Multi-domain coupled numerical modeling and simulation system and method

Abstract

The present invention provides a high-performance multi-domain coupled numerical modeling and simulation system, including a data layer, a discrete layer, an application layer and a system layer; the data layer is used to store data required for data simulation calculations; the discrete layer is used for numerical simulation calculations to build physical models; the application layer is used to provide simulation applications based on numerical physical models; the system layer is used to provide a user interface for users to perform simulations. The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention can effectively solve the technical problems existing in the prior art, and provide a new type of simulation software architecture that meets the requirements of digital twins in terms of computing scale, efficiency, and user usage.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a high-performance multi-domain coupled numerical modeling and simulation system. Background Art

[0002] Numerical simulation is the main technical means in the field of industrial simulation. It corresponds to computer-aided design (CAD) and computer-aided manufacturing (CAM), and is also called computer-aided engineering (CAE). Due to historical reasons, the current mainstream commercial CAE software is mainly single-discipline general analysis software, such as Fluent for fluid analysis and Abaqus for structural analysis. Through some collaborative simulation platforms (AnsysWorkbench, etc.) and tools, multi-software coupling can be achieved to solve the problem of multi-component and multi-discipline coupling simulation calculation in engineering. However, this coupling calculation has many defects. First, the software of each discipline is implemented independently, and it is impossible to flexibly control the coupling solution process of multiple solution domains, and it is difficult to perform deep coupling, such as strong coupling solution between multiple calculation domains (forming a unified algebraic equation solution) and boundary correction of intermediate processes (such as correction of gradient field on the boundary of the solution domain), which affects the calculation accuracy and stability. Secondly, it is impossible to load balance and optimize multi-program multi-data stream (MPMD) processes such as multi-domain and multi-discipline coupling, and computing resources cannot be optimally configured, making it difficult to achieve efficient parallel computing.

[0003] In 2012, NASA gave a conceptual description of digital twins: Digital twins refer to the integration of multidisciplinary and multi-scale simulation processes by making full use of data such as physical models, sensors, and operation history. As a mirror image of physical products in virtual space, it reflects the entire life cycle of the corresponding physical products. Digital twins are deeply developed on the basis of MBD. In the process of implementing model-based system engineering (MBSE), enterprises have generated a large number of physical and mathematical models, which have laid the foundation for the development of digital twins. With the deepening of MBSE in enterprises and the gradual implementation of digital twins, a collaborative modeling and simulation platform from a system perspective has become necessary, and performance and accuracy are important indicators for measuring the maturity of modeling and simulation platforms and digital twins. Accuracy is the core meaning of twins. The digital twin model tends to be consistent with physical reality, so as to achieve the purpose of prediction and optimization; computing performance that exceeds physical time is a necessary guarantee for the twin model to achieve the purpose of prediction and optimization. Meeting high indicators in terms of accuracy and performance at the same time will bring about a significant increase in computing demand. Under the existing industrial simulation software architecture system, it is difficult to meet the requirements of digital twins in terms of computing scale, efficiency, and user usage. Summary of the invention

[0004] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention solves the deficiencies in the prior art and meets the requirements of digital twins in terms of computing scale, computing efficiency and user usage.

[0005] The technical solution provided by the present invention is:

[0006] A high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention comprises a data layer, a discrete layer, an application layer and a system layer;

[0007] The data layer is used to store data required in data simulation calculation;

[0008] The discrete layer is used for numerical simulation calculation and construction of physical model;

[0009] The application layer is used to provide simulation applications built based on physical models;

[0010] The system layer is used to provide a user interface for users to perform simulation.

[0011] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention comprises: the data layer comprises a data container and an iterator module for storing discrete data; the discrete layer comprises a discrete method module and a physical model module; the discrete method module comprises a discrete operator module; the application layer comprises a simulation application and a simulation operating environment; the system layer comprises a user interface;

[0012] Among them, the discrete operator module implements global traversal calculation on the data in the data container through the iterator module; the physical model module constructs a physical model through the discrete operator module in the discrete method module; the simulation application module constructs a simulation application based on the physical model constructed by the physical model module; the simulation running environment is used as a support environment for the running of simulation applications.

[0013] In the high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention, preferably, the data container is a distributed data container; the distributed data container includes a grid module storing a set of geometric elements for discretizing space, a field module storing a sequence of physical quantities defined on a grid element, a computational domain of a spatial region determined by the geometry of a numerical simulation object, and an aggregate computational domain; the aggregate computational domain includes a computational domain set formed by coupling computational domain data stored in a plurality of the computational domain modules;

[0014] The iterator module is used to traverse the data stored in the grid module and the field module.

