Digital parallel system and method for key algorithm verification

Abstract

The invention provides a digital parallel system and method for key algorithm verification, and the method comprises the steps: achieving the construction of an algorithm verification platform through the servitization design thought of algorithm resources, and constructing an algorithm slot and a goods shelf type data system under the constraint of algorithm specifications; an object-oriented, componentization and parameterization thought is adopted, a model system containing different levels of red, blue and white parties is built, task elements are flexibly configured, model granularity is matched according to task requirements, a physical world is mapped, and an algorithm credibility verification environment structure is built; aiming at the evaluation and efficiency verification problem of the algorithm in a changeable scene, completing the establishment of an evaluation system; comprising the steps of multi-means evaluation data acquisition service, construction of a configurable evaluation system, evaluation implementation and algorithm verification, and implementation of an algorithm performance evaluation closed loop. In order to cope with different business scenes, the modularized and parameterized model system can quickly establish corresponding deduction scenes through different combinations and parameter configurations.

Description

Technical Field

[0001] The present invention relates to the field of simulation system design, and in particular to a digital parallel system and method for key algorithm verification. Background Art

[0002] In the demonstration of aerospace equipment systems and verification of core capabilities, the confidence and reliability of core algorithms such as single-satellite payload scheduling algorithms and satellite constellation resource scheduling algorithms are crucial to the development of single-type equipment and the empowerment of the entire business system. However, due to the complexity and diversity of actual business scenarios, the experimental verification of these algorithms in real environments has problems such as high testing costs and small data samples. Therefore, algorithm verification by building a fully digital parallel system is an important method.

[0003] Therefore, in order to further improve the scalability and authenticity of algorithm verification scenarios, a digital parallel system is needed that can simulate the complex business environment required for key algorithm verification, evaluate the information chain, time chain, and precision chain of the entire equipment system that supports the actual business for the core algorithm, and improve the comprehensiveness of algorithm verification and the confidence of the algorithm itself. It can not only explore the effectiveness of future equipment systems in advance, but also test and verify the algorithm before the actual task is executed. Summary of the invention

[0004] In view of the defects in the prior art, the object of the present invention is to provide a digital parallel system and method for key algorithm verification.

[0005] A digital parallel method for key algorithm verification provided by the present invention includes:

[0006] Step S1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and under the constraints of algorithm specifications, the algorithm slot and shelf data system are constructed;

[0007] Step S2: Adopting the idea of ​​object-oriented, component-based, and parameterized, we build a model system of different levels for the red, blue, and white parties, and construct a trusted verification environment for the algorithm;

[0008] Step S3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

[0009] Preferably, in step S1:

[0010] The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems.

[0011] The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

[0012] Preferably, in step S2:

[0013] In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world.

[0014] The Red Force equipment system covers the resource scheduling model, data processing model, and task decision center model in the center layer, the space-based detection equipment model in the single-type equipment layer, the equipment platform model and the main load model in the ground-based detection equipment model subsystem layer, and abstracts the corresponding algorithm services based on the model:

[0015] The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services.

[0016] The Red Force's equipment system covers the resource scheduling model, data processing model, and task decision model in the central layer. It completes task decisions for system tasks based on command information, generates task plans, schedules and commands various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain.

[0017] The space-based detection equipment model and the ground-based detection equipment model in the single-type equipment layer. The space-based detection equipment model includes satellite detection equipment of different orbits and systems. The ground-based detection equipment model includes radar detection equipment and telescope detection equipment. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building.

[0018] The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

[0019] Preferably, in step S2:

[0020] The blue equipment system involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and mission behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level:

[0021] The blue side equipment model system includes a complete mission system and a single equipment layer model;

[0022] The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information.

[0023] The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions.

[0024] The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain;

[0025] The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

[0026] Preferably, in step S3:

[0027] Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification:

[0028] Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators;

[0029] The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction.

[0030] The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan;

[0031] The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage.

[0032] The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm.

