CN117608963A - Control method of intelligent self-monitoring computational fluid dynamics simulation solving system - Google Patents
Control method of intelligent self-monitoring computational fluid dynamics simulation solving system Download PDFInfo
- Publication number
- CN117608963A CN117608963A CN202311595250.XA CN202311595250A CN117608963A CN 117608963 A CN117608963 A CN 117608963A CN 202311595250 A CN202311595250 A CN 202311595250A CN 117608963 A CN117608963 A CN 117608963A
- Authority
- CN
- China
- Prior art keywords
- residual
- convergence
- solver
- solution
- physical quantity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012544 monitoring process Methods 0.000 title claims abstract description 53
- 238000004088 simulation Methods 0.000 title claims abstract description 47
- 239000012530 fluid Substances 0.000 title claims abstract description 38
- 238000004364 calculation method Methods 0.000 claims abstract description 37
- 230000008569 process Effects 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims description 20
- 230000010355 oscillation Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 238000002372 labelling Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 4
- 230000006399 behavior Effects 0.000 claims description 3
- 230000003534 oscillatory effect Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/302—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a software system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3055—Monitoring arrangements for monitoring the status of the computing system or of the computing system component, e.g. monitoring if the computing system is on, off, available, not available
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Analysis (AREA)
- Computer Hardware Design (AREA)
- Fluid Mechanics (AREA)
- Geometry (AREA)
- Algebra (AREA)
- Mathematical Optimization (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a control method of an intelligent self-monitoring computational fluid dynamics simulation solving system, wherein the computational fluid dynamics simulation solving system at least comprises a solver, and the method comprises the steps of enabling the solver to iteratively solve numerical solutions of physical quantities in a computational domain flow field and residual values of the physical quantities according to a preset solving scheme; generating residual curve graphs according to the second step physical quantity based on the residual data sent by the solver, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value; inputting the residual curve graph into a pre-trained convergence feature recognition model aiming at each generated residual curve graph, and obtaining a convergence recognition result corresponding to the residual curve graph; and controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph, and improving the autonomous monitoring management capability of solving calculation so as to reduce the dependence of the solving calculation process on manual monitoring.
Description
Technical Field
The application relates to the technical field of simulation, in particular to a control method of an intelligent self-monitoring computational fluid dynamics simulation solving system.
Background
In the existing CFD (Computational Fluid Dynamics ) simulation solving calculation, the residual curve change is required to be observed manually, whether the solving calculation is converged or not is judged, when the solving calculation is not converged, the solving calculation is required to be stopped manually, and after the solving scheme is adjusted, the solver is restarted. The CFD is usually long in one-time solving and calculating, the time is as long as several hours or even tens of hours, and the manual whole-process monitoring and solving operation efficiency is low. Particularly, when a plurality of states are required to be calculated and a plurality of solving operations are started to carry out batch solving calculation, it is difficult for a person to monitor the convergence condition of the residual curve of each solving operation task at the same time.
Disclosure of Invention
In view of this, the present application aims to provide a control method for an intelligent self-monitoring computational fluid dynamics simulation solving system, so as to automatically monitor and manage the simulation solving calculation process, and reduce the dependence on manual solving monitoring management.
In a first aspect, the present application provides a control method of an intelligent self-monitoring computational fluid dynamics simulation solving system, where the computational fluid dynamics simulation solving system at least includes a solver, and the method includes that according to a preset solving scheme, the solver iteratively solves a numerical solution of each physical quantity in a computational domain flow field and a residual value of each physical quantity according to a first step length; generating residual curve graphs according to the second step physical quantity based on the residual data sent by the solver, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value; inputting the residual curve graph into a pre-trained convergence feature recognition model aiming at each generated residual curve graph, and obtaining a convergence recognition result corresponding to the residual curve graph; and controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph.