[0015] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention preferably comprises the discrete method module further comprising a formulaic programming interface, a discrete equation module and a discrete expression module; the formulaic programming interface is used to convert expressions and equations in the form of discrete mathematical formulas into codes written in a high-level language, so as to realize the compilation and operation of the discrete equation module and the discrete expression module.

[0016] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention preferably comprises an engineering modeling module in the application layer;

[0017] The engineering modeling module performs engineering modeling using the engineering modeling data fed back by the user interface module.

[0018] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention preferably comprises a physical model development module and an application customization development module at the system layer; the customization development module provides a simulation application customization development function based on the physical model module; the simulation application developed based on the customization development module is run on the simulation running environment; the physical model development module provides a physical model development function based on the discrete method module; the physical model developed based on the physical model development module is run on the simulation running environment

[0019] The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention preferably further includes a system function module; the system function module includes a parameter module, a function registration module, a multi-domain load balancing module and a multi-level process control module;

[0020] Among them, the parameter module is used to store all system, model and process configuration parameters other than grid and field data; the function registration module loads discrete operators and physical models into the registration table in an extensible manner according to classification, so that they can be selected according to the parameter module at runtime; the multi-domain load balancing obtains parameter configuration through the parameter module, simulates process scripts, and dynamically allocates computing resources; the multi-level process control module manages and triggers the behavior sequence of the computing domain and the aggregate computing domain through the configuration parameters in the parameter module.

[0021] A high-performance multi-domain coupling numerical modeling and simulation method provided by the present invention preferably comprises a data layer, a discrete layer, an application layer, a system layer, a system function module and a program; the data layer comprises a data container; the data container comprises a computing domain and an aggregation computing domain; the application layer comprises a simulation running environment; the system function module comprises a parameter module, a function registration module, a multi-domain load balancing module and a multi-level process control module;

[0022] Among them, the program includes:

[0023] The simulation running environment is started and initialized;

[0024] The parameter module reads the simulation database, including system parameters, model parameters, simulation process scripts, etc., and initializes the global parameter object;

[0025] The multi-domain load balancing module creates resource configurations based on model parameters and simulation process scripts;

[0026] The simulation running environment reads the engineering model geometry and physical field data according to the model parameters and resource configuration, and creates the calculation domain and the aggregate calculation domain;

[0027] The multi-level process control module creates and configures the physical model of the computational domain and the aggregated computational domain according to the parameters;

[0028] The multi-level process module starts executing the simulation process according to the simulation process script;

[0029] During the simulation process, each physical model is executed serially or concurrently in a certain order; the multi-domain load balancing module performs load balancing and data and task remapping when necessary; and receives interactive commands or parameters when necessary;

[0030] The multi-level process module monitors that the end conditions are met and ends the simulation process;

[0031] The system simulation running environment releases system resources and ends the operation.

[0032] The present invention has the following advantages:

[0033] The present invention provides a high-performance multi-domain coupled numerical modeling and simulation system, including a data layer, a discrete layer, an application layer and a system layer; the data layer is used to store data required for data simulation calculations; the discrete layer is used for numerical simulation calculations to build physical models; the application layer is used to provide simulation applications based on physical models; the system layer is used to provide a user interface for users to perform simulations. The high-performance multi-domain coupled numerical modeling and simulation system provided by the present invention can effectively solve the technical problems existing in the prior art, and provide a new type of simulation software architecture that meets the requirements of digital twins in terms of computing scale, efficiency, and user usage. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention and its features, configurations and advantages will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings. The same reference numerals indicate the same parts throughout the drawings. The drawings are not drawn to scale, but rather to illustrate the subject matter of the present invention.

[0035] Figure 1 It is an architectural design diagram of a high-performance multi-domain coupled numerical modeling and simulation system provided in Example 1 of the present invention;

[0036] Figure 2 is a functional module relationship diagram of a high-performance multi-domain coupled numerical modeling and simulation system provided by Example 1 of the present invention;

[0037] Figure 3 It is a flow chart of the high-performance multi-domain coupling numerical modeling and simulation method provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0038] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. It should be noted that the terms used in the present invention are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application.