[0033] The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

[0034] A digital parallel system for key algorithm verification provided by the present invention includes:

[0035] Module M1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and the algorithm slot and shelf data system are constructed under the constraints of algorithm specifications;

[0036] Module M2: Adopting the idea of ​​object-oriented, component-based and parameterized, we build a model system of different levels for the red, blue and white parties, and construct a trusted verification environment for the algorithm;

[0037] Module M3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

[0038] Preferably, in the module M1:

[0039] The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems.

[0040] The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

[0041] Preferably, in the module M2:

[0042] In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world.

[0043] The Red Force equipment system covers the resource scheduling model, data processing model, and task decision model in the center layer, the space-based detection equipment model in the single-type equipment layer, the equipment platform model and the main load model in the ground-based detection equipment model subsystem layer, and abstracts the corresponding algorithm services based on the model:

[0044] The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services.

[0045] The Red Force's equipment system covers the resource scheduling model, data processing model, and task decision model in the central layer. It completes task decisions for system tasks based on command information, generates task plans, schedules and commands various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain.

[0046] The space-based detection equipment model and the ground-based detection equipment model in the single-type equipment layer. The space-based detection equipment model includes satellite detection equipment of different orbits and systems. The ground-based detection equipment model includes radar detection equipment and telescope detection equipment. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building.

[0047] The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

[0048] Preferably, in the module M2:

[0049] The blue equipment system involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and mission behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level:

[0050] The blue side equipment model system includes a complete mission system and a single equipment layer model;

[0051] The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information.

[0052] The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions.

[0053] The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain;

[0054] The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

[0055] Preferably, in the module M3:

[0056] Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification:

[0057] Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators;

[0058] The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction.

[0059] The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan;

[0060] The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage.

[0061] The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm.

[0062] The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. The present invention supports equipment testing and appraisal by digital means: for equipment that has not yet been delivered, equipment performance analysis can be carried out before actual installation, so that the test and appraisal time point can be moved forward, and the optimization of constellation scheme design can be guided; restrictive tests can be carried out instead of actual installation, the test content can be expanded, and the test data samples can be enriched. The efficiency of performance evaluation can be improved, the test cost can be reduced, and the role of performance evaluation can be maximized;

[0065] 2. The present invention improves the integration of algorithm verification: constructs a digital parallel world for key algorithm verification; couples business processes and simulation deduction lines, designs and implements simple-to-operate and highly portable information processing algorithms and resource scheduling verification platforms. Based on component-based and parameterized design, it facilitates the improvement of existing algorithms and the implantation of new algorithms, and is easy to verify, debug and modify related algorithms;

[0066] 3. The flexibility and reusability of the verification scenario of the present invention: By organizing and managing models of different levels and categories to form a model system, the reusability of the model can be improved. In order to cope with different business scenarios, the modular and parameterized model system can quickly form corresponding deduction scenarios through different combinations and parameter configurations. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0068] Figure 1 It is a schematic diagram of the system of the present invention;

[0069] Figure 2 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0070] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0071] Embodiment 1:

[0072] The present invention relates to a digital parallel system and method for key algorithm verification in actual business, in particular, the design of a digital parallel system for algorithm verification that can support core algorithms such as constellation scheduling and multi-source data fusion processing. The present invention provides a set of solutions for verification and optimization of key algorithms for constellation scheduling in a complete task; the present invention provides an implementation method suitable for a system-level integrated algorithm verification platform, supports the construction of flexible and highly reusable verification scenarios, and can support the design and development of related information systems for algorithm evaluation.

[0073] The present invention is oriented towards the entire actual business process, covers all elements in the business, completes the mapping relationship in the digital space to establish the corresponding parallel system, and constructs an open algorithm verification platform that can support the system performance evaluation of satellite constellations, thereby completing the demonstration, verification and optimization of key algorithms such as payload / satellite scheduling in the equipment system.