Further, a residual plot is generated by: for each physical quantity, generating a residual function corresponding to the physical quantity parameter by taking the iteration step number as an abscissa and taking the residual value of the physical quantity parameter as an ordinate; the physical quantity generates a residual curve graph based on residual functions corresponding to all physical quantity parameters.
Further, a converging feature recognition model is generated by training in the following way: collecting sample residual graphs with non-convergence characteristics, wherein each sample residual graph is drawn with a residual function corresponding to at least one physical quantity parameter; performing feature labeling on the sample residual graph to obtain a residual convergence feature labeling data set, wherein the feature label comprises divergent non-convergence features and oscillation non-convergence features; and training the residual error convergence characteristic recognition model based on the residual error convergence characteristic labeling data set to obtain a convergence characteristic recognition model.
Further, the operating state of the solver is controlled by: determining whether the simulation solving and calculating process is converged according to the convergence characteristic recognition result; if the solution is not converged, stopping the operation of the solver, adjusting the solution according to the adjustment strategy, and restarting the solver to perform iterative solution calculation; if so, the solver is kept running continuously.
Further, judging whether the solving and calculating process is converged or not according to the convergence characteristic recognition result; if the convergence feature identification result does not comprise any one of the divergent non-convergence feature and the oscillation non-convergence feature, solving the convergence of the calculation process; if the convergence feature recognition result comprises any one or combination of the divergent non-convergence feature and the oscillation non-convergence feature, the calculation process is solved to be non-convergence.
Preferably, the solver is solution tuned by: selecting one of higher quality mesh data, reducing the kurron CFL, reducing the relaxation factor, optimizing the boundary condition setting is performed according to the priority.
Preferably, the method further comprises the steps of recording all executed monitoring management actions, wherein the monitoring management actions comprise monitoring time, identified non-convergence characteristics and executed solution scheme adjustment strategies; and forming and displaying a monitoring management log.
In a second aspect, the present application provides an intelligent self-monitoring computational fluid dynamics simulation solution system, the system comprising:
the solver is used for carrying out iterative solution according to a preset solution scheme and a first step length so as to solve the numerical solution of each physical quantity in the computational domain flow field and the residual value of each physical quantity;
the solving controller generates residual curve graphs based on the residual data sent by the solver according to the second step length, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value; controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph;
the convergence characteristic recognition model is used for outputting a convergence characteristic recognition result corresponding to each input residual curve graph and sending the convergence characteristic recognition result to the solving controller;
and the solution scheme adjuster is used for adjusting the solution scheme according to the adjustment strategy when the solution process is not converged.
In a third aspect, the present application further provides an electronic device, including: the system comprises a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, when the electronic equipment runs, the processor and the memory are communicated through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the intelligent self-monitoring computational fluid mechanics simulation solving system control method.
In a fourth aspect, the present application further provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of a method for controlling an intelligent self-monitoring computational fluid dynamics simulation solution system as described above.
The control method of the intelligent self-monitoring computational fluid dynamics simulation solving system comprises the steps that according to a preset solving scheme, the solving device iteratively solves the numerical solution of each physical quantity in a computational domain flow field and the residual value of each physical quantity; generating residual curve graphs according to the second step physical quantity based on the residual data sent by the solver, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value; inputting the residual curve graph into a pre-trained convergence feature recognition model aiming at each generated residual curve graph, and obtaining a convergence recognition result corresponding to the residual curve graph; and controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph. And whether the iterative solution calculation is converged or not is automatically identified by a convergence characteristic identification module based on deep learning, and when the iterative solution calculation is not converged, the solution scheme is adjusted according to an adjustment strategy, so that the automatic monitoring management of the CFD simulation solution system is realized, and the dependence of the simulation solution calculation on manual monitoring management is reduced.
In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
For a clearer description of the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and should therefore not be considered limiting in scope, and that other related drawings can be obtained according to these drawings without the inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an intelligent self-monitoring computational fluid dynamics simulation solution system according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating steps for controlling a solver according to an embodiment of the present application;
FIG. 3 is a schematic diagram of another intelligent self-monitoring computational fluid dynamics simulation solution system according to an embodiment of the present application;
FIG. 4 is a flow chart of a control method of an intelligent self-monitoring computational fluid dynamics simulation solution system according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment that a person skilled in the art would obtain without making any inventive effort is within the scope of protection of the present application.
