-dimensional construction of the experimental equipment and experimental scene based on the semi-physical simulation model and the data during the experiment.

[0015] As an improvement, the steps of time synchronization between the digital simulation system and the semi-physical simulation system include: Clock synchronization between simulation nodes within the S101 semi-physical simulation system; S102 uses a UTC clock source to align the start time of the digital simulation system and the semi-physical simulation system; S103 performs clock synchronization based on step length verification when the simulation step length of the digital simulation system is consistent with the simulation step length of the semi-physical simulation system, specifically including: S1031 initialization, setting the synchronization period, setting the digital simulation system preparation synchronization flag and the synchronization flag to 0, and returning the simulation step counter to zero; When receiving the clock synchronization information pushed back by the semi-physical simulation system, S1032 parses the simulation number step value and assigns it to the step counter, and sets the synchronization ready flag to 1, when both the synchronization ready flag and the synchronization ready flag are 0; When the preparation synchronization flag is 1 and the synchronization flag is 0, the simulation step number value is parsed and compared with the current value of the step counter. If they are equal, the synchronization flag is set to 1, the preparation synchronization flag is set to 0, and step S1033 is executed; if they are not equal, the preparation synchronization flag is set to 0 and step S1032 is executed repeatedly; S1033 waits until the synchronization cycle ends; S1034 uses a UTC clock source to align current times of the digital simulation system and the semi-physical simulation system; S1035 enters the next synchronization cycle and executes steps S1032 to S1035 in a loop.

[0016] As an improvement, the steps of time synchronization between the digital simulation system and the semi-physical simulation system include: S111 uses the clock of the semi-physical simulation system as the master clock, uses the clock of the digital simulation system as the slave clock, and sets a correction period; The S112 digital simulation system receives the clock synchronization message sent from the semi-physical simulation system, extracts the master clock time of the semi-physical simulation system sending the synchronization message from the clock synchronization message and records it as T1; S113 digital simulation system records the arrival time T2 of the clock synchronization message from the clock; S114 The digital simulation system sends a message, sets a flag in the message to start transmission error measurement, and records the sending time T3 from the clock; S115 After the semi-physical simulation system receives the message with the start transmission error measurement mark, the master clock records the delivery time T4; and sets the mark representing the completion of the transmission error measurement in the message pushed back to the digital simulation system, and records the delivery time T4 in the pushed back message; After receiving the message with the transmission error measurement completion mark, the S116 digital simulation system calculates the clock deviation and time delay of the master and slave clocks according to T1, T2, T3, and T4, and adjusts the digital simulation system slave clock according to the clock deviation and time delay.

[0017] Beneficial effects: 1. Efficient system integration and process connection. The present invention realizes the seamless integration of the key parameter module of the experimental equipment, the pneumatic analysis module and the equipment simulation model in the environment. The data interaction and collaborative work between different modules are smoother, avoiding the inefficiency caused by compatibility issues between modules or poor data transmission. For example, the results of the pneumatic analysis module can be directly used as the input parameters of the equipment simulation model, reducing the time for manual intervention and data processing, making the entire experimental process more compact and efficient. From the construction of the experimental environment of the experimental equipment, the generation of random configuration parameters, the rapid construction of the model to the statistical analysis of the test results, the full process simulation is realized. This means that all aspects of the experiment can be completed within one system without switching between multiple different systems or tools, which greatly shortens the experimental cycle. For example, the Monte Carlo equipment test platform can quickly generate random configuration parameters of experimental equipment, and directly apply them to model construction and experimental operation in the semi-physical simulation system, which is done in one go and improves the overall experimental efficiency.

[0018] 2. Rapid design and systematic large-sample simulation. The Monte Carlo equipment test platform developed and designed in the present invention can complete the rapid design of equipment Monte Carlo experiments. The platform adopts a layered module design concept, and the modules at different levels have clear division of labor, and have multiple functions such as experimental parameter configuration, data generation, and process control. Experimenters can use the platform to quickly set the number of experimental tasks, the number of threads, and various experimental data, such as the injection type and injection time of random interference, so as to quickly start the experiment and save experimental preparation time. The present invention can complete systematic large-sample simulation experiments and verify the performance and reliability of experimental equipment through a large amount of experimental data. The Monte Carlo equipment test platform extracts digital simulation key sample data from large sample experiments. These data are more representative and authentic, and can more accurately reflect the actual operation of the experimental equipment. Experiments are carried out using these key sample data on a semi-physical simulation system, and random interference parameters are loaded to simulate various complex actual scenarios, thereby obtaining more comprehensive and reliable experimental results.

[0019] 3. Advanced technical support and optimization. A variety of clock synchronization methods are used between the digital simulation system and the semi-physical simulation system, such as step-length verification and satellite clock cycle alignment, and reflection memory and master clock verification. High-precision clock synchronization ensures the time consistency of the two systems during the experiment, avoids experimental errors and data inconsistencies caused by time asynchrony, and improves the reliability of experimental results. For example, during real-time simulation, each simulation node can perform data interaction and calculation under the same time reference, ensuring the accuracy of the experiment.

[0020] The system has optimized designs in terms of communication architecture, data integration standards, and data storage. In terms of communication, the Monte Carlo equipment experiment platform and the semi-physical simulation integrated management system exchange data through the DDS communication protocol, and use a unified semi-physical framework integration standard to send and receive data in the form of Json strings, ensuring the efficiency and accuracy of data transmission. In terms of data integration, clear interaction interfaces and processes are defined to ensure smooth interaction of data between different systems. In terms of data storage, based on database design patterns and transaction mechanisms, concurrent access control of large data samples is achieved, effectively controlling the load of CPU and I / O resources. At the same time, data visualization tools are used to provide data support for experiments, making it convenient for experimenters to manage and analyze experimental data.

[0021] The semi-physical simulation system uses reflective memory network and RTX real-time system to solve the problems of traditional Ethernet and Windows system in terms of real-time and clock stability. Reflective memory network has strict transmission determinism and predictability, small data transmission delay, fast speed and large capacity, which can meet the requirements of distributed real-time simulation system for real-time and large-scale data transmission. RTX real-time system has excellent real-time controllability, scalability and stability, ensuring the reliability and accuracy of the simulation process.

[0022] 4. Visualization and data analysis support. The digital simulation system uses virtual engines and physical engines to construct experimental equipment and experimental scenes in three dimensions, realizing the visualization of the experimental process. Experimenters can intuitively observe the operating status and performance of experimental equipment in different scenarios, find problems in time and make adjustments. For example, in an aircraft simulation experiment, experimenters can observe the aircraft's attitude, trajectory and other parameters in real time through a three-dimensional visualization interface, and evaluate the experimental results more intuitively.