[0039] Embodiment 1:

[0040] The architecture of the high-performance multi-domain coupled numerical modeling and simulation system provided in Example 1 of the present invention is as follows: Figure 1 As shown, the overall structure is divided into four layers, from low to high: data layer, discrete layer, application layer and system layer.

[0041] The data layer mainly includes distributed data containers, which store the geometry, mesh, physical field, and computational domain data required for numerical simulation calculations in distributed memory or virtual memory. Physical field and mesh data provide iterators for data traversal. Through iterators, users do not need to care about the structure of distributed data and the details of parallel algorithms. They only need to define the operations that users care about. This achieves the purpose of serial programming and parallel implementation.

[0042] The discrete layer mainly includes numerical computing elements built on the data layer, including various discrete operators, as well as discrete expressions and discrete equations implemented based on the formulaic programming interface. Through numerical computing elements, physical models including solution algorithms, boundary conditions and material models can be constructed.

[0043] The application layer mainly includes simulation applications and engineering models built based on discrete layer physical models, as well as the system simulation operating environment of the master control system functions. With the assistance of CAD and CAE pre- and post-processing tools, combined with discrete layer physical models, simulation applications for specific scenarios can be built; combined with CAD and geometric data such as meshes, using discrete layer physical models, specific physical products or components can be described and digital engineering models can be established; the system simulation operating environment achieves system-level simulation by loading one or more engineering models with coupling relationships.

[0044] The system layer uses the underlying components to provide friendly interfaces and management functions for different developers and users, including an integrated development environment that facilitates physical model development and application customization, as well as a user interface that facilitates user application subscription, engineering modeling, model reference and system simulation.

[0045] like Figure 2 As shown, the high-performance multi-domain coupled numerical modeling and simulation system provided by Embodiment 1 of the present invention includes the following modules:

[0046] Grid module: stores a collection of geometric elements that discretize space, including points, lines, surfaces, volumes and their connection relationships. The grid module supports, but is not limited to, structured grids, unstructured grids, Cartesian grids, adaptive grids and spatial multitrees. Grid data can be stored in a distributed manner and provides parallel data indexing and operation interfaces.

[0047] Field module: stores a sequence of physical quantities defined on a grid element, which is used to characterize the physical state of the computational domain corresponding to the grid. Field data can be stored in a distributed manner, and provides parallel data indexing and operation interfaces.

[0048] Computational domain: A spatial region determined by the geometry of a numerical simulation object. The computational domain determines the range and boundaries through the grid, and the physical properties and states through the field. The computational domain also stores the necessary parameters of the numerical simulation object and provides parallelized data indexing and operation interfaces.

[0049] Aggregate computational domain: An aggregate computational domain is an organic collection of multiple computational domains. The computational domains are coupled through boundaries and overlaps, providing a data basis for realizing code-level multi-domain coupling.

[0050] Iterator module: It is mainly used to traverse the grid and field data and apply defined operations. Iterators shield the distributed data structure and parallel details, and users implement global operations on distributed grid and field data in a serial programming manner.

[0051] Discrete operator module: Discrete partial differential operators or other operators used to calculate terms with certain independence and generality in the control equations. Operators include but are not limited to gradient, divergence, curl, Laplace, and various weak forms of linear and bilinear integrals. Discrete methods include but are not limited to finite element (FVM), finite volume (FVM), and finite difference (FDM).

[0052] Formula programming interface: Automatically convert expressions and equations in the form of discrete mathematical formulas into codes written in general high-level languages ​​so that they can be compiled and run. This function is implemented based on source-to-source interpreters for different discrete domain programming languages, thereby forming DSLs (domain programming languages) for different discrete methods. Supported mathematical discrete expressions include but are not limited to partial differential conservation forms, partial differential weak forms, etc. DSL is implemented in a high-level language embedded manner, that is, the main program is written in a high-level language, and DSL statements such as discrete expressions and discrete equations are embedded in the main high-level language program.

[0053] Discrete Expression Module: Discrete expressions written using a formulaic programming interface. Discrete methods include but are not limited to finite element (FVM), finite volume (FVM), and finite difference (FDM).

[0054] Discrete Equation Module: Discrete form equations written using a formal programming interface. Discrete methods include but are not limited to finite element (FVM), finite volume (FVM) and finite difference (FDM).

[0055] Physical model module: physical quantity (field) solution update process and boundary correction process realized by numerical calculation means such as discrete expression calculation and discrete equation solution.

