A solution process control method and related equipment

Abstract

This application discloses a solution process control method. In this method, a cloud management platform can determine the first physics discipline type of the first physical equation, and then intelligently determine one or more of the first auxiliary equation, first coupling strategy, and first numerical solution strategy corresponding to the first physical equation based on the first physics discipline type. This allows for the configuration of the mathematical equation set and / or the solution method for each first solution task in the current physics field service, thereby solving the first physical equation. As can be seen, in this solution, the cloud management platform can intelligently determine one or more of the first auxiliary equation, first coupling strategy, and first numerical solution strategy corresponding to the first physical equation, thus conveniently and efficiently determining the mathematical equation set and / or the solution method for each first solution task for various physical scenarios, and solving the first physical equation with a suitable solution process.

Description

Technical Field

[0001] This application relates to the field of simulation solution technology, specifically to a solution process control method and related equipment. Background Technology

[0002] Computer-aided engineering (CAE) refers to the use of computers to help solve and analyze physical problems in complex engineering and products, thereby improving product design or assisting in solving engineering problems in various industries.

[0003] When using CAE simulation to solve physical problems, traditional CAE simulation software typically pre-defines a fixed CAE solution process for each specific pre-defined CAE simulation scenario. Taking fluid mechanics CAE simulation scenarios as an example, the simulation scenario can be divided into multiple sub-simulation scenarios, and a specific CAE solution process can be pre-defined for each sub-simulation scenario. This specific CAE simulation process includes fixed basic physical equations, auxiliary equations, coupling strategies, and numerical calculation methods.

[0004] It is evident that traditional methods require pre-customizing numerous solution processes for the physical problem to be solved. Each customized solution process is only applicable to a specific physical scenario, with a very narrow scope of application. Furthermore, it is difficult for users to modify and adjust these processes, which may result in users being unable to obtain a solution process that meets simulation requirements, thus affecting the solution results. Summary of the Invention

[0005] This application provides a solution flow control method that can conveniently and efficiently determine a suitable solution flow for a task in various physical scenarios. This application also provides corresponding apparatus, devices, computer-readable storage media, and computer program products.

[0006] The first aspect of this application provides a solution process control method, which is executed by a cloud management platform. The cloud management platform is used to manage the infrastructure providing cloud services. The infrastructure includes multiple servers, and the servers are used to deploy virtual instances that implement the solution process. The method includes: the cloud management platform obtaining a first physical equation for a physical field business from an equation input interface; the cloud management platform determining a first physics discipline type corresponding to the first physical equation based on the first physical equation; and the cloud management platform determining first configuration information corresponding to the first physical equation based on the first physics discipline type. The first configuration information includes one or more of the following information corresponding to the first physical equation: a first auxiliary equation, a first coupling strategy, and a first numerical calculation strategy. The first configuration information is used to determine the set of mathematical equations contained in the first solution task and / or the solution method of each first solution task. The first solution task is used to solve the first physical equation.

[0007] In the first aspect, the cloud management platform can determine the first physics discipline type of the first physical equation, such as the fundamental physical equation. Based on this first physics discipline type, it intelligently determines one or more of the first auxiliary equation, first coupling strategy, and first numerical solution strategy corresponding to the first physical equation. This obtains the first configuration information for the first solution process of the current physics field service, thereby configuring the set of mathematical equations and / or the solution method for each first solution task in the current physics field service to solve the first physical equation. Therefore, in the first aspect, the cloud management platform can intelligently determine one or more of the first auxiliary equation, first coupling strategy, and first numerical solution strategy corresponding to the first physical equation, thus conveniently and efficiently determining the set of mathematical equations and / or the solution method for each first solution task for various physical scenarios, and solving the first physical equation with a suitable solution process.

[0008] In one possible implementation of the first aspect, the physics field service involves a first time step and a second time step, the second time step being prior to the first time step, and the first physical equation being the physical equation corresponding to the physics field service in the first time step. The method further includes: a cloud management platform acquiring second configuration information corresponding to the second physical equation of the physics field service in the second time step, and / or second solution information corresponding to the second physical equation. The second configuration information includes one or more of the following information corresponding to the second physical equation: a second auxiliary equation, a second coupling strategy, and a second numerical calculation strategy. The second solution information includes one or more of the following information: the solution result of the second solution task and the solution performance of the second solution task. The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: the cloud management platform determining the first configuration information based on the first physics discipline type, the second configuration information, and / or the second solution information.

[0009] In this possible implementation, the second time step can be one or more time steps preceding the first time step. In this implementation, during the CAE simulation of the physical field service, the solution process of the current time step can be dynamically adjusted using the configuration information and / or solution information of the solution process of the historical time step (i.e., the second time step). This optimizes the CAE simulation performance, ensures high accuracy, efficiency, and stability of the CAE simulation at the current time step, shortens the end-to-end CAE simulation time, accelerates the CAE simulation, and directly reduces the resource cost of the CAE simulation.

[0010] In one possible implementation of the first aspect, the first configuration information includes a first set of physical equations determined based on the first physical equation and the first auxiliary equation, and the second solution information includes the solution results of the second solution task; the cloud management platform determines the first configuration information based on the first physics discipline type, and the second configuration information and / or the second solution information, including: the cloud management platform determines the first auxiliary equation based on the physical state reflected by the solution results of the second solution task and the first physics discipline type; the cloud management platform determines the first set of physical equations based on the first physical equation and the first auxiliary equation.

[0011] In this possible implementation, the physical state reflected by the solution result of the second solution task can be described based on the value information of one or more physical quantities involved in the solution result. For example, the physical state reflected by the solution result of the second solution task may include the values ​​of one or more physical quantities; or, the corresponding physical state can be determined based on the range of values ​​of one or more physical quantities in the solution result of the second solution task. For instance, if the solution result of the second solution task includes the values ​​of one or more fluid physical quantities, the physical state of the flow field, such as the range of values ​​of the one or more fluid physical quantities, can be used to determine whether the flow field is turbulent.

[0012] In some examples, the first auxiliary equation corresponding to the first physical equation can be determined by using a machine learning model corresponding to the first physics subject type and / or a preset database corresponding to the first physics subject type, based on the physical state reflected by the solution results of the second solution task and the first physics subject type.

[0013] In one possible implementation of the first aspect, the physics field service is a multiphysics field service, the number of first physical equations is multiple, the number of second physical equations is multiple, the first configuration information includes a first coupling strategy corresponding to each of the multiple first physical equations, the second configuration information includes a second coupling strategy corresponding to each of the multiple second physical equations, and the solution performance in the second solution information includes coupling performance information corresponding to the second solution task using the second coupling strategy; the method further includes: a cloud management platform obtaining multiple sets of first physical equations based on multiple first physical equations; the cloud management platform determines the first configuration information according to the first physics discipline type, and the second configuration information and / or the second solution information, including: if there is a set of second physical equations in the multiple sets of second physical equations that is the same as one or more sets of first physical equations, then the cloud management platform determines the first coupling strategy corresponding to one or more sets of first physical equations according to the second coupling strategy and coupling performance information corresponding to the same set of second physical equations; wherein, the first coupling strategy is used to indicate the set of mathematical equations included in the first solution task and the solution order of the first solution task, and the set of mathematical equations included in the first solution task is obtained by discretizing one or more sets of first physical equations based on the first coupling strategy.

[0014] One possible implementation provides a method for determining the first coupling strategy when the same historical set of physical equations exists. Specifically, after obtaining the coupling performance information of the second coupling strategy for the same second set of physical equations, if the coupling performance information of the second coupling strategy for the same second set of physical equations meets preset performance requirements, one or more first coupling strategies corresponding to the first set of physical equations can be determined based on the specific content of the second coupling strategy for the same second set of physical equations.

[0015] For example, coupling performance parameter thresholds (such as coupling accuracy thresholds and / or coupling speed thresholds) can be preset. If the value of the coupling parameter in the coupling performance information of the second coupling strategy of the same second physical equation set is greater than the corresponding coupling performance parameter threshold, then the first coupling strategy corresponding to one or more first physical equation sets can be determined according to the second coupling strategy of the same second physical equation set.

[0016] In one possible implementation of the first aspect, the physics field service is a multiphysics field service, the number of first physical equations is multiple, the number of second physical equations is multiple, the first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations, and the second solution process configuration information includes the solution result of the second solution task; the method further includes: a cloud management platform acquiring multiple sets of first physical equations based on multiple first physical equations; the cloud management platform determining the first configuration information according to the first physics discipline type, and the second configuration information and / or the second solution information, including: if there is no second physical equation set in the multiple sets of second physical equations that is the same as one or more sets of first physical equations, then the cloud management platform determines the first coupling strategy corresponding to one or more sets of first physical equations according to the physical state reflected by the solution result of the second solution task; wherein, the first coupling strategy is used to indicate the mathematical equation set included in the first solution task and the solution order of the first solution task, and the mathematical equation set included in the first solution task is obtained by discretizing one or more sets of first physical equations based on the first coupling strategy.

[0017] In this possible implementation, in the absence of the same historical set of physical equations, a first coupling strategy corresponding to one or more sets of first physical equations can be determined based on the physical state reflected by the solution results of the second solution task, using a machine learning model corresponding to the first physics discipline type and / or a preset database corresponding to the first physics discipline type.

[0018] In one possible implementation of the first aspect, the first configuration information includes a first numerical calculation strategy corresponding to the first physical equation, and the second solution information includes the solution performance of the second solution task; the method further includes: a cloud management platform acquiring a first solution task, the first solution task including a set of mathematical equations obtained based on the first physical equation; the cloud management platform determining the first configuration information according to the first physics discipline type, and the second configuration information and / or the second solution information, including: if the second solution task matches the first solution task, the cloud management platform determining the first numerical calculation strategy of the set of mathematical equations contained in the first solution task according to the solution performance of the second solution task.

[0019] In this possible implementation, if any first solving task has a matching second solving task in the historical solving tasks (i.e., one or more second solving tasks), after obtaining the solving performance information of the second numerical calculation strategy of the matching second solving task, the first numerical calculation strategy corresponding to the first solving task can be determined according to the specific content of the second numerical calculation strategy of the matching second solving task, provided that the solving performance information of the second numerical calculation strategy of the matching second solving task meets the preset performance requirements.

[0020] For example, a threshold for solution performance parameters (such as a threshold for numerical computation accuracy and / or a threshold for numerical computation speed) can be preset. If the value of the numerical computation performance parameter in the solution performance information of the second numerical computation strategy of the matching second solution task is greater than the corresponding threshold for numerical computation performance parameters, then the first numerical computation strategy corresponding to the first solution task can be determined according to the second numerical computation strategy of the matching second solution task.

[0021] In one possible implementation of the first aspect, the physical field service involves a first time step, and the first physical equation is the physical equation corresponding to the physical field service in the first time step; the cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: the cloud management platform determines the first configuration information based on the first physics discipline type and the initial physical state corresponding to the first time step.

[0022] In one possible implementation, the initial physical state can be either a default physical state or a user-configured physical state, which can be described by the values ​​of variables. For example, when the physics field service involves fluids, the initial physical state indicates that the fluid velocity is 0, or a certain velocity preset by the user. The first time step involved in the physics field service can be the initial time step in the CAE simulation solution process. In this case, the first configuration information can be determined based on the first physics subject type and the initial physical state corresponding to the first time step, using a machine learning model corresponding to the first physics subject type and / or a preset database corresponding to the first physics subject type.

[0023] In one possible implementation of the first aspect, the cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: the cloud management platform determines the first configuration information through the machine learning model corresponding to the first physics discipline type and / or the preset database corresponding to the first physics discipline type, wherein the preset database corresponding to the first physics discipline type includes one or more of the following information corresponding to the first physics discipline type: preset auxiliary equation, preset coupling strategy, preset numerical calculation strategy.

[0024] In this possible implementation, the content that the first configuration information needs to include can be determined according to the user's process control preferences (specifically, it may include one or more of the first auxiliary equation, the first coupling strategy, and the first numerical calculation strategy). Thus, the specific content of the first configuration information can be determined according to the user's process control preferences through the machine learning model corresponding to the first physics discipline type and / or the preset database corresponding to the first physics discipline type.

[0025] A second aspect of this application provides a cloud management platform for solving process control, which has the functionality to implement the method described in the first aspect or any possible implementation of the first aspect. This functionality can be implemented in hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described functionality, such as an interface module and a processing module.

