Intelligent test design method and system based on distributed network architecture

Abstract

The invention relates to an intelligent test design method and system based on a distributed network architecture. The method comprises the following steps: loading a historical dialogue record, an input file of a test design algorithm and user request information; and verifying the legality of the input file at the rear-end middleware layer, if the legality of the input file is abnormal, taking the historical dialogue record and the user request information as a training sample data set, triggering the large model of the business layer to perform decision training on the training sample data set according to the input file, and outputting a test design strategy, and the dialogue is output to the front-end interaction layer. Otherwise, reconstructing the input parameters in the input file, calling a calculation core of the test design algorithm to start an independent thread so as to enable the large model to execute the calculation task, and outputting a calculation result. And after the calculation result is loaded into a large model dialogue, an analysis report is output to a front-end interaction layer. By adopting the method, the development efficiency and universality of the virtual computing test can be improved.

Description

Technical Field

[0001] The present application relates to the field of virtual computing experiment technology, and in particular to an intelligent experiment design method and system based on a distributed network architecture. Background Art

[0002] Traditional experimental design methods are widely used in industrial optimization and scientific research, but their parameter configuration relies on manual experience and is difficult to adapt to the optimization needs of complex scenarios. Existing experimental design systems often use localized software or fixed-function modules, resulting in poor flexibility and insufficient intelligence. While large model technology demonstrates strong capabilities in data analysis and pattern prediction, it has yet to be deeply integrated with experimental design methods.

[0003] Currently, traditional experimental design methods require manual setting of parameter ranges and sample sizes, which is inefficient and easily limited by subjective experience, making it difficult to ensure the optimal parameter combination. Existing systems are unable to dynamically adjust experimental plans based on historical data or real-time feedback, resulting in duplicate experiments or insufficient coverage. Most experimental design systems use fixed-function modules. Adding new algorithms or modifying existing logic requires code refactoring, resulting in high development costs and long development cycles. Rapid integration of new methods through plug-in approaches is also impossible, making it difficult to meet the needs of multiple scenarios. Summary of the Invention

[0004] Based on this, it is necessary to provide an intelligent experiment design method and system based on a distributed network architecture that can improve the development efficiency and versatility of virtual computing experiments in response to the above technical problems.

[0005] An intelligent experiment design method based on a distributed network architecture is applied to the distributed network architecture, which includes a front-end interaction layer, a back-end middleware layer, and a business layer. The method includes: Loads historical session records, input files for DOE algorithms, and user request information.

[0006] The backend middleware layer verifies the validity of the input file. If the input file is not legitimate, the historical conversation records and user request information are used as a training sample dataset. This triggers the business layer's large model to perform decision training based on the input file and output the experimental design strategy. The conversation is then transmitted to the frontend interaction layer. Otherwise, the input parameters in the input file are reconstructed, and the computational core of the experimental design algorithm is called to start an independent thread, allowing the large model to execute the computation task and output the results.

[0007] After the calculation results are loaded into the large model dialogue, an analysis report is output to the front-end interaction layer.

[0008] An intelligent experiment design system based on a distributed network architecture, the system comprising: The information loading module is used to load historical conversation records, input files of the experimental design algorithm, and user request information.

[0009] The experiment execution module verifies the validity of input files at the backend middleware layer. If the input files are not valid, the module uses historical conversation records and user request information as a training sample dataset, triggering the business layer's large model to perform decision training on the training sample dataset based on the input files, output the experiment design strategy, and transmit the conversation to the frontend interaction layer. Otherwise, the module reconstructs the input parameters in the input file, calls the computation core of the experiment design algorithm, and starts an independent thread to enable the large model to execute the computation task and output the results.

[0010] The report display module is used to load the calculation results into the large model dialogue and then output the analysis report to the front-end interaction layer.