[0056] Simulation application development module: With the assistance of external CAD and CAE pre- and post-processing tools, combined with discrete layer physical models, simulation tools for specific scenarios are built, including single-discipline simulation tools, multi-discipline simulation tools, and customized simulation tools for specific industrial fields.

[0057] Engineering modeling module: Combine CAD and mesh and other geometric data, and use discrete layer physical models to describe specific physical products or parts, thereby establishing a digital model with complete geometric and physical properties. Engineering models can be created, referenced, and coupled for simulation under this system. Engineering models are reflected in memory as computational domains and aggregated computational domains, and in external storage as configuration parameters, geometry, and physical field data files.

[0058] System simulation operating environment: A supporting environment based on system function-related components that can complete the loading of functional components, reading of model data and simulation process scripts, and execution of simulation processes. The manifestation includes but is not limited to an executable program that accepts parameters, a set of scripts that accept parameters, or a console that accepts command interactions.

[0059] Parameter module: A collection of hierarchically organized key-value pairs used to store all system, model, and process configuration parameters other than grid and field data.

[0060] Function registration module: automatically loads defined discrete operators, physical models and other functions into the registry in an extensible manner according to certain classifications, so that they can be selected through parameters and interactions at runtime.

[0061] Multi-domain load balancing module: By reading parameters, it obtains the geometry, physical model and other configuration parameters of multiple simulated engineering models, as well as the simulation process script. By analyzing the multi-domain geometry, physical model and simulation process script, it dynamically allocates computing resources to maximize resource utilization at all runtimes.

[0062] Multi-level process control module: configure recursively nested physical models by organizing parameters by level. Manage and trigger the behavior sequence (physical model execution sequence) of simulation objects (computational domains and aggregated computational domains) according to the configuration parameters in the simulation process script.

[0063] Third party: The third-party libraries that the framework implementation depends on, including but not limited to mathematical libraries, high-performance middleware, graph libraries, I / O libraries, etc.

[0064] Integrated Development Environment module: a programming, compiling and debugging environment that can load necessary development dependencies (including compilers, dependency libraries and debugging tools, etc.), while supporting the development of physical models and customized application development.

[0065] User interface: A set of user interfaces that can be used to subscribe to and use simulation applications, create engineering models, reference engineering models and perform system-level simulations.

[0066] In this embodiment, the data layer includes a data container and an iterator module for storing discrete data; the discrete layer includes a discrete method module and a physical model module; the discrete method module includes a discrete operator module; the application layer includes a simulation application and a simulation operating environment; the system layer includes a user interface;

[0067] Among them, the discrete operator module realizes global traversal calculation of the data in the data container through the iterator module; the physical model module constructs the physical model through the discrete operator module in the discrete method module; the simulation application module constructs the simulation application based on the physical model constructed by the physical model module; the simulation running environment is used as a support environment for the operation of the simulation application.

[0068] In this embodiment, the data container is a distributed data container; the distributed data container includes a grid module storing a set of geometric elements for discretizing space, a field module storing a sequence of physical quantities defined on a grid element, a computational domain of a spatial region determined by the geometry of a numerical simulation object, and an aggregate computational domain; the aggregate computational domain includes a computational domain set formed by coupling computational domain data stored in a number of the computational domain modules;

[0069] The iterator module is used to traverse the data stored in the grid module and the field module.

[0070] In this embodiment, the discrete method module also includes a formulaic programming interface, a discrete equation module and a discrete expression module; the formulaic programming interface is used to convert expressions and equations in the form of discrete mathematical formulas into codes written in a high-level language to realize the compilation and operation of the discrete equation module and the discrete expression module.

[0071] In this embodiment, the application layer also includes an engineering modeling module;

[0072] The engineering modeling module performs engineering modeling by using the engineering modeling data fed back by the user interface module.

[0073] In this embodiment, the system layer includes a physical model development module and an application customization development module; the customization development module provides simulation application customization development functions based on the physical model module; the simulation application developed based on the application customization development module runs on the simulation running environment; the physical model development module provides physical model development functions based on the discrete method module; the physical model developed based on the physical model development module runs on the simulation running environment.

[0074] In this embodiment, a system function module is also included; the system function module includes a parameter module, a function registration module, a multi-domain load balancing module and a multi-level process control module;

[0075] Among them, the parameter module is used to store all system, model and process configuration parameters other than grid and field data; the function registration module loads discrete operators and physical models into the registration table in an extensible manner according to classification for selection according to the parameter module at runtime; multi-domain load balancing obtains parameter configuration and simulation process scripts through the parameter module to dynamically allocate computing resources; the multi-level process control module manages and triggers the behavior sequence of the computing domain and the aggregate computing domain through the configuration parameters in the parameter module.

