Automatic analysis system and method for test data of sonic nozzle products

Abstract

The invention relates to an automatic analysis system and method for test data of sonic nozzle products, and belongs to the technical field of data processing. The method comprises the following steps: importing a test data file and displaying an original data curve; in the multi-mode data analysis mode, working conditions needing to be calculated and analyzed are selected; automatically calling stored historical parameter configuration; calculating parameters are manually adjusted; carrying out data processing and analysis by adopting an inflection point judgment algorithm based on dynamic threshold adjustment, inputting the obtained data into a correlation analysis model, and calculating to obtain a test data curve of each channel and a data correlation analysis result; the test data analysis curve and the data analysis result are visually displayed; and automatically generating a test report according to a preset template. According to the method, human errors in the data processing process can be eliminated, the visualization effect of the data is enhanced, the labor cost is reduced, and the working efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology and relates to an automated analysis system and method for test data of sonic nozzle products. Background Technology

[0002] Sonic nozzles are crucial regulating components in liquid rocket engine system piping, and their flow resistance characteristics directly affect the engine's combustion ratio. The flow resistance characteristics of sonic nozzles are obtained through experimental parameters such as pressure, temperature, and flow rate acquired during processor airflow tests.

[0003] Traditional airflow test data processing involves manually selecting stable data segments, extracting the average parameters for those segments, and then substituting them into formulas for table lookup calculations and report writing. This method has several drawbacks: manual data selection and processing cannot eliminate human error, potentially leading to biased results; the sequential processing of manually importing, selecting, analyzing, configuring parameters, calculating results, and writing reports involves numerous steps and a long chain, increasing overall processing time and efficiency; and the lack of visualization of the correspondence between calculated results and raw data makes it difficult for users to intuitively, accurately, and promptly obtain information or identify problems. Summary of the Invention

[0004] The technical problem solved by this invention is to overcome the shortcomings of the prior art and propose an automated analysis system and method for test data of sonic nozzle products.

[0005] The solution of this invention is: an automated analysis system for test data of sonic nozzle products, comprising:

[0006] (1) Import the test data files of sonic nozzle products into the server system;

[0007] (2) Select the test data file in the server system and display the raw data curve;

[0008] (3) Among the three data analysis modes of single pressure condition calculation, three pressure condition calculation, and recovery coefficient calculation, select the condition that needs to be calculated and analyzed;

[0009] (4) The system automatically retrieves the stored historical parameters of the operating conditions that need to be calculated and analyzed;

[0010] (5) Manually adjust the calculation parameters to ensure they meet the calculation requirements;

[0011] (6) Determine whether the test data file and calculation mode have been selected. If not selected, return to (2); otherwise, execute (7).

[0012] (7) Data processing and analysis are performed using an inflection point determination algorithm based on dynamic threshold adjustment. The obtained data are then subjected to correlation analysis to calculate the experimental data curves and data correlation analysis results for each channel.

[0013] (8) Visualize the experimental data analysis curves and data analysis results;

[0014] (9) Automatically generate test reports according to preset templates.

[0015] Furthermore, the inflection point determination algorithm based on dynamic threshold adjustment includes:

[0016] (71) The experimental data were denoised and smoothed;

[0017] (72) Calculate the rate of change of the trend Δx;

[0018] (73) Set the sliding window size N, and calculate the statistic of the trend change rate Δx within the sliding window;

[0019] (74) Update the upper threshold U and lower threshold L in real time;

[0020] (75) Then determine whether Δx > U or Δx < L. If the condition is met, execute (76); otherwise, return to (74).

[0021] (76) Determine whether the local trend slopes before and after the potential inflection point are opposite. If they are, execute (77); otherwise, return to (74).

[0022] (77) Input a valid inflection point, record the inflection point index and the direction of trend change. If there is more data to be processed, return (74); otherwise, end the inflection point determination.

