Civil aircraft test flight data preprocessing method and system, computer equipment and storage medium

Abstract

The invention relates to a civil aircraft test flight data preprocessing method and system, computer equipment and a storage medium. The method comprises the following steps of S1, importing data; s2, preprocessing the data; s3, storing the data and performing standardized fine adjustment; and S4, exporting the data. By using an Electron architecture and integrating various algorithms, a unified processing platform is provided for data from different sources, and standardized management can be performed according to a time sequence when civil aircraft test flight data is stored, so that subsequent evaluation, analysis, tracing and query processes are simplified, the system supports cross-platform data import, and the system is suitable for popularization and application. The flight data processing system has high customizable data preprocessing capability, realizes efficient data storage and query functions, further provides intuitive data display and export functions and data exchange and cooperation capability among different modules, and provides a comprehensive and efficient solution for processing and analysis of flight data.

Description

Technical Field

[0001] The present invention relates to the field of civil aircraft data processing, and in particular to a civil aircraft flight test data pre-processing method, system, computer equipment and storage medium. Background Art

[0002] During the civil aircraft test flight, a huge amount of data is generated. These data are not only huge in quantity but also of various types, including flight speed, altitude, heading, engine speed, performance parameters, aerodynamic characteristics and other parameter data. These data are not only crucial for real-time monitoring of flight status, but also have important significance for subsequent flight performance analysis and optimization.

[0003] Since test flight data is directly related to the performance evaluation and safety verification of the aircraft, the accuracy of this data is also crucial. In the process of data collection and processing, errors and interference must be strictly controlled to ensure the accuracy and reliability of the data.

[0004] However, in the actual test flight process, it will inevitably be affected by various factors, such as sensor errors and signal interference, which may cause the original data to have noise, outliers or missing values.

[0005] The existing technology usually requires aircraft engineers to install data operation software (such as Excel spreadsheets), Python, MATLAB and other tools on their local computers to cooperate with each other to pre-process different data in a decentralized manner. Therefore, when processing large amounts of complex test flight data, engineers need to manually perform multiple steps. The tools are diverse and not unified, and there are also differences in data formats and processing methods between different tools, which makes the conversion and migration of data between different tools very complicated and increases the possibility of errors. In addition, using different tools for data processing may cause the processed data to be stored in different locations in a decentralized manner, which is not only difficult to manage in a unified manner, but also when the data needs to be updated or modified, due to the lack of a unified version control mechanism, the data version will be chaotic and difficult to track and manage.

[0006] Due to the lack of unified data pre-processing tools, the decentralized processing methods will cause engineers to process data in different or even completely different ways. The lack of unified standards will affect the efficiency and accuracy of data processing, and will be more detrimental to the unified management and integrated analysis of data. When processing large amounts of data, manual operations are not only time-consuming, but also prone to omission or incorrect processing of certain data items. With the continuous increase in test flight missions for various aircraft models, the amount of data is also increasing. The existing data processing tools are diverse and non-uniform, lack standardization, and are difficult to update, maintain and expand, which limits the efficiency and accuracy of data processing and can no longer meet the ever-changing needs of contemporary aircraft engineers and test flight data. Summary of the invention

[0007] In view of the above-mentioned defects of the prior art, a first aspect of the present invention provides a method for pre-processing civil aircraft flight test data, which includes the following steps:

[0008] Step S1: data import; in a cross-platform application built based on the Web technology stack and the Electron framework, a Node.js file system is used to import civil aircraft flight test data from a local or online platform, and data of different time periods are manually selected from the civil aircraft flight test data as raw data to be processed, wherein the raw data includes raw time series data and raw information data;

[0009] Step S2: data preprocessing; calling the preprocessing algorithm of the preset algorithm library to perform a preprocessing process including at least data cleaning and / or noise removal and / or outlier detection and / or feature extraction and / or data transformation and / or resampling and / or normalization on the original time series data to obtain result time series data;

[0010] Step S3: storing data and normalizing and fine-tuning; storing result data through a database and providing a query interface, the result data includes the result time series data and result information data, the result information data includes the original information data and information for calling the query interface; the database includes a time series database IoTDB and a relational database MySQL; the time series database IoTDB is used to store the result time series data and perform normalization and fine-tuning on the result time series; the relational database MySQL is used to store the result information data;

[0011] Step S4: data export; exporting the result data or exporting it after visualization to the desktop of the cross-platform application built based on the Web technology stack and the Electron framework.