[0026] A third aspect of this application provides a computing device cluster including at least one computing device, the at least one computing device including a processor and a memory, the memory of the at least one computing device storing computer-executable instructions that can run on the processor, and when the computer-executable instructions are executed by the processor, the processor executes a method as described in the first aspect or any possible implementation of the first aspect.

[0027] The fourth aspect of this application provides a computer-readable storage medium storing one or more computer-executable instructions, wherein when the computer-executable instructions are executed by a processor, the processor performs a method as described in the first aspect or any possible implementation thereof.

[0028] The fifth aspect of this application provides a computer program product that stores one or more computer-executable instructions, wherein when the computer-executable instructions are executed by a processor, the processor executes a method as described in the first aspect or any possible implementation thereof.

[0029] A sixth aspect of this application provides a chip system including a processor for supporting the processor in implementing the functions involved in the first aspect or any possible implementation thereof. In one possible design, the chip system may further include a memory for storing necessary program instructions and data. This chip system may be composed of chips or may include chips and other discrete devices.

[0030] The technical effects of the second to sixth aspects or any of their possible implementations can be found in the first aspect or the technical effects of its related possible implementations, and will not be repeated here. Attached Figure Description

[0031] Figure 1 This is an exemplary schematic diagram of multiphysics coupling provided in an embodiment of this application;

[0032] Figure 2 This is an exemplary schematic diagram of a data center provided in an embodiment of this application;

[0033] Figure 3 This is a schematic diagram of an exemplary system framework provided in an embodiment of this application;

[0034] Figure 4 This is a schematic diagram of an embodiment of the solution flow control method provided in this application;

[0035] Figure 5 This is an exemplary schematic diagram of the solution process control provided in the embodiments of this application;

[0036] Figure 6 This is an exemplary schematic diagram of the solution process control provided in the embodiments of this application;

[0037] Figure 7 This is a schematic diagram of an embodiment of a cloud management platform for solving process control provided in this application;

[0038] Figure 8 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application;

[0039] Figure 9 This is a schematic diagram of a computing device cluster provided in an embodiment of this application;

[0040] Figure 10 This is a schematic diagram of a computing device cluster provided in an embodiment of this application. Detailed Implementation

[0041] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0042] As will be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0043] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not expressly listed or inherent to those processes, methods, products, or apparatus.

[0044] The following section will first provide an exemplary introduction to some of the concepts involved in this application.

[0045] 1. Physical equations

[0046] A physical equation, also known as a physical formula, is a way of expressing physical quantities using symbols.

[0047] The physical equations in this application may be in the form of partial differential equations (PDEs).

[0048] In physics-related business, the physical model describing a physical field can include basic equations and auxiliary equations.

[0049] 2. Basic Equations

[0050] The fundamental equations in physical field operations can be the basic equations describing physical phenomena. The fundamental equations involved in different physical fields can be determined based on relevant physics research. For example, the fundamental equations of electromagnetic fields may include Maxwell's equations, while the fundamental equations of flow fields may include the Navier-Stokes equations.

[0051] 3. Auxiliary equations

[0052] In physics field operations, in addition to the basic equations that mainly describe physical phenomena, auxiliary equations may also be included to supplement and improve the basic equations, providing more physical details and / or boundary conditions to ensure the accuracy and reliability of the physical model in physics field operations.

[0053] For example, auxiliary equations may include physical equations describing constitutive relations in physical field operations. Constitutive relations refer to the relationship between a material subjected to external excitation and the resulting response. They are typically used to describe the properties of a substance, such as elasticity, plasticity, electrical properties, magnetic properties, thermal properties, etc., and how a substance responds to external forces, electric fields, magnetic fields, temperature, and other factors.

[0054] For example, in electromagnetism, Maxwell's equations are the fundamental equations describing electromagnetic fields. However, in order to more accurately describe the electromagnetic properties of matter, it is necessary to supplement them with constitutive relations such as those between electric displacement vector D and electric field strength E, magnetic induction intensity B and magnetic field strength H, and current density J and electric field strength E as auxiliary equations to help improve Maxwell's equations and enable them to describe the behavior of different substances in electromagnetic fields.

[0055] 4. Mathematical equations

[0056] In this application, mathematical equations can also be called algebraic equations, which refer to equations composed of algebraic expressions of unknowns.

[0057] Computer-aided engineering (CAE) refers to the process of analyzing and solving engineering problems through numerical calculations and simulations using computer-aided engineering techniques. CAE simulation is a computer-based engineering analysis method that allows for virtual testing of products during the design phase to evaluate their performance, reliability, and safety, thereby reducing the cost and time of actual testing.

[0058] The simulation process in CAE generally includes steps such as geometric modeling, mesh generation, physical modeling, numerical computation, and post-processing of solution results. Among these, the numerical computation part provides a powerful computing engine for CAE simulation, including discretizing the physical model, handling coupling between physical models, and numerical computation, which has a significant impact on the simulation performance, such as accuracy, efficiency, and stability.

[0059] Real-world engineering problems typically involve physical fields such as electromagnetics, fluid dynamics, structures, and acoustics. In many scenarios, multiple disciplines and multiple physical fields are involved, with these fields coupling to varying degrees in different computational domains. In CAE simulation, the goal of numerical computation is to obtain the distribution of each physical field in each computational domain over time, thereby simulating the evolution of the physical world in the digital world.

[0060] It is evident that in many CAE numerical computing scenarios, the specific solution process proceeds from near to far along the time series [t0, t1, t2, t3, ...].

[0061] like Figure 1 As shown, the time series includes multiple time steps such as [t0, t1, t2, t3, ...].

[0062] In numerical computation, a time step is a concept used to represent the discretization of a simulated system over time. In numerical computation, the entire simulation time process is discretized into several smaller processes, that is, the simulation time range is divided into continuous small time intervals, each corresponding to a time step. A time step can describe a specific point in time within a time interval, or it can correspond to an entire time interval. The step size of a time step can be determined based on the actual simulation scenario.

[0063] In numerical computation scenarios, a computational domain can be a spatial region, which can be described by factors such as boundary conditions. For example... Figure 1 As shown, the computational regions involved in the entire simulation process can include computational region 1, computational region 2, computational region 3, and computational region 4, etc.

[0064] And, as Figure 1 The simulation scenario shown involves multiphysics, specifically including electromagnetics, fluid dynamics, structural physics, and acoustics. These different physics disciplines can include the following fundamental equations:

[0065] 1) Fundamental equations of electromagnetism: Maxwell's equations;

[0066] 2) Fundamental equations of fluid dynamics: The Navier-Stokes equations;

[0067] 3) Structural fundamental equations: equilibrium differential equations, geometric equations, strain compatibility equations, etc.;

[0068] 4) Fundamental acoustic equation: the sound wave equation;

[0069] For example, it can include continuity equations, state equations, and motion equations.

[0070] Furthermore, it can include auxiliary equations corresponding to different physics disciplines, such as one or more of electromagnetic, fluid, structural, or acoustic auxiliary equations. In multiphysics operations, multiple sets of physical equations under different physics disciplines can be constructed based on the fundamental equations and corresponding auxiliary equations of different physics disciplines. These sets of equations can be coupled using a coupling strategy to obtain multiple numerical computation tasks, which are then solved according to the numerical computation strategy to obtain the solution results. For example, Figure 1 In the example shown, the numerical computation strategy may, by way of example, include one or more of the following: linear direct method, linear iterative method, nonlinear solution method, or heterogeneous parallel algorithm.

[0071] In traditional CAE simulation, commonly used CAE simulation software typically pre-defines a fixed CAE solution process for each specific pre-defined CAE simulation scenario. Taking fluid mechanics CAE simulation scenarios as an example, simulation scenarios can be divided into several categories, such as basic flow, incompressible flow, compressible flow, multiphase flow, direct numerical simulation, and heat transfer. Each category is further subdivided based on flow intensity, transient and steady-state conditions, boundary conditions, etc., and a specific CAE solution process is pre-defined for each subdivided simulation scenario. This specific CAE solution process includes fixed basic models, auxiliary models, coupling strategies, and numerical calculation methods.

[0072] As can be seen, this method requires the pre-configuration of numerous CAE solution processes. Each customized CAE solution process is only applicable to a specific simulation scenario, resulting in a very narrow scope of application, and it is difficult for users to modify and adjust. Furthermore, users need to accurately identify the applicable CAE solution process, which presents a significant barrier to entry for general users. Moreover, since the CAE solution process remains fixed and does not dynamically adjust as the time series progresses, problems such as solution oscillations, divergence, or even computational crashes may occur at certain time steps, significantly impacting solution performance.

[0073] Alternatively, another common CAE approach involves users configuring and setting up the CAE solution process based on their experience. However, expert experience often requires a long period of accumulation and frequently necessitates user involvement during the solution process, which can impact processing efficiency.

[0074] Based on this, this application provides a solution process control method that can efficiently determine a suitable and accurate solution process for physical field services, thereby improving the solution performance of physical field service solution tasks.

[0075] The method described in this application embodiment can be applied to a cloud management platform. The cloud management platform manages the infrastructure providing cloud services. This infrastructure includes multiple servers, which are used to deploy virtual instances that implement the solution process. These virtual instances can take the form of virtual machines, containers, bare-metal servers, etc.

[0076] The infrastructure can be located in one or more data centers to provide cloud resources through one or more data centers.

[0077] The following is combined Figure 2 The schematic diagram shown illustrates an example of a data center architecture.

[0078] Figure 2 In this context, the cloud management platform connects to one or more servers (such as...) via the data center's internal network. Figure 2 Servers 1 and 2 interact with each other. A server consists of a hardware layer and a software layer. The hardware layer includes the server's hardware configuration, such as PCI devices like network interface cards (NICs), graphics processing units (GPUs), and offloading cards, which can be plugged into peripheral component interconnect (PCI) or peripheral component interconnect express (PCIe) slots. The software layer includes the operating system installed and running on the server (the operating system relative to a virtual machine can be called the host operating system). The host operating system contains a virtual machine manager (also called a hypervisor), whose role is to implement computational virtualization, network virtualization, and storage virtualization of the virtual machine, and to manage the virtual machine. A virtual machine (VM) refers to a complete computer system simulated by software, possessing full hardware system functionality, and running in a completely isolated environment. Figure 2 In the system architecture shown, the infrastructure includes multiple servers that can run virtual machines. These virtual machines can have the same or different specifications. These virtual machines may also be called Elastic Compute Service (ECS), Elastic Instances, etc., depending on the cloud service provider.

[0079] In one example of an embodiment of this application, the cloud management platform can be a public cloud platform. In this case, cloud service providers such as individuals or software developers with cloud resource development capabilities can provide cloud services to users. Users obtain cloud services through the Internet but do not own cloud computing resources. In other embodiments of this application, the cloud management platform can be a private cloud platform or a hybrid cloud platform, and this application does not impose any restrictions on this.

[0080] Specifically, in Figure 2 In the example shown, the cloud management platform can provide an access interface (such as a user interface or application programming interface (API)). Users of the cloud management platform and cloud service providers can operate the client to remotely access the access interface to register a cloud account and password on the cloud management platform. After the cloud account and password are successfully authenticated on the cloud management platform, they can log in to the cloud management platform to create, manage, log in to and operate virtual machines in the cloud data center.

[0081] For example, when it is necessary to perform a solution task for a physical field business, some enterprises, organizations or individuals can use the cloud resources of the cloud management platform to execute the solution flow control method of this application embodiment, obtain the configuration information of the solution flow corresponding to the solution task, and then determine the specific method of the first solution flow according to the configuration information, so as to execute the solution task through cloud resources according to the first solution flow, or execute the solution task through local processing resources according to the determined first solution flow, and obtain the corresponding solution result.

[0082] Of course, the cloud management platform can also be other types of cloud management platforms, such as private cloud or hybrid cloud, and this application embodiment does not limit this.

[0083] like Figure 3 The diagram illustrates an example of a service within a cloud management platform. Figure 3 In the examples shown, the cloud management platform can provide users with solution process configuration services. Furthermore, in some examples, the cloud management platform can also provide solution services.

[0084] Specifically, the solution process configuration service may include one or more of the following modules: physics discipline type identification module, auxiliary equation control module, coupling strategy control module, numerical calculation strategy control module, and management module.

[0085] The CAE simulation solution process for physical field services specifically includes a simulation solution process for one or more time steps.