[0023] Furthermore, the correlation analysis involves: after obtaining the upper and lower inflection points of the pressure measurement channel data curve and several pressure stability segment data index intervals; within the upper and lower inflection point index intervals, for several pressure stability segment index intervals, extracting the temperature channel and flow channel data under the corresponding time dimension of each interval, and simultaneously calculating the average temperature and average flow rate within each index interval.

[0024] Furthermore, the manually adjusted calculation parameters include: nominal value 1 of pressure Pi, nominal value 2 of pressure Pi, nominal value 3 of pressure Pi, stability error of pressure Pi, inflection point determination coefficient, stability time of pressure Pi, and inlet pipe diameter and throat diameter of sonic nozzle products; the data analysis results include average pressure, average temperature, average flow rate, recovery coefficient, and flow coefficient, providing accurate quantitative basis for product performance evaluation.

[0025] Furthermore, the experimental data analysis curve can be visualized, integrating the original curves of each channel's experimental data with the data analysis results to obtain the experimental data analysis curve. The curve's inflection points and stable segments are dynamically marked on it, and interactive events are set so that clicking on the marked points can display specific data information.

[0026] Furthermore, the experimental data file format is Excel or TDMS.

[0027] Furthermore, the historical parameters are stored in a local file in a standardized JSON format. When a user selects a calculation mode, the historical parameters in the JSON file can be read directly and automatically populated into the parameter configuration interface. Users can then directly use the loaded historical parameters or make certain adjustments for data processing and analysis.

[0028] An automated analysis system for test data of sonic nozzle products includes: a file management module, a data analysis and processing module, an interactive visualization module, a multi-mode selection module, a parameter input and historical parameter management module, and an automated report generation module;

[0029] File management module: Enables the import and export of experimental data files, as well as the selection and reading of experimental data files;

[0030] The data analysis and processing module: On the one hand, based on the inflection point determination algorithm with dynamic threshold adjustment, it processes the pressure measurement channel data curves in the test of rocket engine sonic nozzle products, automatically determines the inflection points of the curves and identifies the pressure stability segment; on the other hand, it constructs a correlation analysis model between the pressure stability segment and temperature and flow rate data, and calculates and obtains the data correlation results.

[0031] Interactive visualization module: Utilizes dynamic plotting technology to integrate and visualize the curves of each channel of experimental data with the algorithm analysis results;

[0032] Multi-mode selection module: Provides three data analysis modes: single test condition, three test conditions, and coefficient of recovery calculation, which users can switch flexibly as needed;

[0033] Parameter input and historical parameter management module: Users can manually input calculation parameters for subsequent calculation applications; and complete the calling and configuration of historical parameters in various modes;

[0034] Automated report generation module: After data analysis is completed, the experimental data, analysis results and chart content are automatically filled into specific locations in the report according to the preset template to generate an experimental analysis report.

[0035] Furthermore, the multi-mode selection module is based on the state pattern and designs a working condition management class. By inheriting the base class State, it realizes the switching of multi-mode analysis algorithms, avoids hard coding, and can be quickly developed and adapted to new test scenarios and working conditions.

[0036] Furthermore, the automated analysis system adopts a four-layer architecture design, including: interface layer, data layer, business logic layer and application layer;

[0037] The interface layer is the human-computer interaction interface used to complete the data analysis process.

[0038] The data layer processes and interprets imported experimental data files, as well as loads historical parameter management files;

[0039] The business logic layer is responsible for managing test conditions, executing dynamic threshold algorithms, and building and performing correlation analysis models.

[0040] At the application layer, the visualization of analytical data and the automated generation of reports are completed.

[0041] The advantages of this invention compared to the prior art are:

[0042] (1) Based on the data curve algorithm and data correlation analysis of dynamic threshold adjustment inflection point, this invention realizes the automated analysis of test data of liquid rocket engine sonic nozzle products, replacing manual screening. The processing time of a single set of data is reduced from hours to minutes, which significantly improves the processing efficiency and accuracy, and provides accurate quantitative basis for product performance evaluation.

