Multidisciplinary optimization analysis method and device, terminal equipment and storage medium

Abstract

The invention is suitable for the technical field of multidisciplinary optimization analysis, and provides a multidisciplinary optimization analysis method and device, terminal equipment and a storage medium, and the method comprises the steps: obtaining a target task of a user; according to a business process of a preset target task, at least one function analysis module is selected from a plurality of preset function analysis modules to serve as a target analysis module, and the target analysis module is a function analysis module needing to be used for completing the target task; and calling the target analysis module to analyze the target task according to the function of the target analysis module to obtain an optimal result of the target task. According to the method, different analysis modules can be flexibly selected and combined according to different user requirements and different calculation tasks, the adaptability is high, and the problem that interdisciplinary cooperation is difficult can be solved.

Description

Technical Field

[0001] The present application belongs to the technical field of multidisciplinary optimization analysis, and in particular, relates to a multidisciplinary optimization analysis method, apparatus, terminal device and readable storage medium. Background Art

[0002] Multidisciplinary optimization is a system engineering technology that comprehensively utilizes knowledge and methods from multiple disciplines to solve complex problems. It aims to improve the efficiency and quality of problem solving and promote the development of scientific research and technological innovation by integrating expertise and techniques from different disciplines.

[0003] However, methodological differences between different disciplines have led to difficulties in interdisciplinary collaboration, and multidisciplinary joint simulation scenarios are facing severe challenges in terms of solution time. Therefore, there is an urgent need to establish a unified theoretical framework and methodological system to solve the difficulties in interdisciplinary collaboration. Summary of the invention

[0004] The embodiments of the present application provide a multidisciplinary optimization analysis method, apparatus, terminal device and computer-readable storage medium, which can flexibly select and combine different analysis modules to analyze user needs according to different user needs and computing tasks, and have strong adaptability.

[0005] In a first aspect, an embodiment of the present application provides an optimization analysis method, comprising:

[0006] Get the user's target task;

[0007] According to the business process of the preset target task, at least one function analysis module is selected from a plurality of preset function analysis modules as a target analysis module, where the target analysis module is a function analysis module required to complete the target task;

[0008] According to the function of the target analysis module, the target analysis module is called to analyze the target task and obtain the optimal result of the target task.

[0009] In the embodiment of the present application, the system needs to clarify the performance indicators of the target task from the user. These indicators may be such as speed, efficiency, cost, accuracy, etc., which are key parameters for measuring the quality of task completion. The system defines a set of related functional analysis modules according to the performance indicators set by the user, and determines the execution order of these functional analysis modules. According to the planned execution order, the system runs each functional analysis module in turn, conducts an in-depth analysis of the target task, and provides the optimal solution for the target task. In the above method, different analysis modules can be flexibly selected and combined according to different user needs and computing tasks, and the adaptability is strong.

[0010] In a possible implementation manner of the first aspect, when there are multiple target analysis modules;

[0011] According to the function of the target analysis module, the target analysis module is called to analyze the target task and obtain the optimal result of the target task, including:

[0012] According to the functions of the target analysis modules, plan the first execution order of the multiple target analysis modules;

[0013] Acquire configuration parameters corresponding to each of the multiple function analysis modules according to the target task, wherein the configuration parameters include parameter type, parameter format and algorithm type of the function analysis module corresponding to the configuration parameters;

[0014] The target task is executed according to the configuration parameters corresponding to each of the multiple target analysis modules and the first execution order to obtain the optimal result of the target task.

[0015] In the embodiment of the present application, planning the first execution order of multiple target analysis modules helps to ensure the efficiency and accuracy of the decision-making process. By reasonably arranging the execution order, unnecessary calculations can be avoided, resource waste can be reduced, and key steps can be ensured to be executed first. The configuration parameters of each functional analysis module are obtained according to the target task, including parameter type, parameter format and algorithm type, so that the system can flexibly adjust the analysis strategy according to different task requirements. This flexibility means that the system can adapt to different tasks and scenarios, improving its versatility and applicability.

[0016] In a possible implementation manner of the first aspect, executing a target task according to configuration parameters corresponding to each of the plurality of target analysis modules and a first execution order to obtain an optimal result of the target task includes:

[0017] sending the plurality of configuration parameters to each target analysis module respectively according to the first execution order;

[0018] According to the first execution order, multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task.

[0019] In the embodiment of the present application, the configuration parameters are sent to each target analysis module in the first execution order, ensuring that each module starts working only after receiving the correct configuration information. This helps to reduce the module waiting time, thereby improving the efficiency of the entire processing flow.