[0086] During the simulation and solution process at each time step, based on the method of any embodiment of this application, and according to the user-input physical equation information such as the basic equations and the user's control flow preferences, one or more modules of the solution flow configuration service—the physics subject type identification module, the auxiliary equation control module, the coupling strategy control module, the numerical calculation strategy control module, or the management module—can obtain the configuration information of the solution flow for that time step through a machine learning model and / or a preset database. The specific functions of the physics subject type identification module, the auxiliary equation control module, the coupling strategy control module, the numerical calculation strategy control module, or the management module can be found in the subsequent descriptions of related method embodiments and will not be repeated here.

[0087] After obtaining the solution process configuration information for the current time step through the solution process configuration service, the solution process configuration information for that time step can be sent to the solution service. The solution service can then use the appropriate solution process for that time step to execute the solution task and obtain the solution result for that time step based on the solution process configuration information for that time step.

[0088] In this way, the solution process configuration service and the solution service solve each time step in sequence to obtain the final solution result and feed it back to the user.

[0089] It should be noted that, Figure 3 The services described are merely one example of the services provided by the cloud management platform, and are not limited thereto.

[0090] For example, in another scenario, the cloud management platform does not provide solution services, but only solution process configuration services. In this example, the user sends a request to the cloud management platform's solution process configuration service via a client to obtain the configuration information for the solution process. After receiving the request, the cloud management platform can invoke the solution process configuration service to obtain the configuration information and then send this information to the user, enabling the user to solve the physical field service's solution task via the client based on the obtained configuration information.

[0091] Furthermore, in other examples provided in this application, the aforementioned solution service and / or solution process configuration service can be deployed holistically within the user's server cluster; and the functional division of each service can be synchronized with... Figure 3 The services shown may differ, and the deployment methods for each service may also differ. Each service can be provided independently, embedded in other services, or multiple services can be combined for deployment; this application does not impose any restrictions on this.

[0092] Based on the aforementioned cloud management platform, such as Figure 4As shown, a solution flow control method according to an embodiment of this application may include steps 401-403.

[0093] Step 401: The cloud management platform obtains the first physical equation for the physical field business from the equation input interface.

[0094] In this embodiment, the physical field service can be a single physical field service or a multi-physical field service. The physical fields involved in the physical field service can be of various types, and no limitation is made here. For example, the physical field service may include one or more of the following physical fields: magnetic field, electric field, fluid, acoustic, mechanical, chemical, and heat transfer.

[0095] The first physical equation can specifically be a partial differential equation (PDE).

[0096] There can be one or more first physical equations. When there are multiple first physical equations, they can belong to the same physical field or to different physical fields.

[0097] The first physical equation may include the fundamental equations of the physical field, and in some examples, it may also include auxiliary equations of the physical field.

[0098] The fundamental equations for different physical fields can be determined based on the specific type of the physical field, and no restrictions are imposed here.

[0099] For example, the fundamental equations of electromagnetic fields include Maxwell's equations, while the fundamental equations of fluids include the Navier-Stokes equations.

[0100] In this embodiment of the application, there are multiple ways to obtain the first physical equation from the equation input interface.

[0101] In one example, the user inputs the first physical equation into the equation input interface so that the cloud management platform can obtain the first physical equation through the equation input interface.

[0102] For example, refer to Figure 5 As shown in the example, the solution process configuration service can provide a configuration interface, which includes equation input interfaces such as application programming interfaces (APIs). Users can input the first physical equation into the equation input interface of the configuration interface through the client, so that the cloud management platform can receive the first physical equation input by the user.

[0103] In another example, the first physical equation can be obtained from historical time steps of the current time step. For instance, if the current time step is the first time step, the first physical equation required for the current first time step can be determined based on information such as the physical state reflected in the calculation results of the previous time step.

[0104] Step 402: The cloud management platform determines the first physics discipline type corresponding to the first physics equation based on the first physics equation.

[0105] In this embodiment of the application, considering that the physical characteristics involved in different types of physics disciplines are significantly different, resulting in significant differences in the content of the corresponding solution tasks, the solution order of the solution tasks, and other solution methods, this embodiment of the application can determine the first physics discipline type corresponding to the first physical equation, so as to reasonably determine the corresponding first solution process based on the relevant characteristics of the first physics discipline type.

[0106] The first physics discipline type can have multiple specific types, and the classification methods for these physics discipline types can also vary. For example, the first physics discipline type can be one or more of the discipline types such as magnetic field, electric field, fluid dynamics, acoustics, mechanics, chemical engineering, and heat transfer. In this case, the classification method for the physics discipline types involved can be considered a coarse-grained classification method. Alternatively, the first physics discipline type can be further refined based on the coarse-grained classification method. For example, the first physics discipline type can be one or more of the discipline types such as fluid statics, fluid dynamics, particle fluid dynamics, aerodynamics, and hydrodynamics within fluid mechanics. In this case, the classification method for the physics discipline types involved can be considered a more fine-grained classification method.

[0107] There are multiple ways to determine the type of the first physics discipline corresponding to the first physical equation.

[0108] In one example, the user can input the first physics subject type through a configuration interface provided by the cloud management platform.

[0109] In another example, the preset database may include information for storing one or more preset physical equations and their corresponding preset physical discipline types.

[0110] After obtaining the first physical equation, through methods such as Figure 3 The physics subject type identification module shown identifies the first physics subject type of the first physical equation. Specifically, the physics subject type identification module can identify the first physics subject type of the first physical equation in various ways.

[0111] For example, in one instance, the first physical equation can be matched with at least one preset physical equation in a preset database, and the preset physics subject type corresponding to the matched preset physical equation can be used as the first physics subject type.

[0112] In matching the first physical equation with preset physical equations, matching can be performed based on the characteristic information of the physical equations. For example, matching can be performed based on the characteristics of the mathematical operators (e.g., partial derivatives, differential order, power series), operation symbols (e.g., addition, subtraction, multiplication, division), and equation structure of the first and preset physical equations. For instance, if a first physical equation is a partial differential equation, its characteristics can be obtained, such as one or more of the partial derivatives, differential order, coefficients, and initial boundary conditions of the partial differential equation with respect to the time variable t and the space variables x / y / z. Then, it can be compared with at least one preset physical equation. If the characteristics of a preset physical equation match those of the partial differential equation, the preset physics discipline type corresponding to the preset physical equation can be obtained from the preset database as the first physics discipline type corresponding to the first physical equation.

[0113] In another example, a machine learning model can be used to identify the first physics discipline type of the first physical equation.

[0114] In this example, the machine learning model can be considered a classification model. The specific type of the machine learning model is not limited here. For example, the machine learning model can be a convolutional neural network (CNN), a feedforward neural network (FNN), etc.

[0115] In addition, a pre-set database and machine learning model can be combined to determine the first physics discipline type corresponding to the first physics equation.

[0116] For example, a machine learning model (e.g., a feature extraction model) can be used to identify the features of a preset physical equation and a first physical equation in a preset database. Then, the similarity between the features of the preset physical equation and the features of the first physical equation can be calculated. In this way, a preset physical equation whose similarity to the first physical equation meets a specified condition (e.g., the highest similarity) can be determined from one or more preset physical equations in the preset database. Thus, the preset physics subject type corresponding to the preset physical equation can be obtained from the preset database as the first physics subject type corresponding to the first physical equation.

[0117] Alternatively, if no matching preset physical equation exists in the preset database, a machine learning model can be used to determine the first physics discipline type corresponding to the first physical equation.

[0118] Furthermore, in some examples, the physics subject type identification module can also identify and correct errors in the physical equations input by the user, outputting a complete and correct first physical equation. Additionally, in some examples, the physics subject type identification module can also output the fundamental variables to be solved in the first physical equation; for example, in the electromagnetic domain, this includes electric field strength, magnetic field strength, electric flux density, and magnetic flux density; in the fluid domain, it includes velocity, pressure, and temperature.

[0119] Step 403: The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type.

[0120] The first configuration information includes one or more of the following information corresponding to the first physical equation: first auxiliary equation, first coupling strategy, first numerical calculation strategy. The first configuration information is used to determine the set of mathematical equations contained in the first solution task and / or the solution method of each first solution task. The first solution task is used to solve the first physical equation.

[0121] In CAE simulation scenarios, the physical equations need to be discretized into mathematical equations before a solution task is created to obtain the results. The first configuration information configures the first solution process, which determines the specific content of one or more first solution tasks and how to perform the solution. In other words, the first solution process can indicate the specific numerical calculation method, thereby controlling the solver to perform numerical calculations and obtain the solution results of one or more first solution tasks. The solution results of these one or more first solution tasks include, but are not limited to, numerical solutions corresponding to the physical parameters to be solved in the first physical equations. For example, in some examples, the solution results of these one or more first solution tasks include numerical solutions corresponding to the physical parameters to be solved in the first physical equations, as well as numerical solutions corresponding to one or more physical parameters in the first auxiliary equations.

[0122] In this embodiment, first configuration information can be intelligently obtained based on a first physics discipline type. The specific content of the first configuration information may include one or more of the following: a first auxiliary equation, a first coupling strategy, and a first numerical calculation strategy. For example, the content of the first configuration information may be determined based on the specific type of the physics service (e.g., a single-physics service or a multi-physics service). Alternatively, the content of the first configuration information may be pre-configured by the user.

[0123] For example, in one example, reference Figure 5 As shown in the example, the solver configuration service can receive flow control preference information input by the user. This flow control preference information may include one or more of the following:

[0124] (1) Indication information indicating whether to start automatic process control;

[0125] (2) When automatic process control is started, indicate whether to automatically control one or more of the following information: physics discipline type identification, auxiliary equations, coupling strategy, numerical calculation strategy;

[0126] (3) For each type of automatic control information (i.e., physics discipline type identification, auxiliary equation, coupling strategy or numerical calculation strategy), set specific control preferences. For example, automatic control can be enabled at every time step / iteration step in the solution process of the physics field business, or every few time steps / iteration steps.

[0127] (4) The contents of the process control preference information in (1), (2) and / or (3) above can be set for any specific physical equation.

[0128] In this way, the user can determine the start of the solution process configuration service based on the flow control preference information input by the user, and determine the specific module called in the solution process configuration service, thereby determining the content contained in the first configuration information, so as to determine one or more aspects in the first solution process based on the first configuration information.

[0129] It is understood that, depending on the content included in the first configuration information, the first configuration information may contain complete information from the first solution process, or it may contain only partial information from the first solution process. When the first configuration information contains only partial information from the first solution process, the other information in the first solution process may be determined in other ways, such as by the user or by using pre-configured information, without any restrictions.

[0130] In this application embodiment, there can be multiple ways to determine the first configuration information, and it can be affected by multiple scenario factors.

[0131] For example, refer to Figure 5 As shown in the example, the physical field service may involve multiple time steps. For each time step, the solution flow control method in this embodiment can be executed to determine the respective first solution flow. However, the specific methods for determining the corresponding first configuration information in different time steps, such as the initial time step (e.g., time step t0) and subsequent time steps, may differ.

[0132] For example, in some embodiments, in each time step after the initial time step, for each time step, the solution status of the historical time steps of the current time step can be referenced (e.g., obtaining the historical configuration information of the solution process of the historical time steps and / or the solution information of the historical solution results through the management module, etc.) to determine the first configuration information corresponding to the current time step. For the initial time step, the initial physical state, etc., can be referenced to determine the first configuration information corresponding to the initial time step, so as to serve as the configuration information of the initial solution process of the initial time step.

[0133] The following provides an example of how to determine the first configuration information in different time step scenarios.

[0134] 1. Based on the solution results and / or solution performance information of the historical time steps, determine the first configuration information of the first solution process for the current time step.

[0135] Specifically, in some embodiments, the physical field service involves a first time step and a second time step, the second time step being prior to the first time step, and the first physical equation being the physical equation corresponding to the physical field service in the first time step. The method further includes:

[0136] The cloud management platform obtains the second configuration information corresponding to the second physical equation of the physical field service at the second time step, and / or the second solution information corresponding to the second physical equation. The second configuration information includes one or more of the following information corresponding to the second physical equation: second auxiliary equation, second coupling strategy, second numerical calculation strategy. The second solution information includes one or more of the following information: solution result of the second solution task, solution performance of the second solution task.

[0137] Step 403 includes:

[0138] The cloud management platform determines the first configuration information based on the first physics subject type, as well as the second configuration information and / or the second solution information.

[0139] In this embodiment of the application, the second time step can be one or more time steps located before the first time step.