[0012] In the above-mentioned civil aircraft flight test data pre-processing method, optionally, the step S2 includes the following steps:

[0013] Step S2.1: Selecting the pre-processing algorithm from the algorithm library;

[0014] Step S2.2: Determine whether the selected pre-processing algorithm meets the current processing requirements; if so, execute step S2.3; otherwise, supplement or recompile the pre-processing algorithm and add it to the algorithm library, and then return to step S2.1;

[0015] Step S2.3: calling the pre-processing algorithm in the algorithm library to perform pre-processing calculations on the original data and obtain the result time series data;

[0016] Step S2.4: perform a conformity assessment on the result time series data; if conforming, execute step S3; otherwise, return to step S2.1 or directly return to step S1 to reselect a time period as the original data to be processed.

[0017] In the civil aircraft flight test data pre-processing method as described above, optionally, the algorithm library includes at least a mathematics library, a machine learning library and a self-compiled algorithm library; the pre-processing calculation includes at least frequency conversion and / or outlier detection and / or interpolation calculation.

[0018] In the civil aircraft flight test data pre-processing method as described above, optionally, in step S2.2, Numba is used to compile a three-dimensional interpolation algorithm and add it to the self-compiled algorithm library.

[0019] In the civil aircraft flight test data pre-processing method as described above, optionally, in step S3, the result time series data is normalized and fine-tuned by normalization or standardization so that the timestamps of each result time series data are aligned to a fixed time interval.

[0020] In the civil aircraft flight test data pre-processing method as described above, optionally, each of the result time series data is unified to the same time axis and the data is interpolated or filtered.

[0021] In the civil aircraft flight test data pre-processing method as described above, optionally, in step S4, Matplotlib is used to plot the result data and transmit it to the desktop of the cross-platform application through Django.

[0022] In order to achieve the above-mentioned object, the second aspect of the present invention provides a civil aircraft flight test data pre-processing desktop system, which includes:

[0023] An import module, which is used to import civil aircraft flight test data from a local or online platform into the desktop system;

[0024] A pre-processing module, wherein the pre-processing module has a built-in algorithm library to perform pre-processing calculations on the data; the algorithm library at least includes a database, a machine learning library, and a self-compiled algorithm library;

[0025] A storage module, wherein the storage module has a built-in database to store various data, and the database includes a time series database IoTDB and a relational database MySQL;

[0026] A display export module, which exports the processed data in the storage module or exports it to the desktop after visual drawing;

[0027] The communication module is connected to each module to realize data exchange between the modules.

[0028] In order to achieve the above-mentioned purpose, the third aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor runs the program, the steps of the civil aircraft flight test data pre-processing method as described in any one of the first aspects above are implemented.

[0029] In order to achieve the above-mentioned purpose, the fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions or computer programs, and when the computer-executable instructions or computer programs are processed and executed, the civil aircraft flight test data pre-processing method as described in any one of the first aspects above is implemented.

[0030] The civil aircraft flight test data pre-processing method, system, computer equipment and storage medium provided by the present invention support the input and conversion of multiple data formats. By using the Electron architecture and integrating multiple algorithms, a unified processing platform is provided for data from different sources, ensuring that civil aircraft flight test data can be standardized and managed in chronological order when stored, thereby simplifying the subsequent evaluation, analysis, tracing and query process. The system supports cross-platform data import, has highly customizable data pre-processing capabilities, and realizes efficient data storage and query functions. In addition, it also provides intuitive data display and export functions, as well as data exchange and collaboration capabilities between different modules. Overall, this invention provides a comprehensive and efficient solution for the processing and analysis of flight data.