[0020] In a possible implementation manner of the first aspect, the multiple target analysis modules are respectively a sample generation module, a key information extraction module, a candidate result output module, and a result optimization module. According to a first execution order, the solvers in the multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task, including:

[0021] In the case where the first execution order is the sample generation module, the key information extraction module, the candidate result output module and the result optimization module;

[0022] Input the target task into the sample generation module, use the sample generation module to solve the target task, and obtain multiple combined sample parameters;

[0023] Inputting multiple combined sample parameters into a key information extraction module, and using the key information extraction module to screen the combined sample parameters to obtain a key sample that has the greatest impact on the optimal result of the target task;

[0024] Input the key sample into the candidate result output module, find the task processing model matching the key sample, and use the task processing model matching the key sample as the target task processing model;

[0025] The target task is processed using the target processing model and result optimization module to obtain the optimal result corresponding to the target task.

[0026] In an embodiment of the present application, by decomposing the problem-solving process into different modules, each stage can have clear tasks and goals. This multi-module, multi-stage target analysis process helps to improve the efficiency, accuracy and flexibility of problem solving, and is also beneficial to the maintenance and expansion of the system.

[0027] In a possible implementation manner of the first aspect, the target task is input into a sample generation module, and the sample generation module is used to solve the target task to obtain multiple combined sample parameters, including:

[0028] Input the target task into the sample generation module, and use the solver in the sample generation module to solve the target task to obtain multiple sample parameters;

[0029] Multiple sample parameters are combined to generate multiple groups of different combined sample parameters.

[0030] In the embodiments of the present application, by generating and combining multiple sample parameters, the diversity, robustness, and generalization ability of problem solving can be improved, and more efficient decision-making and experimental design can be promoted, which is an important means to improve the quality and efficiency of solving complex problems.

[0031] In a possible implementation of the first aspect, a key sample is input into a candidate result output module, a task processing model matching the key sample is searched, and the task processing model matching the key sample is used as a target task processing model, including:

[0032] The key sample is input into the candidate result output module, and a task processing model matching the key sample is selected as the target task processing model from a plurality of task processing models stored in the candidate result output module.

[0033] In the embodiment of the present application, by matching key samples with multiple task processing models, it is possible to ensure that the most appropriate model is selected to process the target task, which helps to improve the accuracy of the processing results.

[0034] In a possible implementation manner of the first aspect, the target task is processed by using the target processing model and the result optimization module to obtain an optimal result corresponding to the target task, including:

[0035] Use the target processing model to process the target task and obtain the first candidate result;

[0036] The first candidate result is processed by the result optimization module to obtain the first candidate result corresponding to the target task;

[0037] The result optimization module is used to process the first candidate result to obtain the second candidate result corresponding to the target task. The second candidate result is processed by the result optimization module until the i-th candidate result meets the preset requirements. The i-th candidate result is used as the optimal result of the target task.

[0038] In the embodiment of the present application, by continuously using the result optimization module, each candidate result can be iteratively optimized, thereby improving the quality and accuracy of the final result.

[0039] In a second aspect, an embodiment of the present application provides an optimization analysis device, comprising:

[0040] An acquisition unit, used to acquire a user's target task;

[0041] A selection unit, configured to select at least one function analysis module from a plurality of pre-set function analysis modules as a target analysis module according to a business process of a pre-set target task, wherein the target analysis module is a function analysis module required to complete the target task;

[0042] The analysis unit is used to call the target analysis module to analyze the target task according to the function of the target analysis module to obtain the optimal result of the target task.

[0043] In a third aspect, an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, an optimization analysis method as described in any one of the first aspects above is implemented.

[0044] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the optimization analysis method as described in any one of the first aspects above.

[0045] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute any one of the optimization analysis methods of the first aspect.

[0046] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0048] Figure 1 It is a schematic diagram of the architecture of a multidisciplinary optimization analysis system provided in an embodiment of the present application;

[0049] Figure 2 It is a schematic diagram of module data definition provided in an embodiment of the present application;

[0050] Figure 3 It is a flow chart of the multidisciplinary optimization analysis method provided in the embodiment of the present application;

[0051] Figure 4 This is a schematic diagram of the process of obtaining the optimal result provided in the embodiment of the present application. Figure 1 ;

[0052] Figure 5 This is a schematic diagram of the process of obtaining the optimal result provided in the embodiment of the present application. Figure 2 ;

[0053] Figure 6 It is a structural schematic diagram of the optimization analysis method provided in the embodiment of the present application;

[0054] Figure 7 It is a schematic diagram of the structure of an optimization analysis device provided in an embodiment of the present application;

[0055] Figure 8 It is a structural diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0057] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0058] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0059] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0060] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0061] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. appearing in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways.

[0062] Multidisciplinary optimization is a system engineering technology that uses knowledge and methods from multiple disciplines to solve complex problems. It aims to improve the efficiency and quality of problem solving by integrating expertise and technology from different disciplines, and to promote the development of scientific research and technological innovation. In the past few decades, multidisciplinary optimization has been widely used in aerospace, automotive engineering, power systems, petroleum and petrochemicals, and has achieved remarkable results.

