Method and system for preprocessing binary offline data of aircraft engine
By organizing parallel data processing in binary form, the problem of slow data processing speed in aircraft engine field monitoring and analysis is solved, efficient data analysis and quality improvement is achieved, and multi-source data collaborative adaptation is supported.
Patent Information
- Application Number
- CN202510505919.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-22
- Publication Date
- 2025-07-25
AI Technical Summary
The offline data processing system for field monitoring and analysis of existing aircraft engines has non-stationary data, a lot of empty value invalid data, and the processing speed is slow, making it difficult to meet the needs of efficient diagnosis and prediction.
Data is organized and processed in binary form, and efficient data analysis is achieved through multiple parallel processing, lock-free ring buffer segmentation, shared memory pool storage, combined with data verification, cleaning and outlier value replacement.
It accelerates the processing process of binary offline data of aircraft engines, improves data quality and processing efficiency, supports multi-source data collaborative adaptation, and meets the needs of high concurrency collaborative processing.
Smart Images
Figure CN120372165A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the technical field of aircraft engine binary offline data processing, and specifically relates to a method and system for preprocessing aircraft engine binary offline data. Background Technique
[0002] The offline data rapid processing system is an important working system of aircraft engines. By systematically analyzing the offline data, internal faults of aircraft engines can be detected in advance, providing a strong basis for predicting the service life and reliability of aircraft engine components, and having an important impact on the safety and reliability of aircraft engines.
[0003] The offline data for aircraft engine outfield monitoring and analysis comes from flight parameter data, controller data, bench test data, and manual inspection data. In practice, due to the influence of many interference factors such as harsh external environments, the offline data for aircraft engine outfield monitoring and analysis generally has phenomena such as non-stationarity and null invalid data, which brings certain difficulties to the diagnostic prediction and maintenance work of aircraft engines.
[0004] In order to improve the quality of the offline data for aircraft engine outfield monitoring and analysis and save processing time, it is necessary to preprocess the received data, clean, detect, and eliminate errors in the data, identify or delete outliers, smooth the noise data, and compensate for the possibly missing data to improve the data quality.
[0005] Data preprocessing is generally the first step before the operation of each algorithm, providing preprocessed data for each algorithm. The input of the discrete data preprocessing for aircraft engine outfield monitoring and analysis is the massive transmission data from flights or tests, and the output is the preprocessed data. After removing data outliers and filling in null values, it is then passed to modules such as key component fault prediction, system performance prediction, and remaining life prediction for use.
[0006] For the data preprocessing of aircraft engine outfield monitoring and analysis, it is first necessary to check the integrity of the input data, and then perform data cleaning and outlier replacement.
[0007] Missing data is a problem often encountered in data cleaning, generally divided into missing values and null values. A missing value means that the value actually exists but is not stored in the field to which the value belongs, and a null value means a value that is empty because it actually does not exist. The common processing methods for missing data are to ignore the tuples, predict the missing values of attributes through the dependency relationship between attributes, expert experience values, and use the available average value, median value, maximum value, minimum value, or more complex probability statistical function values (for example, it can be determined by regression, tools using Bayesian formalization methods based on derivation, or decision tree induction) for replacement.
[0008] Outliers in the data can be identified using statistical analysis methods. For example, calculate the mean, standard deviation, range of values, the number and frequency of null values, maximum value, minimum value, etc. of a certain field. Based on these statistical values and relevant heuristic rules, outliers in the data can be discovered, and values outside the confidence interval of the field are considered outliers. Data mining techniques can also be used to discover outliers in the data.
[0009] In the current aircraft engine offline data rapid processing system, when processing.PHY and.DAT files, there is a lack of uniformity in separation, and the processing speed is slow. The processing time for 100 megabytes of stored data exceeds 100 seconds, especially highlighting the slow data processing in scenarios where a large amount of data is waiting to be processed. In addition, the predictive analysis of the traditional offline data processing system is mainly achieved through the oil debris monitor and oil spectroscopy analysis. When the aircraft engine is operating, a large number of metal chips are generated by the relatively moving components inside it, mixed with the lubricating oil and participating in the lubricating oil circulation. Larger particles will be intercepted by the oil debris monitor, but only a small part of the fine particles are captured by the oil debris monitor, and most of the fine particles still remain in the lubricating oil. Using the oil debris monitor to judge the engine wear rate and conduct lubricating oil predictive analysis often has a large error.