[0031] It meets the processing requirements of data from different sources and ensures that civil aircraft test flight data is stored in a standardized manner according to time sequence while being stored in the database, providing convenience for subsequent evaluation and analysis, traceability management, and call query. It realizes cross-platform, diversified data import, powerful and customizable data pre-processing, efficient data storage and query, intuitive data display and export, and data exchange and collaboration between modules, providing a comprehensive and efficient solution for the processing and analysis of flight data.

[0032] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flow chart of a method for pre-processing civil aircraft flight test data of the present invention;

[0034] Figure 2 yes Figure 1Detailed flow chart of the civil aircraft flight test data pre-processing method;

[0035] Figure 3 It is a structural schematic diagram of a civil aircraft flight test data pre-processing desktop system of the present invention. DETAILED DESCRIPTION

[0036] In order to make the technical means, creative features, objectives and effects of the invention easier to understand, the invention is further described below with reference to specific diagrams. However, the invention is not limited to the following implementation cases.

[0037] It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantive technical significance. Any structural modification, change in proportion or adjustment of size, without affecting the effects and purposes that can be achieved by the present invention, should still fall within the scope of the technical contents disclosed by the present invention.

[0038] Terms such as “comprise” and “include” indicate that in addition to the components directly and explicitly stated in the specification and claims, the technical solution of the present invention does not exclude the situation where it has other components that are not directly or explicitly stated.

[0039] Electron is a framework for building front-end desktop applications using JavaScript, HTML, and CSS. Its core modules are all Node.js event triggers, following Node's asynchronous event-driven architecture. The application module is one of these triggers.

[0040] The desktop program built on the Electron framework enables the desktop system to run on multiple operating systems, achieving cross-platform compatibility. At the same time, since the Electron framework makes full use of the Web technology stack, users can enjoy a consistent user experience regardless of which operating system they are on. The Web technology stack generally refers to a set of technologies, tools, and programming languages ​​required to build and maintain Web applications, such as:

[0041] (1) Front-end technology

[0042] HTML / CSS / JavaScript: The basic technologies for building web pages, HTML for structure, CSS for style, and JavaScript for interactivity.

[0043] Front-end frameworks / libraries: React, Angular, Vue.js, etc., used to build user interfaces and single-page applications (SPAs).

[0044] CSS preprocessors: such as Sass, LESS, etc., provide advanced features such as variables and nesting.

[0045] Front-end build tools: such as Webpack, Gulp, Grunt, etc., for modular development, compression, and optimization of resources.

[0046] (2) Backend Technology

[0047] Server-side languages: Node.js, Python (Django, Flask), Ruby (Ruby on Rails), PHP, Java (Spring Boot), etc.

[0048] Database technology: such as MySQL, PostgreSQL, MongoDB, Redis, etc., for data storage and management.

[0049] API design: RESTful API, GraphQL, etc., for communication with separated front-end and back-end.

[0050] Figure 1 and Figure 2 The flow of the civil aircraft flight test data pre-processing method is shown.

[0051] As mentioned above, Node.js is an open source, cross-platform JavaScript runtime environment that allows developers to run JavaScript code on the server side. The Electron framework uses JavaScript, HTML, and CSS to build desktop applications. The Electron framework embeds Chromium (an open source browser project) and Node.js into its binary, allowing developers to maintain a JavaScript code base to create cross-platform desktop applications that can run on Windows, macOS, and Linux. Electron is closely related to Node.js because Electron is essentially a Node.js application, but it focuses on the development of desktop applications rather than the web server side. Electron applications can use web pages as user interfaces and can access native resources of the operating system. In addition, when developing Electron applications, Node.js provides a wealth of modules and APIs (full name Application Programming Interface, application programming interface), allowing developers to easily perform tasks such as network requests and file system operations. Electron extends these capabilities, allowing developers to use JavaScript to directly interact with the desktop environment and build desktop applications with rich user interfaces.

[0052] like Figure 1 As shown, the civil aircraft flight test data pre-processing method of the present invention can mainly include the following steps:

[0053] Step S1: Data import.

[0054] In step S1, in a cross-platform application built based on the Web technology stack and the Electron framework, a Node.js file system is used to import civil aircraft flight test data from a local or online platform, and data of different time periods are manually selected from the civil aircraft flight test data as the original data to be processed, wherein the original data includes original time series data and original information data.