[0063] However, multidisciplinary optimization faces some challenges and problems. First, the methodological differences between different disciplines lead to difficulties in interdisciplinary cooperation, and a unified theoretical framework and methodological system need to be established. Secondly, for multidisciplinary joint simulation scenarios, there are severe challenges in solution time. To address this problem, many supporting sub-functional modules have been developed. It is necessary to uniformly manage the architecture of the multifunctional modules and improve the flexibility and scalability of the system.

[0064] In order to solve the problems in the above-mentioned related technologies, the embodiment of the present application provides a multidisciplinary optimization analysis method. In the present application, by performing modular design and defining the data structure of the calling interface of each submodule, it is necessary to define the general parameters of the multidisciplinary optimization system, and to introduce the setting parameters of the specific module algorithm, etc., and adopt the standardized json data format to facilitate data exchange and sharing between different modules, and define the combined calling order and loading logic of each submodule to realize the dynamic loading mechanism of the module. The order is visualized according to the optimization process created by the user, and the corresponding modules are flexibly selected and loaded according to the different user needs and computing tasks. The communication between modules adopts the message queue and event-driven mechanism to realize asynchronous communication between modules and improve the response speed and processing power of the system.

[0065] In order to solve the problem of difficulties in interdisciplinary cooperation caused by different methodologies between different disciplines, this application establishes a unified theoretical framework, see Figure 1 , is a schematic diagram of the multidisciplinary optimization analysis system architecture provided in the embodiment of the present application, such as Figure 1 As shown in the figure, based on the modular design concept, the overall system architecture of multidisciplinary optimization analysis includes functional component modules and platform framework basic functional modules, among which the functional component modules are the units that perform specific functions of the system, and the platform framework basic functional modules are the infrastructure that supports the operation of the entire system.

[0066] The above functional component modules include modules such as Design of Experiments (DOE), surrogate models, sensitivity analysis, optimization algorithms, and icon controls for viewing results. DOE is used to design experiments to optimize system parameters; surrogate models are used to simulate the behavior of real systems and are usually used to simplify complex systems; they are used to analyze the sensitivity of system output to changes in input parameters; they are used to find the best combination of system parameters to achieve specific goals; and icon controls for viewing results are used to display analysis results, usually in the form of charts. These functional components together constitute the core tool set for multidisciplinary optimization analysis, and they work together to provide the following functions:

[0067] 1) Multidisciplinary data integration: Ability to collect and analyze data from different disciplines;

[0068] 2) Interdisciplinary collaboration: Allows analysis and optimization work from different disciplines to be carried out in collaboration;

[0069] 3) Iterative optimization: gradually improve the design solution through continuous iterative optimization process;

[0070] 4) Result analysis and verification: Provide tools to analyze and verify the optimized design solution.

[0071] The above functional component modules provide a comprehensive foundation for multidisciplinary optimization analysis, enabling engineers to effectively deal with complex design problems and find the best design solutions that meet multiple disciplines constraints and objectives.

[0072] In addition, the basic functional modules of the platform framework provide the following basic capabilities: workflow orchestration allows users to design, define and analyze workflows; workflow operation and status control manages the execution process and status of workflows; data lifecycle management and services are used to manage the entire life cycle of data, including storage, processing and analysis; organization and services are used to provide user organizational structure and related services.

[0073] In using the above Figure 1 When conducting multidisciplinary optimization analysis based on the theoretical framework shown in the figure, since different functional modules need to exchange and share data, it is necessary to clarify the data format of each functional module to ensure the correct transmission and interpretation of data between modules and avoid problems caused by inconsistent data formats or transmission errors. Figure 2 , is a schematic diagram of module data definition provided in an embodiment of the present application, which is as follows:

[0074] like Figure 2As shown in (a), it is a schematic diagram of DOE experimental design data definition, where the DOE configuration file adopts JSON format, which contains a series of keywords and their corresponding values, which are used to define the parameter settings and algorithm settings of the DOE solver. The parameter settings include input parameters, output parameters, constraints, optimization objectives and value range settings, and the algorithm settings include: method, operator, sample size, number of iterations, random seed and other settings. The scheduling engine passes the DOE configuration file to the DOE solver, and generates the DOE result file after calculation and solution.

[0075] like Figure 2 (b) is a schematic diagram of the sensitivity analysis model data definition. The scheduling engine passes the sensitivity analysis model configuration file to the proxy model solver, which generates a sensitivity analysis model result file after calculation and solution. The parameter settings in the sensitivity analysis configuration file include: input parameter ID, output parameter ID, input parameter data, and output parameter data. The algorithm settings include: method, resampling, confidence, and sensitivity threshold, etc.