[0011] The aforementioned intelligent experiment design method and system, based on a distributed network architecture, utilizes a modular architecture for its functional loading middleware, transforming functional components such as the experiment design algorithm and large-scale model decision-making into dynamically loadable modules. This allows for rapid integration of new methods via plug-ins without modifying the core code, significantly reducing development costs and cycles while enabling flexible adaptation to multiple scenarios. A fully automated process encompasses input file validity verification, training sample generation, large-scale model decision-making training, computational task execution, and analysis report output, forming a closed "design-verify-improve" loop and avoiding efficiency losses caused by manual intervention. Large-scale model-driven intelligent interaction and decision support enable the system to automatically generate training samples based on historical conversation records and user request information when input file validity is abnormal, driving the large-scale model to conduct decision-making training and output strategies. When the input is valid, the system reconfigures parameters to initiate an independent thread to call the algorithm computation core, enabling the large-scale model to efficiently execute computational tasks. Combined with a flexible multi-task scheduling mechanism, this effectively improves experiment efficiency and reduces computational costs. This design not only enables the system to quickly adapt to the needs of different industries, but also provides a standardized interface for the subsequent expansion of new methods and new scenarios through the functional loading characteristics of the middleware, achieving the unity of the flexibility of the technical architecture and the adaptability of business needs, and improving the development efficiency and versatility of virtual computing experiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a diagram of an application scenario of an intelligent experiment design method based on a distributed network architecture in one embodiment; Figure 2 1 is a flow chart of an intelligent experiment design method based on a distributed network architecture in one embodiment; Figure 3 A system function flow chart in one embodiment; Figure 4The figure is a structural block diagram of an intelligent experiment design system based on a distributed network architecture in one embodiment. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0014] The intelligent experiment design method based on distributed network architecture provided by the present invention can be applied to Figure 1 The framework structure shown in the figure includes the front-end, back-end, and business layers. The front-end and back-end provide function buttons for experimental design methods (e.g., Latin Hypercube design), support user parameter input, and visualize results. The back-end includes function loading middleware (e.g., get_crazy_function) and business control middleware (e.g., latinHypercube). These middleware can load algorithm modules for various experimental design methods, parse and call the specific implementation logic of the business layer, and coordinate the interaction between the large model and the experimental design methods. The business layer primarily implements the experimental design methods, including the large model and the Latin Hypercube calculation implementation (LHS).

[0015] In one embodiment, Figure 2 As shown in the figure, an intelligent experiment design method based on distributed network architecture is provided. Figure 1 The framework structure in is used as an example to illustrate the following steps: Step 202: Load historical conversation records, input files of the experimental design algorithm, and user request information.

[0016] Specifically, the system first loads the input information, accepts the user's request, loads the historical conversation records, and can also load the input file of the experimental design method and some user-defined parameters (such as dimension, number of samples, etc.).

[0017] In step 204, the backend middleware layer verifies the validity of the input file. If the input file is not legitimate, the historical conversation records and user request information are used as a training sample dataset. This triggers the business layer's large model to perform decision training on the training sample dataset based on the input file, output the experimental design strategy, and transmit the conversation to the frontend interaction layer. Otherwise, the input parameters in the input file are reconstructed, and the computational core of the experimental design algorithm is called to start an independent thread, allowing the large model to execute the computation task and output the calculation results.

[0018] Specifically, check whether the loaded file is complete, including checking the parameter file, parameter requirements, whether the values ​​are reasonable, reference standards and other related information.

[0019] Furthermore, if the parameter file fails the validity check, for example, if the parameter file handle does not exist, the system triggers a large model prompt to the front-end. Alternatively, if the parameter file handle exists but the parameters are unreasonable or illegal, the system also prompts the front-end with a natural language prompt stating: "Input parameters not detected, unreasonable or illegal." The user is asked to provide relevant information and explain the input parameters in conjunction with a project case. The conversation is then saved and returned.

