Rapid retrieval method and system based on aero-engine fault statistics

Abstract

The invention discloses a quick retrieval method and system based on aero-engine fault statistics, and the method comprises the following steps: S1, creating an aero-engine database by using SQLite, creating independent data tables for different components of an aero-engine in the aero-engine database, and constructing a connection relation table of each component; s2, a data table in an aero-engine database is imported into a visualization tool, a node is created for each assembly, and a preliminary incidence relation graph is formed; s3, collecting basic data, maintenance data and health monitoring data of the aero-engine; s4, adding additional attributes to the nodes in the incidence relation graph to form a complete incidence relation graph; s5, realizing the linkage of the incidence relation graph and the user interface in the Qt software; and S6, retrieving the component ID, the component type or the component description in the Qt software to obtain a component retrieval result so as to realize fault judgment. The method can effectively improve the query efficiency, reduce the error rate and accelerate the fault positioning speed.

Description

Technical Field

[0001] The present invention relates to the field of aeroengines, and in particular to a rapid retrieval method and system based on aeroengine fault statistics. Background Art

[0002] As an important part of the national strategic industry, the aviation industry is of great significance to national security and development. With the rapid development of the civil aviation industry, the increase in the number of aircraft and the frequency of flight missions, the safety and reliability of aircraft have become the focus of the aviation industry. As a key link in ensuring flight safety, the efficiency and accuracy of aircraft troubleshooting directly affect the normal operation of flights and the economic benefits of airlines. As the core part of the aircraft, the troubleshooting of aircraft engines is even more important. However, there are still some problems that need to be solved in the existing engine troubleshooting technology.

[0003] In the existing engine troubleshooting process, process personnel are faced with a tedious and complex task. They need to query and analyze a variety of information, including design information, finished product manuals, circuit diagrams, Catia digital and analog data, etc., to accurately locate the location of the fault. This process is not only time-consuming, but also due to the complex cross-linking relationships between finished products and within the finished products inside the engine, the query efficiency is low and it is prone to errors. Especially when facing the LRU and SRU level troubleshooting of the engine, this complexity is more obvious, and aircraft engines are prone to false alarms and false alarms during troubleshooting. It is unreliable and inaccurate to rely solely on current data to judge faults. The establishment of a sequential fault database is very necessary, which directly affects the troubleshooting efficiency and engine availability.

[0004] Although modern aircraft engine maintenance technology has made some progress, such as using high-precision equipment to locate faults and building dynamic information systems and maintenance databases to assist in fault analysis, these technical means have not yet completely solved the efficiency and accuracy issues in troubleshooting. In actual operations, technicians still need to rely on a lot of manual queries and experience judgments, which not only increases the workload, but also increases the risk of errors caused by human factors.

[0005] In view of this, the aviation maintenance field urgently needs further technological innovation and management optimization to improve the automation and intelligence level of troubleshooting, reduce reliance on manual operations, and thus improve the work efficiency and safety standards of the entire aviation maintenance industry. It can effectively improve the efficiency and accuracy of engine troubleshooting, ensure the safe operation of civil aircraft, and promote the healthy development of the entire aviation industry. Summary of the invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a rapid retrieval method and system based on aircraft engine fault statistics, which can improve query efficiency, reduce error rate and speed up fault location.

[0007] The object of the present invention is achieved through the following technical scheme: a rapid retrieval method based on aircraft engine fault statistics, comprising the following steps:

[0008] S1. Use SQLite to create an aircraft engine database. In the aircraft engine database, create independent data tables for different components of the aircraft engine, describe component information through description fields, and build a connection relationship table for each component;

[0009] The different components in step S1 include: finished product, cables, pins, and plugs;

[0010] When using SQLite to create a database, multiple data tables were designed to store the finished product information, cable information, pin information, and plug information of aircraft engines.

[0011] When describing component information through the description field, each data table needs to add fields for component ID, component type, and component description;

[0012] The connection relationship table of each component is used to save the connection relationship between components.

[0013] S2. Import the data table of the aircraft engine database into the visualization tool, create a node for each component, and the node contains the description field of the corresponding component. Create edges through the connection relationship between the components to form a preliminary association relationship diagram;

[0014] The visualization tool in step S2 is Qt software;

[0015] Edges are created through the connection relationship between components. That is, for nodes with connection relationships, the nodes are connected by lines to form edges.