[0055] It should be noted here that time series data refers to data points recorded in chronological order, and each data point is associated with a timestamp, which is often used to record the exact time when an event occurs. Time series data exists in the form of time series, which can be spaced in seconds, minutes, hours, days and other time units. Specifically, the time series data in this embodiment can be flight data collected by aircraft sensors at a certain frequency, which is professional data that requires pre-processing calculations. Information data refers to non-time series data such as object files that contain various information such as parameter names, user names, file formats, file names, or call upper-level interface information.

[0056] Since the civil aircraft test flight data imported from local or online platforms are usually all the data of a route flown in general, not all of this data needs to be calculated. Engineers will select data from a certain period of time for pre-processing calculations, such as selecting 1 minute or 5 minutes for calculation. Because the selection of data segments for a certain period of time is actually based on the standards that aircraft engineers need to evaluate. For example, there is a professional term called "strong crosswind". When an aircraft leaves the factory, it must undergo a verified crosswind flight to verify the aircraft's ability to resist crosswinds. During this process, the aircraft will have a certain degree of tilt. Aircraft engineers will see that the time period corresponds to a strong crosswind based on the changes in certain parameters, and can manually select a data segment that needs pre-processing calculations. The data after pre-processing calculations are usually used as the basis for evaluating aircraft performance.

[0057] In this embodiment, the online platform refers to a platform for sharing data and mass aggregation. Different engineers can upload flight data to the same online platform to share data. When internal engineers need to process flight data, they can directly retrieve and import it from the online platform, without having to copy data from different sources through a USB flash drive, thus eliminating the need for cloning and migrating a large amount of data, improving the efficiency of data transmission and reducing the probability of errors caused by human factors.

[0058] Step S2: data pre-processing.

[0059] In step S2, the aircraft engineer performs pre-processing calculations on the original time series data by calling the pre-processing algorithm in the algorithm library to obtain the result time series data.

[0060] As mentioned above, the original time series data can be the flight data collected by the aircraft sensor at a certain frequency. This is professional data that needs to be pre-processed and calculated, because the result time series data obtained by the processing and calculation will be used as the basis for engineers to evaluate the performance of the aircraft. The present invention only pre-processes these time series data stored in the time series database IoTDB, and other information data will be stored in the relational database MySQL together with the automatically synchronized and updated interface information.

[0061] Specifically, if Figure 2 As shown, step S2 may specifically include the following steps:

[0062] Step S2.1: Select a pre-processing algorithm from an algorithm library;

[0063] Step S2.2: The engineer determines whether the selected pre-processing algorithm meets the current processing requirements; if so, execute step S2.3; otherwise, since some libraries are not available in Python itself, the engineer needs to compile some algorithms by himself to meet the data processing requirements. Specifically, the filter model can be modified in real time on the front end according to the desired algorithm logic, and Numba can be used to compile, supplement or recompile pre-processing algorithms such as three-dimensional interpolation algorithms and add them to the algorithm library to perform real-time calculations on the data, and then return to step S2.1;

[0064] In this step, Numba can convert Python and NumPy code into optimized machine code, which is particularly suitable for scientific computing and numerically intensive tasks, and can significantly improve the speed and efficiency of data computing and processing. This is because Numba compiles Python code into local machine code at runtime, thereby increasing execution speed. In addition, Numba provides parallel execution capabilities, which can use multi-core CPUs to accelerate calculations.

[0065] Step S2.3: Call the pre-processing algorithm in the algorithm library to perform pre-processing calculations on the original data and obtain the resulting time series data;

[0066] Step S2.4: The engineering staff conducts a compliance assessment on the result time series data; if it complies, the subsequent steps are continued; otherwise, the process returns to step S2.1 or directly returns to step S1 to reselect the time period as the original data to be processed.