[0140] For example, in some examples, the second time step can be multiple consecutive time steps preceding the first time step. This allows for a more accurate determination of the physical field state before the current time step, avoiding interference from temporary anomalies that may occur in a single historical time step on the first solution process of the current first time step. Alternatively, in some examples, the second time step can be a single time step preceding the first time step, or multiple non-consecutive time steps.

[0141] The specific content of the second configuration information corresponding to the second time step can be found in the description of the relevant embodiments of the first configuration information, and will not be repeated here. It should be noted that the information types included in the second configuration information can be the same as or different from the first configuration information. For example, the second configuration information may include a second auxiliary equation and a second numerical calculation strategy, while the first configuration information may include a first auxiliary equation, a first coupling strategy, and a first numerical calculation strategy.

[0142] The second solution information corresponding to the second time step may include the solution result of the second solution task and / or the solution performance of the second solution task. The solution result of the second solution task may include the solutions to the relevant variables obtained after completing the second solution task. The solution performance of the second solution task may reflect the accuracy and other performance characteristics of the solution result itself, and / or the coupling solution performance (e.g., number of coupling iterations, coupling accuracy) and numerical computation solution performance (e.g., number of numerical computation iterations, numerical computation solution accuracy) in the solution process for obtaining the solution result of the second solution task.

[0143] When determining the first configuration information, depending on the specific content of the first configuration information, one or more of the following can be used: the second configuration information of the second time step, the solution result of the second solving task, or the solution performance of the second solving task.

[0144] Below, for reference Figure 5 The examples shown illustrate how, using the auxiliary equation control module, coupling strategy control module, and numerical computation strategy control module as examples, the methods for determining the first auxiliary equation, first coupling strategy, and first numerical computation strategy in the first configuration information are exemplified based on the solution results and / or solution performance information of historical time steps. It is understood that the auxiliary equation control module, coupling strategy control module, and numerical computation strategy control module can be software modules and functional modules. In practical application scenarios, other module division methods can also be used. For example, the functions of one or more modules in the auxiliary equation control module, coupling strategy control module, and numerical computation strategy control module can be split or merged. This is only presented as an example and is not intended to limit the scope.

[0145] 1. Auxiliary Equation Control Module

[0146] In this example, the auxiliary equation control module is used to determine the functions related to the first auxiliary equation. That is to say, in this example, the auxiliary equation control module can be activated based on the user's flow control preferences to automatically determine the first auxiliary equation and related physical equations.

[0147] Furthermore, in this example, the first auxiliary equation and the first set of physical equations obtained based on the solution information of the second time step can be determined at the first time step.

[0148] Specifically, in some embodiments, the first configuration information includes a first set of physical equations determined based on the first physical equation and the first auxiliary equation, and the second solution information includes the solution results of the second solution task;

[0149] Step 403 includes:

[0150] The cloud management platform determines the first auxiliary equation based on the physical state reflected in the solution results of the second solution task and the type of the first physics discipline.

[0151] The cloud management platform determines the first set of physical equations based on the first physical equation and the first auxiliary equation.

[0152] In this embodiment, the physical state reflected by the solution result of the second solving task can be described based on the value information of one or more physical quantities involved in the solution result of the second solving task. For example, the physical state reflected by the solution result of the second solving task may include the values ​​of one or more physical quantities; or, the corresponding physical state can be determined based on the range of the values ​​of one or more physical quantities in the solution result of the second solving task. For example, if the solution result of the second solving task includes the values ​​of one or more fluid physical quantities, the physical state of the flow field can be determined based on the range of the values ​​of the one or more fluid physical quantities, etc., to determine whether the flow field is turbulent. When the second time step is the time step preceding the first time step, the physical state reflected by the solution result of the second solving task in the second time step can be used as the initial state corresponding to the first time step.

[0153] In addition, in some examples, the first auxiliary equation corresponding to the first physical equation can be determined based on the physical state reflected by the solution result of the second solution task and other information other than the first physics discipline type. For example, the other information may include the initial value conditions and / or boundary value conditions corresponding to the first time step.

[0154] After obtaining the solution results of the second solution task and information such as the first physics discipline type, the first auxiliary equation corresponding to the first physical equation can be determined based on the physical state reflected by the solution results of the second solution task and the first physics discipline type.

[0155] In this embodiment of the application, there may be multiple ways to determine the first auxiliary equation corresponding to the first physical equation, which are not limited here.

[0156] In some embodiments, a first auxiliary equation corresponding to the first physical equation can be determined by using a machine learning model and / or a preset database corresponding to the first physics discipline type, based on the physical state reflected by the solution result of the second solving task and the first physics discipline type.

[0157] For example, in one instance, the preset database may include preset auxiliary equations corresponding to preset physics disciplines.

[0158] In this example, relevant experts and developers can pre-collect pre-defined auxiliary equations corresponding to various pre-defined physical states under one or more pre-defined physics disciplines from a pre-defined database.

[0159] In this way, the first physics subject type can be matched with one or more preset physics subject types in the preset database to determine the target preset physics subject type that matches the first physics subject type. Then, the physical state reflected by the solution result of the second solution task can be matched with one or more preset physical states corresponding to the target preset physics subject type in the preset database to determine the target preset physical state that matches the physical state reflected by the solution result of the second solution task. The preset auxiliary equation corresponding to the target preset physical state in the preset database is then used as the first auxiliary equation corresponding to the first physical equation.

[0160] In some examples, a first machine learning model can process information such as the physical state reflected in the solution results of the first physical equation and the first physics discipline type, and output a first auxiliary equation corresponding to the first physical equation. This first machine learning model can be trained on first training data from a pre-defined database, or it can be trained on other third-party data. For example, pre-defined physics discipline types and their associated pre-defined physical states can be obtained from the pre-defined database as first training data, and the pre-defined auxiliary equations corresponding to these pre-defined physics discipline types and their associated pre-defined physical states can be used as labels for the first training data to train the machine learning model. After training, a first machine learning model for determining the auxiliary equation can be obtained.

[0161] Alternatively, the first auxiliary equation can be determined by combining the first machine learning model with pre-set database matching. For example, the first initial auxiliary equation corresponding to the first physical equation can be obtained first through pre-set database matching. Then, the first initial auxiliary equation, along with the first physical equation and the first physics discipline type, can be input into the first machine learning model to supplement its reasoning ability and obtain a complete and accurate first auxiliary equation.

[0162] For different types of physics disciplines, the specific content of the first auxiliary equation corresponding to the first physical equation can vary and can be determined according to the actual application scenario. No restrictions are imposed here.

[0163] For example, in some examples, the physics field service is a multiphysics field service that includes electromagnetics and fluids, with the first physics discipline type including electromagnetic fields and fluids.

[0164] The first set of physical equations includes Maxwell's equations, which form the basis of the electromagnetic field, and Navier-Stokes' equations, which form the basis of the flow field.

[0165] Therefore, in the preset database, the first auxiliary equation may include auxiliary equations describing electromagnetic constitutive relations (e.g., auxiliary equations describing the relationship between electric displacement vector and electric field strength, magnetic induction intensity and magnetic field strength, current density and electric field strength), and auxiliary equations describing fluid constitutive relations (e.g., auxiliary equations describing Newtonian fluids, incompressible fluids, ideal fluids, etc.). In addition, it may also include auxiliary equations corresponding to auxiliary models related to the first physical equation (e.g., if the Reynolds number corresponding to the physical state indication velocity of the solution result of the second solution task is greater than 3200, then the auxiliary equations corresponding to the auxiliary models related to the first physical equation are obtained, for example, the auxiliary equations corresponding to turbulence models such as Spalart-Allmaras, k-epsilon, or k-omega can be obtained).

[0166] In the embodiments of this application, the number and specific form of the first auxiliary equations can be varied. For example, there can be one or more first auxiliary equations, and a certain first auxiliary equation can be a physical equation or a term in a physical equation (e.g., a certain first physical equation).

[0167] Furthermore, in this embodiment, not only can the first auxiliary equation corresponding to the first physical equation be determined, but also the first set of physical equations corresponding to the first auxiliary equation can be determined.

[0168] The first set of physical equations can be determined by the form of the first auxiliary equations output during the matching of a preset database and / or the inference of a machine learning model. Alternatively, after obtaining all the first auxiliary equations and the first physical equations, the first physical equations and the first auxiliary equations can be grouped according to the relationship between them to obtain one or more sets of first physical equations.

[0169] For example, the first auxiliary equation describing the constitutive relation corresponding to a certain first physical equation is contained in the same set of first physical equations as the first physical equation. The auxiliary equations in the auxiliary model (e.g., the turbulence model) corresponding to the first physical equation can be considered as a set of first physical equations; that is, the auxiliary equations in the auxiliary model are considered as a set of first physical equations.

[0170] Based on any of the above embodiments, one or more sets of first physical equations can be obtained through the auxiliary equation module.

[0171] After obtaining the first set of physical equations, if the control flow preference information indicates that only this auxiliary equation module should be used, then it can be done through methods such as Figure 5 In the example shown, the management module sends the information of the first set of physical equations to the solution service as the first configuration information for the current first time step. Alternatively, in other examples, the auxiliary equation module may directly send the information of the first set of physical equations to the solution service as the first configuration information for the current first time step.

[0172] If the control flow preference information indicates the use of the coupling strategy control module, the auxiliary equation module can send the information of the first physical equation set to the coupling strategy control module to execute the subsequent steps related to determining the first coupling strategy.

[0173] 2. Coupling Strategy Control Module

[0174] In this example, the coupling strategy control module is used to determine the first coupling strategy. That is, in this example, the coupling strategy control module can be activated based on the user's flow control preferences to automatically determine the first coupling strategy.

[0175] Furthermore, in this example, the first coupling strategy can be determined at the first time step based on the second configuration information and / or the second solution information at the second time step.

[0176] Specifically, in some embodiments, the physical field service is a multi-physics service, the number of first physical equations is multiple, the number of second physical equations is multiple, the first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations, the second configuration information includes the second coupling strategy corresponding to each of the multiple second physical equations, and the solution performance in the second solution information includes the coupling performance information corresponding to the second solution task using the second coupling strategy for solution.

[0177] The method also includes:

[0178] The cloud management platform acquires multiple sets of first physical equations based on multiple first physical equations;

[0179] Step 403 includes:

[0180] The cloud management platform determines the first coupling strategy based on multiple sets of first physical equations, as well as second configuration information and / or second solution information.

[0181] The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more first physical equations based on the first coupling strategy.

[0182] In this embodiment of the application, in the case of multiphysics services, it is necessary to determine the coupling strategy between multiple sets of first physical equations and / or within any set of first physical equations in the multiphysics service.

[0183] The plurality of first physical equations can be obtained by using the relevant functions of the auxiliary equation control module in any of the above embodiments, that is, they can be output by the auxiliary equation control module in any of the above embodiments. Alternatively, when the user's flow control preference information indicates that the functions of the auxiliary equation control module are not used, the plurality of first physical equations can also be input by the user; in this case, the user can directly input a plurality of first physical equations including the first physical equation, or the user can input the first physical equation and the first auxiliary equation separately, and then combine them to obtain a plurality of first physical equations, etc.

[0184] After obtaining multiple sets of first physical equations, a coupling strategy can be determined between any two sets of first physical equations and / or between physical equations within a set of first physical equations as the first coupling strategy.

[0185] Specifically, the method for determining the first coupling strategy corresponding to each first physical equation set can differ depending on the specific circumstances. For example, the method for determining the first coupling strategy corresponding to a certain first physical equation set differs depending on whether a historical physical equation set identical to a certain first physical equation set exists in the historical time step or whether no historical physical equation set identical to that first physical equation set exists.

[0186] The following are examples of each.

[0187] (1) The method for determining the first coupling strategy when there is the same set of historical physics equations.

[0188] Specifically, in some embodiments, step 403 includes:

[0189] If among multiple sets of second physical equations, there exists a set of second physical equations that is identical to one or more sets of first physical equations, then the cloud management platform determines the first coupling strategy corresponding to one or more sets of first physical equations based on the second coupling strategy and coupling performance information corresponding to the identical set of second physical equations.

[0190] In this embodiment of the application, if one or more first physical equation sets have the same second physical equation set in the historical physical equation sets (i.e., multiple second physical equation sets), the first coupling strategy corresponding to one or more first physical equation sets can be determined by referring to the specific situation of the second coupling strategy of the same second physical equation set.

[0191] Specifically, coupling performance information of the second coupling strategy for the same second set of physical equations can be obtained. For example, this coupling performance information may include one or more of the following: information reflecting the coupling solution speed, such as the number of coupling iterations; and information reflecting the coupling solution accuracy, such as coupling accuracy and solution accuracy.