[0016] S3. Collect basic data, maintenance data and health monitoring data of aircraft engines;

[0017] The basic data is obtained from the manufacturer of the aircraft engine, including the detailed technical specifications of the aircraft engine;

[0018] The maintenance data is obtained from the historical maintenance data records of the aircraft engine, including the maintenance manual and historical fault records of the aircraft engine;

[0019] The health monitoring data is obtained from the engine health monitoring system, including: sensor data, high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR and enthalpy value Ht.

[0020] S4. Add additional attributes to the nodes in the association graph based on the data collected in step S3 to form a complete association graph;

[0021] The step S4 comprises:

[0022] Add basic data for each node;

[0023] According to the historical fault records of aircraft engines, at the node where the fault occurs, a data structure is designed to store the fault type, sensor data, test data and timestamp, and save the high-pressure rotor speed N2, turbine exhaust temperature T45, unit fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR, enthalpy value Ht, and maintenance records corresponding to the fault; due to the timestamp, the data formed by the designed data structure is time series data;

[0024] Write the time series data into the node, and process user interactions by writing slot functions to implement data addition, deletion, modification and query operations, and complete the establishment of the association relationship graph.

[0025] S5. Realize the linkage between the association diagram and the user interface in Qt software;

[0026] S6. Obtain component search results by searching through component ID, component type or component description in Qt software, and determine which components the component search results are specifically connected to, so as to achieve precise positioning, and realize fault determination by comparing historical data and real-time data.

[0027] Retrieve component ID, component type or component description: If you search by component ID, you will get a unique component search result; if you search by component type or component description, you will get one or more component search results; for each component search result, you can find out which components the search result is connected to from the association diagram to achieve accurate positioning;

[0028] For each component search result, data comparison and false alarm investigation are performed: real-time data is compared with the historical data of the component node. The data to be compared include real-time data and historical data, including high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow rate W, combustion chamber oil-gas ratio FAR and enthalpy value Ht;

[0029] The threshold is set by historical data. If the threshold is exceeded, a warning is given. If the warning lasts for a preset time, a fault is considered. If the warning lasts for less than the preset time, it is considered a false alarm. The threshold is the mean of the fault data plus 2 times the mean square error.

[0030] A rapid retrieval system based on aircraft engine fault statistics, comprising:

[0031] SQLite database creation module, used to create an aircraft engine database using SQLite. In the aircraft engine database, independent data tables are created for different components of the aircraft engine, and component information is described through description fields to build a connection relationship table for each component.

[0032] The relationship graph construction module imports the data table in the aircraft engine database into the visualization tool, creates a node for each component, and this node contains the description field of the corresponding component. It creates edges through the connection relationship between components to form a preliminary relationship graph.

[0033] Data collection module, used to collect basic data, maintenance data and health monitoring data of aircraft engines;

[0034] The additional data adding module is used to add additional attributes to the nodes in the association relationship graph to form a complete association relationship graph;

[0035] Linkage module, used to realize the linkage between database and user interface in Qt;

[0036] The retrieval and determination module obtains component retrieval results by searching the component ID, component type or component description in the Qt software, and determines which components the component retrieval results are specifically connected to, thereby achieving precise positioning, and achieving fault determination by comparing historical data and real-time data.

[0037] The beneficial effects of the present invention are as follows: the present invention eliminates potential false alarms through comparative analysis with time series data, preliminarily locates the possible location of the fault, and can further predict the trend of fault development through the cross-linking relationship diagram, thereby improving query efficiency, reducing error rate, and accelerating fault location. A fault database is established based on historical fault data, and the database is updated in real time during the actual operation and maintenance of the aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a schematic diagram of the overall principle of the present invention. DETAILED DESCRIPTION

[0039] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0040] The present invention establishes an efficient, reliable and easy-to-use cross-linking relationship search platform for complex industrial products (i.e., a rapid retrieval system based on aviation engine failure statistics) to ensure the integrity and consistency of cross-linking relationship data.

[0041] By combining relational and non-relational databases, multiple relational data tables are designed through SQLite to store non-relational data and optimize data storage and query efficiency.