[0043] (2) The present invention uses dynamic drawing technology to construct an interactive experimental data curve analysis visualization system, which enables users to intuitively obtain key information before and after experimental data analysis, thereby improving the convenience of data analysis.

[0044] (3) The present invention integrates an automated report generation function. Compared with traditional manual operation, the software can automatically generate an experimental analysis report by one click based on the test data, data analysis results and charts, etc., according to the preset template, which improves the efficiency of report preparation and ensures the standardization of report format and the completeness of content. Attached Figure Description

[0045] Figure 1 This is the human-computer interaction interface of the system of the present invention;

[0046] Figure 2 This is a system structure diagram of the present invention;

[0047] Figure 3 This is a flowchart of the inflection point determination algorithm based on dynamic threshold adjustment of the present invention;

[0048] Figure 4 This is a flowchart illustrating the system design of the present invention.

[0049] Figure 5 The curves represent the data analysis results of the system of the present invention under three operating conditions. Detailed Implementation

[0050] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0051] An automated analysis system for test data of sonic nozzle products can be implemented through programming, and is developed using the Python programming language.

[0052] First, use the PyQT5 framework to build a graphical user interface (GUI), such as... Figure 1 As shown, the interface includes the application main window title bar 1, the test data file import / export button 2, the test data file selection 3, the selected test data file display bar 4, the calculation mode selection button 5, the manual parameter input bar 6, the start data analysis button 7, the data analysis results bar 8, the test analysis report generation button 9, and the calculation mode selection window 10. Users can complete the entire test data analysis process from data import, parameter configuration, analysis execution to result output through the orderly operation of the above functional modules.

[0053] like Figure 2 As shown, the overall architecture of the system consists of an interface layer 11, a data layer 12, a business logic layer 13, and an application layer 14. The interface layer 11 is designed based on PyQt5 to provide a user interface through which users can complete the entire data analysis process.

[0054] Data layer 12 processes and interprets the data input by the software. After the user performs corresponding operations in the GUI interface, the data layer reads Excel or TDM experimental data files and historical parameter management files. The historical parameter management module stores key parameters in a local file in a standardized JSON format. When the user selects a calculation mode, the program reads the parameters from the JSON file and automatically populates them into the parameter configuration interface. The user can directly use the loaded historical parameters or make certain adjustments before data processing and analysis.

[0055] The business logic layer 13 is mainly responsible for modules such as test condition management, dynamic threshold algorithm, and correlation analysis model. Among them, the test condition management module is designed based on the state pattern. It defines a base class State, and then creates subclasses for single condition, three condition, and recovery coefficient calculation conditions. These subclasses inherit from the base class State and implement specific data analysis algorithms. The system switches the condition state according to the mode selected by the user and calls the algorithm of the corresponding subclass to perform data analysis.

[0056] In the data processing, the calculation modules such as the dynamic threshold algorithm and correlation analysis model utilize a dynamic threshold adjustment-based inflection point determination algorithm. This algorithm calculates the derivative of the Pi pressure measurement channel data curve and determines the inflection point position based on the dynamically adjusted threshold. For example, an initial threshold is set, and then the threshold is dynamically adjusted according to data fluctuations to adapt to the characteristics of different experimental data. The correlation analysis model establishes mathematical formulas to correlate the pressure value in the stable pressure range with the corresponding temperature and flow rate values, obtaining the calculated average of multiple physical quantities.

[0057] Application layer 14 includes an interactive visualization module and an automated report generation module. The interactive visualization module utilizes the dynamic plotting capabilities of the Matplotlib library to display the raw curves of the experimental data after the user selects the data, facilitating intuitive observation and judgment. During experimental data analysis, multi-channel data curves are plotted in real time, and inflection points and stable segments are marked on the curves. Interactive events are set up; clicking on a marked point displays specific data information, enhancing user interaction with the data.