[0067] It should be noted here that, since the flight states corresponding to the data segments selected by different engineers are different, in step S2.4, since the parameter types of the result time series data are various and the corresponding normal value ranges are different, the conformity assessment of the result time series data can be manually assessed by the engineers based on the historical flight conditions and personal experience. The engineers who need to process the flight data will definitely know whether the parameters corresponding to each result time series data meet the expectations. These do not affect the key contents of the present invention. For example, a value that should have been between 10 and 100 becomes 10,000, and the engineers can intuitively judge that the data is abnormal. The above description is only used as an example to illustrate that the engineers can intuitively make conformity judgments on the result time series data, and does not limit the scope of the data itself.

[0068] In addition, in response to the problem that the result time series data may not meet the evaluation criteria but the data pre-processing algorithm in the algorithm library meets the current data processing requirements, it should be noted that in this case, the re-selected time period in step S2.4 will be executed as the original data to be processed. In short, the engineering personnel re-select a period of test flight data for data analysis and pre-processing.

[0069] The algorithm library of the present invention is an algorithm library that integrates various pre-processing algorithms, which may include a mathematical library, a machine learning library, and a self-compiled algorithm library, which can realize frequency conversion, outlier detection, interpolation calculation and other operations on data. The pre-processing algorithm refers to a series of methods for preliminarily processing the raw data of the flight test in the process of data analysis, machine learning or signal processing to improve the efficiency of subsequent processing. The pre-processing calculation of the data is based on the Django backend and uses Python's rich data processing ecology to complete the calculation.

[0070] In an optional embodiment, the pre-processing algorithm may include the following functions:

[0071] (1) Data Cleaning:

[0072] Remove invalid or erroneous data records;

[0073] Fill missing values;

[0074] Correct inconsistent data.

[0075] (2) Data Integration:

[0076] Merge data from different data sources;

[0077] Resolve inconsistencies between data sources.

[0078] (3) Data Selection: Select the data subset that is most relevant to the analysis task.

[0079] (4) Data Transformation:

[0080] Normalization: adjust the data to the same scale;

[0081] Standardization: Make the data have zero mean and unit variance;

[0082] Logarithmic transformation, power law transformation, etc.: reduce the skewness of the data.

[0083] (5) Feature Extraction:

[0084] Principal Component Analysis (PCA): dimensionality reduction technique to extract the main features of the data;

[0085] Independent Component Analysis (ICA): Used to separate statistically independent sources in a signal.

[0086] (6) Feature Selection: Select the most informative features and reduce the number of features.

[0087] (7) Data Discretization: Convert continuous attributes into discrete attributes.

[0088] (8) Noise Reduction: Use filters to remove random noise in the data.

[0089] (9) Anomaly Detection: Identify and process anomalies or outliers in data.

[0090] (10) Data Resampling: Adjusting the sampling rate of data, for example in time series analysis.

[0091] (11) Data Smoothing: Use smoothing techniques to reduce data fluctuations.

[0092] (12) Data Rescaling: Adjusting the range of data, for example, scaling the data to the [0,1] interval.

[0093] (13) Data Encoding: Convert non-numeric data to numeric data, for example, using One-Hot Encoding.

[0094] (14) Time Series Analysis: Specific preprocessing methods for time series data, such as sliding average, time decomposition, etc.

[0095] Step S3: Store the data and normalize the fine-tuning.

[0096] The result data can be stored in a database and a query interface can be provided. The result data can include result time series data and result information data. The result information data can include original information data and call interface information. The database can specifically include a time series database IoTDB and a relational database MySQL.

[0097] In this embodiment, the time series database IoTDB is used to store the result time series data, the relational database MySQL is used to store the result information data, and the query interface is an API interface.

[0098] It should be noted here that step S2 of the present invention will directly and synchronously store the resulting time series data in the time series database IoTDB after the data pre-processing calculation, rather than first storing it in the database and then reading it and then pre-processing the data. The storage mechanism of the time series database IoTDB adopts column storage and partitioning technology, which helps to improve the performance of data storage and query. Its main advantage lies in efficient data storage and query. In this embodiment, the time series database IoTDB is directly connected to the online platform for importing data, which is convenient for different engineering personnel to directly call the data for professional subject analysis and flight test report generation.