[0042] Specifically, a rapid retrieval system based on aircraft engine fault statistics includes:

[0043] SQLite database creation module, used to create an aircraft engine database using SQLite. In the aircraft engine database, independent data tables are created for different components of the aircraft engine, and component information is described through description fields to build a connection relationship table for each component.

[0044] The relationship graph construction module imports the data table in the aircraft engine database into the visualization tool, creates a node for each component, and this node contains the description field of the corresponding component. It creates edges through the connection relationship between components to form a preliminary relationship graph.

[0045] Data collection module, used to collect basic data, maintenance data and health monitoring data of aircraft engines;

[0046] The additional data adding module is used to add additional attributes to the nodes in the association relationship graph to form a complete association relationship graph;

[0047] Linkage module, used to realize the linkage between database and user interface in Qt;

[0048] The retrieval and determination module obtains component retrieval results by searching the component ID, component type or component description in the Qt software, and determines which components the component retrieval results are specifically connected to, thereby achieving precise positioning, and achieving fault determination by comparing historical data and real-time data.

[0049] The specific construction process is as follows:

[0050] 1. Database establishment

[0051] In order to determine the cross-linking relationship between finished products and within finished products in the engine, a unified association database needs to be established. The data comes from aircraft engine design data, finished product manuals, circuit diagrams, digital simulation data, etc. By sorting out the data, the corresponding finished product name, finished product drawing number, and finished product model are obtained.

[0052] The database uses a relational database. A relational database uses a relational model to organize data. It stores data in the form of rows and columns to facilitate user understanding. This series of rows and columns in a relational database is called a table. As shown below, a group of tables constitute a database. There is a certain cross-linking relationship between tables, and a database with a cross-linking relationship needs to be built.

[0053] Rowid (component ID) Cable Type Number (Component Type) Cable Name (Component Description) 1 1001 High frequency cable 2 1002 Low frequency cable … … … n 3001 powered by

[0054] SQLite is a database that does not require a server or any configuration. All database elements such as data tables and indexes are stored in a single file. The SQLite database can be used as a single file in an application. SQLite is a cross-platform database that can be copied between different platforms at will. The driver library file of SQLite is very small, and a driver with complete functions can be as small as 500KB. SQLite Expert is a visual management tool for SQLite databases. The latest installation file can be downloaded from its official website. The SQLite Expert installation file comes with the SQLite database driver, so there is no need to download and install the SQLite database driver after installing SQLite Expert.

[0055] Build a cross-linked relationship database and use SQLite to design data tables to store cross-linked relationships, finished product information, pins, plugs, and cables. Each table has its own attributes. Cross-linked relationships allow multiple tables to be associated with each other. Here, the association between finished product information, pins, plugs, and cables is established. The intermediate table, the association table, contains four foreign keys, which point to the primary keys of the finished product information, pins, plugs, and cable tables respectively. The primary key is used to index data and is the only identifier for each record in the database. The information of the primary key is referenced through the foreign key.

[0056] Non-relational data is stored through multiple relational data tables. Finished products, plugs, and pins all have their own attribute values, and relational data tables can be used to easily query their respective attribute values. There are pins at both ends of the cable, the pins belong to the plug, and the plug belongs to the finished product. The so-called relational data table means that a field in the main table is stored as a code field, and the specific meaning of the code is in another data table. The use of relational data tables has the following three advantages: it can reduce the storage capacity of the data table. A large data table uses much less storage space to store codes than to store specific text; the text represented by the code may be modified, and in this case, only one record in the code table needs to be modified; the code table is more in line with the actual attribute value storage method. For example, the finished product table stores information such as the finished product drawing number, finished product name, and finished product model.

[0057] Due to the large number of design data, finished product manuals, circuit diagrams, digital and analog data, the actual operation status is often inconsistent with the historical data, so the database needs to be updated and corrected in real time. In addition, there will be "unexpected" reasons for failures during the actual operation and maintenance process, and traditional data cannot be deduced, so a real-time update module needs to be added to the database design to update, correct, and expand a larger association relationship structure.