[0010] Currently, there are mainly two ways to conduct research on aircraft engine offline data rapid processing technology. One is to use the MATLAB platform to process matrices / arrays through an interpreted language, which includes control statements, functions, data structures, input, output, and object-oriented programming features. Users can synchronize input statements with execution commands in the command window, or they can write a large and complex application program first and then run it together. The syntax features of the MATLAB platform are very similar to those of the C++ language and are simpler, conforming to the writing format of mathematical expressions by scientific and technical personnel, facilitating the use of non-computer professional scientific and technical personnel, and having good portability and extremely strong scalability. The other is to use the efficient high-level data structures provided by Python and simple object-oriented programming. The syntax and dynamic typing of Python, as well as the nature of the interpreted language, make it a programming language for writing scripts and rapid application development on most platforms. With continuous version updates and the addition of new language features, it has gradually been used in the development of independent and large projects. The Python interpreter is easy to expand and can also be used as an extended programming language in customizable software. Python has a rich standard library, providing source code or machine code applicable to various major system platforms.
[0011] The MATLAB platform has efficient numerical and symbolic calculation functions, which can free users from complex mathematical operation analyses. Its friendly user interface and natural language close to mathematical expressions are easy to learn and master. It also has rich application toolboxes, such as the signal processing toolbox and communication toolbox, which can provide users with a large number of convenient and practical processing tools for handling matrix data. However, it has a slow running speed and serious memory consumption. Python is a language representing the idea of simplicity. It is easy to learn, read, and maintain, and is free and open-source. In some cases, its processing speed is better than that of MATLAB, but there is still a considerable gap from being used on-site for aircraft engines.
[0012] Both MATLAB and Python show the characteristics of being convenient and easy to use in dealing with strings. However, they both have low processing efficiency and are not suitable for high-speed, high-frequency, and large-data-volume situations, making it difficult to apply to the offline data processing for aircraft engine outfield monitoring and analysis. To effectively improve data processing efficiency and make full use of the computer's CPU resources, data is organized and processed in binary form. Binary is naturally compatible with logical operations. Binary number operations are simple, greatly simplifying the structure of the operation components in calculations. Implementing in binary form is technically easy. It is very easy to represent binary digits 0 and 1 with bistable circuits. Since binary only uses two digits, 0 and 1, it is not easy to make mistakes when transmitting and processing data. Therefore, it can ensure that the computer has high reliability. Compared with decimal numbers, the operation rules of binary are simpler, which can simplify the structure of the arithmetic unit and is conducive to improving the operation speed.
[0013] With the progress of technology, the multi-source data collaborative adaptation technology has emerged. The multi-source data collaborative adaptation technology uses high-concurrency collaborative processing technology, dynamic allocation of virtual memory pools, and lightweight service technology, etc., to achieve the multi-source data collaborative adaptation of the intelligent diagnosis technology based on the multi-source intelligent monitoring system. For the binary offline data used for aircraft engine outfield monitoring and analysis, which comes from flight parameter data, controller data, bench test data, and manual inspection data, and has characteristics such as diverse sampling frequencies and field strength drift of surge data, vibration data, and combustion oscillation data, the multi-mode service method can be used to achieve the switching of the diagnostic prediction service mode to meet the requirements of the multi-algorithm module scheduling and execution integration of the aircraft engine health management system. Also, the extensible general markup technology can be used to provide a unified method description and process the structured data exchange independent of application programs or algorithms, providing adaptation technology support for the data exchange and collaboration of each subsystem and algorithm module of the aircraft engine health management, and at the same time opening up the indexing, sorting, searching, and related consistency channels to the database.
[0014] The acceleration of the off-site monitoring and analysis of aircraft engine binary offline data through high-concurrency collaborative processing and the rapid processing of binary offline data are of great significance for subsequent early warning, alarm, taking proactive maintenance measures, reducing fault losses, and accident incidence rates. Since the maturity of high-concurrency collaborative processing technology in offline applications is relatively low, and the amount of data involved in the off-site monitoring and analysis of aircraft engine binary offline data is large, how to effectively increase the data processing efficiency, improve the parsing speed, eliminate invalid data, clean dirty data, and supplement missing values while ensuring the service life and reliability prediction of aircraft engine components is an important issue currently faced. In view of this, this application is proposed. Summary of the Invention
[0015] The purpose of this application is to provide a method and system for preprocessing aircraft engine binary offline data to improve the processing efficiency and parsing speed of aircraft engine off-site monitoring and analysis of binary offline data.