[0099] Since the timestamp of flight data is different from the regulations of IoTDB, the source data of the business end or test flight may not necessarily change with the version of IoTDB, so the engineering staff will make adaptive modifications to the built-in functions of IoTDB. For example, the pre-processing algorithm in the algorithm library integrated into IoTDB can perform normalized fine-tuning on the result time series data input into IoTDB as the final output data.

[0100] In this embodiment, after the original time series data is processed by the data pre-processing calculation of step S2, it is stored in the time series database IoTDB and all data segments with non-equal time intervals are normalized again through the algorithm library. The time series data can be timestamp aligned by normalization or standardization. Specifically, the time axis can be unified through interpolation or filtering processing methods to ensure that the civil aircraft test flight data is stored in time series while being stored in the database, which can reflect the changes in the parameters of the aircraft during this period of time, which provides convenience for subsequent evaluation and analysis, traceability management, and call query.

[0101] Step S4: Data export.

[0102] Export the result data or visualize the plot to the desktop side of the application.

[0103] In step S4, Matplotlib can be used to visualize the result data and transmit it to the desktop of a cross-platform application built on the Electron framework through Django. Matplotlib is a drawing library for the Python programming language, which is widely used to generate charts and graphs in various formats. It provides a drawing framework similar to MATLAB based on NumPy, supports multiple formats such as PNG, PDF, SVG, EPS, and rich chart types such as line charts, scatter plots, bar charts, pie charts, histograms, box plots, etc., making it very suitable for scientific computing and data visualization.

[0104] In order to achieve the above-mentioned purpose, the present invention also provides a civil aircraft flight test data pre-processing desktop system.

[0105] Based on the Electron framework, the present invention proposes a desktop system for pre-processing civil aircraft flight test data. The desktop program is built based on the Electron framework, making full use of Electron's cross-platform characteristics and rich Web technology stack, so that the program can adapt to multiple operating systems and achieve a consistent customer experience. The details have been described in detail above, so I will not go into details here.

[0106] Figure 3 It is a structural schematic diagram of a civil aircraft flight test data pre-processing desktop system of the present invention.

[0107] The civil aircraft flight test data pre-processing desktop system is equivalent to a civil aircraft flight test data pre-processing platform, which includes functions such as data import, data pre-processing, data storage, result display and data export.

[0108] like Figure 3 As shown, the desktop system may specifically include: an import module, a pre-processing module, a storage module, a display export module and a communication module.

[0109] The import module is used to import civil aircraft test flight data from local or online platforms into the desktop system. It implements file functions using the Node.js file system API under the Electron framework. It can support the input of multiple data formats and realize the conversion of multiple data formats to meet the processing requirements of data from different sources, such as txt, CSV, Excel, JSON, etc.

[0110] The pre-processing module has a built-in algorithm library to perform pre-processing analysis and calculation on the data. The algorithm library can include a database, a machine learning library, and a self-compiled algorithm library. It can perform frequency conversion, outlier detection, interpolation calculation, etc. After selecting the corresponding processing algorithm, you can perform pre-processing calculation on the relevant data. The pre-processing calculation module uses Python's rich data processing ecology based on the Django backend to complete the calculation.

[0111] The storage module has a built-in database to store various data and provide a data query interface, allowing users to conditionally retrieve and obtain data. The database includes the time series database IoTDB and the relational database MySQL.

[0112] The display export module exports the processed data in the storage module or exports it to the desktop after visual drawing. Engineers can use Python's built-in Matplotlib library to draw the data and transfer it to the front-end desktop for display through Django.

[0113] The communication module communicates with each module to realize data exchange between modules. Engineers can make HTTP requests in browsers and Node.js based on the Axios network request library. When data is exchanged, such as when the front-end and back-end communicate with each other, a GET or POST request is sent so that data can be transmitted through the interface.

[0114] In order to achieve the above-mentioned purpose, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor runs the program, the steps of the civil aircraft flight test data pre-processing method as described in any of the aforementioned embodiments can be implemented.

[0115] The processor and the memory may be provided separately or integrated together, for example, integrated into a system on chip (SOC) of a terminal device.

[0116] In order to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are processed and executed, the civil aircraft flight test data pre-processing method as described above is implemented.