[0192] For example, the coupling log of the second time step can be obtained, and the second coupling strategy and corresponding coupling performance information of the same second set of physical equations can be obtained from the coupling log.

[0193] After obtaining the coupling performance information of the second coupling strategy of the same second physical equation set, if the coupling performance information of the second coupling strategy of the same second physical equation set meets the preset performance requirements, one or more first coupling strategies corresponding to the first physical equation set can be determined according to the specific content of the second coupling strategy of the same second physical equation set.

[0194] For example, coupling performance parameter thresholds (such as coupling accuracy thresholds and / or coupling speed thresholds) can be preset. If the value of the coupling parameter in the coupling performance information of the second coupling strategy for the same second physical equation set is greater than the corresponding coupling performance parameter threshold, then the second coupling strategy for the same second physical equation set can be considered a more suitable second coupling strategy. Therefore, based on the second coupling strategy for the same second physical equation set, the first coupling strategy corresponding to one or more first physical equation sets can be determined. For example, the second coupling strategy for the same second physical equation set can be used as the first coupling strategy corresponding to one or more first physical equation sets. Alternatively, the information of the second coupling strategy for the same second physical equation set and the information of one or more first physical equation sets can be input into the corresponding machine learning model (such as the second machine learning model in subsequent embodiments) to assist in reasoning and output the first coupling strategy corresponding to one or more first physical equation sets.

[0195] If the value of the coupling parameter in the coupling performance information of the second coupling strategy of the same second physical equation set is not greater than the corresponding coupling performance parameter threshold, then in another embodiment, the method for determining the first coupling strategy in the absence of the same historical physical equation set can be used to determine the first coupling strategy corresponding to one or more first physical equation sets.

[0196] (2) How to determine the first coupling strategy when there is no identical set of historical physics equations.

[0197] Specifically, in some embodiments, step 403 includes:

[0198] If none of the multiple sets of second physical equations are the same as one or more sets of first physical equations, the cloud management platform determines the first coupling strategy corresponding to one or more sets of first physical equations based on the physical state reflected by the solution results of the second solution task.

[0199] In this embodiment, when one or more first physical equation sets do not share the same second physical equation sets among multiple second physical equation sets, it is generally difficult to directly determine the first coupling strategy of the first physical equation set based on the historical coupling strategy (i.e., the second coupling strategy of the second physical equation set). However, the current scene situation can be determined based on the physical state reflected by the solution results of the second solving task, thereby determining a first coupling strategy that conforms to the current scene for the first physical equation set at the current time step. The physical state reflected by the solution results of the second solving task can be referred to in the relevant descriptions of any of the embodiments such as the auxiliary equation control module described above, and is not limited here.

[0200] There are multiple ways to determine the specific coupling strategy corresponding to one or more sets of first physical equations based on the physical state reflected by the solution results of the second solution task.

[0201] For example, in some embodiments, a first coupling strategy corresponding to one or more first physical equation sets can be determined based on the physical state reflected by the solution results of the second solving task, using a machine learning model and / or a preset database corresponding to the first physics discipline type.

[0202] For example, in one instance, the preset database may include preset coupling strategies and their corresponding scope of application. This scope may include preset physical equations and their corresponding preset physical states, and may also include preset physics disciplines. The preset physical states may include the values ​​of one or more physical variables or other parameters, or the range of values ​​for one or more physical variables or other parameters.

[0203] In this way, the first set of physical equations can be matched with one or more preset physical equation sets in a preset database. After matching the target preset physical equation set corresponding to the first set of physical equations, the physical state reflected by the solution result of the second solution task can be matched with one or more preset physical states corresponding to the target preset physical equation set in the preset database. This determines the target preset physical state that matches the physical state reflected by the solution result of the second solution task, and the preset coupling strategy corresponding to the target preset physical state in the preset database is used as the first coupling strategy corresponding to the first set of physical equations. Alternatively, in other examples, other preset database matching methods can be used to determine the first coupling strategy corresponding to the first set of physical equations; this is not limited here.

[0204] In some examples, a second machine learning model can also process the physical state and the first set of physical equations reflected in the solution results of the second solution task to output a first coupling strategy corresponding to the first set of physical equations. This second machine learning model can be trained on second training data from a pre-defined database, or it can be trained on other third-party data. For example, information such as the applicable scope of a pre-defined set of physical equations and the corresponding pre-defined coupling strategy (including the corresponding pre-defined physical state) can be obtained from a pre-defined database as second training data. The pre-defined coupling strategy corresponding to the pre-defined set of physical equations can be used as the label for the second training data to train the machine learning model, resulting in a second machine learning model used to determine the coupling strategy.

[0205] Alternatively, a combination of methods, such as a second machine learning model and pre-defined database matching, can be used to determine the first coupling strategy. For example, different priorities can be set among these methods, and a lower priority method can be used to determine the first coupling strategy when a higher priority method fails to do so.

[0206] The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task. It can be seen that the solution form (e.g., the corresponding solution matrix) of the solution task is different for different coupling strategies. Furthermore, the first coupling strategy is used to indicate the solution order and other solution methods of the first solution task.

[0207] The first coupling strategy can specifically adopt existing or future-developed coupling strategies, such as one or more of the following: fully coupled strategy, decoupled coupling strategy, weakly coupled iterative strategy, etc.

[0208] The following examples illustrate the specifics of the first coupling strategy and the exemplary determination principles.

[0209] In one example, the first coupling strategy may include one or more of the following:

[0210] (1) Coupling strategies between different physics disciplines (disciplinary coupling): For example, when the number of external iterations of electromagnetic field and flow field is large (e.g., when the number of external iterations exceeds the specified threshold), a full coupling strategy is adopted; otherwise, a separation coupling strategy is adopted.

[0211] Among them, the first set of physical equations related to the electromagnetic field and the corresponding first set of physical equations related to the flow field, which adopt a fully coupled strategy, can be combined into a large set of physical equations. This large set of physical equations can be discretized to obtain a set of mathematical equations for the first solution task, thereby obtaining the corresponding solution matrix, etc.

[0212] The electromagnetic field-related first physical equations and the corresponding flow field-related first physical equations, employing a decoupling strategy, are each discretized. That is, the electromagnetic field-related first physical equations can be discretized to obtain a set of mathematical equations for one primary solution task, thus yielding the corresponding solution matrix; similarly, the flow field-related first physical equations can be discretized to obtain another set of mathematical equations for the primary solution task, also yielding the corresponding solution matrix. Furthermore, the solution order between the primary solution tasks corresponding to these two sets of first physical equations can be determined based on the variables and the relationship between them.

[0213] (2) Coupling strategy (variable coupling) within the first set of physical equations in the same physics discipline: When electromagnetic force dominates relative to inertial force, the source term of the momentum equation is too large and it is not easy to converge. A coupling solution strategy can be adopted in the basic physical equations of the flow field (Navier-Stokes equations). Conversely, a separation solution strategy can be adopted to split the basic physical equations and then discretize them to obtain multiple first solution tasks and the mathematical equations of each of the multiple first solution tasks. The solution order among the multiple first solution tasks can be determined according to the relationship between variables and physical equations.

[0214] 3. Numerical Calculation Strategy Control Module

[0215] In this example, the numerical calculation strategy control module is used to determine the first numerical calculation strategy. That is, in this example, the numerical calculation strategy control module can be activated based on the user's flow control preferences to automatically determine the first numerical calculation strategy.

[0216] Furthermore, in this example, a first numerical calculation strategy can be determined at the first time step based on the second configuration information and / or the second solution information at the second time step.

[0217] Specifically, in some embodiments, the first configuration information includes a first numerical calculation strategy corresponding to the first physical equation, and the second solution information includes the solution performance of the second solution task;

[0218] The method also includes:

[0219] The cloud management platform acquires the first solution task, which includes a set of mathematical equations derived from the first physical equation.

[0220] Step 403 includes:

[0221] The cloud management platform determines the first numerical computation strategy for the system of mathematical equations contained in the first solution task.

[0222] In this embodiment, the specific content of the first solution task can be obtained by using the relevant functions of the coupling strategy control module in any of the above embodiments, that is, it can be output by the coupling strategy control module in any of the above embodiments. Alternatively, when the user's flow control preference information indicates that the functions of the coupling strategy control module are not used, the first solution task can also be input by the user.

[0223] After obtaining the first solution task, the numerical calculation strategy control module can determine the first numerical calculation strategy for each first solution task.

[0224] The specific method for determining the first numerical computation strategy corresponding to each first solution task can differ depending on the specific circumstances of that first solution task. For example, the method for determining the first numerical computation strategy corresponding to that first solution task differs depending on whether there is a historical solution task identical to that first solution task in the historical time step or whether there is no historical solution task identical to that first solution task.

[0225] The following are examples of each.

[0226] (1) When a first solution task has a matching historical solution task, the method for determining the first numerical calculation strategy corresponding to the first solution task.

[0227] Specifically, in some embodiments, step 403 includes:

[0228] If the second solution task matches the first solution task, the cloud management platform determines the first numerical calculation strategy for the set of mathematical equations contained in the first solution task based on the solution performance of the second solution task.

[0229] In this embodiment of the application, if any first solving task has a matching second solving task in the historical solving tasks (i.e., one or more second solving tasks), the first numerical calculation strategy corresponding to the first solving task can be determined by referring to the specific situation of the second numerical calculation strategy of the matching second solving task.

[0230] The matching of the second solution task with the first solution task can be that the type of the mathematical equation set contained in the second solution task is the same as that contained in the first solution task, for example, both are mathematical equation sets obtained by discretization of the turbulence model; and / or, the characteristics of the mathematical equation set contained in the second solution task are the same as those of the mathematical equation set contained in the first solution task, for example, the corresponding coefficient matrices are both large sparse matrices.

[0231] In determining the first numerical computation strategy for the system of mathematical equations contained in the first solution task, solution performance information of the second numerical computation strategy for the matching second solution task can be obtained. For example, this solution performance information may include one or more of the following: information reflecting the numerical computation speed, such as the number of iterations during numerical computation; and information reflecting the numerical computation accuracy, such as the solution accuracy of the numerical computation result.

[0232] For example, the numerical computation log of the second time step can be obtained, and the second numerical computation strategy and corresponding solution performance information of the matching second solution task can be obtained from the numerical computation log.

[0233] After obtaining the solution performance information of the second numerical calculation strategy of the matching second solution task, if the solution performance information of the second numerical calculation strategy of the matching second solution task meets the preset performance requirements, the first numerical calculation strategy corresponding to the first solution task can be determined according to the specific content of the second numerical calculation strategy of the matching second solution task.

[0234] For example, a threshold for solution performance parameters (e.g., a numerical computation accuracy threshold and / or a numerical computation speed threshold) can be preset. If the value of the numerical computation performance parameter in the solution performance information of the second numerical computation strategy of the matching second solution task is greater than the corresponding numerical computation performance parameter threshold, then the first numerical computation strategy corresponding to the corresponding first solution task can be determined based on the second numerical computation strategy of the matching second solution task. For example, the second numerical computation strategy of the matching second solution task can be used as the first numerical computation strategy corresponding to the corresponding first solution task. However, if the value of the numerical computation parameter in the solution performance information of the second numerical computation strategy of the matching second solution task is not greater than the corresponding numerical computation performance parameter threshold, then in another embodiment, when a first solution task does not have the same historical solution task, the method for determining the first numerical computation strategy corresponding to the first solution task can be used to determine the first numerical computation strategy corresponding to the first solution task.

[0235] (2) The method for determining the first numerical calculation strategy corresponding to a first solution task when there is no matching historical solution task for a certain first solution task.

[0236] Specifically, in some embodiments, a first numerical calculation strategy for the set of mathematical equations contained in the first solution task can be determined by a machine learning model and / or a preset database corresponding to the first physics discipline type.

[0237] For example, in one instance, the preset database may include information about preset mathematical equation sets and corresponding preset numerical calculation strategies. The information about the preset mathematical equation sets may include information about the preset mathematical equation sets themselves (e.g., one or more of the objective function, coefficient matrix, constraints, etc.), and may also include scenario information about the preset mathematical equation sets (e.g., one or more of the corresponding physical equation sets, physical state information, etc.).

[0238] In this way, for a certain first solution task, the information of the mathematical equation set of the first solution task can be matched with the information of one or more sets of preset mathematical equation sets in the preset database to determine the information of the target preset mathematical equation set that matches the information of the mathematical equation set of the first solution task. The preset numerical calculation strategy corresponding to the information of the target preset mathematical equation set in the preset database is used as the first numerical calculation strategy of the mathematical equation set of the corresponding first solution task.