First, application scenarios applicable to the present application will be described. The method and the device can be applied to the CFD simulation system.
In the use of the CFD simulation system, a user needs to set a fluid solving scheme according to the simulation purpose and the solving working condition, specifically including setting the adopted calculation domain grid data, the applicable physical model, the boundary condition, the solving control parameters (the Brownian number and the relaxation factor) and the like, and then starting a solver to carry out solving calculation.
Because of the iterative solution adopted by the CFD solver, there may be a problem of misconvergence in solving the computation when the fluid solution is set unreasonably. If the setting of physical model parameters, boundary conditions and the like related to a physical actual scene is unreasonable, calculation is often not converged. Too large a time step for transient calculations may also cause the calculations to not converge. When the solution calculation is not converged, the current solution calculation needs to be terminated, and the solver is restarted after the corresponding solution scheme parameters are adjusted. At present, after the calculation of a solver is started, the shape of a residual curve in the solving process is observed manually to judge whether the solving calculation is converged, when the situations of divergence, oscillation and the like of the residual curve occur, the current solving calculation is stopped, and the solver is restarted after the corresponding solving scheme is adjusted. However, the manual whole-process monitoring solution has low operation efficiency, and particularly when a plurality of solution operations are operated simultaneously, the plurality of solvers are difficult to monitor at the same time by manpower.
Based on the above, the embodiment of the application provides a control method of an intelligent self-monitoring computational fluid dynamics simulation solving system.
Referring to fig. 1, fig. 1 is a schematic structural diagram of an intelligent self-monitoring computational fluid dynamics simulation solving system according to an embodiment of the present application. As shown in fig. 1, an intelligent self-monitoring computational fluid dynamics simulation solution system provided in an embodiment of the present application at least includes:
the solver 10 is configured to iteratively solve according to a preset solution scheme according to a first step to solve a numerical solution of each physical quantity in the computational domain flow field, and calculate a residual value of each physical quantity in a statistical manner, where a manner of calculating the residual is not limited.
The solution scheme is determined by a user according to the actual needs of an industrial project, and needs to determine computational domain grid division, a physical model, boundary conditions, solution control parameters (including a first step length) and the like corresponding to the simulation project, and then select a corresponding numerical solution calculation mode to obtain fluid field simulation data of the computational domain, wherein the numerical solution generally adopts an iterative solution.
The physical quantity may be one or more physical parameters, specifically, a pressure value, a speed value, a temperature value, and the like.
The solving controller 20 is configured to generate residual graphs based on the residual data sent by the solver according to the second step size, where an abscissa of each residual graph is the iteration step number and an ordinate is the corresponding physical quantity residual value; and controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph.
After the solver 10 starts to operate, a numerical solution and a statistical residual value of each physical quantity are output to the controller according to a first step, and the solver controller 20 generates a residual curve chart according to a second step and sends the residual curve chart to the convergence feature recognition model.
Specifically, solution controller 20 may generate a residual plot by:
for each physical quantity, generating a residual function corresponding to the physical quantity parameter by taking the iteration step number as an abscissa and taking a residual value corresponding to the physical quantity parameter as an ordinate; and generating a residual curve graph based on residual functions corresponding to all the physical quantity parameters.
The solving controller 20 receives the residual data output by the solver, constructs a residual function, and the abscissa of the residual function is the number of solving steps and the ordinate is the corresponding residual value. The residual functions corresponding to all physical quantities may all be plotted in the same residual graph for convergence detection.
The convergence feature recognition model 30 is configured to output, for each input residual graph, a convergence feature recognition result corresponding to the residual graph, and send the convergence feature recognition result to the solution controller.