[0076] like Figure 2 As shown in (c), it is a schematic diagram of the proxy model data definition. The scheduling engine passes the proxy model configuration file to the proxy model solver, which generates a proxy model result file after calculation and solution. The parameter settings in the proxy model configuration file include input parameter ID, output parameter ID and training set samples, and the algorithm settings include method, kernel function and validation set samples.

[0077] like Figure 2 (d) is a schematic diagram of the agent model data definition. The scheduling engine passes the optimization algorithm configuration file to the optimization algorithm solver, which generates the optimization algorithm result file after calculation and solution. The parameter settings in the optimization algorithm configuration file include optimization variables, constraints and optimization objectives, and the algorithm settings include methods, population size, number of iterations, and convergence accuracy.

[0078] Through the above data interface definition, the solver corresponding to each functional module can understand how to receive input, perform calculations, and generate result files that meet the requirements. This definition method enables the solver corresponding to each functional module to flexibly adapt to different experimental design requirements and algorithm settings.

[0079] Use the above analysis system framework and defined module data to perform multidisciplinary optimization analysis, see Figure 3 , is a flow chart of the multidisciplinary optimization analysis method provided in the embodiment of the present application. As an example but not a limitation, the method may include the following steps:

[0080] S101, obtaining the user's target task.

[0081] In the embodiment of the present application, the target task is the goal that the user hopes to achieve or the problem that is to be solved. Communication with the user can be conducted to ensure a complete understanding of their goals and task requirements, which involves asking specific questions, collecting relevant information, or conducting in-depth discussions.

[0082] For example, taking an online travel planning service, a user wants to find the best route from the starting point to the destination. The user's starting point, destination, and travel preferences (such as budget, time, mode of transportation, etc.) can be asked through an online form or chat interface to understand the user's needs, and then the user's needs are converted into specific solution requests, such as "find the best driving route from New York to Los Angeles with a budget of US$2,000 and a travel time of 7 days". The user's solution needs are the target tasks.

[0083] S102, according to the business process of the preset target task, at least one function analysis module is selected from a plurality of preset function analysis modules as a target analysis module, where the target analysis module is a function analysis module required to complete the target task.

[0084] In the embodiment of the present application, the business process is set to achieve the user's target task. According to the specific needs of the target task, at least one required functional analysis is selected from modules such as pre-set proxy models, sensitivity analysis, optimization algorithms, and icon controls for viewing results to analyze the target task. The selected functional analysis module, such as the sensitivity analysis module, is the target analysis module.

[0085] Specifically, understand the specific requirements of the target task in detail, evaluate whether each functional module can meet the requirements of the target task, and based on the evaluation results, select at least one functional module that can complete the target task and integrate the selected module into the business process.

[0086] S103, according to the function of the target analysis module, calling the target analysis module to analyze the target task and obtain the optimal result of the target task.

[0087] In an embodiment of the present application, the scheduling engine in the platform framework basic function module is used to call the target analysis module to analyze and process the target task.

[0088] In one embodiment, see Figure 4 , is a schematic diagram of the process of obtaining the optimal result provided in the embodiment of the present application Figure 1 ,like Figure 4 As shown, when there are multiple target analysis modules, step S103 includes:

[0089] S201, planning a first execution order of multiple target analysis modules according to the functions of the target analysis modules.

[0090] In an embodiment of the present application, if the target task of the user to be analyzed corresponds to multiple functional modules in the icon control of the DOE model, proxy model, sensitivity analysis, optimization algorithm, and result viewing, it is necessary to plan the execution order of each functional model, that is, it is necessary to decide which module should be executed first and how to arrange their order according to the function of each module.

[0091] Specifically, based on the goals and analysis requirements, determine which modules are most critical to achieving the goals. For example, in the case of finding the best driving route, sample points along the possible driving route may be analyzed, and the execution order of the modules may be planned based on module functions and priorities. Through such planning, it can be ensured that the target analysis modules are executed in the most effective and logical order, thereby efficiently achieving the analysis goals.

[0092] S202, obtaining configuration parameters corresponding to each of the plurality of function analysis modules according to the target task, wherein the configuration parameters include parameter type, parameter format and algorithm type of the function analysis module corresponding to the configuration parameters.

[0093] In an embodiment of the present application, in order to obtain the configuration parameters of each functional module corresponding to the target task, it is necessary to perform a detailed analysis of the target task, such as finding the objective function under constraints, and obtaining parameters provided by the optimization algorithm such as particle swarm optimization, or when obtaining sensitive parameters corresponding to the objective function, it is necessary to obtain configuration parameters such as input parameters and parameter variables in the sensitivity analysis module.