[0076] like Figure 3 As shown, when the high-performance multi-domain coupled numerical modeling and simulation system provided by this embodiment is running, Figure 3 As shown,

[0077] S101: The simulation running environment is started and initialized;

[0078] S102: The parameter module reads the simulation database, including system parameters, model parameters, simulation process scripts, etc., and initializes the global parameter object;

[0079] S103: The multi-domain load balancing module creates resource configuration according to the model parameters and the simulation process script;

[0080] S104: The simulation running environment reads the engineering model geometry and physical field data according to the model parameters and resource configuration, and creates a calculation domain and an aggregate calculation domain;

[0081] S105: The multi-level process control module creates and configures the physical model of the computational domain and the aggregate computational domain according to the parameters;

[0082] S106: The multi-level process module starts executing the simulation process according to the simulation process script;

[0083] S107: During the execution of the simulation process, each physical model is executed serially or concurrently in a certain order; the multi-domain load balancing module performs load balancing and data and task remapping when necessary; and receives interactive commands or parameters when necessary;

[0084] S108: The multi-level process module monitors that the end condition is met and ends the simulation process;

[0085] S109: The system simulation running environment releases system resources and ends the running.

[0086] In Example 1 of the present invention, distributed containers are set up to encapsulate data in a distributed parallel mode, including grids, fields, computing domains, aggregate computing domains, etc., and shield data structures; user operation definitions are received through an iterator module, and related elements of the distributed data container are traversed to achieve automatic parallel computing; numerical computing development is separated from parallel computing details to reduce development difficulty.

[0087] In this implementation 1, by setting up a formulaic programming interface, it is stipulated that a high-level language is used as the main body, and embedded formulaic programming statements are used to describe discrete expressions and discrete equations; at the same time, it is stipulated that the code is interpreted and generated by a source-to-source interpreter, and the loops inside the expression or equation are decomposed, fused and optimized during the processing, so that the numerical calculation process and the physical model definition are close to the mathematical formula, reducing the development difficulty and improving the calculation efficiency.

[0088] In this embodiment 1, the engineering model is embodied as an object of a computing domain or an aggregate computing domain at runtime, and the simulation process script clarifies the order of each object's behavior (execution of the physical model). By obtaining the object data scale and the complexity of the behavior, the behavior calculation weight can be obtained; the behavior dependency can be clarified through the script sequence. With the behavior weight and dependency as input, the multi-domain load balancing function module can formulate a resource allocation schedule according to a certain mechanism, that is, the computing resources that need to be matched when each object behavior occurs, thereby achieving dynamic load balancing. The main purpose of this method is to achieve load balancing in the MPMD calculation process of multi-domain coupling simulation.

[0089] The multi-level precise process control module in this embodiment 1 performs process encapsulation on the physical model implementation, and the process can be nested and configured through multi-level parameters. The multi-level precise process configuration is reflected in: within the physical model, the physical model process can be configured with multi-level nesting through tree-organized parameters; according to the configuration parameters in the simulation process script, the behavior order (physical model execution order) of the simulation object (computational domain and aggregated calculation domain) can be configured. The main purpose of this method is to facilitate the precise configuration of the simulation process.

[0090] The high-performance multi-domain coupled numerical modeling and simulation system provided in Embodiment 1 of the present invention can effectively solve the technical problems existing in the prior art and can achieve:

[0091] (1) Strong coupling calculation at the code level in multiple computing domains to improve the calculation accuracy and stability of coupled simulation;

[0092] (2) Dynamic load balancing of multi-domain coupled multi-program multi-data (MPMD) computing processes to improve the scale and efficiency of parallel computing;

[0093] (3) Formulated numerical computing programming interface to automatically and efficiently implement physical model development and parallel computing;

[0094] (4) Parallel details are hidden, enabling performance portability across hardware platforms;

[0095] (5) Object-oriented system modeling and simulation interface, providing flexible and rich component modeling configuration, as well as graphical and scripted multi-component coupling behavior modeling;

[0096] (6) A multi-level open physical and application customization environment to meet the needs of developers at different levels;

[0097] (7) Multiple user interfaces such as application subscription, engineering modeling, and system simulation to meet the needs of users at different levels.