[0023] The Monte Carlo equipment experiment platform collects statistics on the operating parameters of the experimental equipment and the equipment evaluation results, and obtains the statistical characteristic values ​​of the key parameters of the experimental equipment. Through the statistical analysis of a large amount of experimental data, we can deeply understand the performance and reliability of the experimental equipment under different random interferences, and provide a scientific basis for the research and development and design of the equipment. For example, by calculating the statistical characteristic values ​​such as the mean and variance of the key parameters of the experimental equipment, we can evaluate the stability and consistency of the equipment, so as to optimize and improve the equipment in a targeted manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, each element or part is not necessarily drawn according to the actual scale. Obviously, the drawings described below are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without paying creative labor.

[0025] Figure 1 This is a schematic diagram of the architecture of the online verification system for semi-physical simulation experiments; Figure 2 This is a schematic diagram of the Monte Carlo equipment experimental platform architecture; Figure 3 Provide a data interaction flow chart for the Monte Carlo equipment experimental platform; Figure 4 This is a schematic diagram of the architecture of the semi-physical simulation integrated management system; Figure 5 It is the topology diagram of the semi-physical simulation node system resources; Figure 6 Data communication flow chart of online verification system for semi-physical simulation experiment; Figure 7 This is a schematic diagram of simulation time synchronization for the online verification system of semi-physical simulation experiments; Figure 8 A flowchart for completing clock synchronization for a digital simulation system to receive clock synchronization information sent by a semi-physical simulation system; Fig. 9 It is a synchronization flow chart based on the IEEE1588 protocol principle; Fig.10 Data interaction diagram of the Monte Carlo equipment experimental platform and management software; Fig.11 Integrate schematic diagrams for simulation models based on hardware-in-the-loop integrated management systems; Fig.12 A technical roadmap for storing semi-physical large sample simulation result data; Fig.13 It is an interactive link for the semi-physical large sample simulation result database. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] Herein, the suffixes such as "module", "component" or "unit" used to represent elements are used only to facilitate the description of the present invention and have no specific meaning in themselves. Therefore, "module", "component" or "unit" can be used in a mixed manner. Herein, the orientation or position relationship indicated by the terms "upper", "lower", "inner", "outer", "front", "back", "one end", "the other end" and the like are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first" and "second" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance. Herein, unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected" and the like should be understood in a broad sense, for example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, an indirect connection through an intermediate medium, or a communication between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood in specific circumstances. Herein, "and / or" includes any and all combinations of one or more of the listed related items. Herein, "plurality" means two or more, that is, it includes two, three, four, five, etc.

[0028] The basic architecture of the semi-physical simulation experiment online verification system provided by the present invention consists of three parts: user layer, model layer and hardware layer. Figure 1 As shown. Among them, the hardware layer, as the bottom layer of the environment, is composed of high-performance servers or client PCs, network communication equipment, information security equipment, etc. It is the hardware environment foundation that supports the interaction between simulation and model data. The model layer is mainly composed of simulation models of each subsystem (including equipment models, pneumatic models, control models and visualization models) and data modules, and is integrated and built through the semi-physical simulation integrated management system. The top layer of the environment is the user layer, including the semi-physical simulation integrated management system, the Monte Carlo equipment experiment platform and the digital simulation system. Among them, the semi-physical simulation integrated management system uses fiber reflection memory combined with Ethernet to build a heterogeneous communication network with the Monte Carlo equipment experiment platform and the digital simulation system to realize the interaction between users and the environment and the visualization analysis of simulation results.

[0029] The following introduces the Monte Carlo equipment experiment platform, semi-physical simulation integrated management system, and digital simulation system respectively.

[0030] 1. Monte Carlo equipment experimental platform.

[0031] The Monte Carlo equipment experiment platform is used to configure experimental parameters, provide key sample data for digital simulation and generate random interference parameters; it performs statistics on the operating parameters of the experimental equipment and the equipment evaluation results to obtain the statistical characteristic values ​​of the key parameters of the experimental equipment.

[0032] Specifically, the platform uses a layered module design concept to complete the functional division of the platform framework, and arranges different modules in their respective appropriate levels. Different levels construct different task modules, and modules use specific structures for data exchange. The system architecture diagram is shown below: Figure 2 As shown in the figure, the lower layer provides services for the upper layer, and the upper layer completes the functions it needs to call by calling the lower layer interface. The experimental platform interaction layer has functions such as various experimental parameter configurations, experimental data generation, experimental process control, test progress monitoring, and experimental log management. The functional layer has functions such as thread management, data management, data encapsulation\analysis, data storage, and internal communication management to support the various functional requirements of the application layer. The general interface layer provides communication interfaces and interface protocols for external interactions to achieve standardization of data interaction. It opens up real-time communication between the experimental platform and the semi-physical integration framework.

[0033] More specifically, the experimental parameter configuration function is implemented by the experimental parameter configuration module, which is used to configure the number of experimental tasks, set the number of threads to be started, and the experimental data to be configured for each task experiment; the experimental data includes the injection type and injection time of random interference.

[0034] The purpose of the experiment is to obtain the data generated by the experimental equipment under various interferences. Therefore, in addition to some basic parameters such as the number of tasks and the number of threads, the experimental parameter configuration module also needs to set the target location, deployment location, and injection type and injection time of random interference (including faults and deviations).

[0035] The experimental data generation function is implemented by the experimental data generation module, which is used to extract digital simulation key sample data from large sample experiments for semi-physical simulation experiments and generate random interference parameters. Large sample experiments are digital simulation experiments, and their data are randomly generated, so some of the data do not conform to reality. In order to improve the authenticity of the experiment, the experimental data generation module is responsible for extracting digital simulation key sample data that meets the requirements from the data, and focusing the digital simulation key sample data on the semi-physical simulation model of the experimental equipment for experiments.

[0036] Parameters of random disturbances such as faults and deviations are also randomly generated by the experimental data generation module.

[0037] The process of starting the experiment on the Monte Carlo equipment experimental platform is as follows: Figure 3As shown in the figure, first start the equipment experiment platform, set the number of cycle sample tests, start the random generation button of experimental data, and wait for the data to be generated and then close the data generation. According to the generated data, the tasks are configured in detail according to the actual situation of each task. After the data configuration is completed, set the number of threads to be started, and the semi-physical integrated management system can be controlled to schedule the relevant simulation nodes to participate in the experiment.