[0016] The technical solution of this application is as follows:
[0017] On the one hand, a method for preprocessing aircraft engine binary offline data is provided, including:
[0018] Step 1: Receive aircraft engine binary offline data;
[0019] Segment the aircraft engine binary offline data and send them into a lock-free circular buffer respectively. Through multi-channel parallel processing, perform binary conversion, and store each block of aircraft engine binary offline data after binary conversion into a large shared memory pool for multi-channel multiplexing output;
[0020] Step 2: Perform binary data verification on the aircraft engine binary offline data;
[0021] Step 3: Extract valid data from the aircraft engine binary offline data;
[0022] Step 4: Parse the valid data in the aircraft engine binary offline data;
[0023] Step 5: Perform data cleaning and outlier replacement on the valid data in the aircraft engine binary offline data.
[0024] According to at least one embodiment of this application, in the above method for preprocessing aircraft engine binary offline data, in Step 1, receiving aircraft engine binary offline data includes QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc., and perform file format adaptation on QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data.
[0025] According to at least one embodiment of the present application, in the above-mentioned aircraft engine binary offline data preprocessing method, in step 2, the data header, data tail, data length and check word of the aircraft engine binary offline data are checked to check the integrity of the aircraft engine binary offline data, and incomplete aircraft engine binary offline data are discarded, and only complete aircraft engine binary offline data are retained for subsequent processing.
[0026] According to at least one embodiment of the present application, in the above-mentioned aircraft engine binary offline data preprocessing method, in step five, data cleaning is performed on the valid data in the aircraft engine binary offline data, missing data is identified, tuples are ignored, missing attribute values and expert experience values are predicted through the dependency relationship between attributes, and average values, median values, maximum values, minimum values or more probabilistic statistical function values can be used to replace them;
[0027] By using statistical analysis methods or data mining technology, outliers in valid data in the engine binary offline data are identified and replaced.
[0028] According to at least one embodiment of the present application, the above-mentioned aircraft engine binary offline data preprocessing method further includes:
[0029] Step 6: Supplement the time column with the valid data in the aircraft engine binary offline data to obtain the aircraft engine binary offline data of each time column.
[0030] On the other hand, there is provided an aircraft engine binary offline data preprocessing system, including a data receiving module, a data checking module, a data extracting module, a data parsing module, a data cleaning and outlier replacement module;
[0031] The data receiving module is used to receive the aircraft engine binary offline data, and divide the aircraft engine binary offline data, respectively send them into the lock-free ring buffer, perform binary conversion through multi-channel parallel processing, and store each block of aircraft engine binary offline data after binary conversion into a large shared memory pool for multiplexing output;
[0032] The data verification module is used to perform binary data verification on the aircraft engine binary offline data;
[0033] The data extraction module is used to extract valid data from the aircraft engine binary offline data;
[0034] The data analysis module is used to analyze the valid data in the aircraft engine binary offline data;
[0035] The data cleaning and outlier replacement module is used to clean the valid data and replace outliers in the aircraft engine binary offline data.
[0036] According to at least one embodiment of the present application, in the above-mentioned aircraft engine binary offline data preprocessing system, the data receiving module receives aircraft engine binary offline data, including QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc., and adapts the file formats of the QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data.
[0037] According to at least one embodiment of the present application, in the above-mentioned aircraft engine binary offline data preprocessing system, the data verification module verifies the data header, data tail, data length, and check word of the aircraft engine binary offline data to check the integrity of the aircraft engine binary offline data, eliminates the incomplete aircraft engine binary offline data, and only retains the complete aircraft engine binary offline data for subsequent processing.
[0038] According to at least one embodiment of the present application, in the above-mentioned aircraft engine binary offline data preprocessing system, in the data cleaning and outlier replacement module, data cleaning is performed on the valid data in the aircraft engine binary offline data, defective data is identified, tuples are ignored, missing attribute values are predicted through the dependency relationship between attributes, expert experience values, and available average values, median values, maximum values, minimum values, or more probability statistical function values are used for substitution;
[0039] Outliers in the valid data of the aircraft engine binary offline data are identified by means of statistical analysis methods or data mining techniques, and the outliers are substituted.
[0041] According to at least one embodiment of the present application, the above-mentioned aircraft engine binary offline data preprocessing system further includes a data time column supplement module;
[0042] The data time column supplement module is used to supplement the time column for the valid data in the aircraft engine binary offline data to obtain the aircraft engine binary offline data of each time column.