[0058] 2. Design of Time Series Data

[0059] The time series database built by InfluxDB is used to construct the measurement data of each sensor under the fault mode and record the time series fault data of a certain fault type of the engine. In the InfluxDB line protocol, a line break is used to separate one piece of data from another, so one line is one piece of data. In addition, in the time series database, a line of data is composed of the following four elements. The measurement name (data source name), label set (unique id, attribute), field set, timestamp, test name and label set form a search together. By calling the time series database built with QT software, the fault mode is connected to the cross-link relationship diagram. As shown in the following table.

[0060]

[0061] 3.QISQL database development

[0062] In order to realize the operation of the data interface, you need to use the QISQL module of the Qt software.

[0063] The QISQL module provides support for database programming. Qt supports a variety of common databases, such as MySQL, Oracle, MSSQL Server, SQLite, etc. The QtSQL module includes multiple classes that can implement functions such as database connection, SQL statement execution, data acquisition and interface display. The Model / View architecture is used between data and the interface, so that the interface display and operation of data can be easily realized. The brief functional description of the main classes provided by QISQL is shown in Table 2.

[0064] 4. Cross-linking relationship search platform

[0065] Use programming language to write programs, build a general platform, and implement the functional modules required for parameter correlation analysis. The main components and functional processes of the platform are as follows: Figure 1 shown.

[0066] The cross-link display module can be used to display the cross-link relationships within the aircraft engine system. There are two display modes. The topological cross-link diagram displays the topological cross-link diagram representing all relationships, where green nodes represent finished products, yellow nodes represent plugs, and red nodes represent pins. The lines between pins represent cables. The cross-link table displays cross-link relationship information.

[0067] like Figure 1 As shown, a rapid retrieval method based on aircraft engine fault statistics includes the following steps:

[0068] S1: Design an aircraft engine database and create separate data tables for different components, such as finished products, cables, pins, and plugs. These tables contain specific fields to describe the respective component information, such as component ID, component type, and component description. Construct a connection relationship table for each component to establish the connection relationship between each component. In this way, using SQL queries, especially JOIN operations, you can query and display the cross-linking relationship between components. For example, you can query which other components a specific component is connected to.

[0069] S2: Import the above data table into the visualization tool (Qt), create a node for each component, and this node can contain the description field of the corresponding component. Create edges through the connection relationship table of each component. Edges represent a relationship, that is, for nodes with connection relationships, connect the nodes through lines to form edges.

[0070] S3. Collect basic data, maintenance data and health monitoring data of aircraft engines;

[0071] The basic data is obtained from the manufacturer of the aircraft engine, including the detailed technical specifications of the aircraft engine;

[0072] The maintenance data is obtained from the historical maintenance data records of the aircraft engine, including the maintenance manual and historical fault records of the aircraft engine;

[0073] The health monitoring data is obtained from the engine health monitoring system, including: sensor data, high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR and enthalpy value Ht.

[0074] The step S4 comprises:

[0075] Add basic data for each node;

[0076] According to the maintenance manual and historical fault records of aircraft engines, at the node where the fault occurs, a data structure is designed to store the fault type, sensor data, test data and timestamp, and save the high-pressure rotor speed N2, turbine exhaust temperature T45, unit fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR, and enthalpy value Ht; due to the timestamp, the data formed by the designed data structure is time series data;

[0077] Write the time series data into the node, and process user interactions by writing slot functions to implement data addition, deletion, modification and query operations, and complete the establishment of the association relationship graph.

[0078] S5. Realize the linkage between the association diagram and the user interface in Qt software;

[0079] S6. Obtain component search results by searching through component ID, component type or component description in Qt software, and determine which components the component search results are specifically connected to, so as to achieve precise positioning, and realize fault determination by comparing historical data and real-time data.

[0080] Retrieve component ID, component type or component description: If you search by component ID, you will get a unique component search result; if you search by component type or component description, you will get one or more component search results; for each component search result, you can find out which components the search result is connected to from the association diagram to achieve accurate positioning;

[0081] For each component search result, data comparison and false alarm investigation are performed: real-time data is compared with the historical data of the component node. The data to be compared include real-time data and historical data, including high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow rate W, combustion chamber oil-gas ratio FAR and enthalpy value Ht;

[0082] In addition, for each component search result, you can also view its description field, basic information and maintenance records;

[0083] The threshold is set by historical data. If the threshold is exceeded, a warning is issued. After the warning lasts for a preset time, a fault is determined. If the duration (of the warning of all component retrieval results) does not reach the preset time (or all the real-time data retrieved does not exceed the threshold), it is a false alarm. The threshold is the mean of the fault data plus 2 times the mean square error.