[0038] After the configuration is completed, the verification experiment can be started. The experimental platform will send DDS command data packets according to the parameter configuration of each experiment. After issuing the experimental task instructions, the semi-physical simulation integration system will call the key node resource model according to the instructions to complete the node task and report the task completion status. The equipment experiment platform subscribes to the simulation result information and parses it. After all simulation nodes of the current group of tasks complete the tasks, the platform will send the next group of equipment experiment tasks. Similarly, when the test progress completion rate is 100%, the equipment experiment ends. If the experimental task data is abnormal, you can also manually stop the experimental task sending thread to end the experiment. After the experiment is completed, the data can be visualized and analyzed. The evaluation results of the large sample Monte Carlo equipment test are given.

[0039] 2. Semi-physical simulation integrated management system.

[0040] The semi-physical simulation integrated management system is used for scheduling and managing the semi-physical simulation system; the semi-physical simulation system is used for establishing a semi-physical simulation model of the experimental equipment according to the motion characteristics and motion laws of the experimental equipment; performing experimental planning according to the experimental parameters; performing semi-physical simulation experiments on the semi-physical simulation system using digital simulation key sample data according to the experimental planning; and loading random interference parameters during the experimental operation to obtain the experimental equipment operation parameters and equipment evaluation results after the disturbance.

[0041] Specifically, with the goal of integrating the semi-physical simulation resources of industrial departments and flexibly building a semi-physical simulation environment, this embodiment constructs Figure 4 The semi-physical simulation integrated management system shown in the figure realizes the scheduling management of the semi-physical simulation system through the interactive application layer and the functional layer, and realizes the data interaction with the Monte Carlo equipment experiment platform and the digital simulation system as well as various semi-physical simulation test systems through the general interface layer and the hardware interface layer, providing a bridge for the simulation model integrated management system to schedule and control the participating semi-physical simulation test systems, and supporting the joint simulation experiments of diversified scenarios combining virtual and real.

[0042] In the design of the semi-physical simulation integrated management system, the layering and division of the framework is completed with the layered modular design concept, and different modules are reasonably arranged in their respective appropriate layers. The same layer has different functional modules. The lower layer provides services for the upper layer, and the upper layer completes the functions required by this layer by calling the interface of the lower layer. The interactive application layer includes the experiment planning module, the node scheduling control module, the experiment monitoring module, and the experiment node management module to realize the experiment planning, node scheduling control, experiment monitoring, and experiment node management, so as to schedule, control, and monitor the overall operation process of the semi-physical simulation system.

[0043] For example, the implementation planning module is used to plan experiments according to the experimental parameters (set by the Monte Carlo equipment experiment platform) and send the experimental data and key sample data of digital simulation to the semi-physical simulation node for experiment. The Monte Carlo equipment experiment platform provides relevant experimental parameters including the number of experiments. The node scheduling module is used to schedule the corresponding semi-physical simulation nodes to participate in the experiment according to the number of threads set in the Monte Carlo equipment experiment platform.

[0044] The functional layer has functions such as clock synchronization management, internal communication management, data transmission and reception management, data storage management, thread management and dynamic filling management, supporting various functional requirements of the application layer. The general interface layer provides interface protocols with digital simulation systems and Monte Carlo equipment experimental platforms, as well as interface protocols with semi-physical simulation test systems, to achieve standardization of data interaction. The hardware interface layer provides Ethernet, reflective memory network cards and other general intelligent I / O devices to support the access of heterogeneous hardware and realize diversified semi-physical simulation.

[0045] In this embodiment, the semi-physical simulation system used for semi-physical simulation of experimental equipment is a distributed system. The distributed real-time simulation system is a complex simulation system composed of multiple subsystems (i.e., semi-physical simulation nodes). During the simulation process, the subsystems need to exchange data in real time. Therefore, the real-time simulation system has relatively strict requirements on the ability to exchange data between the semi-physical simulation nodes. In order to obtain real-time simulation effects, the real-time requirements for data transmission during the simulation process are very high, and the traditional Ethernet based on TCP / IP protocol cannot meet the requirements. Compared with the TCP / IP network used in the past, the real-time network based on optical front reflection memory has the characteristics of small data transmission delay, fast transmission speed, large transmission data volume, simple and easy-to-use communication protocol, light load on the simulation machine, strong adaptability of the software and hardware platform, reliable transmission error correction capability, support for interrupt signal transmission, etc. These advantages of the reflection memory network can meet the requirements for real-time and large-scale data transmission in the distributed real-time simulation system.

[0046] In addition, in order to solve the problem that the Windows system clock is unstable and cannot be used for real-time simulation system, a clock control mechanism based on RTX real-time system is adopted. RTX is a universal real-time extension system based on Windows operating system. RTX has excellent real-time controllability, efficient scalability and stability, and is the only software-based hard real-time solution on Windows platform.

[0047] like Figure 5 As shown in the figure, the semi-physical simulation system built based on the above technology is a distributed simulation system composed of multiple semi-physical simulation nodes. The semi-physical simulation nodes are composed of digital simulation systems, simulation management machines, real-time simulation machines, satellite navigation simulators, general intelligent devices, etc., and communicate through Ethernet and reflective memory networks. The main control machine and the computing node machine run the Windows+RTX system. In the real-time simulation mode, the simulation model is compiled into an RTSS real-time executable program and runs in the RTX real-time subsystem. Windows and the RTX real-time subsystem use shared memory for communication. Simulation nodes are divided into lightweight nodes and fully quantized simulation nodes. The entire semi-physical integrated management system can dynamically schedule any simulation node to participate in the construction of the Monte Carlo equipment experiment in the entire process according to the requirements of the assumed scenario.

[0048] 3. Digital simulation system.

[0049] The digital simulation system is used to construct the experimental equipment and experimental scenes in three dimensions based on the semi-physical simulation model and the data in the experimental process, so as to realize the visualization of the experimental process. Specifically, the digital simulation system includes: a virtual engine, which is used to construct a three-dimensional geometric model of the experimental equipment and create a visual experimental scene, thereby realizing the visual reconstruction of the equipment status; a physical engine, which is used to simulate the physical experimental scene and provide physical interaction.

[0050] More specifically, in order to further increase the authenticity of the entire experiment, the digital virtual engine and the physical engine are combined to construct the experimental scene, and the entire semi-physical simulation experimental scene is visualized, and the real-time status of the equipment is reconstructed more realistically in the three-dimensional scene. That is, the semi-physical simulation model of the experimental equipment is established using the semi-physical simulation system, and the semi-physical simulation experiment is carried out according to the key sample data of the digital simulation; on the basis of time synchronization with the semi-physical simulation system, the digital simulation system constructs the experimental equipment and the experimental scene in three dimensions according to the semi-physical simulation model and the data during the experiment.