[0043] The present application has at least the following beneficial technical effects:
[0044] Provided is a method and system for preprocessing aircraft engine binary offline data. It is designed to, after receiving the aircraft engine binary offline data, perform data segmentation, send them into a lock-free circular buffer respectively, conduct binary conversion through multi-channel parallel processing, store each block of the aircraft engine binary offline data after binary conversion into a large shared memory pool respectively, and then respectively carry out operations such as binary data verification, valid data extraction, data parsing, data cleaning and outlier replacement, and supplementing the time column. Finally, the result is output through multiplexing to complete the preprocessing of the aircraft engine binary offline data, which can effectively accelerate the realization of the fast processing flow of the aircraft engine binary offline data. Description of the Drawings
[0045] Figure 1 is a schematic diagram of the method for preprocessing aircraft engine binary offline data provided by an embodiment of the present application;
[0046] Figure 2 is a schematic diagram of adapting the file format of the received aircraft engine binary offline data file provided by an embodiment of the present application;
[0047] Figure 3 is a schematic diagram of segmenting the aircraft engine binary offline data provided by an embodiment of the present application;
[0048] Figure 4 is a schematic diagram of the system for preprocessing aircraft engine binary offline data provided by an embodiment of the present application.
[0049] For better illustration of this embodiment, some contents in the drawings will be omitted, enlarged or reduced, which are only for exemplary illustration and should not be construed as a limitation to the present application. Detailed Embodiments
[0050] To make the technical solutions and their advantages of the present application clearer, the technical solutions of the present application will be further described clearly and completely below in conjunction with the drawings. It can be understood that the specific embodiments described herein are only part of the embodiments of the present application, which are only used to explain the present application and not to limit the present application. It should be noted that for the convenience of description, only the parts related to the present application are shown in the drawings, and other related parts can refer to the general design.
[0051] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of the present application should have the ordinary meanings understood by those of ordinary skill in the art to which the present application belongs. The "including" used in the description of the present application means that the concept appearing before this word covers the concepts listed after this word and their equivalents, without excluding other related concepts.
[0052] A method for preprocessing aircraft engine binary offline data is as Figure 1 shown.
[0053] Step 1: Receive the binary offline data of the aircraft engine.
[0054] The offline data for outfield monitoring and analysis of the aircraft engine is sourced from flight parameter data, controller data, bench test data, and manual inspection data, etc., and usually includes QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc.
[0055] Receiving the binary offline data file of the aircraft engine means receiving QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc. The QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data are in different formats and need to be adapted to the file format before performing file content parsing and processing, as Figure 2 shown.
[0056] To effectively accelerate the processing of the binary offline data of the aircraft engine and shorten the data processing time, the received original binary data is innovatively processed as follows:
[0057] The binary offline data of the aircraft engine is segmented and sent to a lock-free circular buffer respectively. Through multi-channel parallel processing, binary conversion is performed. Each block of the binary offline data of the aircraft engine after binary conversion is stored in a large shared memory pool respectively, and multi-channel multiplexing output is carried out, and they wait for subsequent file content parsing and processing respectively, as Figure 3 shown.
[0058] Step 2: Perform binary data verification on the binary offline data of the aircraft engine.
[0059] Verify the data header, data tail, data length, and checksum of the binary offline data of the aircraft engine to check the integrity of the binary offline data of the aircraft engine. Discard the incomplete binary offline data of the aircraft engine and only retain the complete binary offline data of the aircraft engine for subsequent processing.
[0060] Step 3: Extract the valid data from the binary offline data of the aircraft engine.
[0061] Extract the valid data from the binary offline data of the aircraft engine, that is, extract the valid data content for subsequent processing.
[0062] Step 4: Parse the valid data in the binary offline data of the aircraft engine.
[0063] Step 5: Perform data cleaning and outlier replacement on the valid data in the binary offline data of the aircraft engine.
[0064] Clean the valid data in the binary offline data of aircraft engines, identify missing data, ignore tuples, predict missing attribute values, expert experience values, and replace them with available average values, median values, maximum values, minimum values, or more complex probability statistical function values through the dependency relationships between attributes.
[0065] Use statistical analysis methods or data mining techniques to identify outliers in the valid data of the binary offline data of aircraft engines and replace the outliers. For the specific replacement method, refer to the replacement method for missing data.