[0058] File management module: Enables the import and export of test data files between the local computer and the server, and allows for the selection and reading of standard data files.

[0059] Data analysis and processing module: Used to realize the automated analysis of test data. It adopts an inflection point determination algorithm based on dynamic threshold adjustment to process the Pi pressure measurement channel data curve in the test of rocket engine sonic nozzle products, automatically determine the inflection point of the curve and identify the pressure stability segment.

[0060] like Figure 3 The diagram shows the flowchart of the inflection point determination algorithm based on dynamic threshold adjustment. The specific steps are as follows:

[0061] (1) The experimental data were denoised and smoothed;

[0062] (2) Calculate the rate of change of the trend Δx;

[0063] (3) Set the sliding window size N and calculate the statistic of the trend change rate Δx within the sliding window;

[0064] (4) Update the upper threshold U and lower threshold L in real time;

[0065] (5) Then determine whether Δx > U or Δx < L. If the condition is met, execute (6); otherwise, return to (4).

[0066] (6) Determine whether the local trend slope after the potential inflection point Zener is reversed. If it is, execute (7); otherwise, return to (4).

[0067] (7) Input a valid inflection point, record the inflection point index and the direction of trend change. If there is more data to be processed, return (4); otherwise, end the inflection point determination.

[0068] Simultaneously, a correlation analysis model was constructed between the pressure stabilization segment and temperature and flow rate data. The average values ​​of multiple physical quantities within the pressure stabilization segment interval were calculated concurrently to form multi-dimensional data correlation results. The correlation analysis process involved obtaining the upper and lower inflection points of the Pi pressure measurement channel data curve and multiple pressure stabilization segment data index intervals. Within these index intervals, the temperature and flow rate data for each interval were extracted according to the corresponding time dimension, and the average temperature and flow rate within each index interval were calculated concurrently.

[0069] This module reduces the analysis time for a single set of data from hours to minutes, replacing the traditional manual screening method. Furthermore, based on the analyzed pressure, temperature, and flow data, the flow coefficient of the test product is calculated, providing precise quantitative evidence for product performance evaluation.

[0070] Interactive visualization module: Utilizing dynamic plotting technology, this module integrates and visualizes multi-channel curves of experimental data with algorithm analysis results. Users can intuitively observe the dynamic markings of inflection points and stable segments of the original data curves and the analyzed curves. Combined with the real-time calculation results displayed by the software, this forms an interactive "data-analysis-visualization" system, significantly improving the portability of data analysis.

[0071] Multi-mode selection module: Provides three data analysis modes: single-condition test, three-condition test, and coefficient of recovery calculation, which users can switch flexibly as needed. The software architecture of this module is based on the State Pattern for designing the condition management class. By inheriting the base class State, the switching of multi-mode analysis algorithms is realized, avoiding hard coding. It enables rapid development and adaptation optimization for new test scenarios, reducing time and manpower costs.

[0072] Parameter Input and Historical Parameter Management Module: Users can manually input parameters such as "inlet pipe diameter," "nominal Pi value," and "inflection point determination coefficient" in the parameter input interface for subsequent calculations. Furthermore, the software has a parameter memory function. By constructing an automated data storage mechanism, key parameters for multi-mode analysis are stored in standard JSON data format. When a user selects a calculation type, the system can quickly retrieve and configure the parameters.

[0073] Automated report generation module: This module integrates automated report generation functionality. After completing data analysis, users can use the software to automatically fill in experimental data, analysis results, and chart content into specific locations in the report according to preset templates, generating standardized and rigorous experimental analysis reports. This improves report preparation efficiency and ensures the completeness and professionalism of the report content.

[0074] like Figure 4 As shown, an automated analysis method for test data of liquid rocket engine sonic nozzle products includes the following steps:

[0075] 1) Import data files: Users can click the data file import / export button 2 to import the test data files from their local computer to the server computer.