[0051] In the experiment, the communication between the digital simulation system and the semi-physical simulation system adopts the scheme combining DDS and optical fiber. In order to realize the communication function between the two types of systems, according to the specific needs of data transmission in the experiment, a virtual-real system data interaction scheme consisting of three communication classes and two communication modules is designed. The data interaction scheme is as follows: Figure 6shown.

[0052] The Monte Carlo equipment experiment platform controls the digital simulation system and the semi-physical simulation integrated management system simultaneously through DDS instructions. The digital simulation system constructs the equipment three-dimensional geometric model and physical experimental scene through the virtual engine and the physical engine. The three-dimensional equipment geometric model and the physical environment model data of the experimental scene interact with the Monte Carlo equipment experiment platform and the semi-physical simulation integrated management system through the DDS communication plug-in and the fiber reflection memory communication plug-in. The Monte Carlo equipment experiment platform is responsible for the configuration of various experimental task parameters. The experimental tasks include but are not limited to the key parameters of the experimental equipment, the key parameters of the experimental environment, and the key parameters of interference / fault injection. After configuring the relevant parameters, they are distributed to the digital system and the semi-physical simulation system at the same time through the DDS instruction method. After receiving the relevant simulation experiment scene configuration parameters, the digital system initializes the experimental scene and the key parameters of the corresponding equipment geometric model to perform the simulation experiment initialization stage. After receiving the configuration parameters and fault / interference injection parameters related to the relevant simulation model, the semi-physical simulation system also initializes the corresponding semi-physical simulation node.

[0053] The equipment geometry model corresponds to the semi-physical simulation node one by one. The data interaction between them is carried out through the fiber reflection memory communication plug-in for real-time data sharing. During the simulation experiment, the virtual system and the virtual / real system clock of the semi-physical simulation system will be strictly synchronized through the distributed clock synchronization algorithm.

[0054] The simulation time synchronization between the semi-physical simulation system and the digital simulation system, and between the two systems, is a key technology for realizing virtual-real joint simulation experiments. Figure 7 shown.

[0055] The running time inside the semi-physical simulation system takes the moment when the model receives the running instruction issued by the control software as time 0, and uses a fixed time interval as the time verification to start stepping.

[0056] The real-time fiber optic reflective memory network is a real-time network based on high-speed fiber optic network shared storage technology. Compared with traditional networking technology, in addition to strict transmission determinism and predictability, it also has the characteristics of high data transmission speed, simple communication protocol, light host load, and strong adaptability of software and hardware platforms. The delay of reflective memory communication nodes is less than 1us; the fiber optic communication rate can reach 2.125Gbps; independent DMA channels, write speed is not less than 100Mbps; the real-time fiber optic reflective memory network is connected by fiber optic interface boards inserted in the computer through fiber optic lines to form a ring network. The onboard memory of the fiber optic interface board of each node has a shared data copy of other nodes. Logically, all nodes in the entire network share the same memory. Data is written at one point and updated at multiple points at the same time. In this way, high-speed data transmission and sharing are achieved. The semi-physical simulation system completes clock synchronization through the reflective memory fiber optic network within the system, and the clock error is within 0.5ms, so we can regard the internal time of the semi-physical simulation system as precise time.

[0057] The running time inside the digital system takes the moment when the running instruction data starts to be sent to the DDS channel as the 0 time, and adopts the Tick time stepping method (there will be a situation where the step length is set to 5ms, but the actual step length is 4.99ms). The internal time of the digital system is imprecise and the step length is controllable. And the time acquisition method of the UE digital system is the Windows computer system time. The Windows operating system uses a time slice rotation mechanism to schedule processes, and the imbalance of the network is uncertain and uncontrollable. It is impossible to control when the time synchronization process is scheduled and when it stops, resulting in unpredictable delays in the transmission of information packets in the network, affecting the accuracy of clock synchronization. Therefore, the digital simulation system under the Windows operating system cannot achieve synchronization with sub-millisecond accuracy. Therefore, the simulation time synchronization between the digital simulation system and the semi-physical simulation system is a bridge connecting the physical hardware clock and the virtual logic clock, and is the key to the clock synchronization solution of the entire virtual-real joint simulation system.

[0058] The computer's own clock usually cannot provide stable high-precision step division for the digital system, and it is difficult to achieve high-precision clock synchronization with the hardware clock of the semi-physical simulation system. Based on the idea of ​​step verification and satellite clock cycle alignment, this scheme designs a low-cost and high-precision virtual-real member clock synchronization method. After the semi-physical simulation system completes fine synchronization, it outputs clock synchronization information such as simulation step and related verification information. The digital system receives the semi-physical output step pulse information as the digital system logic clock simulation step frequency, receives the parsed related verification step through the network interface as the digital system logic clock simulation step, and uses the satellite clock cycle to verify the clock of the semi-physical simulation system and the digital simulation system. In this way, the logic clock of the virtual member and the hardware clock of the physical member have highly consistent simulation steps and values, that is, the digital simulation system and the semi-physical simulation system complete high-precision clock synchronization.

[0059] Based on the above ideas, Figure 8 As shown, the steps of time synchronization between the digital simulation system and the semi-physical simulation system include: The S101 semi-physical simulation system achieves clock synchronization between the simulation nodes based on the reflective memory system's high transmission efficiency, low latency, and simple protocol. The reflective memory system has the advantages of high transmission efficiency, low latency, and simple protocol. In the semi-physical simulation system, each simulation node uses these characteristics to achieve clock synchronization. This step lays the foundation for high-precision clock synchronization of the entire system, ensuring that the various parts of the semi-physical simulation system are consistent in time, just like soldiers in the army first calibrate the time within their respective teams to prepare for subsequent coordinated actions.

[0060] Before starting the joint simulation, S102 uses the UTC clock source to align the start time of the digital simulation system and the semi-physical simulation system. UTC (Coordinated Universal Time) is a high-precision time standard. By using it as a benchmark, the two systems can be at the same starting line at the initial moment, ensuring that the time difference is within a very small range, providing initial consistency for subsequent precise clock synchronization.