[0094] For example, if the target task is "find the best driving route from New York to Los Angeles with a budget of $2,000 and a travel time of 7 days", we need to configure multiple functional analysis modules to achieve this goal. The configuration parameters needed are: constraint variables, such as $2,000, 7 days, etc., obtain the corresponding functional analysis module according to the target task, and obtain the configuration parameters corresponding to the target task.

[0095] S203, executing the target task according to the configuration parameters corresponding to each of the multiple target analysis modules and the first execution order to obtain the optimal result of the target task.

[0096] In the embodiment of the present application, the target tasks are finally executed in sequence according to the planned order.

[0097] In the above method, planning the first execution order of multiple target analysis modules helps to ensure the efficiency and accuracy of the decision-making process. By reasonably arranging the execution order, unnecessary calculations can be avoided, resource waste can be reduced, and key steps can be ensured to be executed first. The configuration parameters of each functional analysis module are obtained according to the target task, including parameter type, parameter format and algorithm type, so that the system can flexibly adjust the analysis strategy according to different task requirements. This flexibility means that the system can adapt to different tasks and scenarios, improving its versatility and applicability.

[0098] In one embodiment, step S203 includes:

[0099] According to the first execution order, multiple configuration parameters are sent to each target analysis module respectively, and according to the first execution order, multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task.

[0100] In the embodiment of the present application, the configuration parameters are variables or settings of the target task as described above. These parameters will be sent to different "target analysis modules" one by one. When executing the task, each analysis module will be activated and process data in a predetermined order. By using these analysis modules in sequence, the entire task will be gradually decomposed and solved. The processing results of each functional analysis module may affect the input and output of subsequent modules. Ultimately, through the collaborative work of these modules, an optimal solution (optimal result) that meets all constraints (such as budget and time) will be obtained.

[0101] In the above method, the configuration parameters are sent to each target analysis module in the first execution order, ensuring that each module starts working only after receiving the correct configuration information. This helps to reduce the module waiting time, thereby improving the efficiency of the entire processing flow.

[0102] In one embodiment, see Figure 5 , is a schematic diagram of the process of obtaining the optimal result provided in the embodiment of the present application Figure 2 ,like Figure 5 As shown, the multiple target analysis modules are respectively a sample generation module, a key information extraction module, a candidate result output module and a result optimization module. When the first execution order is the sample generation module, the key information extraction module, the candidate result output module and the result optimization module, the solvers in the multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task, including:

[0103] S301, input the target task into the sample generation module, use the sample generation module to solve the target task, and obtain multiple combined sample parameters.

[0104] In the embodiment of the present application, the sample generation module is also the above-mentioned DOE model. The DOE model is a statistical method for determining the optimal design of an experiment to efficiently test the impact of multiple factors on the results. This is the first execution module. The target task is input into the sample generation module, which uses the DOE model to design experiments to generate multiple combined sample parameters. These parameters represent different experimental conditions or settings, which will be used for subsequent analysis.

[0105] In one embodiment, step S301 includes:

[0106] The target task is input into the sample generation module, and the solver in the sample generation module is used to solve the target task to obtain multiple sample parameters; the multiple sample parameters are combined to generate multiple groups of different combined sample parameters.

[0107] In an embodiment of the present application, the task objective is clearly defined, including the problem to be solved or the goal to be achieved. The sample generation module includes one or more solvers, which can process the target task based on different algorithms or models. The solvers solve according to the requirements of the target task and generate multiple candidate sample parameters. These sample parameters are possible solutions or experimental settings for the target task. The sample generation module combines these individual sample parameters to generate multiple sets of different combined sample parameters.

[0108] Specifically, the parameter combination method can generate groups by randomly selecting or systematically scanning different parameter values ​​in the parameter space, or can generate combinations through an iterative optimization process according to certain optimization goals or constraints.

[0109] For example, let the target task be to solve two different functions and find their optimal solutions under certain conditions. For example, the target is considered as an optimization problem, which can be expressed as:

[0110] Objective function: f(x,y)=x 2 +y 2

[0111] Constraint function: g(x,y)=x+y-5=0

[0112] Our goal is to find the minimum value of the objective function under the constraints.

[0113] The above optimization problem is submitted as input to the sample generation module, that is, the corresponding variables are input into the defined data, and the solver corresponding to the sample generation module solves it to generate sample parameters (x, y) that meet the constraints, such as:

[0114] Sample 1: (x1, y1) = (1, 4)

[0115] Sample 1: (x2, y2) = (2, 3)

[0116] Sample 3: (x3, y3) = (3, 2)

[0117] ……etc

[0118] The generated sample parameters are combined to generate different combination samples. For example, two samples can be combined by linear combination, for example (x1c, y1c) = α(x1, y1) + (1-α)(x2, y2), and the following possible combination samples can be obtained:

[0119] Combined sample 1: (x1c, y1c) = 0.6(1, 4) + 0.4(3, 2) = (1.8, 3.2)

[0120] Combined sample 2: (x2c, y2c) = 0.5(1,3) + 0.5(3,2) = (2,2.5)

[0121] Combined sample 3: (x3c, y3c) = 0.7(3, 2) + 0.3(2, 3) = (2.7, 2.3)

[0122] In the above method, by generating and combining multiple sample parameters, the diversity, robustness, and generalization ability of problem solving can be improved, and more efficient decision-making and experimental design can be promoted. It is an important means to improve the quality and efficiency of solving complex problems.