[0076] 2) Data File Selection and Display: After clicking the experimental data file selection button 3, the user can select a standard Excel or TDMS experimental data file for reading. Once selected, the file path and filename will be fully displayed in the selected experimental data file display bar 4. Simultaneously, the interactive visualization module uses dynamic plotting to display the original curves of the experimental data, facilitating intuitive observation and judgment by the user.

[0077] 3) Calculation Mode Selection: When the user clicks the calculation mode selection button 5, a calculation mode selection pop-up window 10 appears, allowing the user to select from three calculation modes: single-condition, three-condition, and recovery coefficient calculation. Clicking the "OK" button completes the selection. Simultaneously, the system automatically reads the historical parameter file and fills the parameters into the manual input parameter field 6. The user can directly use the historical parameters or adjust them before data processing and analysis.

[0078] 4) Data Analysis Execution: After data and parameter preparation is complete, the user clicks the "Start Data Analysis" button 7. If the user's input calculation parameters are unreasonable or the experimental data file / calculation mode is not selected, the software will pop up a prompt box, prompting the user to perform corresponding operations. If the parameters are reasonable and the data file and calculation mode are both selected, the software will process the data based on a series of algorithm modules. After processing, the data analysis results will be displayed in the data analysis results column 8. Simultaneously, if... Figure 5 As shown, the interactive visualization module uses dynamic plotting to draw multi-channel data curves in real time, marking inflection points and stable segments on the data curves. Clicking on a marked point displays specific data information.

[0079] 5) Parameter Adjustment and Reanalysis: If the user is not satisfied with the data analysis results after the analysis is completed, the parameters can be adjusted in the manual input parameter column 6, and the data analysis operation can be performed again until a satisfactory result is obtained.

[0080] 6) Test Analysis Report Generation: Finally, users can click the Test Analysis Report Generation button 9. The software uses the automated report generation module to automatically fill in the data analysis results and charts into the template and generate a Word format test data analysis report.

[0081] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications to the technical solutions of the present invention by utilizing the methods and techniques disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention shall fall within the protection scope of the technical solutions of the present invention.

Claims

1. An automated analysis method for test data of sonic nozzle products, characterized in that, include: (1) Import the test data files of sonic nozzle products into the server system; (2) Select the test data file in the server system and display the raw data curve; (3) Among the three data analysis modes of single pressure condition calculation, three pressure condition calculation, and recovery coefficient calculation, select the condition that needs to be calculated and analyzed; (4) The system automatically retrieves the stored historical parameters of the operating conditions that need to be calculated and analyzed; (5) Manually adjust the calculation parameters to ensure they meet the calculation requirements; (6) Determine whether the test data file and calculation mode have been selected. If not selected, return to (2); otherwise, execute (7). (7) Data processing and analysis are performed using an inflection point determination algorithm based on dynamic threshold adjustment. The obtained data are then subjected to correlation analysis to calculate the experimental data curves and data correlation analysis results for each channel. (8) Visualize the experimental data analysis curves and data analysis results; (9) Automatically generate test reports according to preset templates.

2. The automated analysis method for test data of sonic nozzle products according to claim 1, characterized in that, The inflection point determination algorithm based on dynamic threshold adjustment includes: (71) The experimental data were denoised and smoothed; (72) Calculate the rate of change of the trend Δx; (73) Set the sliding window size N, and calculate the statistic of the trend change rate Δx within the sliding window; (74) Update the upper threshold U and lower threshold L in real time; (75) Then determine whether Δx > U or Δx < L. If the condition is met, execute (76); otherwise, return to (74). (76) Determine whether the local trend slopes before and after the potential inflection point are opposite. If they are, execute (77); otherwise, return to (74). (77) Input a valid inflection point, record the inflection point index and the direction of trend change. If there is more data to be processed, return (74); otherwise, end the inflection point determination.