[0020] Furthermore, if the parameter file is legal, the system will reconstruct the input parameters read from the file, including conversion types, splicing and packaging, etc. After reading is completed, the independent thread is started by calling the calculation core of the experimental design method to perform the calculation task. The present invention supports multiple experimental design methods, including but not limited to Latin hypercube sampling, factorial design, central composite design, etc. The following is a detailed introduction to the embodiment: Example 1: Experimental Design Method for Latin Hypercube Sampling Latin Hypercube Sampling (LHS) is a statistical method for generating approximately random samples of parameter values ​​from a multidimensional distribution. This method is often used to construct computer experiments or Monte Carlo integration. The following is its mathematical formula: Assume that n samples need to be extracted from d-dimensional space, and the value range of each dimension is normalized to [0,1). For each sample and each dimension , and its coordinate calculation formula is: ; in, Indicates the dimensional random permutation function, Randomly shuffle the order, represents independent uniform random numbers, .

[0021] Example 2: Execution of a central composite experimental design method Central Composite Design (CCD) is a commonly used experimental design method, mainly used in Response Surface Methodology (RSM) to optimize the process. CCD mainly consists of the following parts: factor points (standard running points in full factorial or fractional factorial designs, usually performed at high and low levels of each factor), center points (repeated runs when all factors are set at their intermediate values, which are helpful for estimating pure errors and detecting curvature), star points (expanding outward from the center point along each factor axis, with a distance of The decision may be a specific value chosen to achieve orthogonality, rotation, or other properties).

[0022] For CCD, some key formulas are as follows: Total number of trials ( ): The total number of trials can be calculated using the following formula: ; in, is the number of center points, is the number of factors.

[0023] Value: used to determine the position of the star point. Different strategies can be used to determine Values, such as rotation, then The value can be calculated by the following formula: ; in is the number of factor points.

[0024] Second-order regression model formula: The relationship between the response variable y and k factors: ; in, represents a constant term, represents the linear effect coefficient, represents the quadratic effect coefficient, represents the interaction effect coefficient, Represents random error.

[0025] Example 3: Execution of factorial experimental design method Factorial Design is a statistical experimental design method used to study how a response variable is affected by one or more factors, each factor has at least two levels in the experiment. For a completely randomized factorial design, if there is factors, each factor has levels, the total number of experimental runs It can be calculated by the following formula: ; Each experimental combination corresponds to a response value , the following model can be established: ; in, represents the overall mean response, Indicates the The main effect of the factors, Indicates the and The interaction effect of factors, Indicates the The coded value of each factor (usually -1 or 1), Indicates the The coded values ​​of the factors, Indicates the The coded values ​​of the factors, Indicates the and and The interaction effect of factors, represents experimental error.

[0026] During the calculation process, the main thread keeps interacting with the front-end, which displays a progress bar and pushes the task status in real time (for example, "Calculating, 30% completed").

[0027] Furthermore, while waiting for the calculation to complete, the system will start a new thread and support proactive messaging. For example, users can learn about the application of a certain experimental design method in a particular evaluation and the system's main features. Interactive messaging also allows users to proactively ask questions. After the calculation is complete, the user can abort the task by clicking the "Stop" button, and the system will destroy the corresponding thread and release resources.

[0028] Step 206: After the calculation results are loaded into the large model dialogue, an analysis report is output to the front-end interaction layer.

[0029] Specifically, after the calculation is complete, the system accepts the raw result data and converts it into a standard format for storage, ensuring compatibility and interpretability. The data is then sent to the large model, triggering the result interpretation module on the large model side. Based on historical data and the rationality of the calculation results combined with the engineering context, the system analyzes the rationality of the result data and provides suggestions for its application. An analysis report is then generated and sent to the user. Furthermore, the user side can display the data visually, drawing scientific calculation charts, scatter plots, layered scatter plots, and other methods to help users intuitively assess the uniformity of the sample distribution.

[0030] Furthermore, after all operations and calculations are complete, the raw data, visualizations, and large-scale model analysis reports can be packaged into a standardized file (e.g., a ZIP archive) and stored locally. The front-end interface provides a download link, allowing users to access the complete results package in multiple formats with a single click. A notification message (e.g., "Task completed, click here to download") is also displayed.