[0117] The computer-readable storage medium is, for example, a memory. The memory may be a volatile memory or a nonvolatile memory, or the memory may include both a volatile memory and a nonvolatile memory. Among them, the nonvolatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM).

[0118] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which can be personal computers, servers or network devices, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.

[0119] The preferred specific embodiments of the present invention are described in detail above. It should be understood that ordinary technicians in the field can make many modifications and changes based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by technicians in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for pre-processing civil aircraft flight test data, characterized in that: The following steps are involved: Step S1: data import; in a cross-platform application built based on the Web technology stack and the Electron framework, a Node.js file system is used to import civil aircraft flight test data from a local or online platform, and data of different time periods are manually selected from the civil aircraft flight test data as raw data to be processed, wherein the raw data includes raw time series data and raw information data; Step S2: data pre-processing; Calling a pre-processing algorithm of a preset algorithm library to perform a pre-processing process including at least data cleaning and / or noise removal and / or outlier detection and / or feature extraction and / or data transformation and / or resampling and / or normalization on the original time series data to obtain result time series data; Step S3: storing data and normalizing and fine-tuning; storing result data through a database and providing a query interface, the result data includes the result time series data and result information data, the result information data includes the original information data and information for calling the query interface; the database includes a time series database IoTDB and a relational database MySQL; the time series database IoTDB is used to store the result time series data and perform normalization and fine-tuning on the result time series data; The relational database MySQL is used to store the result information data; Step S4: data export; exporting the result data or exporting it after visualization to the desktop of the cross-platform application built based on the Web technology stack and the Electron framework.

2. The civil aircraft flight test data pre-processing method according to claim 1, characterized in that: The step S2 comprises the following steps: Step S2.1: Selecting the pre-processing algorithm from the algorithm library; Step S2.2: Determine whether the selected pre-processing algorithm meets the current processing requirements; if so, execute step S2.3; otherwise, supplement or recompile the pre-processing algorithm and add it to the algorithm library, and then return to step S2.1; Step S2.3: calling the pre-processing algorithm in the algorithm library to perform pre-processing calculations on the original data and obtain the result time series data; Step S2.4: perform a conformity assessment on the result time series data; if conforming, execute step S3; otherwise, return to step S2.1 or directly return to step S1 to reselect a time period as the original data to be processed.

3. The civil aircraft flight test data pre-processing method according to claim 2, characterized in that: The algorithm library at least includes a mathematics library, a machine learning library and a self-compiled algorithm library; the pre-processing calculation at least includes frequency conversion and / or outlier detection and / or interpolation calculation.

4. The civil aircraft flight test data pre-processing method according to claim 3, characterized in that: In step S2.2, Numba is used to compile the three-dimensional interpolation algorithm and add it to the self-compiled algorithm library.

5. The civil aircraft flight test data pre-processing method according to claim 1, characterized in that: In the step S3, the result time series data is normalized or standardized so as to align the timestamps of each result time series data to a fixed time interval.

6. The civil aircraft flight test data pre-processing method according to claim 5, characterized in that: The result time series data are unified to the same time axis and the data are interpolated or filtered.

7. The civil aircraft flight test data pre-processing method according to claim 1, characterized in that: In step S4, the result data is plotted using Matplotlib and transmitted to the desktop of the cross-platform application via Django.

8. A civil aircraft flight test data pre-processing desktop system, characterized in that: include: An import module, which is used to import civil aircraft flight test data from a local or online platform into the desktop system; A pre-processing module, wherein the pre-processing module has a built-in algorithm library to perform pre-processing calculations on the data; the algorithm library at least includes a database, a machine learning library, and a self-compiled algorithm library; A storage module, wherein the storage module has a built-in database to store various data, and the database includes a time series database IoTDB and a relational database MySQL; A display export module, which exports the processed data in the storage module or exports it to the desktop after visual drawing; The communication module is connected to each module to realize data exchange between the modules.

9. A computer 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 runs the program, the steps of the civil aircraft flight test data pre-processing method as described in any one of claims 1-7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions or a computer program, and when the computer-executable instructions or the computer program are processed and executed, the civil aircraft flight test data pre-processing method as described in any one of claims 1 to 7 is implemented.

Images