[0239] In some examples, a third machine learning model can also process the information of the mathematical equations for the first solution task to output a first numerical computation strategy for the mathematical equations of the first solution task. This third machine learning model can be trained on third training data from a pre-defined database, or it can be trained on other third-party data. For example, information about a pre-defined mathematical equation set can be obtained from a pre-defined database as third training data, and the pre-defined numerical computation strategy corresponding to the pre-defined mathematical equation set can be used as the label for the third training data to train the machine learning model. After training, a third machine learning model is obtained to determine the numerical computation strategy.

[0240] Alternatively, a third machine learning model and pre-defined database matching can be combined to determine the first numerical computation strategy for the first solution task.

[0241] For example, multiple candidate first numerical calculation strategies for the first solution task can be obtained by using a third machine learning model and a preset database matching, respectively. Then, based on the weights of the third machine learning model and the preset database matching, and / or the confidence levels of the multiple candidate first solution strategies, the first numerical calculation strategy can be determined from the multiple candidate first numerical calculation strategies for the first solution task.

[0242] As can be seen from the relevant embodiments of the auxiliary equation control module, coupling strategy control module and numerical calculation strategy control module mentioned above, step 403 may specifically include: the cloud management platform determines the first configuration information of the first solution process of the first solution task corresponding to the first physical equation through the machine learning model corresponding to the first physics discipline type and / or the preset database corresponding to the first physics discipline type.

[0243] The first configuration information may include one or more of the following: a first auxiliary equation, a first coupling strategy, and a first numerical calculation strategy. The preset database corresponding to the first physics discipline type includes one or more of the following information corresponding to the first physics discipline type: a preset auxiliary equation, a preset coupling strategy, and a preset numerical calculation strategy. The specific methods for determining different first configuration information through machine learning models and / or the preset database corresponding to the first physics discipline type can be found in the relevant embodiments of the auxiliary equation control module, coupling strategy control module, and numerical calculation strategy control module described above, and will not be repeated here.

[0244] The first numerical computation strategy is used to indicate the numerical computation method of the mathematical equation system contained in the first solution task. It can be seen that the solution form (e.g., the corresponding solution matrix) of the solution task is different for different coupling strategies; and the first coupling strategy is used to indicate the solution order and other solution methods of the first solution task.

[0245] The first numerical computation strategy can specifically adopt existing or future-developed numerical computation strategies.

[0246] The following examples illustrate some possible types of the first numerical computation strategy and exemplary determination principles.

[0247] For example, when the first solution task is a small to medium-sized solution task in the structural domain, mathematical solution strategies such as Cholesky decomposition and Gaussian elimination can be used. When the solution matrix of the mathematical equation system in the first solution task is a large sparse matrix, mathematical solution strategies such as the preconditional conjugate gradient method (PCG), generalized geometric / algebraic multigrid (GAMG), and generalized minimal residual algorithm (GMRES) can be used. The first solution task obtained by discretizing the first physical equation system with strong coupling between structure and fluid can be solved using mathematical solution strategies such as GMRES, field split preprocessor, and algebraic multigrid (AMG).

[0248] As can be seen, in the embodiments of this application, during the CAE simulation solution of physical field services, the solution process of the current time step can be dynamically adjusted by utilizing the configuration information and / or solution information of the solution process of the historical time step, thereby optimizing the performance of CAE simulation solution, ensuring high accuracy, efficiency and stability of CAE simulation solution of the current time step, shortening the time consumption of end-to-end CAE simulation solution, thereby accelerating CAE simulation solution and directly reducing the resource cost of CAE simulation solution.

[0249] 2. Based on the initial conditions corresponding to the current time step, determine the first configuration information of the first solution process for the current time step.

[0250] Specifically, in some embodiments, step 403 includes:

[0251] The cloud management platform determines the first configuration information based on the first physics subject type and the initial physical state corresponding to the first time step.

[0252] In this embodiment of the application, the first time step involved in the physics field service can be the initial time step in the CAE simulation solution process, for example, it can be... Figure 1The time step t0 in the time series shown. At this point, the first configuration information can be determined based on the initial physical state corresponding to this initial time step.

[0253] For example, the initial physical state can be a default physical state or a user-configured physical state, which can be described by the value of a variable. For instance, when the physics field involves fluids, the initial physical state indicates that the fluid velocity is 0, or a certain velocity preset by the user.

[0254] Then, the first configuration information can be determined based on the first physics subject type and the initial physical state corresponding to the first time step, using machine learning models and / or a preset database corresponding to the first physics subject type. The specific method for determining the first configuration information based on the first physics subject type and the initial physical state corresponding to the first time step using machine learning models and / or a preset database corresponding to the first physics subject type can be found in the descriptions of determining the first auxiliary equation, first coupling strategy, and / or first numerical calculation strategy in the first configuration information using machine learning models and / or a preset database corresponding to the first physics subject type in the relevant embodiments described above, and will not be repeated here.

[0255] For example, refer to Figure 6 The example shown is a schematic diagram illustrating an exemplary process for solving a CAE simulation task by using a physics discipline type identification module, an auxiliary equation control module, a coupling strategy control module, a numerical calculation strategy control module, or a management module to determine the solution process.

[0256] Figure 6 In this process, the flow control preference indication uses the physics discipline type identification module, auxiliary equation control module, coupling strategy control module, and numerical calculation strategy control module to determine the first configuration information. Then, the management module activates the functions of the physics discipline type identification module, auxiliary equation control module, coupling strategy control module, and numerical calculation strategy control module.

[0257] During the solution process at the first time step, the first physics discipline type of the first physical equation can be determined by the physics discipline type identification module, and this first physics discipline type is output to the auxiliary equation control module. Based on this first physics discipline type, the auxiliary equation control module determines the first auxiliary equation of the first physical equation, and then determines one or more sets of first physical equations based on the first auxiliary equation.

[0258] Then, the coupling strategy control module determines the first coupling strategy based on the first physics discipline type output from the physics discipline type identification module and one or more first physics equation sets from the auxiliary equation control module, thereby determining the mathematical equation sets contained in one or more first solution tasks and the solution order of each first solution task.

[0259] Next, the numerical computation strategy control module determines the numerical solution strategy for each first solution task based on the first physics subject type output from the physics subject type identification module and the mathematical equation set contained in one or more first solution tasks from the coupling strategy control module.

[0260] Then, the management module can obtain from the coupling strategy control module the mathematical equations contained in one or more first solution tasks indicated by the first coupling strategy, the solution order of each first solution task, and the numerical solution strategy of each first solution task output by the numerical calculation strategy control module, as the first configuration information and output to the CAE simulation solution service, so that the CAE simulation solution service can determine the first solution process of the first time step according to the first configuration information, and perform CAE simulation solution for the first time step through the calculation engine to obtain the solution result of the first time step.

[0261] In determining the first configuration information, the auxiliary equation control module, coupling strategy control module, and numerical calculation strategy control module can obtain simulation history from the CAE simulation solution service from the management module. They then combine this simulation history with relevant data from the preset database to execute their respective functions and obtain their output data. Specifically, the simulation history includes configuration information for the solution process at historical time steps and / or solution information for those historical time steps.

[0262] Thus, for reference Figure 1 As shown in the example, at each time step, the configuration information of the corresponding solution process can be determined by referring to the method of determining the first configuration information of the first time step. Then, according to the configuration information of the corresponding solution process, the corresponding solution process is executed sequentially at each time step to obtain the numerical solution of the corresponding solution task. Once the numerical solution of the current time step meets the corresponding accuracy requirements, the calculation of the next time step is entered until the solution of all time steps is completed and the final solution result is obtained.

[0263] In this embodiment, the cloud management platform can determine the first physics discipline type of the first physical equation, such as the fundamental physical equation. Based on this first physics discipline type, it intelligently determines one or more of the following: a first auxiliary equation, a first coupling strategy, and a first numerical solution strategy corresponding to the first physical equation. This obtains the first configuration information for the first solution process of the current physics field service, thereby configuring the set of mathematical equations and / or the solution method for each first solution task in the current physics field service to solve the first physical equation. Therefore, in this embodiment, the cloud management platform can intelligently determine one or more of the first auxiliary equation, the first coupling strategy, and the first numerical solution strategy corresponding to the first physical equation, thereby conveniently and efficiently determining the set of mathematical equations and / or the solution method for each first solution task for various physical scenarios, and solving the first physical equation with a suitable solution process.

[0264] The following example illustrates an exemplary implementation of determining the solution process.

[0265] Electromagnetic-fluid coupling problems, a multiphysics application, are commonly found in many complex engineering and scientific scenarios, such as electromagnetically driven fluid pumps, thermal convection in electromagnetic environments, and electromagnetic stirring of conductive fluids. In these scenarios, the electromagnetic field interacts with the flow field, forming a complex dynamic system. Therefore, studying electromagnetic-fluid coupling problems has significant scientific value and practical implications.

[0266] In solving electromagnetic-fluid coupling problems, at the current time step, the user can input the fundamental equations of the electromagnetic field and the fundamental equations of the flow field as the first physical equations.

[0267] The fundamental equations of electromagnetic fields are Maxwell's equations:

[0268] Current density, E is electric field strength, B is magnetic flux density, and ρ is free charge density.

[0269] Among them, among them, Let be the del operator, H be the magnetic field strength, J be the conduction current density, D be the electric displacement, and t be the time. displacement

[0270] The fundamental equations for the flow field are the Navier-Stokes equations:

[0271]

[0272] Where V is the fluid velocity, ρ is the fluid density, and σ is the fluid pressure.

[0273] The Maxwell and Navier-Stokes equations input by the user can be fed into the physics subject type recognition module. This module uses pre-stored information from a database to identify the first physics subject type for each equation: electromagnetic field and fluid field, respectively. Furthermore, in some examples, the module can identify and correct errors in the user-input fundamental equations, outputting complete and correct equations. Additionally, in some examples, the module can output the fundamental variables to be solved in the equations; for example, in the electromagnetic domain, this includes electric field strength, magnetic field strength, electric flux density, and magnetic flux density; in the fluid domain, it includes velocity, pressure, and temperature.

[0274] Then, the auxiliary equation control module can determine the auxiliary equation of the first physical equation based on the initial boundary conditions of the electromagnetic field and the flow field, the solution results of the historical time step, and the first physical discipline type output by the physics discipline type identification module, through the information of the preset auxiliary equations stored in the preset database, thereby obtaining one or more sets of first physical equations.

[0275] For example, in this example, the auxiliary equations of Maxwell's equations include those describing electromagnetic constitutive relations, used to describe the relationships between electric displacement vector and electric field strength, magnetic induction and magnetic field strength, and current density and electric field strength. The auxiliary equations of Navier-Stokes' equations include those describing fluid constitutive relations, used to describe Newtonian fluids, incompressible fluids, or ideal fluids. A first set of physical equations can be obtained from Maxwell's equations and their auxiliary equations, and another set can be obtained from Navier-Stokes' equations and their auxiliary equations. Furthermore, when the Reynolds number corresponding to the flow velocity is greater than 3200, a turbulence model can be obtained as a first set of physical equations. For example, a suitable turbulence model, such as Spalart-Allmaras, k-epsilon, or k-omega, can be selected based on the actual physical conditions.

[0276] After obtaining the first set of physical equations, the coupling strategy control module can determine the coupling methods between and within one or more sets of first physical equations based on the coupling logs of historical time steps and the first set of physical equations, using information from preset coupling strategies pre-stored in a preset database. This discretizes one or more first solution tasks, determines the mathematical equations contained in each first solution task, and determines the execution order of each first solution task. For example, the electromagnetic field is solved using the magnetic vector potential or magnetic scalar potential method, while the first set of physical equations corresponding to the flow field can be solved using either a coupled solution method or a separate solution method, depending on the situation. The coupling between multiple sets of first physical equations between the electromagnetic field and the flow field employs either a fully coupled method or a weakly coupled iterative method.

[0277] Next, the numerical calculation strategy for the mathematical equation set contained in each first solution task indicated by the first coupling strategy can be determined by the numerical calculation control module and the information of the preset numerical calculation strategy stored in the preset database, based on the numerical calculation log of the historical time step.