[0031] Furthermore, if a computing core is interrupted due to insufficient resources or a logical error, the system automatically logs the error, triggers the large model to generate repair suggestions (such as "Number of samples exceeds memory limit, recommend batch calculation"), and notifies the user to resubmit the task. Regardless of whether the task succeeds or fails, the system automatically releases occupied computing resources to ensure efficient resource utilization.

[0032] In the aforementioned intelligent experiment design method based on a distributed network architecture, the functional loading middleware is designed using a modular architecture, transforming functional components such as the experiment design algorithm and large-scale model decision-making into dynamically loadable modules. New methods can be quickly integrated through plug-ins without modifying the core code, significantly reducing development costs and cycles, and enabling flexible adaptation to multiple scenario requirements. A full-process automation mechanism runs through input file validity verification, training sample generation, large-scale model decision training, computational task execution, and analysis report output, forming a closed loop of "design-verification-improvement" to avoid efficiency losses caused by manual intervention. Large-scale model-driven intelligent interaction and decision support enable it to automatically generate training samples based on historical conversation records and user request information when input file validity is abnormal, drive the large-scale model to conduct decision training, and output strategies. When the input is valid, the algorithm computation core is called by reconstructing parameters to enable the large-scale model to efficiently execute computational tasks. Combined with a flexible multi-task scheduling mechanism, it effectively improves experiment efficiency and reduces computational costs. This design not only enables the system to quickly adapt to the needs of different industries, but also provides a standardized interface for the subsequent expansion of new methods and new scenarios through the functional loading characteristics of the middleware, achieving the unity of the flexibility of the technical architecture and the adaptability of business needs, and improving the development efficiency and versatility of virtual computing experiments.

[0033] In one embodiment, at the front-end interaction layer, historical conversation records, input files of the experimental design algorithm, and user request information are loaded according to the user input information in the system interface.

[0034] In one embodiment, at the backend middleware layer, the input file is verified for validity based on preset verification conditions. The specific steps are as follows: If the parameter file handle in the input file does not exist, the input file is considered invalid. If the parameter file handle exists and the parameters are incorrect, the input file is considered invalid. Except for the above two exceptions, the input file is considered valid.

[0035] In one embodiment, if the input file's legality is abnormal, historical conversation records and user request information are used as a training sample dataset, triggering the business layer's large model to conduct decision training on the training sample dataset based on the input file and output an experimental design strategy. After saving the conversation, it is exported to the front-end interaction layer, generating natural language to prompt the user to adjust the experimental design input information.

[0036] In one embodiment, if the legality of the input file is normal, the input parameters in the input file are reconstructed, and the computing core of the experimental design algorithm is called to start an independent thread. Within the independent thread, the large model of the business layer is triggered to use the reconstructed input parameters as training samples, execute the computing tasks corresponding to the experimental design algorithm, and output the calculation results.

[0037] It is worth noting that the entire process, from parameter file loading and legality verification to calculation task submission, result analysis and visualization, is automated without the need for human intervention. The system implements one-click operation through middleware scheduling and algorithm encapsulation, which improves test efficiency, reduces computing costs, and avoids efficiency bottlenecks caused by traditional reliance on manual experience.

[0038] In one embodiment, while waiting for the calculation to be completed, an independent new thread is started. In the independent new thread, the application of the experimental design algorithm in the engineering operation experiment and the platform functions are generated into a visual natural language through the front-end interaction layer and displayed to the user, so that the user can control the dialogue output of the independent new thread through the function button. If the user selects the stop function button, the dialogue output is terminated and the independent new thread is destroyed.

[0039] It's worth noting that the big model drives intelligent interaction and decision support. Based on historical data and user feedback, the big model continuously iterates, generating parameter optimization strategies and interpretation reports, truly achieving an automated closed-loop "design-verify-improve." Furthermore, the big model is deeply embedded in the experimental design process, proactively providing natural language prompts at key points like parameter input, lowering the learning curve and facilitating user experience.