[0061] S103 performs clock synchronization based on step length verification when the simulation step length of the digital simulation system is consistent with the simulation step length of the semi-physical simulation system, specifically including: S1031 is initialized, the synchronization cycle is set, the digital simulation system preparation synchronization flag and synchronization flag are set to 0, and the simulation step counter is reset. The preparation synchronization flag is a status flag used to indicate whether the digital simulation system is ready for clock synchronization. When the preparation synchronization flag is 0, it indicates that the system is not ready and no key synchronization-related processing is performed; when a specific semi-physical simulation system pushes back the clock synchronization information and meets certain conditions, it is set to 1, indicating that the system begins to prepare for synchronization operations and prepare for subsequent key synchronization operations such as comparison of step values.

[0062] The synchronization flag is also a status flag, which is used to clarify whether the clocks of the digital simulation system and the semi-physical simulation system have been synchronized. The initial value is 0, which means that the synchronization is not completed. Under the premise that the synchronization flag is prepared to be 1, when the simulation step number value of the semi-physical simulation system parsed by the digital simulation system is equal to the current value of its own step counter, the synchronization flag is set to 1, which means that at the current moment, the digital simulation system confirms that the step lengths of the two match based on the clock synchronization information provided by the semi-physical simulation system, and completes a clock synchronization process.

[0063] The simulation step counter is a variable used to record and track the simulation step value in a digital simulation system.

[0064] When receiving the clock synchronization information pushed back by the semi-physical simulation system, S1032 parses the simulation number step value and assigns it to the step counter, and sets the synchronization ready flag to 1, when both the synchronization ready flag and the synchronization ready flag are 0; When the preparation synchronization flag is 1 and the synchronization flag is 0, the simulation step number value is parsed and compared with the current value of the step counter. If they are equal, the synchronization flag is set to 1, the preparation synchronization flag is set to 0, and step S1033 is executed; if they are not equal, the preparation synchronization flag is set to 0 and step S1032 is executed; S1033 waits until the synchronization cycle ends. The system waits in this step until the preset synchronization cycle ends. The synchronization cycle is a time interval set for periodic clock synchronization verification. During this waiting process, the system maintains the current state to ensure that related operations are completed within a complete cycle.

[0065] S1034 uses the UTC clock source to align the current time of the digital simulation system and the semi-physical simulation system. After a synchronization cycle, the time of the two systems is calibrated again to eliminate the time deviation that may occur during the synchronization process, so that the time of the two systems always remains highly consistent.

[0066] S1035 enters the next synchronization cycle and executes steps S1032 to S1035 in a loop. Through continuous looping, the clock synchronization check and calibration of the digital simulation system and the semi-physical simulation system are continuously performed to ensure that the two systems always maintain high-precision clock synchronization throughout the simulation process.

[0067] In the above-mentioned clock synchronization method based on step length verification, the digital simulation system needs to perform more system blocking and clock adjustment behaviors. Considering the instability of the digital system, the above-mentioned method is improved. Based on the verification principle of the IEEE1588 protocol, a clock synchronization method based on reflective memory and master clock verification is proposed. The semi-physical simulation system clock with high accuracy and stability is used as the master clock. Its main task is to send time-related information as a benchmark for clock correction. The digital system acts as a slave clock and returns information to the master clock after receiving the message. The deviation between the semi-physical simulation system clock and the digital system clock is determined by exchanging clock messages between the two, and time correction is completed in the digital simulation system.

[0068] Based on the above ideas, Fig. 9 As shown, this embodiment further proposes a method for time synchronization between a digital simulation system and a semi-physical simulation system, including: S111 uses the clock of the semi-physical simulation system as the master clock and the clock of the digital simulation system as the slave clock, and sets the correction cycle. The correction cycle is set to periodically adjust the slave clock to ensure that the slave clock can continue to synchronize with the master clock during long-term operation. The length of the correction cycle needs to be determined according to the specific needs and stability of the system. If the cycle is too short, the system overhead will increase, and if it is too long, it may cause the synchronization error to accumulate.

[0069] The S112 digital simulation system receives the clock synchronization message sent from the semi-physical simulation system, extracts the master clock time of the semi-physical simulation system sending the synchronization message from the clock synchronization message and records it as T1; S113 digital simulation system records the arrival time T2 of the clock synchronization message from the clock; S114 The digital simulation system sends a message, sets a flag in the message to start transmission error measurement, and records the sending time T3 from the clock; S115 After the semi-physical simulation system receives the message with the start transmission error measurement mark, the master clock records the delivery time T4; and sets the mark representing the completion of the transmission error measurement in the message pushed back to the digital simulation system, and records the delivery time T4 in the pushed back message; After receiving the message with the transmission error measurement completion mark, the S116 digital simulation system calculates the clock deviation and time delay of the master and slave times according to T1, T2, T3, and T4, and adjusts the digital simulation system slave clock according to the clock deviation and time delay.

[0070] The deviation between T2 and T1 is the sum of the clock deviation and time delay of the semi-physical and digital systems, that is, ; Among them, Offset is the clock deviation between the master and slave clocks, and Delay is the time delay.

[0071] The semi-physical simulation system receives a delay signal from the digital simulation system at each simulation step, wherein the system delay signal includes the sending time T3 of the delay signal; the master clock records the arrival time T4 of the system delay signal, and the deviation between T4 and T3 is the difference between the clock deviation and the time delay.

[0072] The deviation between T4 and T3 is the difference between the clock deviation and the time delay of the semi-physical and digital systems, that is, ; By integrating the above formulas, we can get: ; ; Thus, the clock deviation and time delay between the master and slave clocks can be obtained.

[0073] 4. Online verification system for semi-physical simulation experiments. The online verification system for semi-physical simulation experiments realizes the seamless integration of the key parameter modules of experimental equipment, the aerodynamic analysis module and the equipment simulation model in the environment, and develops and designs the Monte Carlo equipment test platform to complete the rapid design of the equipment Monte Carlo experiment and then complete the systematic large-sample simulation test. In order to realize the simulation of the entire process from the construction of the experimental environment of experimental equipment, the random generation of random configuration parameters of experimental equipment, the rapid construction of experimental equipment models, to the statistical analysis of experimental equipment test results, accelerate the efficiency and quality of equipment experimental tests, and significantly improve the efficiency and quality of equipment research and development design. The following introduces the system from the aspects of communication construction, data integration standards, experimental method integration, data storage, etc.

[0074] 1. Communication architecture.

[0075] The main task of the Monte Carlo equipment experiment platform is to extract key sample data of digital simulation from large sample experiments as data for sample-level simulation experiments of semi-physical simulation systems. By configuring the sample-level experimental situation data on the Monte Carlo equipment experiment platform and introducing factors such as relevant environments, system deviations, and system failures, multiple batches of semi-physical simulation system combat scenario verification can be performed, further improving the fidelity of experimental equipment in digital scenarios.