[0278] In this way, the management module can obtain the set of mathematical equations contained in the first solution task, as well as the solution method information such as the execution order and numerical calculation strategy of each first solution task, as the first configuration information. Based on the specified interface, the management module sends the first configuration information to the CAE simulation solution service, so that the CAE simulation solution service can determine the solution process of the current time step according to the first configuration information, so as to perform the solution and obtain the solution result.

[0279] In another example, a fluid-structure coupling problem is involved.

[0280] Fluid-structure coupling problems involve the dynamic phenomena of the interaction between a fluid and a solid structure. This coupling scenario is widely applicable in nature and engineering practice, such as vibrations caused by wind blowing across bridges, the interaction between an aircraft wing and airflow during flight, and the impact of waves on offshore platforms in marine engineering. In these cases, the flow of the fluid is not only affected by the shape and motion of the solid structure, but the dynamic action of the fluid also, in turn, affects the deformation and motion of the solid structure.

[0281] The configuration of the fundamental equations for fluids and the corresponding solution process can be found in the relevant examples of electromagnetic-fluid coupling problems mentioned above. The fundamental equations for structures include equilibrium differential equations, geometric equations, strain compatibility equations, etc., as shown below.

[0282]

[0283] The fundamental variables in the above basic equations include stress σ, strain ε, and displacement d.

[0284] The auxiliary equation control module determines the fundamental equations of the structure, including the physical equations describing the constitutive relations of the material (i.e., the relationship between stress σ and strain ε under stress). Material constitutive relations can be categorized based on material properties, and these relations often vary significantly between different materials. Common types include linear elastic constitutive relations, nonlinear elastic constitutive relations, ideal elastoplastic constitutive relations, linearly strengthening elastoplastic constitutive relations, and rigid-plastic constitutive relations. For linear elastic materials, the constitutive relation is typically:

[0285] σ=Dε

[0286] The automated control scheme for solving fluid-structure coupling problems is similar to that for electromagnetic-fluid coupling problems. For details, please refer to the automated control scheme for solving electromagnetic-fluid coupling problems. The specific difference lies in the data in the preset database. In fluid-structure coupling problems, it is necessary to use preset auxiliary equations, preset coupling strategies, and preset numerical calculation strategies related to the two physics disciplines of fluid and structure in the preset database.

[0287] The following provides an exemplary introduction to the pre-defined auxiliary equations, pre-defined coupling strategies, and pre-defined numerical calculation strategies involved in the structural physics discipline.

[0288] (1) The database is pre-stored with the basic equations of the structural field and the specific equations of actual engineering cases, which can automatically identify the corresponding physics discipline based on the basic equations of the structural field input by the user.

[0289] (2) In the pre-set database, auxiliary equations describing the constitutive relations of the structural domain are stored in advance to ensure that the corresponding constitutive relations can be selected according to the material properties of the computational domain. For example, different constitutive relations can be switched as the material changes. For example, the constitutive relations of each stage of the material from linear elasticity to elastoplasticity to plasticity and finally to fracture are different.

[0290] (3) In the pre-set database, for multi-physics coupling problems within structural mechanics, corresponding pre-set coupling strategies can be stored in advance. For example, the pre-set coupling strategy can instruct the deriving of physical equations about nodal displacements from equilibrium differential equations, geometric equations, strain compatibility equations, constitutive relations, etc. After solving for the nodal displacements, the solution tasks corresponding to other physical equations can be calculated to calculate the results of other basic variables. For the coupling between structure and fluid, the pre-set coupling strategy can instruct the processing of physical equations of the structural and fluid regions separately from the boundary surface. Then, the physical equations of the structural and fluid regions can be coupled into a strongly coupled equations to obtain the solution task. Alternatively, it can instruct the separation of the physical equations of the structural and fluid regions and use an external iteration for weak coupling.

[0291] (4) In the preset database, the matrix system properties and the corresponding preset numerical calculation strategies for various preset solution tasks are stored in advance. For example, small and medium-sized solution tasks in the structural domain can adopt preset numerical calculation strategies such as Cholesky decomposition and LU decomposition, while large sparse matrices can adopt preset numerical calculation strategies such as PCG, GAMG, and GMRES. Solution tasks obtained based on the strongly coupled physical equations between structure and fluid can adopt preset numerical calculation strategies such as GMRES and FieldSplit preprocessor, AMG.

[0292] The content of the aforementioned preset database is merely an example and limitation. For different physics disciplines, a large amount of physical field business data can be collected based on existing large-scale engineering cases and experience, and can be stored in the preset database according to different physics discipline types for the purpose of determining the configuration information of the solution process.

[0293] The above describes the solution process control method provided by the embodiments of this application from multiple aspects. The following, in conjunction with the accompanying drawings, describes the cloud management platform for solution process control provided by the embodiments of this application.

[0294] like Figure 7 As shown in the figure, this application embodiment provides a cloud management platform 70 for solving process control, the cloud management platform 70 including:

[0295] Interface module 701 is used to obtain the first physical equation for the physical field service from the equation input interface;

[0296] Processing module 702 is used for:

[0297] Based on the first physical equation, determine the first physics discipline type corresponding to the first physical equation;

[0298] Based on the first physics discipline type, determine the first configuration information corresponding to the first physical equation. The first configuration information includes one or more of the following information corresponding to the first physical equation: first auxiliary equation, first coupling strategy, first numerical calculation strategy. The first configuration information is used to determine the set of mathematical equations contained in the first solution task and / or the solution method of each first solution task. The first solution task is used to solve the first physical equation.

[0299] Optionally, the physics field service involves a first time step and a second time step, with the second time step preceding the first time step, and the first physics equation being the physics equation corresponding to the physics field service in the first time step.

[0300] The interface module 701 is used to: obtain the second configuration information corresponding to the second physical equation of the physical field service at the second time step, and / or the second solution information corresponding to the second physical equation. The second configuration information includes one or more of the following information corresponding to the second physical equation: second auxiliary equation, second coupling strategy, second numerical calculation strategy. The second solution information includes one or more of the following information: solution result of the second solution task, solution performance of the second solution task.

[0301] The processing module 702 is used to: determine the first configuration information based on the first physics subject type, and the second configuration information and / or the second solution information.

[0302] Optionally, the first configuration information includes a first set of physical equations determined based on the first physical equation and the first auxiliary equation, and the second solution information includes the solution results of the second solution task;

[0303] Processing module 702 is used for:

[0304] Based on the physical state reflected by the solution results of the second solution task and the type of the first physics discipline, determine the first auxiliary equation;

[0305] Based on the first physical equation and the first auxiliary equation, determine the first set of physical equations.

[0306] Optionally, the physics field service is a multiphysics field service, with multiple first physics equations and multiple second physics equations. The first configuration information includes the first coupling strategy corresponding to each of the multiple first physics equations, the second configuration information includes the second coupling strategy corresponding to each of the multiple second physics equations, and the solution performance in the second solution information includes the coupling performance information corresponding to the second solution task using the second coupling strategy for solution.

[0307] Interface module 701 is used to: obtain multiple sets of first physical equations based on multiple first physical equations;

[0308] Processing module 702 is used for:

[0309] If among multiple sets of second physical equations, there exists a set of second physical equations that is identical to one or more sets of first physical equations, then based on the second coupling strategy and coupling performance information corresponding to the identical set of second physical equations, the first coupling strategy corresponding to one or more sets of first physical equations is determined.

[0310] The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more first physical equations based on the first coupling strategy.

[0311] Optionally, the physics field service is a multiphysics field service, with multiple first physics equations and multiple second physics equations. The first configuration information includes the first coupling strategy corresponding to each of the multiple first physics equations, and the second solution process configuration information includes the solution results of the second solution task.

[0312] Interface module 701 is used to: obtain multiple sets of first physical equations based on multiple first physical equations;

[0313] Processing module 702 is used for:

[0314] If there is no second physical equation set that is the same as one or more first physical equation sets among multiple second physical equation sets, then the first coupling strategy corresponding to one or more first physical equation sets is determined based on the physical state reflected by the solution results of the second solution task.

[0315] The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more first physical equations based on the first coupling strategy.

[0316] Optionally, the first configuration information includes a first numerical calculation strategy corresponding to the first physical equation, and the second solution information includes the solution performance of the second solution task;

[0317] Interface module 701 is used to: obtain a first solution task, the first solution task including a set of mathematical equations obtained based on a first physical equation;

[0318] The processing module 702 is used to: if the second solving task matches the first solving task, determine the first numerical calculation strategy for the set of mathematical equations contained in the first solving task based on the solving performance of the second solving task.

[0319] Optionally, the physics field service involves the first time step, and the first physics equation is the physics equation corresponding to the physics field service at the first time step;

[0320] The processing module 702 is used to: determine the first configuration information based on the first physics subject type and the initial physical state corresponding to the first time step.

[0321] Optionally, the processing module 702 is used to: determine the first configuration information through the machine learning model corresponding to the first physics subject type and / or the preset database corresponding to the first physics subject type, wherein the preset database corresponding to the first physics subject type includes one or more of the following information corresponding to the first physics subject type: preset auxiliary equation, preset coupling strategy, and preset numerical calculation strategy.

[0322] Both the processing module and the interface module can be implemented in software or hardware. For example, the implementation of the processing module will be described below. Similarly, the implementation of the interface module can be referenced from that of the processing module.

[0323] As an example of a software functional unit, a processing module may include code running on a computing instance. A computing instance may include at least one of a physical host (computing device), a virtual machine, or a container. Furthermore, the aforementioned computing instance may be one or more. For example, a processing module may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed within the same region or in different regions. Further, the multiple hosts / virtual machines / containers used to run the code may be distributed within the same availability zone (AZ) or in different AZs, each AZ comprising one or more geographically proximate data centers. Typically, a region may include multiple AZs.

[0324] Similarly, multiple hosts / virtual machines / containers used to run this code can be distributed within the same Virtual Private Cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Communication between two VPCs within the same region, as well as between VPCs in different regions, requires a communication gateway to be set up within each VPC to enable interconnection between VPCs.

[0325] As an example of a hardware functional unit, a processing module may include at least one computing device, such as a server. Alternatively, a processing module may be implemented using a central processing unit (CPU), an application-specific integrated circuit (ASIC), or a programmable logic device (PLD). The aforementioned PLD may be implemented using a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), a data processing unit (DPU), a neural network processing unit (NPU), a system-on-chip (SoC), an offload card, an accelerator card, or any combination thereof.

[0326] The processing module comprises multiple computing devices that can be distributed within the same region or in different regions. Similarly, the processing module can be distributed within the same Availability Zone (AZ) or in different AZs. Likewise, the processing module can be distributed within the same Virtual Private Cloud (VPC) or multiple VPCs. These computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, GALs, DPUs, NPUs, SoCs, offloading cards, and accelerator cards.

[0327] It should be noted that, in other embodiments, the processing module can be used to execute any step in the solution process control method, and the interface module can be used to execute any step in the solution process control method. The steps implemented by the processing module and the interface module can be specified as needed. By implementing different steps in the solution process control method through the processing module and the interface module respectively, all functions of the cloud management platform for solution process control can be realized.

[0328] This application also provides a computing device 80. For example... Figure 8As shown, the computing device 80 includes a bus 82, a processor 84, a memory 86, and a communication interface 88. The processor 84, the memory 86, and the communication interface 88 communicate with each other via the bus 82. The computing device 80 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in the computing device 80.

[0329] Bus 82 can be a Peripheral Component Interconnect Express (PCIe) bus, an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (Ubus or UB), a Compute Express Link (CXL) bus, a Cache Coherent Interconnect for Accelerators (CCIX) bus, etc. The Unified Bus is also known as the Lingqu Bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 8 The bus 84 is represented by a single line, but this does not mean that there is only one bus or one type of bus. The bus 84 may include a path for transmitting information between various components of the computing device 80 (e.g., memory 86, processor 84, communication interface 88).

[0330] The processor 84 may include any one or more computing devices such as a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MP) or a digital signal processor (DSP), an ASIC, an FPGA, a CPLD, an NPU, a SoC, an offload card, or an accelerator card.

[0331] Memory 86 may include volatile memory, such as random access memory (RAM). Memory 86 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD) or one or more of these. Furthermore, memory 86 may also be implemented using storage class memory (SCM), phase change memory (PCM), or other types of storage media.

[0332] It is worth noting that the same type of storage medium can be configured in the same computing device to realize the function of memory 86, or two or more types of storage media can be configured to realize the function of memory 86. This application does not limit this.