3. The automated analysis method for test data of sonic nozzle products according to claim 2, characterized in that, The correlation analysis described above involves: after obtaining the upper and lower inflection points of the pressure measurement channel data curve and several pressure stability segment data index intervals; within the upper and lower inflection point index intervals, for several pressure stability segment index intervals, extracting the temperature channel and flow channel data under the corresponding time dimension of each interval, and simultaneously calculating the average temperature and average flow rate within each index interval.

4. The automated analysis method for test data of sonic nozzle products according to claim 1, characterized in that, The manually adjusted calculation parameters include: nominal value 1 of pressure Pi, nominal value 2 of pressure Pi, nominal value 3 of pressure Pi, stability error of pressure Pi, inflection point determination coefficient, stability time of pressure Pi, and inlet pipe diameter and throat diameter of sonic nozzle products; the data analysis results include average pressure, average temperature, average flow rate, recovery coefficient, and flow coefficient, which provide accurate quantitative basis for product performance evaluation.

5. The automated analysis method for test data of sonic nozzle products according to claim 1, characterized in that: The aforementioned test data analysis curve can be visualized, integrating the original curves of each channel's test data with the data analysis results to obtain the test data analysis curve. The curve's inflection points and stable segments are dynamically marked on it, and interactive events are set so that clicking on the marked points can display specific data information.

6. The automated analysis method for test data of sonic nozzle products according to claim 1, characterized in that: The test data file format is Excel or TDMS.

7. The automated analysis method for test data of sonic nozzle products according to claim 1, characterized in that: The historical parameters are stored in a local file in a standardized JSON format. After the user selects the calculation mode, the historical parameters in the JSON file can be read directly and automatically populated into the parameter configuration interface. The user can directly use the loaded historical parameters or make certain adjustments for data processing and analysis.

8. An automated analysis system for test data of sonic nozzle products, characterized in that, include: The module includes a file management module, a data analysis and processing module, an interactive visualization module, a multi-mode selection module, a parameter input and historical parameter management module, and an automated report generation module. File management module: Enables the import and export of experimental data files, as well as the selection and reading of experimental data files; The data analysis and processing module: On the one hand, based on the inflection point determination algorithm of dynamic threshold adjustment, it processes the pressure measurement channel data curves in the test of rocket engine sonic nozzle products, automatically determines the inflection points of the curves and identifies the pressure stability segment; on the other hand, it constructs a correlation analysis model between the pressure stability segment and temperature and flow rate data, and calculates and obtains the data correlation results. Interactive visualization module: Utilizes dynamic plotting technology to integrate and visualize the curves of each channel of experimental data with the algorithm analysis results; Multi-mode selection module: Provides three data analysis modes: single test condition, three test conditions, and coefficient of recovery calculation, which users can switch flexibly as needed; Parameter input and historical parameter management module: Users can manually input calculation parameters for subsequent calculation applications; and complete the calling and configuration of historical parameters in various modes; Automated report generation module: After data analysis is completed, the experimental data, analysis results and chart content are automatically filled into specific locations in the report according to the preset template to generate an experimental analysis report.

9. The automated analysis system for test data of sonic nozzle products according to claim 8, characterized in that, The multi-mode selection module is based on the state pattern and uses a working condition management class. It implements multi-mode analysis algorithm switching by inheriting the base class State, avoiding hard coding and enabling rapid development and adaptation optimization for new test scenarios.

10. The automated analysis system for test data of sonic nozzle products according to claim 8, characterized in that, The automated analysis system adopts a four-layer architecture design, including: interface layer (11), data layer (12), business logic layer (13) and application layer (14); The interface layer (11) is the human-computer interaction interface used to complete the data analysis process. The data layer (12) processes and interprets the imported experimental data files, and loads the historical parameter management files; The business logic layer (13) is responsible for completing the test condition management, dynamic threshold algorithm execution, and correlation analysis model construction and correlation analysis; The application layer (14) completes the visualization of the analyzed data and the automated generation of reports.

Images