[0076] The semi-physical simulation integrated management system is in the form of software and hardware, relying on the semi-physical simulation model management software to schedule and manage various heterogeneous semi-physical model hardware equipment. In order to carry out joint simulation experiments, the Monte Carlo equipment experiment platform needs to interact with the semi-physical simulation integrated management system for data and instructions, and report the model data, status, and events to the experiment platform to realize data interaction between the experimental task platform and the semi-physical simulation nodes in the semi-physical integration framework.

[0077] The Monte Carlo equipment experiment platform uses DDS communication to communicate with the semi-physical integrated management system. The data communication method uses a unified semi-physical framework integration standard to send and receive data in the form of Json strings. The specific data communication logic is as follows: Fig.10 First, the number of experimental tasks, the number of threads to be started, and the relevant data to be configured for each task experiment are configured through the Monte Carlo equipment experimental platform, including the target location, deployment location, fault, deviation injection type, and injection time.

[0078] The Monte Carlo equipment experiment platform sends the configured experimental tasks to the semi-physical model integrated management system in batches through DDS. The semi-physical model integrated management system performs resource scheduling and task segmentation according to the experimental tasks. After each semi-physical simulation node task is completed, the task execution result is reported to the semi-physical model integrated management system. The management software then reports the task completion status to the Monte Carlo experiment platform through DDS. The experimental platform starts the next experimental task according to the completion progress of the experimental task. The main task of the Monte Carlo equipment experiment platform is to parse each configured experimental task, convert the data into a specific data structure, and further encapsulate the configuration key data in the data structure into the semi-physical integrated framework communication standard, and send the data to the management software using the DDS communication protocol. The management software parses the protocol, subdivides the task and sends it to each simulation node. After the task is completed, the management software encapsulates the task result into a string of semi-physical integration standards, and then uses DDS to send the data back to the Monte Carlo equipment experiment platform. Complete the closed-loop data communication link of the entire sample experiment.

[0079] 2. Data Integration Standards The semi-physical model integrated management system and the Monte Carlo equipment experiment platform communicate through the DDS communication mechanism. Based on the pre-defined topic names, communication channels and data standards, high-quality data interaction between the experiment platform and the semi-physical integrated management system can be achieved.

[0080] The semi-physical model integrated management system uses control software to manage and schedule various semi-physical simulation nodes. Taking the aircraft as an example, the semi-physical model integrated management system and the Monte Carlo equipment experiment platform exchange data through the DDS communication protocol. The interaction interface and process are as follows: ①Interface Table 1 Interaction interface ② Interaction process Interaction implementation process: After the experimental data configuration of the Monte Carlo equipment experimental platform is completed, the initialization data packet is sent to the semi-physical integrated management software through the DDS sending interface function. The data content is described in detail as follows: Table 2 Initialization data packet After receiving the initialization JSON data packet, the management software parses the JSON data packet, sends the parameters to the real-time simulation machine for model initialization, and reports the initialization success event package. The content of the event package is described as follows: Table 3 Initialization event feedback After receiving the initialization success event package, the digital proxy model feeds back to the digital integration framework, waiting to receive and send the operation instruction package to the semi-physical model integration framework. The content of the event package is described as follows: Table 4 Launch event package The management software receives the transmission instruction package, parses it and issues the model operation instruction to the semi-physical model. The model returns the operation success event and starts the real-time simulation. The returned event package is described as follows: Table 5 Launch event feedback packet Subsequently, the semi-physical digital proxy model will send the semi-physical model running JSON data packet in real time through the time stepping Tick data packet, which is described as follows: Table 6 Tick operation data packet After receiving the running JSON data packet, the management software parses the JSON data packet and sends the parameters to the real-time simulator for real-time solution. The management and control software receives the latest state of the model calculated by the real-time simulator. The specific contents are described as follows: Table 7 Tick report data packet If the model task is completed / equipment fails, the control software will report the task completion / failure event. After receiving the hit event, the equipment experiment platform will determine whether the experiment task is completed based on the task event. The content is described as follows: Table 8 Task completion / fault event reporting data packet If it is necessary to inject deviation or fault into the model, the equipment experiment platform sends the fault data packet to the semi-physical integrated management software through the DDS sending interface function. The content is described as follows: Table 9 Fault injection event running data package 3. Monte Carlo test method integration During the operation of experimental equipment in the actual environment, due to the influence of environmental factors and system errors, its attitude, trajectory and control parameters deviate from the ideal attitude, trajectory and control parameters. The Monte Carlo equipment test aims to analyze the various random interference factors that special equipment is subject to in practice, provide treatment methods for various interference factors, and establish a six-degree-of-freedom Monte Carlo equipment simulation model. A large number of simulation experiments were carried out through the semi-physical simulation experiment online verification system, and the accuracy of the key parameter data of the experimental equipment was studied to verify the accuracy of the simulation model and the effectiveness of the simulation method.

[0081] In this embodiment, the online verification method of semi-physical simulation experiment using Monte Carlo method includes the following steps: S1 establishes a semi-physical simulation model of the experimental equipment based on its motion characteristics and motion laws.

[0082] The semi-physical simulation system builds a semi-physical simulation model based on the motion characteristics of the experimental equipment (such as whether the equipment is in linear motion, curved motion, or rotational motion, etc.) and the laws of motion (such as the dynamic equations followed, the periodicity of motion, etc.). This model will simulate the motion of the experimental equipment in the actual environment as realistically as possible, providing an operable virtual object for subsequent experiments, just like building a "stage" that simulates the operation of the actual equipment. All subsequent experimental operations will be carried out based on this model.

[0083] S2 extracts key sample data of digital simulation from large sample experiments.

[0084] After conducting a large number of digital simulation experiments (i.e., large sample experiments), the Monte Carlo equipment experiment platform selects key sample data that are more in line with the actual situation from a large number of data. These key sample data are representative and can reflect the typical operating status of the experimental equipment under different conditions. For example, in a large sample experiment simulating aircraft flight, data from key stages such as aircraft takeoff, cruising, and landing may be selected as key sample data. These data are of great guiding significance for conducting experiments on semi-physical simulation systems and can help simulate actual conditions more accurately.

[0085] S3 determines the random interference factors and their distribution patterns during the operation of the experimental equipment, and generates random interference parameters based on the random interference factors and their distribution patterns.