[0123] S302, inputting multiple combined sample parameters into a key information extraction module, and using the key information extraction module to screen the combined sample parameters to obtain a key sample that has the greatest impact on the optimal result of the target task.

[0124] In the embodiment of the present application, as in the above example, it is necessary to extract key information (combined samples) from the generated samples from multiple groups of combined samples. This information is crucial for understanding the characteristics of the samples and subsequent analysis. This is the second execution module, specifically: the multiple combined sample parameters output by the sample generation module are input into the key information extraction module (sensitivity analysis module). This module may involve multidisciplinary optimization analysis, feature extraction or pattern recognition and other technologies to identify and extract key information from the samples.

[0125] As in the above example, since the objective function is to find the minimum value of f1(x, y) under g(x, y), the sensitive combination parameters can be obtained by inputting the combination samples into the sensitivity analysis module (key information extraction module) for screening:

[0126] Combined sample 3: (x3c, y3c) = 0.7(3, 2) + 0.3(2, 3) = (2.7, 2.3)

[0127] S303, input the key sample into the candidate result output module, find the task processing model matching the key sample, and use the task processing model matching the key sample as the target task processing model.

[0128] In an embodiment of the present application, the key sample is input into the candidate result output module (proxy model), and the task processing model matching the key sample is selected as the target task processing model from the multiple task processing models stored in the candidate result output module.

[0129] S304, using the target processing model and the result optimization module to process the target task and obtain the optimal result corresponding to the target task.

[0130] In the embodiment of the present application, step S304 may be implemented as follows:

[0131] The key sample is input into the candidate result output module, and the task processing model matching the key sample is selected as the target task processing model from the multiple task processing models stored in the candidate result output module; the target task is processed using the target processing model and the result optimization module to obtain the optimal result corresponding to the target task.

[0132] In an embodiment of the present application, we input the key sample into the candidate result output model, i.e., the above-mentioned proxy model, which stores multiple pre-trained task processing models. According to the key sample, the model that best matches the key sample characteristics is selected as the target task processing model. Once the target task processing model is selected, we use it and the result optimization module to process the key sample to obtain the optimal output result.

[0133] In the above method, by matching key samples with multiple task processing models, it is possible to ensure that the most appropriate model is selected to process the target task. This helps to improve the accuracy of the processing results.

[0134] In one embodiment, the target task is processed using the target processing model and the result optimization module to obtain the optimal result corresponding to the target task, including:

[0135] The target task is processed using the target processing model to obtain the first candidate result; the first candidate result is processed using the result optimization module to obtain the first candidate result corresponding to the target task;

[0136] The result optimization module is used to process the first candidate result to obtain the second candidate result corresponding to the target task. The second candidate result is processed by the result optimization module until the i-th candidate result meets the preset requirements. The i-th candidate result is used as the optimal result of the target task.

[0137] In the embodiment of the present application, the key information obtained in the above example is combination sample 3: (x3c, y3c) = 0.7(3, 2) + 0.3(2, 3) = (2.7, 2.3), where the target task is to calculate the target function f(x, y) = x 2 +y 2 The minimum value of g(x,y)=x+y-5=0 under the constraint condition.

[0138] According to the combined sample, if a linear search algorithm is selected from the proxy model as the target processing model, the objective function is calculated according to the key information to obtain a candidate result f1=12.58, and the result optimization module (optimization) such as the gradient descent algorithm is used to improve the first candidate point (2.7, 2.3). After gradient descent iterations, we get an improved second candidate point (2.6, 2.4), and calculate the objective function to get the second candidate result f2=12.52. Then, the structural candidate module is used to improve the second candidate point (2.6, 2.4) in turn to get the third candidate point (2.5, 2.5), and the objective function is calculated to get the third candidate result f2=12.5... and so on, until f(x, y)=x is reached. 2 +y 2 Under the constraint condition that g(x,y)=x+y-5=0 is the minimum value, when the value of f(x,y) corresponding to the i-th candidate result is the smallest, the i-th candidate point is taken as the optimal result of the target task.

[0139] In the above method, by continuously using the result optimization module, each candidate result can be iteratively optimized, thereby improving the quality and accuracy of the final result.