[0084] In the embodiment of the present application, generally, when a warning of a component ID, a component type, or a component description appears, the search and positioning in step S6 is performed in order to accurately locate and perform fault judgment and false alarm judgment.

[0085] Through the above process, a complete retrieval platform is built, integrating all functions, and building a complete retrieval platform. That is, data retrieval, display and comparison functions are integrated into a unified Qt application;

[0086] In the embodiment of the present application, the complete association relationship table may also be updated according to actual operation and maintenance data.

[0087] In summary, a complete search platform is obtained. Search is performed through the QT UI interface. Finally, the corresponding cross-link information is obtained by searching the name, and false alarms are checked by comparing the fault data.

[0088]

[0089]

[0090] In the embodiments of the present application, the embodiments of the present invention are described in combination with actual aircraft engine failure conditions. According to the design data, the engine uses a fiber optic bus to cross-link finished products, and the cross-linking relationship is relatively complex. In daily maintenance, the following types of failures may occur in optical fibers: ① Fiber breakage caused by physical squeezing or excessive bending by external forces; ② Damage to the connector may cause signal loss; ③ Failure at the connection of the fiber optic patch panel or splice tray; ④ Excessive number of connections causes severe fiber attenuation; ⑤ The connector is damaged by dust, fingerprints, scratches, humidity and other factors. At this time, faults such as "XX finished product-XX finished product communication disconnected" will be reported, and a search will be performed based on the constructed database, and key features will be searched in the QT UI interface. The relevant cross-linking relationship in the SQLite database will be retrieved, and the relationship diagram and relationship table will be displayed to intuitively understand the link relationship between the positions of each finished product of the aircraft engine, thereby improving the efficiency of fault location and the speed of troubleshooting when a fault occurs.

[0091] Search for controller-fiber connector in the QT UI interface and get the following connection status. Analyze the fault type based on the connection status. See the following table.

[0092]

[0093]

[0094] In the simulation of actual operation and maintenance, it is assumed that we need to update and back up the components of the optical fiber monitoring system of the aircraft engine to improve the reliability and fault tolerance of the system. The following is the updated connection relationship and the corresponding table:

[0095] Update content:

[0096] Replace sensor 1: Due to frequent failures of sensor 1, it was decided to replace it with a new sensor model.

[0097] Add a backup for sensor group 1: To improve the reliability of the system, add a new backup sensor (sensor 1C) to sensor group 1.

[0098] Replacement of optical fiber distribution tray 1: Due to long-term wear of optical fiber distribution tray 1, it is decided to replace it with a new optical fiber distribution tray (optical fiber distribution tray 1B).

[0099] Update health monitoring system: The health monitoring system software needs to be updated to improve data processing capabilities.

[0100]

[0101]

[0102] The following table is part of the data of the low-pressure compressor failure of an aircraft engine, which is used to demonstrate the function of eliminating false alarms. By searching the measurement name (lpc_failure), the following table is obtained.

[0103]

[0104] High-pressure rotor speed N2, exhaust temperature after turbine T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR, enthalpy value Ht. Select a time stamp data 1725933600000-1725934200000, and retrieve sensor_id=N2 to obtain the following data.

[0105]

[0106]

[0107] The statistical results show that the thresholds of fault N2, T45, Tsfc, W, FAR, and enthalpy value Ht are (2222.028322, 257.3315852, 0.244518683, 10.96558452, 0.002947349, 89.22285975). According to the current aircraft engine data, it is judged whether it is higher than the fault threshold and remains higher than the threshold for a period of time, and the result is judged as a fault. If it is lower than the threshold, it is a false alarm.

[0108] The above is a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A rapid retrieval method based on aircraft engine failure statistics, characterized by: The following steps are involved: S1. Use SQLite to create an aircraft engine database. In the aircraft engine database, create independent data tables for different components of the aircraft engine, describe component information through description fields, and build a connection relationship table for each component; S2. Import the data table of the aircraft engine database into the visualization tool, create a node for each component, and the node contains the description field of the corresponding component. Create edges through the connection relationship between the components to form a preliminary association relationship diagram; S3. Collect basic data, maintenance data and health monitoring data of aircraft engines; S4. Add additional attributes to the nodes in the association graph based on the data collected in step S3 to form a complete association graph; S5. Realize the linkage between the association diagram and the user interface in Qt software; S6. Obtain component search results by searching through component ID, component type or component description in Qt software, and determine which components the component search results are specifically connected to, so as to achieve precise positioning, and realize fault determination by comparing historical data and real-time data.