The solution regulator 40 is configured to perform solution adjustment according to an adjustment policy when the solution process is not converged.
Specifically, the convergence feature recognition model may be generated by training in the following manner:
collecting sample residual graphs with non-convergence characteristics, wherein each sample residual graph is drawn with a residual function corresponding to at least one physical quantity parameter; performing feature labeling on the sample residual graph to obtain a residual convergence feature labeling data set, wherein the feature label comprises divergent non-convergence features and oscillatory non-convergence features; and training the residual error convergence characteristic recognition model based on the residual error convergence characteristic labeling data set to obtain a convergence characteristic recognition model.
Plotted in the sample residual plot herein may be a residual function of two non-converging characteristics of either divergent or oscillatory type.
The initial convergence feature identification model herein may employ a VGG16 model. Randomly selecting 4/5 data from the labeling data set as training sample data, training to obtain a convergence feature recognition model, and evaluating the model accuracy by using the rest 1/5 data. After the model training is finished, whether the accuracy of the obtained model is more than 95% is evaluated, if not, the model is adjusted and the training is continued, and if yes, the obtained model is saved to be used as a convergence feature recognition model.
The input of the convergence feature recognition model 30 is a residual curve graph corresponding to the number of solving steps, and the output result is any one or a combination of null, divergent non-convergence features and oscillation non-convergence features.
The recognition result output by the converging characteristic recognition model 30 may be fed back to the solution controller 20 and the solution regulator 40. Referring to fig. 2, fig. 2 is a flowchart illustrating steps for controlling a solver according to an embodiment of the present application. Solution controller 20 controls the operating state of the solver by:
s100, determining whether the current solving and calculating process is converged or not according to the residual curve convergence characteristic recognition result.
Here, whether the solution calculation process converges or not may be determined based on the convergence feature recognition result. If the convergence feature recognition result does not comprise any one or combination of the divergent non-convergence feature tag and the oscillation non-convergence feature tag, solving the convergence of the calculation process. If the convergence feature recognition result comprises any one or combination of the divergent non-convergence feature and the oscillation non-convergence feature, the calculation process is solved to be non-convergence.
And S101, if the solution is not converged, the control solver stops, the solution scheme is adjusted according to the adjustment strategy, and then the solver is restarted to perform iterative solution calculation.
When it is determined that the residual graph corresponding to the current solution step number has a non-convergence characteristic, the solution adjustment of the solution to the solver 10 may be performed by the solution adjustor 40. Specifically, the solution adjustor 40 performs solution adjustment on the solver by: selecting one of higher quality mesh data, reducing the kurron CFL, reducing the relaxation factor, optimizing the boundary condition setting is performed according to the priority.
In an actual application scene, if a plurality of grid data schemes exist in the current simulation project, grid data with finer grid division and more attached grid shapes and higher quality are preferentially selected. If the grid data no longer has modification space, consider determining if the kurrow number (CFL) is greater than 0.2, if so, the CFL number may be reduced. Second, it may be considered to determine if the slack is greater than 0.1, if so, the slack may be reduced. Finally, it may be considered whether the boundary conditions of the detection arrangement follow the principles of mass, momentum or energy conservation, e.g. in boundary conditions, the outlet pressure value is generally smaller than the inlet pressure value, and if it is detected that the outlet pressure value is greater than the inlet pressure value, a confirmation interface may be displayed to the user to confirm whether the boundary condition settings are accurate.
And S102, if the convergence is carried out, keeping the solver to normally continue to operate.
In step S102, if the solution calculation process converges, no adjustment is required, and the normal operation of the solver 10 is maintained.
According to the control method of the computational fluid dynamics simulation solving system, whether the iterative solution is converged or not is automatically recognized through the convergence feature recognition module based on deep learning, and when the iterative solution is not converged, the solving scheme is adjusted according to the scheme adjustment strategy, so that automatic monitoring management optimization of the CFD simulation solving system is realized, and dependence on manual monitoring management is reduced.