[0140] See also Figure 6 , is a schematic diagram of the structure of the multidisciplinary optimization analysis method provided in the embodiment of the present application, such as Figure 6 As shown, the workflow engine mobilizes the function analysis module to work, and the steps are as follows:

[0141] 1) Send the user's solution request (target task) through the client

[0142] 2) DOE model starts, and the optimization process is designed according to the target task

[0143] 3) DOE model exit

[0144] 4) The DOE model is started to obtain multiple sets of combination parameters corresponding to the target task;

[0145] 5) DOE model exit

[0146] 6) Start the sensitivity analysis module to screen key parameters

[0147] 7) Exit the sensitivity analysis module and output key parameters

[0148] 8) The proxy model is started and the target task processing model is selected based on key parameters

[0149] 9) Agent Model Exit

[0150] 10) The target task processing model is started, the target task is calculated, and the candidate results are obtained;

[0151] 11) The optimization model is started, and the candidate results are iterated using the optimization model to obtain the optimal result;

[0152] 12) The optimization model exits and outputs the final result.

[0153] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0154] Corresponding to the multidisciplinary optimization analysis method of the above embodiment, Figure 6 This is a structural block diagram of the optimization analysis device provided in an embodiment of the present application. For the sake of convenience of explanation, only the parts related to the embodiment of the present application are shown.

[0155] Reference Figure 7 , the device comprises:

[0156] An acquisition unit 71 is used to acquire a user's target task;

[0157] A selection unit 72, configured to select at least one function analysis module from a plurality of pre-set function analysis modules as a target analysis module according to a business process of a pre-set target task, wherein the target analysis module is a function analysis module required to complete the target task;

[0158] The analysis unit 73 is used to call the target analysis module to analyze the target task according to the function of the target analysis module to obtain the optimal result of the target task.

[0159] In one implementation, the analysis unit 73 is further configured to plan a first execution order of the plurality of target analysis modules according to the functions of the target analysis modules;

[0160] Acquire configuration parameters corresponding to each of the multiple function analysis modules according to the target task, wherein the configuration parameters include parameter type, parameter format and algorithm type of the function analysis module corresponding to the configuration parameters;

[0161] The target task is executed according to the configuration parameters corresponding to each of the multiple target analysis modules and the first execution order to obtain the optimal result of the target task.

[0162] In one implementation, the analysis unit 73 is further configured to send the plurality of configuration parameters to each target analysis module respectively according to the first execution order;

[0163] According to the first execution order, multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task.

[0164] In one implementation, the analysis unit 73 is further used for, when the first execution order is the sample generation module, the key information extraction module, the candidate result output module and the result optimization module;

[0165] Input the target task into the sample generation module, use the sample generation module to solve the target task, and obtain multiple combined sample parameters;

[0166] Inputting multiple combined sample parameters into a key information extraction module, and using the key information extraction module to screen the combined sample parameters to obtain a key sample that has the greatest impact on the optimal result of the target task;

[0167] Input the key sample into the candidate result output module, find the task processing model matching the key sample, and use the task processing model matching the key sample as the target task processing model;

[0168] The target task is processed using the target processing model and result optimization module to obtain the optimal result corresponding to the target task.

[0169] In one implementation, the analysis unit 73 is further used to input the target task into the sample generation module, and use the solver in the sample generation module to solve the target task to obtain multiple sample parameters;

[0170] Multiple sample parameters are combined to generate multiple groups of different combined sample parameters.

[0171] In one implementation, the analysis unit 73 is further configured to input the key sample into the candidate result output module, and select a task processing model matching the key sample as the target task processing model from a plurality of task processing models stored in the candidate result output module.

[0172] In one implementation, the analyzing unit 73 is further configured to process the target task using the target processing model to obtain a first candidate result;

[0173] The first candidate result is processed by the result optimization module to obtain the first candidate result corresponding to the target task;

[0174] The result optimization module is used to process the first candidate result to obtain the second candidate result corresponding to the target task. The second candidate result is processed by the result optimization module until the i-th candidate result meets the preset requirements. The i-th candidate result is used as the optimal result of the target task.

[0175] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0176] in addition, Figure 7 The optimization and analysis device shown may be a software unit, a hardware unit, or a combination of software and hardware, which is built into an existing terminal device, or may be integrated into the terminal device as an independent accessory, or may exist as an independent terminal device.

[0177] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0178] Figure 8 Schematic diagram of the structure of the terminal device provided in the embodiment of the present application. Figure 8 As shown, the terminal device 8 of this embodiment includes: at least one processor 80 ( Figure 8 Only one is shown in the figure) a processor, a memory 81, and a computer program 82 stored in the memory 81 and executable on at least one processor 80, and when the processor 80 executes the computer program 82, the steps in any of the above-mentioned optimization analysis method embodiments are implemented.