2. A rapid retrieval method based on aircraft engine fault statistics according to claim 1, characterized in that: The different components in step S1 include: finished products, cables, pins, and plugs; When using SQLite to create a database, multiple data tables were designed to store the finished product information, cable information, pin information, and plug information of aircraft engines. When describing component information through the description field, each data table needs to add fields for component ID, component type, and component description; The connection relationship table of each component is used to save the connection relationship between components.

3. A rapid retrieval method based on aircraft engine fault statistics according to claim 1, characterized in that: The visualization tool in step S2 is Qt software; Edges are created through the connection relationship between components. That is, for nodes with connection relationships, the nodes are connected by lines to form edges.

4. A rapid retrieval method based on aircraft engine fault statistics according to claim 1, characterized in that: The basic data is obtained from the manufacturer of the aircraft engine, including the detailed technical specifications of the aircraft engine; The maintenance data is obtained from the historical maintenance data records of the aircraft engine, including the maintenance manual and historical fault records of the aircraft engine; The health monitoring data is obtained from the engine health monitoring system, including: sensor data, high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR and enthalpy value Ht.

5. The rapid retrieval method based on aircraft engine fault statistics according to claim 1, characterized in that: The step S4 comprises: Add basic data for each node; According to the historical fault records of aircraft engines, at the node where the fault occurs, a data structure is designed to store the fault type, sensor data, test data and timestamp, and save the high-pressure rotor speed N2, turbine exhaust temperature T45, unit fuel consumption rate Tsfc, high-pressure compressor mass flow W, combustion chamber oil-gas ratio FAR, enthalpy value Ht, and maintenance records corresponding to the fault; due to the timestamp, the data formed by the designed data structure is time series data; Write the time series data into the node, and process user interactions by writing slot functions to implement data addition, deletion, modification and query operations, and complete the establishment of the association relationship graph.

6. A rapid retrieval method based on aircraft engine fault statistics according to claim 1, characterized in that: The step S6 comprises: Retrieve component ID, component type or component description: If you search by component ID, you will get a unique component search result; if you search by component type or component description, you will get one or more component search results; for each component search result, you can find out which components the search result is connected to from the association diagram to achieve accurate positioning; For each component search result, data comparison and false alarm investigation are performed: real-time data is compared with the historical data of the component node. The data to be compared include real-time data and historical data, including high-pressure rotor speed N2, turbine exhaust temperature T45, specific fuel consumption rate Tsfc, high-pressure compressor mass flow rate W, combustion chamber oil-gas ratio FAR and enthalpy value Ht; The threshold is set by historical data. If the threshold is exceeded, a warning is given. If the warning lasts for a preset time, a fault is considered. If the warning lasts for less than the preset time, it is considered a false alarm. The threshold is the mean of the fault data plus 2 times the mean square error.

7. A rapid retrieval system based on aircraft engine fault statistics, using the method described in any one of claims 1 to 6, characterized in that: include: SQLite database creation module, used to create an aircraft engine database using SQLite. In the aircraft engine database, independent data tables are created for different components of the aircraft engine, and component information is described through description fields to build a connection relationship table for each component. The relationship graph construction module imports the data table in the aircraft engine database into the visualization tool, creates a node for each component, and this node contains the description field of the corresponding component. It creates edges through the connection relationship between components to form a preliminary relationship graph. Data collection module, used to collect basic data, maintenance data and health monitoring data of aircraft engines; The additional data adding module is used to add additional attributes to the nodes in the association relationship graph to form a complete association relationship graph; Linkage module, used to realize the linkage between database and user interface in Qt; The retrieval and determination module obtains component retrieval results by searching the component ID, component type or component description in the Qt software, and determines which components the component retrieval results are specifically connected to, thereby achieving precise positioning, and achieving fault determination by comparing historical data and real-time data.

Images