[0040] In one embodiment, the calculation results are converted into a standard format and stored in a local system.

[0041] In one embodiment, if the computing core is interrupted, the large model is triggered to generate repair suggestions and spit out a dialogue to the front-end interaction layer, prompting the user to resubmit the computing task.

[0042] In one embodiment, Figure 3 As shown in the system function flow chart, the specific steps are as follows: 1. Start: Load history dialogues, Latin hypercube parameter input files, user requests, and other information.

[0043] 2. Parameter file validity check: 2.1 If the verification fails and the parameter file handle does not exist, the front-end will be prompted: "Input parameters not detected." Alternatively, if the parameter file handle exists, the front-end will be prompted: "Parameters are unreasonable or illegal." This prompt will be fed back to the main model, which will generate a natural language dialog to the front-end, including the following information: the required parameter file, the required number of parameters, the reasonable value range of each parameter, reference standards, and an explanation of the input parameters based on engineering examples. Save the dialog and return to the program.

[0044] 2.2 Verify legality and reconstruct the parameters read from the file, including type conversion, splicing and packaging, etc.

[0045] 2.3 Calling the Latin Hypercube computing core LHS: Starting a new thread and triggering the large model: While waiting for calculation, the front-end introduces the application of Latin Hypercube in engineering operation test design and the functions of the platform to the user. When the dialogue is completed or the user presses the stop button to terminate the dialogue, the thread is destroyed.

[0046] 3. Calculate the results and accept the resulting data.

[0047] 4. Draw scientific calculation charts, including scatter plots and layered scatter plots. Calculation results are loaded into the large model dialog, and a dialog is presented to the user. The content includes: a typical raw data table of the calculation results, an analysis of the rationality of the calculation results in the context of the engineering, how to apply the result data, suggestions for plotting, and a sample scatter plot generated by the backend. The calculation results and the graph are then saved to a file, and a download link is provided to the front-end prompt, and the user returns to the program.

[0048] It should be understood that although Figure 2-Figure 3 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 2-Figure 3 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0049] In one embodiment, Figure 4 As shown, an intelligent experiment design system based on a distributed network architecture is provided, including: an information loading module 402, an experiment execution module 404 and a report display module 406, wherein: The information loading module 402 is used to load historical conversation records, input files of the experimental design algorithm, and user request information.

[0050] Experiment execution module 404 verifies the validity of input files at the backend middleware layer. If the input files are not valid, it uses historical conversation records and user request information as a training sample dataset, triggering the business layer's large model to perform decision training on the training sample dataset based on the input files, output the experiment design strategy, and transmit the conversation to the frontend interaction layer. Otherwise, it reconstructs the input parameters in the input file, calls the computational core of the experiment design algorithm, and starts an independent thread, allowing the large model to execute the computation task and output the results.

[0051] The report display module 406 is used to load the calculation results into the large model dialogue and then output the analysis report to the front-end interaction layer.

[0052] The specific definition of the intelligent experiment design system based on a distributed network architecture can be found in the definition of the intelligent experiment design method based on a distributed network architecture above, and will not be repeated here. The various modules in the above-mentioned intelligent experiment design system based on a distributed network architecture can be implemented in whole or in part through software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0053] Those skilled in the art will understand that Figure 4 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0054] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware using a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0055] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0056] The above-described embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person skilled in the art would be able to make numerous modifications and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. An intelligent experiment design method for a distributed network architecture, characterized in that: Applied to a distributed network architecture, the distributed network architecture includes a front-end interaction layer, a back-end middleware layer, and a business layer, the method includes: Load historical conversation records, experimental design algorithm input files, and user request information; At the back-end middleware layer, the validity of the input file is checked. If the validity of the input file is abnormal, the historical conversation records and the user request information are used as a training sample data set, triggering the large model of the business layer to perform decision training on the training sample data set based on the input file, output the experimental design strategy, and spit out the conversation to the front-end interaction layer; otherwise, the input parameters in the input file are reconstructed, and the computing core of the experimental design algorithm is called to start an independent thread, so that the large model executes the computing task and outputs the computing results; After the calculation results are loaded into the large model dialogue, an analysis report is output to the front-end interaction layer.