[0179] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Figure 8It is only an example of the terminal device 8 and does not constitute a limitation on the terminal device 8. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, it may also include input and output devices, network access devices, etc.

[0180] The processor 80 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0181] In some embodiments, the memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. In other embodiments, the memory 81 may also be an external storage device of the terminal device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device 8. Further, the memory 81 may also include both an internal storage unit of the terminal device 8 and an external storage device. The memory 81 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as program codes of computer programs, etc. The memory 81 may also be used to temporarily store data that has been output or is to be output.

[0182] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0183] An embodiment of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0184] If the integrated unit is implemented in the form of 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 present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the device / terminal device, a recording medium, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electric carrier signal, a telecommunication signal, and a software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0185] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0186] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0187] In the embodiments provided in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0188] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple 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.

[0189] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An optimization analysis method, characterized in that: The method comprises: Get the user's target task; According to the preset business process of the target task, at least one function analysis module is selected from a plurality of preset function analysis modules as a target analysis module, wherein the target analysis module is a function analysis module required to complete the target task; According to the function of the target analysis module, the target analysis module is called to analyze the target task to obtain the optimal result of the target task.

2. The optimization analysis method according to claim 1, characterized in that: When there are multiple target analysis modules; According to the function of the target analysis module, calling the target analysis module to analyze the target task to obtain the optimal result of the target task includes: According to the functions of the target analysis modules, planning a first execution order of the plurality of target analysis modules; Acquire configuration parameters corresponding to each of the plurality of function analysis modules according to the target task, wherein the configuration parameters include parameter type, parameter format and algorithm type of the function analysis module corresponding to the configuration parameters; The target task is executed according to the configuration parameters corresponding to each of the plurality of target analysis modules and the first execution order to obtain an optimal result of the target task.

3. The optimization analysis method according to claim 2, characterized in that: The step of executing the target task according to the configuration parameters and the first execution order corresponding to each of the plurality of target analysis modules to obtain the optimal result of the target task includes: Sending a plurality of configuration parameters to each of the target analysis modules respectively according to the first execution order; According to the first execution order, the plurality of target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task.

4. The optimization analysis method according to claim 3, characterized in that: The multiple target analysis modules are respectively a sample generation module, a key information extraction module, a candidate result output module and a result optimization module. According to the first execution order, the solvers in the multiple target analysis modules are controlled to work in sequence to process the target task and obtain the optimal result of the target task, including: In the case where the first execution order is the sample generation module, the key information extraction module, the candidate result output module and the result optimization module in sequence; Inputting the target task into the sample generation module, solving the target task using the sample generation module, and obtaining a plurality of combined sample parameters; Inputting a plurality of combined sample parameters into the key information extraction module, and using the key information extraction module to screen the combined sample parameters to obtain a key sample that has the greatest impact on the optimal result of the target task; Inputting the key sample into the candidate result output module, searching for a task processing model matching the key sample, and using the task processing model matching the key sample as the target task processing model; The target task is processed using the target task processing model and the result optimization module to obtain the optimal result corresponding to the target task.

5. The optimization analysis method according to claim 4, characterized in that: The step of inputting the target task into the sample generation module and solving the target task using the sample generation module to obtain a plurality of combined sample parameters includes: Inputting the target task into the sample generation module, solving the target task using the solver in the sample generation module, and obtaining a plurality of sample parameters; A plurality of the sample parameters are combined to generate a plurality of different groups of combined sample parameters.

6. The optimization analysis method according to claim 4, characterized in that: The step of inputting the key sample into the candidate result output module, searching for a task processing model matching the key sample, and using the task processing model matching the key sample as a target task processing model includes: The key sample is input into the candidate result output module, and a task processing model matching the key sample is selected as a target task processing model from a plurality of task processing models stored in the candidate result output module.

7. The optimization analysis method according to claim 4, characterized in that: The using the target processing model and the result optimization module to process the target task to obtain the optimal result corresponding to the target task includes: Processing the target task using the target processing model to obtain a first candidate result; Using the result optimization module to process the first candidate result to obtain the first candidate result corresponding to the target task; The first candidate result is processed by the result optimization module to obtain the second candidate result corresponding to the target task, and the second candidate result is processed by the result optimization module until the i-th candidate result meets the preset requirements, and the i-th candidate result is used as the optimal result of the target task.

8. An optimization analysis device, characterized in that: include: An acquisition unit, used to acquire a user's target task; A selection unit, configured to select at least one function analysis module from a plurality of pre-set function analysis modules as a target analysis module according to a pre-set business process of the target task, wherein the target analysis module is a function analysis module required to complete the target task; The analysis unit is used to call the target analysis module to analyze the target task according to the function of the target analysis module to obtain the optimal result of the target task.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images