[0066] Step 6: Supplement the time column for the valid data in the binary offline data of aircraft engines.
[0067] Supplement the time column for the valid data in the binary offline data of aircraft engines to obtain the binary offline data of aircraft engines for each time column, and complete the preprocessing of the binary offline data of aircraft engines.
[0068] In the method for preprocessing binary offline data of aircraft engines disclosed in the above embodiments, it is designed to perform data segmentation after receiving the binary offline data of aircraft engines, send them into a lock-free circular buffer respectively, perform binary conversion through multi-channel parallel processing, store each block of binary offline data of aircraft engines after binary conversion into a large shared memory pool respectively, and then carry out operations such as binary data verification, valid data extraction, data parsing, data cleaning and outlier replacement, and supplementing the time column respectively. Finally, multiplex the output results to complete the preprocessing of the binary offline data of aircraft engines, which can effectively accelerate the realization of the fast processing process of the binary offline data of aircraft engines.
[0069] An aircraft engine binary offline data preprocessing system is used to implement the method for preprocessing binary offline data of aircraft engines disclosed in the above embodiments, can be embedded in an offline data fast processing system, and preprocesses the binary offline data of aircraft engines, including a data reception module, a data verification module, a data extraction module, a data parsing module, a data cleaning and outlier replacement module, and a data time column supplement module, as Figure 4 shown.
[0070] The data reception module is used to receive the binary offline data of aircraft engines, including receiving QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc., adapt the file formats of QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data, and carry out file content parsing and processing.
[0071] And split the binary offline data of the aircraft engine, and send them into the lock-free circular buffer respectively. Through multi-path parallel processing, perform binary conversion, and store each block of the binary offline data of the aircraft engine after binary conversion into the large shared memory pool respectively, perform multi-path multiplexing output, and wait for subsequent file content parsing and processing respectively.
[0072] The data verification module is used to perform binary data verification on the binary offline data of the aircraft engine.
[0073] Perform verification on the data header, data tail, data length, and check word of the binary offline data of the aircraft engine to check the integrity of the binary offline data of the aircraft engine, eliminate the incomplete binary offline data of the aircraft engine, and only retain the complete binary offline data of the aircraft engine for subsequent processing.
[0074] The data extraction module is used to extract the valid data from the binary offline data of the aircraft engine, and obtain the valid data in the binary offline data of the aircraft engine, that is, obtain the valid data content for subsequent processing.
[0075] The data parsing module is used to parse the valid data in the binary offline data of the aircraft engine.
[0076] The data cleaning and outlier replacement module is used to perform data cleaning and outlier replacement on the valid data in the binary offline data of the aircraft engine.
[0077] Perform data cleaning on the valid data in the binary offline data of the aircraft engine, identify missing data, ignore tuples, predict missing attribute values, expert experience values, and available average values, median values, maximum values, minimum values, or more complex probability statistical function values through the dependency relationship between attributes for replacement.
[0078] Use statistical analysis methods or data mining techniques to identify outliers in the valid data of the binary offline data of the aircraft engine, and replace the outliers. The specific replacement method can refer to the replacement method of missing data.
[0079] The data time column supplement module is used to supplement the time column for the valid data in the binary offline data of the aircraft engine.
[0080] Supplement the time column for the valid data in the binary offline data of the aircraft engine to obtain the binary offline data of the aircraft engine with each time column, and complete the preprocessing of the binary offline data of the aircraft engine.
[0081] The aircraft engine binary offline data preprocessing system disclosed in the above embodiments divides the aircraft engine binary offline data, sends it into a lock-free circular buffer, and performs binary conversion through multi-channel parallel processing to accelerate data extraction and integration, avoiding the situations of inconsistent original data formats, inconsistent parameter quantities, and repeated processing of original data by multiple analysis modules. In addition, it has configurable data formats and can output multiple different preprocessing files simultaneously to interface with different data analysis modules. Through multi-source data collaborative adaptation, it innovatively proposes high-concurrency collaborative processing technology, dynamic allocation of virtual memory pools, and lightweight services to improve the efficiency and reliability of the system.
[0082] In addition, those skilled in the art should also be able to realize that each module of the aircraft engine binary offline data preprocessing system disclosed in the embodiments of the present application can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the functions are generally described according to their functions in the present application. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can choose different methods to implement the described functions for each specific application and its actual constraints, but such implementation should not be considered to exceed the scope of the present application.