2. The method according to claim 1, characterized in that Loads historical session records, input files for the DOE algorithm, and user request information, including: In the front-end interaction layer, historical conversation records, input files of the experimental design algorithm, and user request information are loaded according to the user input information in the system interface.

3. The method according to claim 1, characterized in that At the backend middleware layer, the validity of the input file is checked, including: In the back-end middleware layer, the input file is checked for legitimacy according to preset check conditions. The specific steps are: if the parameter file handle does not exist in the input file, the legitimacy of the input file is abnormal; if the parameter file handle exists and the parameters are abnormal, the legitimacy of the input file is abnormal; except for the above two abnormal situations, the legitimacy of the input file is normal.

4. The method according to claim 3, characterized in that If the legality of the input file is abnormal, the historical conversation record and the user request information are used as a training sample data set, triggering the large model of the business layer to perform decision training on the training sample data set based on the input file, output the experimental design strategy, and spit out the conversation to the front-end interaction layer, including: If the legality of the input file is abnormal, the historical conversation records and the user request information are used as a training sample data set, triggering the large model of the business layer to perform decision training on the training sample data set based on the input file and output an experimental design strategy; After saving this dialogue, the dialogue is sent to the front-end interaction layer to generate natural language to prompt the user to adjust the input information of the experimental design.

5. The method according to claim 4, characterized in that If the legality of the input file is normal, reconstruct the input parameters in the input file, call the computation core of the experimental design algorithm to start an independent thread, so that the large model performs the computation task and outputs the computation result, including: If the legality of the input file is normal, reconstruct the input parameters in the input file, call the computing core of the experimental design algorithm to start an independent thread, and in the independent thread, trigger the large model of the business layer to use the reconstructed input parameters as training samples, execute the computing tasks corresponding to the experimental design algorithm, and output the calculation results.

6. The method according to claim 5, characterized in that Before the step of outputting the calculation results, it also includes: While waiting for the calculation to be completed, an independent new thread is started. In the independent new thread, the application of the experimental design algorithm in the engineering operation experiment and the platform function are generated into a visual natural language through the front-end interaction layer and displayed to the user, so that the user can control the dialogue output of the independent new thread through the function button. If the user selects the stop function button, the dialogue output is terminated and the independent new thread is destroyed.

7. The method according to claim 6, characterized in that After the calculation results are loaded into the large model dialogue, and before the step of outputting the analysis report to the front-end interaction layer, the method further includes: The calculation results are converted into a standard format and stored in a local system.

8. The method according to claim 7, characterized in that: After the calculation results are loaded into the large model dialogue and the analysis report is output to the front-end interaction layer, the method further includes: If the computing core is interrupted, the large model is triggered to generate repair suggestions and spit out a dialogue to the front-end interaction layer, prompting the user to resubmit the computing task.

9. An intelligent experiment design system based on a distributed network architecture, characterized in that: The system comprises: Information loading module, used to load historical conversation records, input files of experimental design algorithms and user request information; The experiment execution module is used to verify the validity of the input file at the back-end middleware layer. If the validity of the input file is abnormal, the historical conversation record and the user request information are used as a training sample data set to trigger the large model of the business layer to perform decision training on the training sample data set based on the input file, output the experiment design strategy, and spit out the conversation to the front-end interaction layer. Otherwise, the input parameters in the input file are reconstructed, the computing core of the experiment design algorithm is called to start an independent thread, so that the large model executes the computing task and outputs the computing result. The report display module is used to load the calculation results into the large model dialogue and then output the analysis report to the front-end interaction layer.

Images