[0086] This step is also performed by the Monte Carlo equipment experiment platform. First, it is necessary to clarify which random interference factors will affect the experimental equipment during actual operation. For example, experimental equipment running outdoors may be affected by weather factors (such as wind speed, wind direction, temperature changes, etc.), electromagnetic environment interference, etc. Then, determine the distribution law of these interference factors. For example, the wind speed may conform to a certain probability distribution. According to these distribution laws, a special platform is used to generate the corresponding random interference parameters. These parameters will be loaded into the semi-physical simulation system in subsequent experiments to simulate the random interference of the experimental equipment in the actual environment, making the experiment closer to the real situation.

[0087] S4 configures experimental parameters, makes experimental plans according to the experimental parameters, performs semi-physical simulation experiments on the semi-physical simulation system using key sample data of digital simulation according to the experimental plans, and loads random interference parameters during the experimental operation.

[0088] The Monte Carlo equipment experiment platform configures various experimental parameters according to the purpose and requirements of the experiment, such as the number of experiments, the duration of each experiment, the initial state of the experimental equipment, etc. The semi-physical model integrated management system conducts comprehensive experimental planning based on these experimental parameters and determines the process and steps of the experiment. Afterwards, the semi-physical simulation experiment is carried out on the semi-physical simulation system using the previously extracted key sample data of digital simulation. During the experimental operation, random interference parameters are loaded in a set manner so that the experiment can simulate the situation where the experimental equipment is interfered with in the actual environment, so as to observe and study the operating performance of the experimental equipment under complex conditions.

[0089] S5 constructs the experimental equipment and experimental scene in three dimensions based on the semi-physical simulation model and the data during the experiment to realize the visualization of the experimental process.

[0090] With the help of the semi-physical simulation model and the data generated during the experiment, the digital simulation system performs three-dimensional modeling of the experimental equipment and the experimental scene. Through three-dimensional construction, the appearance, movement state and experimental scene of the experimental equipment can be presented in an intuitive three-dimensional graphical way to realize the visualization of the experimental process. This helps the experimenters to observe the operation of the experimental equipment more clearly and find possible problems in time, such as whether the movement posture of the equipment meets expectations, whether the influence of interference factors in the experimental scene on the equipment is intuitively visible, etc., providing a more intuitive basis for experimental analysis.

[0091] S6 obtains the experimental equipment operating parameters and equipment evaluation results after the disturbance.

[0092] Under the condition of loading random interference parameters, the experimental equipment is run in the semi-physical simulation system. This step aims to obtain various operating parameters of the experimental equipment after being disturbed, such as attitude parameters (pitch angle, yaw angle, roll angle, etc.), trajectory parameters (position coordinates, speed, etc.) and control parameters (such as throttle opening, servo angle, etc.). At the same time, the operating performance of the experimental equipment under such disturbance is evaluated according to certain evaluation criteria to obtain the equipment evaluation results. These results and parameters are crucial for studying the performance of experimental equipment in actual complex environments.

[0093] S7 collects statistics on the operating parameters of the experimental equipment and the equipment evaluation results, thereby obtaining the statistical characteristic values ​​of the key parameters of the experimental equipment.

[0094] After the data is aggregated to the Monte Carlo equipment experiment platform, it performs statistical analysis on the experimental equipment operating parameters and equipment evaluation results obtained from multiple experiments. Through statistical methods, the statistical characteristic values ​​of the key parameters of the experimental equipment, such as mean, variance, standard deviation, etc., are calculated. The mean can reflect the average level of the key parameters, while the variance and standard deviation can reflect the degree of discreteness of the parameters. These statistical characteristic values ​​help to gain a deeper understanding of the overall characteristics of the key parameters of the experimental equipment, evaluate the stability and reliability of the experimental equipment under different random interferences, and provide data support for further optimizing the experimental equipment or improving the simulation method.

[0095] The Monte Carlo method is an experimental mathematical method that uses random numbers for statistical tests to obtain statistical characteristic values ​​(such as mean, variance, probability, etc.) as the numerical solution to the problem to be solved. It is a method that uses probabilistic mathematical models and statistical characteristic calculations of the physical process of the actual problem being studied to reproduce the process. The Monte Carlo equipment test platform is based on the semi-physical integration framework. The Monte Carlo equipment test platform is developed based on C++ as a semi-physical integrated management system for experimental task data configuration, control instruction generation, result reporting and analysis, and unified management and analysis of experimental result statistics. The platform interacts with the semi-physical integration framework through the DDS communication protocol to achieve management and scheduling of simulation models of each simulation node, thereby completing the simulation experiment.

[0096] The semi-physical simulation experiment online verification system provides a complete set of simulation model integration methods based on the semi-physical integration framework. Its basic principles are as follows: Fig.11As shown. The Monte Carlo equipment test platform uses data random generation technology to encapsulate the equipment model input data in a table format file as an ASCII file. First, the platform parses the input parameters from the input file, and sends the input data to the semi-physical simulation integrated management system in the form of instructions through the DDS protocol. The integrated management system dynamically schedules the simulation nodes that need to be started, and the simulation nodes execute the simulation model program for analysis and calculation. Finally, the analysis results are returned to the semi-physical integrated management system. The semi-physical integrated management system then reports the result data to the Monte Carlo equipment test platform for data statistics. Finally, the input and output data of the parts are stored in the specified location file, and the input and output parameters of the simulation experiment are obtained by parsing the output file.

[0097] 4. Data storage.

[0098] Aiming at the problem of semi-physical large sample simulation data storage, a data abstraction link is established based on database design patterns and target data objects, and the library table design that reflects the semi-physical large sample simulation data is realized. Based on the transaction mechanism, concurrent access control of large data samples is realized, and the load of CPU and I / O resources is effectively controlled while ensuring throughput. Based on data visualization, data support is provided for large sample semi-physical model verification, such as Fig.12 shown.