[0074] An algorithm verification platform based on a deduction platform, a trusted verification environment based on a full-factor model system for the entire task process, and an evaluation system that supports algorithm verification. The following steps are included:

[0075] Step 1: By designing algorithm resources as services, we can build an algorithm verification platform. This includes an integrated simulation support and deduction platform for the core functions of simulation operation, and builds algorithm slots and shelf-type data systems under the constraints of algorithm specifications;

[0076] Step 2: In response to the wide-area construction needs of the integrated space-ground algorithm verification environment, an object-oriented, component-based, and parameterized approach is adopted to build a red, blue, and white model system with different levels. Task elements can be flexibly configured, and the model granularity can be matched according to task needs. The physical world is fully and correctly mapped to build a trusted algorithm verification environment.

[0077] Step 3: To address the evaluation and performance verification issues of key algorithms in changing scenarios, we will complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

[0078] According to a digital parallel method for key algorithm verification provided by the present invention, Figure 1-Figure 2 As shown, including:

[0079] Step S1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and under the constraints of algorithm specifications, the algorithm slot and shelf data system are constructed;

[0080] Specifically, in step S1:

[0081] The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems.

[0082] The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

[0083] Step S2: Adopting the idea of ​​object-oriented, component-based, and parameterized, we build a model system of different levels for the red, blue, and white parties, and construct a trusted verification environment for the algorithm;

[0084] Specifically, in step S2:

[0085] In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world.

[0086] The Red Force equipment system covers the resource scheduling model, data processing model, and task decision model in the center layer, the space-based detection equipment model in the single-type equipment layer, the equipment platform model and the main load model in the ground-based detection equipment model subsystem layer, and abstracts the corresponding algorithm services based on the model:

[0087] The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services.

[0088] The Red Force's equipment system covers the resource scheduling model, data processing model, and task decision model in the central layer. It completes task decisions for system tasks based on command information, generates task plans, schedules and commands various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain.

[0089] The space-based detection equipment model and the ground-based detection equipment model in the single-type equipment layer. The space-based detection equipment model includes satellite detection equipment of different orbits and systems. The ground-based detection equipment model includes radar detection equipment and telescope detection equipment. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building.

[0090] The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

[0091] Specifically, in step S2:

[0092] The blue equipment system involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and mission behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level:

[0093] The blue side equipment model system includes a complete mission system and a single equipment layer model;

[0094] The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information.

[0095] The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions.

[0096] The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain;

[0097] The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

[0098] Step S3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

[0099] Specifically, in step S3:

[0100] Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification:

[0101] Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators;

[0102] The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction.

[0103] The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan;

[0104] The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage.

[0105] The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm.

[0106] The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

[0107] Embodiment 2:

[0108] Embodiment 2 is a preferred example of Embodiment 1, and is used to illustrate the present invention in more detail.

[0109] The present invention also provides a digital parallel system for key algorithm verification. The digital parallel system for key algorithm verification can be implemented by executing the process steps of the digital parallel method for key algorithm verification, that is, those skilled in the art can understand the digital parallel method for key algorithm verification as a preferred implementation of the digital parallel system for key algorithm verification.

[0110] A digital parallel system for key algorithm verification provided by the present invention includes:

[0111] Module M1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and the algorithm slot and shelf data system are constructed under the constraints of algorithm specifications;

[0112] Specifically, in the module M1:

[0113] The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems.

[0114] The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

[0115] Module M2: Adopting the idea of ​​object-oriented, component-based and parameterized, we build a model system of different levels for the red, blue and white parties, and construct a trusted verification environment for the algorithm;

[0116] Specifically, in the module M2:

[0117] In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world.

[0118] The Red Force equipment system covers the resource scheduling model, data processing model, and task decision model in the center layer, the space-based detection equipment model in the single-type equipment layer, the equipment platform model and the main load model in the ground-based detection equipment model subsystem layer, and abstracts the corresponding algorithm services based on the model:

[0119] The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services.

[0120] The Red Force's equipment system covers the resource scheduling model, data processing model, and task decision model in the central layer. It completes task decisions for system tasks based on command information, generates task plans, schedules and commands various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain.