[0333] The memory 86 stores executable program code, and the processor 84 executes the executable program code to implement the functions of the aforementioned interface module and processing module, thereby realizing the solution process control method applied to the cloud management platform in the above embodiments. That is, the memory 86 stores instructions for executing the solution process control method applied to the cloud management platform in the above embodiments.

[0334] The communication interface 88 uses transceiver modules, such as, but not limited to, network interface cards and transceivers, to enable communication between the computing device 80 and other devices or communication networks.

[0335] This application also provides a computing device cluster. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0336] like Figure 9 As shown, the computing device cluster includes at least one computing device 80. The memory 86 of one or more computing devices 80 in the computing device cluster may store the same instructions for executing the solution flow control method.

[0337] In some possible implementations, the memory 86 of one or more computing devices 80 in the computing device cluster may also store partial instructions for executing the solution flow control method. In other words, a combination of one or more computing devices 80 can jointly execute the instructions for executing the solution flow control method.

[0338] It should be noted that the memory 86 in different computing devices 80 within the computing device cluster can store different instructions, each used to execute a portion of the functions of the solution flow control method. That is, the instructions stored in the memory 86 of different computing devices 80 can implement the functions of one or more modules within the interface module and processing module.

[0339] In some possible implementations, one or more computing devices in a computing device cluster can be connected via a network. This network can be a wide area network (WAN), a local area network (LAN), or similar. Figure 10 One possible implementation is shown. For example... Figure 10 As shown, two computing devices 80A and 80B are connected via a network. Specifically, they are connected to the network through communication interfaces in each computing device. In this type of possible implementation, the memory 86 in computing device 80A may store instructions for executing the functions of the interface module. Meanwhile, the memory 86 in computing device 80B may store instructions for executing the functions of the processing module. Alternatively, in other examples, the memory 86 in computing device 80A may store instructions for executing part of the functions of the processing module. Meanwhile, the memory 86 in computing device 80B may store instructions for executing another part of the functions of the processing module.

[0340] It should be understood that Figure 10 The functions of the computing device 80A shown can also be performed by multiple computing devices 80. Similarly, the functions of the computing device 80B can also be performed by multiple computing devices 80.

[0341] This application also provides another computing device cluster. The connection relationships between the computing devices in this computing device cluster can be similarly referred to... Figure 9 and Figure 10 The connection method of the computing device cluster. The difference is that the memory 86 of one or more computing devices 80 in the computing device cluster can store the same instructions for executing the solution flow control method.

[0342] In some possible implementations, the memory 86 of one or more computing devices 80 in the computing device cluster may also store partial instructions for executing the solution flow control method. In other words, a combination of one or more computing devices 80 can jointly execute the instructions for executing the solution flow control method.

[0343] It should be noted that the memory 86 in different computing devices 80 within the computing device cluster can store different instructions for executing parts of the solution flow control method. That is, the instructions stored in the memory 86 of different computing devices 80 can implement the functions of one or more modules in the interface module and processing module.

[0344] This application also provides a computer program product containing instructions. The computer program product may be software or program products containing instructions, capable of running on a computing device or stored on any usable medium. When the computer program product runs on at least one computing device, it causes the at least one computing device to execute a solution flow control method.

[0345] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center that includes one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute a solution flow control method.

[0346] This application also provides a chip system including a processor for implementing the steps performed by the aforementioned computing device cluster. In one possible design, the chip system may further include a memory for storing necessary program instructions and data. This chip system may be composed of chips or may include chips and other discrete devices.

[0347] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0348] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0349] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0350] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0351] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for solving flow control, characterized in that, The method is executed by a cloud management platform, which manages the infrastructure providing cloud services. This infrastructure includes multiple servers, which are used to deploy virtual instances implementing the solution process. The method includes: The cloud management platform obtains the first physical equation for the physical field service from the equation input interface; The cloud management platform determines the first physics discipline type corresponding to the first physics equation based on the first physics equation. The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type. The first configuration information includes one or more of the following information corresponding to the first physical equation: first auxiliary equation, first coupling strategy, and first numerical calculation strategy. The first configuration information is used to determine the set of mathematical equations contained in the first solution task and / or the solution method of each first solution task. The first solution task is used to solve the first physical equation.

2. The method according to claim 1, characterized in that, The physical field service involves a first time step and a second time step, the second time step being prior to the first time step, and the first physical equation being the physical equation corresponding to the physical field service at the first time step. The method further includes: The cloud management platform obtains the second configuration information corresponding to the second physical equation of the physical field service at the second time step, and / or the second solution information corresponding to the second physical equation. The second configuration information includes one or more of the following information corresponding to the second physical equation: second auxiliary equation, second coupling strategy, second numerical calculation strategy. The second solution information includes one or more of the following information: solution result of the second solution task, solution performance of the second solution task. The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: The cloud management platform determines the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information.

3. The method according to claim 2, characterized in that, The first configuration information includes a first set of physical equations determined based on the first physical equation and the first auxiliary equation, and the second solution information includes the solution results of the second solution task; The cloud management platform determines the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information, including: The cloud management platform determines the first auxiliary equation based on the physical state reflected by the solution result of the second solution task and the first physics discipline type; The cloud management platform determines the first set of physical equations based on the first physical equation and the first auxiliary equation.

4. The method according to claim 2 or 3, characterized in that, The physical field service is a multi-physics service. There are multiple first physical equations and multiple second physical equations. The first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations. The second configuration information includes the second coupling strategy corresponding to each of the multiple second physical equations. The solution performance in the second solution information includes the coupling performance information corresponding to the second solution task using the second coupling strategy for solution. The method further includes: The cloud management platform acquires multiple sets of first physical equations based on multiple first physical equations; The cloud management platform determines the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information, including: If among the multiple sets of second physical equations, there exists a set of second physical equations that is identical to one or more sets of first physical equations, then the cloud management platform determines a first coupling strategy corresponding to one or more sets of first physical equations based on the second coupling strategy corresponding to the identical sets of second physical equations and the coupling performance information. The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more of the first physical equations based on the first coupling strategy.

5. The method according to claim 2 or 3, characterized in that, The physical field service is a multi-physics service, the number of first physical equations is multiple, the number of second physical equations is multiple, the first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations, and the second solution process configuration information includes the solution result of the second solution task. The method further includes: The cloud management platform acquires multiple sets of first physical equations based on multiple first physical equations; The cloud management platform determines the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information, including: If none of the multiple sets of second physical equations are the same as one or more sets of first physical equations, then the cloud management platform determines a first coupling strategy corresponding to one or more sets of first physical equations based on the physical state reflected by the solution results of the second solving task. The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more of the first physical equations based on the first coupling strategy.

6. The method according to any one of claims 2-5, characterized in that, The first configuration information includes the first numerical calculation strategy corresponding to the first physical equation, and the second solution information includes the solution performance of the second solution task; The method further includes: The cloud management platform obtains the first solution task, which includes a set of mathematical equations based on the first physical equation. The cloud management platform determines the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information, including: If the second solution task matches the first solution task, the cloud management platform determines the first numerical calculation strategy for the set of mathematical equations contained in the first solution task based on the solution performance of the second solution task.

7. The method according to claim 1, characterized in that, The physical field service involves a first time step, and the first physical equation is the physical equation corresponding to the physical field service at the first time step. The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: The cloud management platform determines the first configuration information based on the first physics subject type and the initial physical state corresponding to the first time step.

8. The method according to any one of claims 1-7, characterized in that, The cloud management platform determines the first configuration information corresponding to the first physical equation based on the first physics discipline type, including: The cloud management platform determines the first configuration information through the machine learning model corresponding to the first physics subject type and / or the preset database corresponding to the first physics subject type. The preset database corresponding to the first physics subject type includes one or more of the following information corresponding to the first physics subject type: preset auxiliary equations, preset coupling strategies, and preset numerical calculation strategies.

9. A cloud management platform for solving process control problems, characterized in that, include: The interface module is used to obtain the first physical equations for the physics field service from the equation input interface; Processing module, used for: Based on the first physical equation, determine the first physics discipline type corresponding to the first physical equation; Based on the first physics discipline type, determine the first configuration information corresponding to the first physical equation. The first configuration information includes one or more of the following information corresponding to the first physical equation: first auxiliary equation, first coupling strategy, first numerical calculation strategy. The first configuration information is used to determine the set of mathematical equations contained in the first solving task and / or the solution method of each of the first solving tasks. The first solving task is used to solve the first physical equation.

10. The cloud management platform according to claim 9, characterized in that, The physical field service involves a first time step and a second time step, the second time step being located before the first time step, and the first physical equation being the physical equation corresponding to the physical field service at the first time step. The interface module is used to: obtain the second configuration information corresponding to the second physical equation of the physical field service at the second time step, and / or the second solution information corresponding to the second physical equation. The second configuration information includes one or more of the following information corresponding to the second physical equation: second auxiliary equation, second coupling strategy, second numerical calculation strategy. The second solution information includes one or more of the following information: solution result of the second solution task, solution performance of the second solution task. The processing module is used to: determine the first configuration information based on the first physics subject type, the second configuration information, and / or the second solution information.

11. The cloud management platform according to claim 10, characterized in that, The first configuration information includes a first set of physical equations determined based on the first physical equation and the first auxiliary equation, and the second solution information includes the solution results of the second solution task; The processing module is used for: The first auxiliary equation is determined based on the physical state reflected by the solution result of the second solution task and the first physics discipline type. The first set of physical equations is determined based on the first physical equation and the first auxiliary equation.

12. The cloud management platform according to claim 10 or 11, characterized in that, The physical field service is a multi-physics service. There are multiple first physical equations and multiple second physical equations. The first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations. The second configuration information includes the second coupling strategy corresponding to each of the multiple second physical equations. The solution performance in the second solution information includes the coupling performance information corresponding to the second solution task using the second coupling strategy for solution. The interface module is used to: obtain a set of multiple first physical equations based on multiple first physical equations; The processing module is used for: If among the multiple sets of second physical equations, there exists a set of second physical equations that is the same as one or more sets of first physical equations, then based on the second coupling strategy corresponding to the same set of second physical equations and the coupling performance information, a first coupling strategy corresponding to one or more sets of first physical equations is determined. The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more of the first physical equations based on the first coupling strategy.

13. The cloud management platform according to claim 10 or 11, characterized in that, The physical field service is a multi-physics service, the number of first physical equations is multiple, the number of second physical equations is multiple, the first configuration information includes the first coupling strategy corresponding to each of the multiple first physical equations, and the second solution process configuration information includes the solution result of the second solution task. The interface module is used to: obtain a set of multiple first physical equations based on multiple first physical equations; The processing module is used for: If none of the multiple sets of second physical equations are the same as one or more sets of first physical equations, then a first coupling strategy corresponding to one or more sets of first physical equations is determined based on the physical state reflected by the solution result of the second solution task. The first coupling strategy is used to indicate the set of mathematical equations contained in the first solution task and the solution order of the first solution task. The set of mathematical equations contained in the first solution task is obtained by discretizing one or more of the first physical equations based on the first coupling strategy.

14. The cloud management platform according to any one of claims 10-13, characterized in that, The first configuration information includes the first numerical calculation strategy corresponding to the first physical equation, and the second solution information includes the solution performance of the second solution task; The interface module is used to: obtain the first solution task, wherein the first solution task includes a set of mathematical equations obtained based on the first physical equation; The processing module is configured to: if the second solving task matches the first solving task, determine a first numerical calculation strategy for the set of mathematical equations contained in the first solving task based on the solving performance of the second solving task.

15. The cloud management platform according to claim 9, characterized in that, The physical field service involves a first time step, and the first physical equation is the physical equation corresponding to the physical field service at the first time step. The processing module is used to: determine the first configuration information based on the first physics subject type and the initial physical state corresponding to the first time step.

16. The cloud management platform according to any one of claims 9-15, characterized in that, The processing module is used to: determine the first configuration information through the machine learning model corresponding to the first physics subject type and / or the preset database corresponding to the first physics subject type. The preset database corresponding to the first physics subject type includes one or more of the following information corresponding to the first physics subject type: preset auxiliary equation, preset coupling strategy, and preset numerical calculation strategy.

17. A computing device cluster, characterized in that, It includes at least one computing device, said at least one computing device including a processor and a memory; The processor is configured to execute instructions stored in the memory to cause the computing device cluster to perform the method as described in any one of claims 1-8.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when run on a processor, causes the processor to perform the method as described in any one of claims 1-8.

19. A computer program product containing instructions, characterized in that, When the instructions are executed by the processor, the method described in any one of claims 1-8 is implemented.

Images