[0099] like Fig.13 As shown in the figure, in the semi-physical large sample simulation experiment, the semi-physical simulation system obtains data from multiple data sources (model parameters returned by the simulation model, heartbeat monitoring of the simulation node), calls the database driver in the form of a dynamic link library (DLL), inserts the data into the transaction queue through the API interface, and waits for the concurrent execution of the transaction. The interface accepts SQL execution statements and data. Through the visualization tool Navicat and the model verification platform, it interacts with the server to query and obtain the corresponding data. It can also export the corresponding data through Navicat and provide it to the model verification platform.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a computer terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0101] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A semi-physical simulation experiment online verification system, characterized in that include: Monte Carlo equipment experimental platform, used to configure experimental parameters, provide key sample data for digital simulation and generate random interference parameters; Conduct statistical analysis on the experimental equipment operating parameters and equipment evaluation results to obtain the statistical characteristic values ​​of the key parameters of the experimental equipment; A semi-physical simulation integrated management system is used to schedule and manage the semi-physical simulation system; the semi-physical simulation system is used to establish a semi-physical simulation model of the experimental equipment according to the motion characteristics and motion laws of the experimental equipment; to plan the experiment according to the experimental parameters; and to perform a semi-physical simulation experiment on the semi-physical simulation system using digital simulation key sample data according to the experimental plan; And during the experiment operation, the random interference parameters issued by the Monte Carlo equipment experiment platform are loaded, so as to obtain the experimental equipment operation parameters and equipment evaluation results after the disturbance; The digital simulation system is used to construct the experimental equipment and experimental scenes in three dimensions according to the semi-physical simulation model and the data in the experimental process, so as to realize the visualization of the experimental process.

2. According to claim 1, a semi-physical simulation experiment online verification system is characterized in that The Monte Carlo equipment experimental platform includes: The experimental parameter configuration module is used to configure the number of experimental tasks, set the number of threads to be started, and the experimental data to be configured for each task experiment; the experimental data includes the injection type and injection time of random interference; The experimental data generation module is used to extract key sample data of digital simulation from large sample experiments for semi-physical simulation experiments and generate random interference parameters.

3. The online verification system for semi-physical simulation experiment according to claim 1 is characterized in that: The semi-physical simulation system is a distributed system, including several semi-physical simulation nodes; the three-dimensional model of the experimental equipment constructed by the digital simulation system corresponds one to one with the semi-physical simulation nodes, and during the experiment, the digital simulation system and the semi-physical simulation system are synchronized in time.

4. The online verification system for semi-physical simulation experiment according to claim 2 is characterized in that: The semi-physical simulation nodes transmit data via optical fiber reflection memory.

5. The online verification system for semi-physical simulation experiment according to claim 1 is characterized in that The semi-physical simulation integrated management system includes: The experiment planning module is used to plan the experiment according to the experimental parameters and send the experimental data and key sample data of digital simulation to the semi-physical simulation node for experiment; The node scheduling module is used to schedule the corresponding semi-physical simulation nodes to participate in the experiment according to the number of threads set in the Monte Carlo equipment experiment platform.

6. The online verification system for semi-physical simulation experiment according to claim 1 is characterized in that The digital simulation system comprises: Virtual engine, used to construct 3D geometric models of experimental equipment and create visual experimental scenes, thus realizing visual reconstruction of equipment status; Physics engine, used to simulate physical experiment scenes and provide physical interaction.

7. A semi-physical simulation experiment online verification method, characterized in that include: According to the motion characteristics and motion laws of the experimental equipment, a semi-physical simulation model of the experimental equipment is established; Extract key sample data of digital simulation from large sample experiments; Determine the random interference factors and their distribution patterns during the operation of the experimental equipment, and generate random interference parameters based on the random interference factors and their distribution patterns; Configure experimental parameters, plan experiments according to the experimental parameters, conduct semi-physical simulation experiments on the semi-physical simulation system using key digital simulation sample data according to the experimental plan, and load random interference parameters during the experiment; The experimental equipment and experimental scene are constructed in three dimensions according to the semi-physical simulation model and the data in the experimental process to realize the visualization of the experimental process; Obtain the experimental equipment operating parameters and equipment evaluation results after disturbance; Statistics are collected on the operating parameters of the experimental equipment and the equipment evaluation results to obtain the statistical characteristic values ​​of the key parameters of the experimental equipment.

8. The online verification method for a semi-physical simulation experiment according to claim 7 is characterized in that: A semi-physical simulation model of the experimental equipment is established using the semi-physical simulation system, and semi-physical simulation experiments are carried out based on key sample data of the digital simulation. Based on time synchronization with the semi-physical simulation system, the digital simulation system performs three-dimensional construction of the experimental equipment and experimental scene based on the semi-physical simulation model and data from the experimental process.

9. The online verification method for a semi-physical simulation experiment according to claim 8 is characterized in that The steps for time synchronization between the digital simulation system and the semi-physical simulation system include: Clock synchronization between simulation nodes within the S101 semi-physical simulation system; S102 uses a UTC clock source to align the start time of the digital simulation system and the semi-physical simulation system; S103 performs clock synchronization based on step length verification when the simulation step length of the digital simulation system is consistent with the simulation step length of the semi-physical simulation system, specifically including: S1031 initialization, setting the synchronization period, setting the digital simulation system preparation synchronization flag and the synchronization flag to 0, and returning the simulation step counter to zero; When receiving the clock synchronization information pushed back by the semi-physical simulation system, S1032 parses the simulation number step value and assigns it to the step counter, and sets the synchronization ready flag to 1, when both the synchronization ready flag and the synchronization ready flag are 0; When the preparation synchronization flag is 1 and the synchronization flag is 0, the simulation step number value is parsed and compared with the current value of the step counter. If they are equal, the synchronization flag is set to 1, the preparation synchronization flag is set to 0, and step S1033 is executed; if they are not equal, the preparation synchronization flag is set to 0 and step S1032 is executed; S1033 waits until the synchronization cycle ends; S1034 uses a UTC clock source to align current times of the digital simulation system and the semi-physical simulation system; S1035 enters the next synchronization cycle and executes steps S1032 to S1035 in a loop.

10. The online verification method for a semi-physical simulation experiment according to claim 8, characterized in that The steps for time synchronization between the digital simulation system and the semi-physical simulation system include: S111 uses the clock of the semi-physical simulation system as the master clock, uses the clock of the digital simulation system as the slave clock, and sets a correction period; The S112 digital simulation system receives the clock synchronization message sent from the semi-physical simulation system, extracts the master clock time of the semi-physical simulation system sending the synchronization message from the clock synchronization message and records it as T1; S113 digital simulation system records the arrival time T2 of the clock synchronization message from the clock; S114 The digital simulation system sends a message, sets a flag in the message to start transmission error measurement, and records the sending time T3 from the clock; S115 After the semi-physical simulation system receives the message with the start transmission error measurement mark, the master clock records the delivery time T4; and sets the mark representing the completion of the transmission error measurement in the message pushed back to the digital simulation system, and records the delivery time T4 in the pushed back message; After receiving the message with the transmission error measurement completion mark, the S116 digital simulation system calculates the clock deviation and time delay of the master and slave clocks according to T1, T2, T3, and T4, and adjusts the digital simulation system slave clock according to the clock deviation and time delay.

Images