In one embodiment of the present application, the controller 20 is further configured to record all monitoring management actions performed, including: monitoring time, identifying non-convergence characteristics, and adjusting a strategy by the executed solution scheme to form a monitoring management log.
The monitoring management log can be provided for a user to analyze so as to optimize an adjustment strategy, and the intelligent management level of the operation control of the computational fluid dynamics simulation solving system is improved.
Referring to fig. 3, fig. 3 is a schematic structural diagram of another intelligent self-monitoring computational fluid dynamics simulation solving system according to an embodiment of the present application. In one embodiment of the present application, the intelligent self-monitoring computational fluid dynamics simulation solution system further includes a post-processing module 50 for displaying the solution results output by the solver 10 in the form of a visual image, in response to a user operation, and the like.
The computational fluid dynamics simulation system further comprises a GUI module 60, configured to provide a graphical user interaction interface for a user, and implement a user interaction function, so that the user performs operations such as grid setting, physical model setting, and the like, and displays output information of a solution computation process, a residual curve, a monitoring management log, and the like for the user.
Referring to fig. 4, fig. 4 is a flowchart of a control method of an intelligent self-monitoring computational fluid dynamics simulation solving system according to an embodiment of the present application. In one embodiment of the present application, there is further provided an intelligent self-monitoring computational fluid dynamics simulation solution control method, where the computational fluid dynamics simulation solution system at least includes a solver, the method including:
s201, according to a preset solving scheme, enabling a solver to iteratively solve the numerical solution of each physical quantity in the computational domain flow field and the residual value of each physical quantity according to a first step length.
S202, generating residual curve graphs according to a second step length based on residual data sent by the solver, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value.
S203, inputting the residual curve graph into a pre-trained convergence feature recognition model for each generated residual curve graph, and obtaining a convergence recognition result corresponding to the residual curve graph.
S204, controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph.
Specifically, the residual plot may be generated by: determining a residual value corresponding to each historical iteration step of each physical quantity according to residual data output by the solver so as to generate a residual function corresponding to the physical quantity, wherein the residual function is used for indicating the relation between the solving step number and the residual value of the physical quantity; and generating a residual curve graph based on residual functions corresponding to all the physical quantities.
Specifically, the convergence feature recognition model may be generated by training in the following manner: collecting sample residual graphs with non-convergence characteristics, wherein each sample residual graph is drawn with at least one residual function corresponding to physical quantity; labeling a non-convergence characteristic label on the sample residual curve graph, wherein the label comprises a divergent type and an oscillation type; and training and establishing a convergence characteristic identification model by using the marked residual error convergence characteristic data set.
Specifically, the operating state of the solver may be controlled by: determining whether the current solving calculation is converged or not according to the residual curve convergence characteristic recognition result; if the solution is not converged, the control solver stops, and the solution scheme is adjusted according to the scheme adjustment strategy, and then the solver is restarted to perform iterative solution calculation; if the convergence is carried out, the normal operation of the solver is maintained.
Specifically, whether the solving and calculating process is converged is judged according to the residual curve convergence characteristic recognition result. If the convergence feature recognition result does not comprise any one or combination of the divergent non-convergence feature tag and the oscillation non-convergence feature tag, solving the convergence of the calculation process. If the convergence feature recognition result comprises any one or combination of the divergent non-convergence feature and the oscillation non-convergence feature, the calculation process is solved to be non-convergence.
Specifically, the fluid solution adjustment may be performed on the solver by: selecting one of higher quality mesh data, reducing the kurron CFL, reducing the relaxation factor, optimizing the boundary condition setting is performed according to the priority.
Specifically, the method further comprises recording all the executed monitoring management behavior records, including: monitoring time, identifying non-convergence characteristics, adjusting strategies by the executed solution scheme, forming a monitoring management log, and displaying the monitoring management log to a user so that the user can understand the rationality of monitoring management.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, when the electronic device 500 is running, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the intelligent self-monitoring computational fluid dynamics simulation solution system control method can be executed, and specific implementation manners can be referred to the method embodiments and are not repeated herein.