[0083] So far, the technical solution of the present application has been described in combination with the preferred embodiments shown in the accompanying drawings. Those skilled in the art should understand that the protection scope of the present application is obviously not limited to these specific embodiments. Without departing from the principle of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.
Claims
1. A binary offline data preprocessing method for aircraft engines, characterized in that, include: Step 1, receiving aircraft engine binary offline data; The aircraft engine binary offline data is segmented and sent to a lock-free ring buffer respectively, and binary conversion is performed through multi-channel parallel processing. Each block of aircraft engine binary offline data after binary conversion is stored in a large shared memory pool respectively for multiplexing output; Step 2: Perform binary data verification on the aircraft engine binary offline data; Step 3: extracting valid data from the aircraft engine binary offline data; Step 4: parsing the valid data in the aircraft engine binary offline data; Step 5: Clean the valid data in the aircraft engine binary offline data and replace the outliers.
2. The aircraft engine binary offline data preprocessing method according to claim 1, characterized in that: In step 1, the aircraft engine binary offline data is received, including QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc., and the file format of the QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data is adapted.
3. The aircraft engine binary offline data preprocessing method according to claim 2 is characterized in that: In step 2, the data header, data tail, data length and check word of the aircraft engine binary offline data are checked to check the integrity of the aircraft engine binary offline data, and incomplete aircraft engine binary offline data are discarded, leaving only complete aircraft engine binary offline data for subsequent processing.
4. The aircraft engine binary offline data preprocessing method according to claim 3 is characterized in that: In step 5, the valid data in the aircraft engine binary offline data is cleaned, the missing data is identified, the tuple is ignored, and the missing value of the attribute and the expert experience value are predicted through the dependency relationship between the attributes, and the average value, median value, maximum value, minimum value or more probabilistic statistical function value can be used to replace it; By using statistical analysis methods or data mining technology, outliers in valid data in the engine binary offline data are identified and replaced.
5. The method for preprocessing aircraft engine binary offline data according to claim 4, wherein Also includes: Step 6: Supplement the time column with the valid data in the aircraft engine binary offline data to obtain the aircraft engine binary offline data of each time column.
6. An aircraft engine binary offline data preprocessing system, characterized in that, It includes data receiving module, data checking module, data extraction module, data parsing module, data cleaning and abnormal value replacement module; The data receiving module is used to receive the aircraft engine binary offline data, and divide the aircraft engine binary offline data, respectively send them into the lock-free ring buffer, perform binary conversion through multi-channel parallel processing, and store each block of aircraft engine binary offline data after binary conversion into a large shared memory pool for multiplexing output; The data verification module is used to perform binary data verification on the aircraft engine binary offline data; The data extraction module is used to extract valid data from the aircraft engine binary offline data; The data analysis module is used to analyze the valid data in the aircraft engine binary offline data; The data cleaning and outlier replacement module is used to clean the valid data in the binary offline data of aircraft engines and replace the outliers.
7. The preprocessing system for binary offline data of aircraft engines according to claim 6, characterized in that The data receiving module receives the binary offline data of aircraft engines, including QAR flight parameter data, ECU test data, ECU event data, CEDU fault record data, etc., and adapts the file formats of the QAR flight parameter data, ECU test data, ECU event data, and CEDU fault record data.
8. The preprocessing system for binary offline data of aircraft engines according to claim 7, characterized in that The data verification module verifies the data header, data tail, data length, and check word of the binary offline data of aircraft engines to check the integrity of the binary offline data of aircraft engines, rejects the incomplete binary offline data of aircraft engines, and only retains the complete binary offline data of aircraft engines for subsequent processing.
9. The preprocessing system for binary offline data of aircraft engines according to claim 8, characterized in that In the data cleaning and outlier replacement module, the valid data in the binary offline data of aircraft engines is cleaned, missing data is identified, tuples are ignored, and the missing values of attributes are predicted through the dependency relationship between attributes, expert experience values, and the available average value, median value, maximum value, minimum value, or more probability statistical function values are used for replacement; Using statistical analysis methods or data mining techniques, outliers in the valid data of the binary offline data of aircraft engines are identified and replaced.
10. The aircraft engine binary offline data preprocessing system according to claim 9, wherein, It further includes a data time column supplement module; The data time column supplement module is used to supplement the time column for the valid data in the binary offline data of aircraft engines to obtain the binary offline data of aircraft engines for each time column.