[0121] The space-based detection equipment model and the ground-based detection equipment model in the single-type equipment layer. The space-based detection equipment model includes satellite detection equipment of different orbits and systems. The ground-based detection equipment model includes radar detection equipment and telescope detection equipment. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building.

[0122] The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

[0123] Specifically, in the module M2:

[0124] The blue equipment system involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and mission behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level:

[0125] The blue side equipment model system includes a complete mission system and a single equipment layer model;

[0126] The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information.

[0127] The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions.

[0128] The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain;

[0129] The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

[0130] Module M3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

[0131] Specifically, in the module M3:

[0132] Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification:

[0133] Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators;

[0134] The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction.

[0135] The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan;

[0136] The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage.

[0137] The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm.

[0138] The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

[0139] Embodiment 3:

[0140] Embodiment 3 is a preferred example of Embodiment 1, and is used to illustrate the present invention in more detail.

[0141] The present invention is oriented to the actual full business process, covers all elements in the business, completes the mapping relationship in the digital space to establish the corresponding parallel system, and builds an open algorithm verification platform that can support the system performance evaluation of the satellite constellation, thereby completing the demonstration verification and optimization of key algorithms such as payload / satellite scheduling in the equipment system. The following is a detailed description of the technical solution of the present invention in conjunction with the accompanying drawings.

[0142] like Figure 1 As shown, the present invention solves the above technical problems through the following technical solutions: a digital parallel system design and method for key algorithm verification, which includes the following steps:

[0143] A digital parallel system for key algorithm verification, characterized by: an algorithm verification platform based on a deduction platform, a trusted verification environment based on a full-factor model system for the entire task process, and an evaluation system that supports algorithm verification. It includes the following steps:

[0144] Step 1: By designing algorithm resources as services, we can build an algorithm verification platform, including a deduction platform with core functions to support simulation operations, the formulation of algorithm specifications, algorithm slots, and the construction of a shelf-type data system;

[0145] Step 2: To address the confidence issues in core algorithm verification, we will complete the construction of an algorithm verification environment that covers the entire process and all elements of the mission. This includes service-oriented algorithm resources, a red team equipment model system that fully covers the information chain and time chain, a blue team command-and-control equipment model system, and a white team model system. A configurable, high-confidence digital world is formed based on a componentized model system.

[0146] Step 3: To evaluate and verify the performance of key algorithms in tasks, we will complete the construction of the evaluation system, including evaluation operator collection services, task-oriented evaluation systems, and evaluation display and comparison solutions, to achieve a closed loop of algorithm performance evaluation.

[0147] Preferably, the simulation platform in step one has core functions such as directing, controlling, adjudicating and managing simulation scenarios, and can provide common simulation support services and application support services for different systems.

[0148] Preferably, the algorithm verification platform in step one has all the functions of the deduction platform, and is further developed in a targeted manner on the basis of the deduction platform. It can be oriented to actual business needs, propose interface specifications and process specifications for core algorithms, complete the design of algorithm interface slots based on the core algorithm specifications, and independently design the algorithm interface.

[0149] Preferably, the Red Team equipment system in step 2 covers the resource scheduling model, data processing model, and task decision model in the central layer, the space-based detection equipment model in the single-type equipment layer, and the equipment platform model and main payload model in the ground-based detection equipment model subsystem layer, and abstracts the corresponding algorithm services based on the model.

[0150] Preferably, the blue equipment system in step 2 involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and mission behavior model at the system level, and the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single-type equipment layer.

[0151] Preferably, the white model system in step 2 includes a spatiotemporal computing framework in the cognitive domain and a geophysical environment, an electromagnetic environment, a stellar orbit simulation and background targets in the physical domain.

[0152] Preferably, the evaluation and performance verification of the key algorithm in the task in step three includes evaluation of data collection services, establishment of a task-oriented evaluation system, implementation of an evaluation plan, and algorithm verification.

[0153] Preferably, the problem of establishing a task-oriented evaluation system in step three mainly includes modeling of terminal performance indicators, normalization of evaluation indicators, determination of evaluation indicator weights and aggregation of evaluation indicators.