[0098] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structural transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A multi-domain coupled numerical modeling and simulation system, It is characterized in that Includes data layer, discrete layer, application layer and system layer; The data layer is used to store data required in data simulation calculation; The discrete layer is used for numerical simulation calculation and construction of physical model; The application layer is used to provide simulation applications built based on physical models; The system layer is used to provide a user interface for users to perform simulation; The data layer includes a data container and an iterator module for storing discrete data; the discrete layer includes a discrete method module and a physical model module; the discrete method module includes a discrete operator module; the application layer includes a simulation application module and a simulation operating environment; the system layer includes a user interface; It also includes a system function module; the system function module includes a parameter module, a function registration module, a multi-domain load balancing module and a multi-level process control module; Among them, the parameter module is used to store all system, model and process configuration parameters other than grid and field data; the function registration module loads discrete operators and physical models into the registration table in an extensible manner according to classification, so that they can be selected according to the parameter module at runtime; the multi-domain load balancing module obtains parameter configuration and simulation process scripts through the parameter module to dynamically allocate computing resources; the multi-level process control module manages and triggers the behavior sequence of computing domains and aggregate computing domains through the configuration parameters in the parameter module.

2. The multi-domain coupled numerical modeling and simulation system according to claim 1, It is characterized in that in, The discrete operator module implements global traversal calculation on the data in the data container through the iterator module; the physical model module constructs a physical model through the discrete operator module in the discrete method module; the simulation application module constructs a simulation application based on the physical model constructed by the physical model module; the simulation running environment is used as a support environment for the running of the simulation application.

3. The multi-domain coupled numerical modeling and simulation system according to claim 1, It is characterized in that The data container is a distributed data container; the distributed data container includes a grid module that stores a set of geometric elements that discretize space, a field module that stores a sequence of physical quantities defined on a grid element, a computational domain of a spatial region determined by the geometry of a numerical simulation object, and an aggregate computational domain; the aggregate computational domain includes a computational domain set formed by coupling computational domain data stored in several of the computational domains; the iterator module is used to traverse the data stored in the grid module and the field module.

4. The multi-domain coupled numerical modeling and simulation system according to claim 2, It is characterized in that The discrete method module also includes a formulaic programming interface, a discrete equation module and a discrete expression module; the formulaic programming interface is used to convert expressions and equations in the form of discrete mathematical formulas into codes written in a high-level language to realize the compilation and operation of the discrete equation module and the discrete expression module.

5. The multi-domain coupled numerical modeling and simulation system according to claim 2, It is characterized in that The application layer also includes an engineering modeling module; The engineering modeling module performs engineering modeling using the engineering modeling data fed back by the user interface module.

6. The multi-domain coupled numerical modeling and simulation system according to claim 2, It is characterized in that The system layer includes a physical model development module and an application customization development module; The customized development module provides a simulation application customized development function based on the physical model module; The simulation application developed based on the custom development module runs on the simulation running environment; the physical model development module provides a physical model development function based on the discrete method module; and the physical model developed based on the physical model development module runs on the simulation running environment.

7. A method for performing multi-domain coupled numerical modeling and simulation using the multi-domain coupled numerical modeling and simulation system as claimed in claim 1, It is characterized in that It includes a data layer, a discrete layer, an application layer, a system layer, a system function module and a program; the data layer includes a data container; the data container includes a computing domain and an aggregation computing domain; the application layer includes a simulation running environment; the system function module includes a parameter module, a function registration module, a multi-domain load balancing module and a multi-level process control module; Among them, the program includes: The simulation running environment is started and initialized; The parameter module reads the simulation database, including system parameters, model parameters, configuration parameters, simulation process scripts, and initializes the global parameter objects; The multi-domain load balancing module creates resource configurations based on model parameters and simulation process scripts; The simulation running environment reads the engineering model geometry and physical field data according to the model parameters and resource configuration, and creates the calculation domain and the aggregate calculation domain; The multi-level process control module creates and configures the physical model of the computational domain and the aggregate computational domain according to the configuration parameters; The multi-level process module starts executing the simulation process according to the simulation process script; During the simulation process, each physical model is executed serially or concurrently in a certain order; the multi-domain load balancing module performs load balancing and data and task remapping; and receives interactive commands or parameters; The multi-level process control module monitors the satisfaction of the end condition and ends the simulation process; The system simulation running environment releases system resources and ends the operation.

Images