The embodiment of the application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the intelligent self-monitoring computational fluid dynamics simulation solving method can be executed, and specific implementation manners can be referred to the method embodiments and are not repeated herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the foregoing examples are merely specific embodiments of the present application, and are not intended to limit the scope of the present application, but the present application is not limited thereto, and those skilled in the art will appreciate that while the foregoing examples are described in detail, the present application is not limited thereto. Any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or make equivalent substitutions for some of the technical features within the technical scope of the disclosure of the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An intelligent self-monitoring computational fluid dynamics simulation solving system control method, which is characterized in that the computational fluid dynamics simulation solving system at least comprises a solver, and the method comprises the following steps:
according to a preset solving scheme, enabling a solver to iteratively solve the numerical solution of each physical quantity in the computational domain flow field and the residual value of each physical quantity according to a first step length;
generating residual curve graphs according to the second step physical quantity based on the residual data sent by the solver, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value;
inputting the residual curve graph into a pre-trained convergence feature recognition model aiming at each generated residual curve graph, and obtaining a convergence recognition result corresponding to the residual curve graph;
and controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph.
2. The method of claim 1, wherein the residual plot is generated by:
for each physical quantity, generating a residual function corresponding to the physical quantity parameter by taking the iteration step number as an abscissa and taking the residual value of the physical quantity parameter as an ordinate;
the physical quantity generates a residual curve graph based on residual functions corresponding to all physical quantity parameters.
3. The method of claim 1, wherein the converging characteristic recognition model is generated by training:
collecting sample residual graphs with non-convergence characteristics, wherein each sample residual graph is drawn with a residual function corresponding to at least one physical quantity parameter;
performing feature labeling on the sample residual graph to obtain a residual convergence feature labeling data set, wherein the feature label comprises divergent non-convergence features and oscillatory non-convergence features;
and training the residual error convergence characteristic recognition model based on the residual error convergence characteristic labeling data set to obtain a convergence characteristic recognition model.
4. The method of claim 1, wherein the operating state of the solver is controlled by:
determining whether the simulation solving and calculating process is converged according to the convergence characteristic recognition result;
if the solution is not converged, stopping the operation of the solver, adjusting the solution according to the adjustment strategy, and restarting the solver to perform iterative solution calculation;
if so, the solver is kept running continuously.
5. The method of claim 4, wherein determining whether the solution computing process is converging is based on a converging characteristic recognition result;
if the convergence feature identification result does not comprise any one of the divergent non-convergence feature and the oscillation non-convergence feature, solving the convergence of the calculation process;
if the convergence feature recognition result comprises any one or combination of the divergent non-convergence feature and the oscillation non-convergence feature, the calculation process is solved to be non-convergence.
6. The method of claim 4, wherein the solver is solution tuned by:
selecting one of higher quality mesh data, reducing the kurron CFL, reducing the relaxation factor, optimizing the boundary condition setting is performed according to the priority.
7. The method as recited in claim 6, further comprising:
recording all executed monitoring management behaviors, wherein the monitoring management behaviors comprise monitoring time, recognized non-convergence characteristics and executed solution scheme adjustment strategies;
and forming and displaying a monitoring management log.
8. An intelligent self-monitoring computational fluid dynamics simulation solution system, the system comprising:
the solver is used for carrying out iterative solution according to a preset solution scheme and a first step length so as to solve the numerical solution of each physical quantity in the computational domain flow field and the residual value of each physical quantity;
the solving controller generates residual curve graphs based on the residual data sent by the solver according to the second step length, wherein the abscissa of each residual curve graph is the iteration step number, and the ordinate is the corresponding physical quantity residual value; controlling the running state of the solver according to the convergence identification result corresponding to the residual curve graph;
the convergence characteristic recognition model is used for outputting a convergence characteristic recognition result corresponding to each input residual curve graph and sending the convergence characteristic recognition result to the solving controller;
and the solution scheme adjuster is used for adjusting the solution scheme according to the adjustment strategy when the solution process is not converged.