[0154] Preferably, the normalization of the evaluation index in step 3 can adopt a curve normalization method. If the value range of a function is [0,1] and the function is a monotonic function, then the function is called a utility function of the curve normalization method. There are three forms of utility functions, namely optimistic, most likely, and pessimistic. The mathematical expressions of these three forms are as follows:

[0155] (1) Optimistic utility function

[0156]

[0157] (2) Most Likely Utility Function

[0158]

[0159] (3) Pessimistic Utility Function

[0160]

[0161] In the above formula, M min is the minimum value required to achieve the target indicator attribute value, M rec is the required value to complete the target indicator attribute value, M exp is the value that is expected to be achieved by completing the target indicator attribute value. min 、M req 、M exp When , the corresponding utility function values ​​are 0, 0.5, and 1, respectively. Then, the values ​​of α, β, and θ can be obtained through the point (M min , 0), and (M req , 0) and (M exp , 0) to calculate.

[0162] Preferably, the normalization of the evaluation index in step 3 can adopt a fuzzification algorithm. The calculation result of the fuzzification algorithm mainly serves the fuzzy hierarchical analysis method. In order to obtain the fuzzy vector of the quantitative index, the fuzzy satisfaction of the quantitative index must be obtained first. Let the number of fuzzy satisfaction levels be N, and use X1, X2, ..., X n They represent the levels of satisfaction from high to low, that is, X1 represents the highest satisfaction, X n Indicates the lowest satisfaction. According to the data variation range of the single indicator, the domain experts give the fuzzy set membership function corresponding to each satisfaction level. For example, the first positioning time is an attribute indicator of the smaller the better type. Suppose there are five satisfaction levels, namely {highest, higher, medium, lower, lowest}. In practical applications, the most widely used fuzzy numbers are triangular fuzzy numbers and trapezoidal fuzzy numbers. For the convenience of calculation, trapezoidal fuzzy numbers are used. The smaller the better type indicator membership function is shown in the following formula.

[0163]

[0164] Similarly, for the larger-is-better type and moderate-value indicators, the same method can be used to give the satisfaction level and its membership function. For each indicator, according to the satisfaction level and its membership function, the corresponding parameter value is given by the expert, and the fuzzy vector of each indicator can be calculated by the above formula. For example, for the smaller-is-better type indicator "first positioning time", the satisfaction level is defined as {highest, higher, medium, lower, lowest} on the domain [0, 1], and the parameters of its satisfaction level membership function are given by experts, which are a1=10, a2=21, a3=40, a4=80; b2=15, b3=26, b4=60, b5=100, respectively. The current positioning time is 25 minutes. Substituting it into the formula, the fuzzy satisfaction membership of "first positioning time" can be calculated, and its fuzzy vector is {0, 0.2, 0.8, 0, 0}. The membership and fuzzy vector of the remaining quantitative indicators can be calculated in the same way as above.

[0165] Preferably, the evaluation index weights in step three may be determined by using a subjective weighting method, an objective weighting method, or a combined weighting method.

[0166] Preferably, the evaluation process in step three involves a simulation preparation stage, a simulation implementation stage and an evaluation closed-loop stage, wherein the three stages can be converted into each other: the simulation preparation stage can enter the simulation implementation stage, the simulation implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the simulation preparation stage and the simulation implementation stage.

[0167] Preferably, the deduction preparation stage in the evaluation of step 3 can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data can be pushed to the scenario management to generate an initial situation. After the rapid deduction and evaluation closed loop is completed according to this situation, the scenario management generates data requirements for the constellation design, thereby assisting the optimization closed loop of the constellation design algorithm.

[0168] Preferably, the simulation implementation stage in the evaluation process of step three can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation and performs simulation simulation according to the preset process. The load intelligent scheduling algorithm can be executed in two modes: human-in-the-loop and human-out-of-the-loop, and scheduling decisions can be generated. Through guidance and intervention, the corresponding scheduling algorithm strategy is implemented in the scenario, and the implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

[0169] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0170] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A digital parallel method for key algorithm verification, characterized in that: include: Step S1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and under the constraints of algorithm specifications, the algorithm slot and shelf data system are constructed; Step S2: Adopting the idea of ​​object-oriented, component-based, and parameterized, we build a model system of different levels for the red, blue, and white parties, and construct a trusted verification environment for the algorithm; Step S3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

2. The digital parallel method for key algorithm verification according to claim 1, characterized in that: In step S1: The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems. The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

3. The digital parallel method for key algorithm verification according to claim 1, characterized in that: In step S2: In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world. The Red Force's equipment system covers resource scheduling models, data processing models, and task decision models in the system layer, space-based detection models and ground-based detection models in the single-type equipment layer, equipment platform models and main payload models in the subsystem layer, and abstracts corresponding algorithm services based on the models: The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services. The Red Force's equipment system covers the resource scheduling model, data processing model, and task decision model in the central layer. It completes task decisions for system tasks, generates task plans, schedules and directs various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain. The space-based detection equipment model and the ground-based detection equipment model in the single-type equipment layer. The space-based detection equipment model includes satellite detection equipment of different orbits and systems. The ground-based detection equipment model includes radar detection equipment and telescope detection equipment. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building. The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

4. The digital parallel method for key algorithm verification according to claim 1, characterized in that: In step S2: The blue equipment system involves the command and control link simulation and mission equipment system of the blue equipment, including the command node model, command and control link model, intelligent decision model and mission behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level: The blue side equipment model system includes a complete mission system and a single equipment layer model; The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information. The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions. The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain; The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

5. The digital parallel method for key algorithm verification according to claim 1, characterized in that: In step S3: Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification: Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators; The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction. The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan; The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage. The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm. The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

6. A digital parallel system for key algorithm verification, characterized in that: include: Module M1: By adopting the service-oriented design idea of ​​algorithm resources, the algorithm verification platform is built, including an integrated simulation support deduction platform for the core functions of simulation operation, and the algorithm slot and shelf data system are constructed under the constraints of algorithm specifications; Module M2: Adopting the idea of ​​object-oriented, component-based and parameterized, we build a model system of different levels for the red, blue and white parties, and construct a trusted verification environment for the algorithm; Module M3: Complete the construction of the evaluation system, including multi-means evaluation data collection services, building a configurable evaluation system, evaluation implementation and algorithm verification, to achieve a closed loop of algorithm performance evaluation.

7. The digital parallel system for key algorithm verification according to claim 6, characterized in that: In the module M1: The simulation platform has functions including directing, controlling, judging and managing simulation scenarios, and providing common simulation support services and application support services for different systems. The algorithm verification platform is further developed based on the deduction platform. The algorithm verification platform has all the functions of the deduction platform. According to business needs, it proposes core algorithm interface specifications and process specifications, completes the design of algorithm interface slots based on the core algorithm specifications, and independently designs the algorithm interface.

8. The digital parallel system for key algorithm verification according to claim 6, characterized in that: In the module M2: In order to meet the current business needs and non-functional service and virtualization needs, the equipment model system implements the functions of related models based on the B / S architecture and divides the interaction logic from the business logic based on the MVC framework model design. In view of the wide-area construction needs of the space-ground integrated algorithm verification environment, the object-oriented concept is used to complete the establishment of models at all levels to ensure data validity and fully and correctly map the physical world. The Red Army's equipment system covers the resource scheduling center model, data processing center model, and command decision center model in the center layer, the space-based detection equipment model in the single-type equipment layer, the ground-based detection equipment model, the equipment platform model, and the main load model in the subsystem layer, and abstracts the corresponding algorithm services based on the model: The Red Team's equipment system covers three levels of model systems: the center layer, the single-type equipment layer, and the subsystem layer. It builds a model system based on the simple DEVS modeling theory, supports multi-resolution simulation of various types of equipment on different platforms, and summarizes the core algorithms of the center layer and the single-type equipment layer to condense the corresponding algorithm services. The Red Force's equipment system covers the resource scheduling center model, data processing center model, and command decision center model in the center layer. It completes task decisions for system tasks and based on command information, generates task plans, schedules and commands various detection equipment, and integrates and processes the returned detection data to simulate the entire task chain. The space-based detection model and ground-based detection model in the single-installation layer. The space-based detection model includes detection satellites of different orbits and systems. The ground-based detection equipment model includes radars and telescopes. Each detector model is composed of componentized and parameterized subsystem models to respond to the needs of systematic deduction and scenario building. The equipment platform model and main load model in the subsystem layer are based on the construction ideas of the model system. The platforms, loads and major subsystems at different levels are componentized and parameterized, and the composition and behavior of the equipment are modeled to assemble equipment models of different systems.

9. The digital parallel system for key algorithm verification according to claim 6, characterized in that: In the module M2: The blue equipment system involves the command and control link simulation and task system of the blue equipment, including the command node model, command and control link model, intelligent command decision model and task behavior model at the system level, as well as the equipment behavior model, radiation characteristic model, electromagnetic characteristic model and dynamic model at the single equipment level: The blue side equipment model system includes a complete mission system and a single equipment layer model; The system-level model in the blue equipment model system is oriented to the real-time confrontation between the red and blue forces. It has the ability to simulate the blue command nodes and command links, form intelligent command decisions, and generate behavioral instructions for the blue system based on the corresponding decision information. The single equipment layer in the blue side's equipment model system has dynamic model simulation, electromagnetic characteristics, and radiation characteristics simulation. According to the behavioral instructions of the system layer, it generates behavioral instructions for single-type equipment and executes them according to corresponding conditions. The white model system includes the space-time computing framework in the cognitive domain and the geophysical environment, electromagnetic environment, star orbit simulation and background targets in the physical domain; The space-time calculation framework in the Baifang model system provides time and coordinate calculation functions for other modules in the system that need to use astronomical time-related calculations, and calculates the physical position coordinates of each relevant element in the scene at a specific time under different time coordinates.

10. The digital parallel system for key algorithm verification according to claim 6, characterized in that: In the module M3: Evaluation and performance verification of key algorithms in tasks, including evaluation of data collection services, establishment of task-oriented evaluation systems, implementation of evaluation plans, and algorithm verification: Adopt high-speed cache and memory database technology to collect data in batches and in parallel, and support the adoption of basic evaluation operators; The evaluation data collection service builds the connection between the evaluation system and business data, simulation data and log information. It is based on the idea of ​​callable service and uses database-oriented data compliance, data probes and direct memory interaction. The steps for building and implementing the evaluation system include: focusing on the actual evaluation objectives, decomposing and sorting out the basic evaluation operators that need to be collected, clarifying the evaluation algorithm and the weight of the indicator system, and forming an evaluation plan; The evaluation process involves the game preparation stage, game implementation stage and evaluation closed-loop stage. The three stages can be converted into each other: the game preparation stage can enter the game implementation stage, the game implementation stage can enter the evaluation closed-loop stage, and the evaluation closed-loop stage can enter the game preparation stage and the game implementation stage. The deduction preparation stage in the evaluation can realize the closed loop of constellation optimization design. After the constellation design algorithm is executed, the generated constellation data is pushed to the scenario management to generate the initial situation. After the rapid deduction evaluation closed loop is completed according to this situation, the scenario management generates data demand for the constellation design, which assists the optimization closed loop of the constellation design algorithm. The simulation implementation stage in the evaluation process can realize the real-time design closed loop of the scheduling algorithm. After receiving the initial assumptions in the simulation preparation stage, the trusted environment loads the initial situation, performs simulation simulation according to the preset process, executes the load intelligent scheduling algorithm in two modes: human-in-the-loop and human-out-of-the-loop, generates scheduling decisions, and implements the corresponding scheduling algorithm strategy in the scenario through guidance and intervention. The implementation feedback and real-time evaluation results are fed back to the intelligent scheduling algorithm to complete the scheduling algorithm optimization closed loop.

Images