9. An electronic device, comprising: a processor, a memory and a bus, said memory storing machine readable instructions executable by said processor, said processor in communication with said memory via the bus when the electronic device is running, said processor executing said machine readable instructions to perform the steps of the intelligent self-monitoring computational fluid dynamics simulation solution system control method according to any one of claims 1 to 7.
10. A computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the computer program performs the steps of the method for controlling the intelligent self-monitoring computational fluid dynamics simulation solving system according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311595250.XA CN117608963A (en) | 2023-11-27 | 2023-11-27 | Control method of intelligent self-monitoring computational fluid dynamics simulation solving system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311595250.XA CN117608963A (en) | 2023-11-27 | 2023-11-27 | Control method of intelligent self-monitoring computational fluid dynamics simulation solving system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117608963A true CN117608963A (en) | 2024-02-27 |
Family
ID=89949331
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311595250.XA Pending CN117608963A (en) | 2023-11-27 | 2023-11-27 | Control method of intelligent self-monitoring computational fluid dynamics simulation solving system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117608963A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118052171A (en) * | 2024-04-16 | 2024-05-17 | 中国空气动力研究与发展中心计算空气动力研究所 | Adaptive CFL number adjustment method, apparatus and storage medium for accompanying equations |
-
2023
- 2023-11-27 CN CN202311595250.XA patent/CN117608963A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118052171A (en) * | 2024-04-16 | 2024-05-17 | 中国空气动力研究与发展中心计算空气动力研究所 | Adaptive CFL number adjustment method, apparatus and storage medium for accompanying equations |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20230350355A1 (en) | Heuristic method of automated and learning control, and building automation systems thereof | |
| CN117608963A (en) | Control method of intelligent self-monitoring computational fluid dynamics simulation solving system | |
| JP6933899B2 (en) | Plant operation support equipment, plant operation support method, and plant operation support program | |
| US20160203036A1 (en) | Machine learning-based fault detection system | |
| US20130166243A1 (en) | Test device and method for testing stability of electronic devices | |
| JP6718500B2 (en) | Optimization of output efficiency in production system | |
| CN112066567A (en) | Constant temperature control method and system for gas water heater, electronic equipment and storage medium | |
| US20160097698A1 (en) | Estimating remaining usage of a component or device | |
| TWI388979B (en) | Computer system and monitoring device | |
| WO2010047937A1 (en) | Robust filtering and prediction using switching models for machine condition monitoring | |
| JPWO2020136859A1 (en) | Estimator, estimation method, and program | |
| US20220058099A1 (en) | Device monitoring systems | |
| KR20240071361A (en) | Server, method and computer program for controlling environment in building | |
| CZ2022226A3 (en) | A system, a method, and a non-transient computer-readable medium for mining cryptocurrencies | |
| CN113254153B (en) | Method and device for processing flow task, computer equipment and storage medium | |
| CN115543855B (en) | Semiconductor device parameter testing method, device, electronic device and storage medium | |
| US9507344B2 (en) | Index generation and embedded fusion for controller performance monitoring | |
| KR101360263B1 (en) | Apparatus for controlling a computation job, method of controlling a computation job, and storage medium for storing a software controlling a computation job | |
| CN115081410A (en) | Method and device for automatically generating experiment report | |
| CN114117778A (en) | Control parameter determination method and device, electronic equipment and storage medium | |
| WO2022056916A1 (en) | Controlling method and device for industrial device | |
| EP3904977B1 (en) | Method for generating a process model and support system using the process model | |
| EP4361745A1 (en) | Autonomous operation of modular industrial plants | |
| WO2021188491A1 (en) | Real-time configurator validation and recommendation engine | |
| CN117348605B (en) | Optimization method and system applied to control system of release film tearing machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |