Method for constructing and using aero-engine failure analysis knowledge base

Abstract

The invention provides a construction and use method of an aero-engine failure analysis knowledge base, and relates to the technical field of aero-engine machining. The method specifically comprises the following steps: constructing an aero-engine failure analysis knowledge base framework by analyzing processing technology types and technological processes of different parts in an aero-engine; historical machining data of different parts in the aero-engine are collected; historical processing data of the aero-engine are stored in an aero-engine failure analysis knowledge base framework, and an aero-engine failure analysis knowledge base is obtained; failure parameters of any part in the aero-engine are obtained and input into the aero-engine failure analysis knowledge base for failure analysis, and a failure analysis report is obtained. According to the method, the failure analysis time can be further shortened, the product failure analysis consistency is improved, and then whole-process automatic failure analysis is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of aviation engine machining, and in particular to a construction and use method of an aviation engine failure analysis knowledge base. Background Art

[0002] Aircraft engines are the heart of fighter jets and are at the forefront of international engineering science. They are called the "crown jewel" of modern industry because of their wide technical span, long R&D cycle and large capital demand. As a complex system composed of multiple parts, aircraft engines need to work for a long time in harsh environments such as high temperature, high pressure and high speed. Such harsh working conditions require the engine to have high reliability and long life, which also makes failure analysis complicated and important. At present, when difficult-to-process materials such as titanium alloys, high-temperature alloys and intermetallic compounds fail during processing, there is a lack of effective means to build the relationship between processing data and failure data. At the same time, there is a lack of relevant means for how to use existing failure data to support new failures. This means that every time an aircraft engine is analyzed for failure, technicians need to analyze it based on experience. In addition, aircraft engines often generate a large amount of data during operation, which leads to low efficiency and poor consistency in failure analysis of aircraft engine blades, integral blade disks, casings and other parts, which brings huge challenges to the rapid resolution of product failures. With the increase in the number of aircraft engine products, there is a higher demand for how to quickly perform failure analysis. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention proposes a method for constructing and using an aircraft engine failure analysis knowledge base, aiming to support aircraft engine failure analysis by building an aircraft engine failure analysis knowledge base.

[0004] The present invention proposes a method for constructing and using an aero-engine failure analysis knowledge base, the method comprising the following steps:

[0005] Step 1: Build the knowledge base framework of aircraft engine failure analysis by analyzing the processing types and processes of different parts in aircraft engines;

[0006] Step 2: Collect historical processing data of different parts in aircraft engines;

[0007] Step 3: storing the historical processing data of the aircraft engine into the aircraft engine failure analysis knowledge base framework to obtain the aircraft engine failure analysis knowledge base;

[0008] Step 4: Obtain failure parameters of any part in the aircraft engine, and input them into the aircraft engine failure analysis knowledge base to perform failure analysis and obtain a failure analysis report;

[0009] The aero-engine failure analysis knowledge base framework described in step 1 includes: a failure analysis enabling tool, a machining database, a welding processing database, a special processing database, and a failure database;

[0010] The failure analysis enabling tool is used to call data in a mechanical processing database, a welding processing database, a special processing database, and a failure database, and use a graph neural network algorithm and a genetic algorithm to perform failure analysis on input failure phenomena, processing parameters, and failure metallographic images, and generate a failure analysis report;

[0011] The machining database is used to store machining data of different parts in the aircraft engine;

[0012] The welding processing database is used to store welding processing data of different parts in the aircraft engine;

[0013] The special processing database is used to store special processing data of different parts in the aircraft engine;

[0014] The failure database is used to store historical failure data of different parts in the aircraft engine;

[0015] The mechanical processing database, welding processing database and special processing database all include: material processing basic database, process basic database and process database;

[0016] The historical processing data in step 2 includes: mechanical processing data, welding processing data, special processing data and failure data;

[0017] The mechanical processing data, welding processing data, and special processing data all include: basic material processing data, basic process data, and process data;

[0018] The failure data includes: failed products, failure phenomena, processing parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions and failure analysis report templates;

[0019] The method for collecting historical processing data of different parts in the aircraft engine is as follows: for any part that uses any processing method, different materials and processing tools are selected for the part and process parameters are set, and the surface integrity and fatigue life of the part are obtained by conducting a simulated processing test, and the basic material processing data of the part is recorded; different processing tools are selected for the part and the characteristics and process parameters of the part are set, and the dimension detection data, surface integrity data and fatigue life data of the part are obtained by conducting a simulated processing test, and the basic process data and process data of the part are recorded; for the failure results of different parts in the production site, the failed product, failure phenomenon, processing parameters, metallographic pictures, failure mechanism, failure cause, recommended measures, analysis conclusion and failure analysis report template corresponding to the failure result are obtained by analysis;

[0020] The machining data in step 3 include: material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, cooling method, cutting / grinding speed, feed rate, cutting depth, machining noise, machining force, machining temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, tool wear, specimen fatigue, feature name, feature size, process method, machining equipment model, tool brand, spindle power, vibration, machining temperature, surface roughness, part name, tool number, gauge number, clamping method, machining sequence and dimensional accuracy;

[0021] The welding processing data includes: material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, welding wire type, cooling method, welding current, feed speed, weld width, processing noise, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, specimen fatigue, electrode type, current pulse interval, electrode loss, feature name, feature size, process method, processing equipment model and welding wire brand;

[0022] The special processing data include: feature name, feature size, process method, material model, heat treatment status, processing equipment model, electrode material brand, cooling method, processing current, current pulse interval, feed speed, processing noise, processing temperature, surface roughness, residual stress, surface hardness, electrode loss, plastic deformation layer, specimen fatigue, part name, tool brand, tooling number, gauge number, clamping method, processing sequence, welding current, weld width, dimensional accuracy, tool wear and dimensional accuracy;

[0023] The method for storing the historical processing data of the aircraft engine into the aircraft engine failure analysis knowledge base framework described in step 3 is: storing the material processing basic data, process basic data and process data of the mechanical processing data into the material processing basic database, process basic database and process database in the mechanical processing database respectively; storing the material processing basic data, process basic data and process data of the welding processing data into the material processing basic database, process basic database and process database in the welding processing database respectively; storing the material processing basic data, process basic data and process data of the special processing data into the material processing basic database, process basic database and process database in the special processing database respectively; storing the failure data into the failure database;

[0024] The specific method of step 4 is:

[0025] Step 4.1: Obtain failure parameters of any part in the aircraft engine, including: part name, part damage feature name, processing procedure name, processing parameters, failure phenomenon and metallographic images;

[0026] Step 4.2: Use the graph neural network algorithm to extract features from the failure phenomenon and metallographic images of the part, and match the extracted features with the failure data stored in the failure database to obtain the failure mechanism, failure cause, recommended measures and analysis conclusions of the part;

[0027] Step 4.3: Call the data stored in the mechanical processing database, the welding processing database and the special processing database, and use the genetic algorithm to predict the surface integrity data and fatigue life data of the part based on the part name, part damage feature name, processing procedure name, processing parameters and the called data;

[0028] Step 4.4: Fill in the failure mechanism, failure cause, recommended measures, analysis conclusion, surface integrity data and fatigue life data of the part into the failure analysis report template to obtain a failure analysis report;

[0029] The method for predicting the surface integrity data and fatigue life data of the part described in step 4.3 is: randomly generate an initial population based on the material processing basic data, process basic data and process data called from the mechanical processing database, welding processing database and special processing database, wherein each individual represents a set of surface integrity data and fatigue life data of the part; select a fitness function, calculate the fitness of each individual, perform selection, crossover and mutation operations on the individuals according to the fitness of the individuals, obtain a new population, calculate the fitness of each individual in the new population, and repeat the individual selection, crossover and mutation operations until the maximum number of iterations is reached or the fitness reaches a preset threshold and stops the iteration, and obtains the individual with the highest fitness in the population in the current iteration round as the surface integrity data and fatigue life data of the part.

[0030] The beneficial effects of adopting the above technical solution are:

[0031] Compared with the existing technology, as the demand for the number of new engines increases, the method of the present invention can further reduce the failure analysis time, improve the consistency of product failure analysis, and thus realize full-process automated failure analysis. At the same time, the invention also has broad market demand and application prospects in the failure analysis of products such as gas turbines and marine propulsion.

[0032] The method of the present invention can significantly improve the efficiency of the company's product failure analysis, while significantly improving quality consistency, thereby reducing economic losses caused by production stoppage during failure analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a flow chart of a method for constructing and using an aero-engine failure analysis knowledge base in this embodiment;

[0034] Figure 2 This is a structural diagram of the aircraft engine failure analysis knowledge base in this embodiment;

[0035] Figure 3 A flow chart of a technician using an aircraft engine failure analysis knowledge base in an implementation manner. DETAILED DESCRIPTION

[0036] In order to facilitate the understanding of the present application, the specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thoroughly understood.

[0037] The present embodiment provides a method for constructing and using an aero-engine failure analysis knowledge base, such as Figure 1 As shown, the method comprises the following steps:

[0038] Step 1: Construct the knowledge base framework of aircraft engine failure analysis by analyzing the processing technology types and processes of different parts in aircraft engines.

[0039] The aero-engine failure analysis knowledge base framework includes: a failure analysis enabling tool, a mechanical processing database, a welding processing database, a special processing database and a failure database.

[0040] The failure analysis enabling tool is used to call data from a mechanical processing database, a welding processing database, a special processing database and a failure database, and uses a graph neural network algorithm and a genetic algorithm to perform failure analysis on input failure phenomena, processing parameters and failure metallographic images, and generate a failure analysis report.

[0041] The machining database is used to store the machining data of different parts in the aircraft engine.

[0042] The welding processing database is used to store welding processing data of different parts in the aircraft engine.

[0043] The special processing database is used to store special processing data of different parts in the aircraft engine.

[0044] The failure database is used to store historical failure data of different parts in the aircraft engine.

[0045] The mechanical processing database, welding processing database and special processing database all include: a material processing basic database, a process basic database and a process database.

[0046] In this embodiment, if Figure 2 As shown in the figure, a failure analysis knowledge base including a machining database, a welding database, a special machining database, a failure database, and a failure analysis enabling tool is constructed, and the failure analysis knowledge base is installed on a network server for use by technical personnel. Among them, the machining database, the welding database, and the special machining database are each further subdivided into a material processing basic database, a process basic database, and a process database. The failure database is mainly divided according to products such as blades, blisks, and casings, and mainly includes failure phenomena, machining parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions, and failure analysis report templates. The failure analysis enabling tool mainly performs failure analysis based on the input failure phenomena, machining parameters, and failure metallographic photos.

[0047] Step 2: Collect historical processing data of different parts in aircraft engines.

[0048] The historical processing data includes: mechanical processing data, welding processing data, special processing data and failure data.

[0049] The mechanical processing data, welding processing data, and special processing data all include: basic material processing data, basic process data, and process data.

[0050] The method for collecting historical processing data of different parts in the aircraft engine is as follows: for any part that uses any processing method, different materials and processing tools are selected for the part and process parameters are set, and the surface integrity and fatigue life of the part are obtained by conducting simulated processing tests, and the basic material processing data of the part is recorded; different processing tools are selected for the part and the characteristics and process parameters of the part are set, and the dimension detection data, surface integrity data and fatigue life data of the part are obtained by conducting simulated processing tests, and the basic process data and process data of the part are recorded; for the failure results of different parts in the production site, the failed product, failure phenomenon, processing parameters, metallographic pictures, failure mechanism, failure cause, recommended measures, analysis conclusion and failure analysis report template corresponding to the failure result are obtained through analysis.

[0051] In this embodiment, the generated failure data is stored in a failure database, mainly including the failure phenomena, processing parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions, etc. obtained after the typical failure results of blades, integral blisks, casings, etc. at the production site are analyzed by technical personnel.

[0052] In this embodiment, the machining data, welding data, special processing data and failure data of the parts such as blades, blisks and casings in the aircraft engine are collected. The existing material basic micro-element machining, welding and special processing data are collected, which are mainly the basic processing parameters and corresponding surface integrity data and fatigue life data obtained from basic tests using different processing tools for different materials. The machining, welding and special processing data of the blade profile, flow channel, blisk blade profile, weld, reinforcement rib and other features are collected, which are mainly the processing parameters and corresponding surface integrity data and fatigue life data of the relevant typical features using different processing tools. The machining, welding and special processing data of the blade profile, flow channel, blisk blade profile, weld, reinforcement rib and other features of the blades, integral blisks, casings and other typical parts processed at the production site are collected, which mainly include relevant dimensional detection data, surface integrity data and fatigue life data.

[0053] The machining data include: material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, cooling method, cutting / grinding speed, feed rate, cutting depth, machining noise, machining force, machining temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, tool wear, specimen fatigue, feature name, feature size, process method, machining equipment model, tool brand, spindle power, vibration, machining temperature, surface roughness, part name, tool number, gauge number, clamping method, machining sequence and dimensional accuracy.

[0054] The welding processing data include: material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, welding wire type, cooling method, welding current, feed speed, weld width, processing noise, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, specimen fatigue, electrode type, current pulse interval, electrode loss, feature name, feature size, process method, processing equipment model and welding wire brand.

[0055] The special processing data include: feature name, feature size, process method, material model, heat treatment status, processing equipment model, electrode material brand, cooling method, processing current, current pulse interval, feed speed, processing noise, processing temperature, surface roughness, residual stress, surface hardness, electrode loss, plastic deformation layer, specimen fatigue, part name, tool brand, tool number, gauge number, clamping method, processing sequence, welding current, weld width, dimensional accuracy, tool and grinder wear and dimensional accuracy.

[0056] The failure data includes: failed products, failure phenomena, processing parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions and failure analysis report templates.

[0057] Step 3: Store the historical processing data of the aircraft engine into the aircraft engine failure analysis knowledge base framework to obtain the aircraft engine failure analysis knowledge base.

[0058] The method for storing the historical processing data of an aircraft engine in the aircraft engine failure analysis knowledge base framework is as follows: storing the material processing basic data, process basic data and process data of the mechanical processing data in the material processing basic database, process basic database and process database of the mechanical processing database respectively; storing the material processing basic data, process basic data and process data of the welding processing data in the material processing basic database, process basic database and process database of the welding processing database respectively; storing the material processing basic data, process basic data and process data of the special processing data in the material processing basic database, process basic database and process database of the special processing database respectively; and storing the failure data in the failure database.

[0059] In this embodiment, the existing material basic micro-element machining, welding processing, and special processing data are stored in the material processing basic database. Specifically, for the material processing basic database of the machining database, it is classified according to material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, and cooling method. The process parameters mainly include cutting / grinding speed, feed rate, and cutting depth. The measured data include processing noise, processing force, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, knife grinder wear, specimen fatigue and other information. For the material processing basic database of the welding processing database, it is classified according to material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, welding wire type, and cooling method. The process parameters mainly include welding current, feed speed, weld width, etc., and the measured data include processing noise, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, specimen fatigue and other information. For the material processing basic database of the special processing database, it is classified according to material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, electrode type, and cooling method. The process parameters mainly include processing current, current pulse interval, feed speed, etc. The measured data include processing noise, processing temperature, surface roughness, residual stress, surface hardness, electrode loss, plastic deformation layer, specimen fatigue and other information.

[0060] In this embodiment, the mechanical processing, welding processing, and special processing data of the blade profile, flow channel, blade disc blade profile, weld, reinforcement rib and other features are stored in the process basic database. Specifically, for the material processing basic database of the mechanical processing database, it is classified according to material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, and cooling method. The process parameters mainly include cutting / grinding speed, feed rate, and cutting depth, and the measured data include processing noise, processing force, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, knife grinder wear, specimen fatigue and other information. For the material processing basic database of the welding processing database, it is classified according to material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, welding wire type, and cooling method. The process parameters mainly include welding current, feed speed, weld width, etc., and the measured data include processing noise, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, specimen fatigue and other information. For the material processing basic database of the special processing database, it is classified according to material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, electrode type, and cooling method. The process parameters mainly include processing current, current pulse interval, feed speed, etc. The measured data include processing noise, processing temperature, surface roughness, residual stress, surface hardness, electrode loss, plastic deformation layer, specimen fatigue and other information.

[0061] In this embodiment, the mechanical processing, welding processing, and special processing data of the blade profile, flow channel, blade disc blade profile, weld, reinforcement rib and other features of the blade, integral blade disk, casing and other typical parts processed at the production site are stored in the process database. Specifically, the process database of the mechanical processing database is classified according to the part name, feature name, feature size, process method, material model, heat treatment status, processing equipment model, tool brand, cooling method, tool number, gauge number, clamping method, and processing sequence. The process parameters mainly include cutting / grinding speed, feed rate, and cutting depth. The measured data include dimensional accuracy, processing noise, processing force, spindle power, vibration, processing temperature, surface roughness, tool wear and other information. For the process database of the welding processing database, it is classified according to the part name, feature name, feature size, process method, material model, heat treatment status, processing equipment model, tool brand, cooling method, tool number, gauge number, clamping method, and processing sequence. The process parameters mainly include feed speed, welding current, weld width, and the measured data include dimensional accuracy, processing noise, processing temperature, surface roughness, tool wear and other information. The process database of the special processing database is classified according to part name, feature name, feature size, process method, material model, heat treatment status, processing equipment model, tool brand, cooling method, tool number, gauge number, clamping method, and processing sequence. The process parameters mainly include processing current, current pulse interval, feed speed, and the measured data include dimensional accuracy, processing noise, processing temperature, surface roughness, tool wear and other information.

[0062] In this embodiment, the generated failure data is stored in a failure database, mainly including failure phenomena, processing parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions, etc. obtained after analysis by technical personnel of typical failure results of blades, integral blisks, casings, etc. at the production site.

[0063] Step 4: Obtain the failure parameters of any part in the aircraft engine, input them into the aircraft engine failure analysis knowledge base for failure analysis, and obtain a failure analysis report.

[0064] Step 4.1: Obtain the failure parameters of any part in the aircraft engine, including: part name, part failure feature name, processing procedure name, processing parameters, failure phenomenon and metallographic images.

[0065] Step 4.2: Use the graph neural network algorithm to extract features from the failure phenomenon and metallographic images of the part, and match the extracted features with the failure data stored in the failure database to obtain the failure mechanism, failure cause, recommended measures and analysis conclusions of the part.

[0066] In this embodiment, by developing relevant enabling tools, relevant data can be automatically matched from the failure database according to the input failure phenomenon description, metallographic images and other query conditions according to the image neural network algorithm, and relevant failure mechanisms, failure causes, recommended measures, analysis conclusions and other contents can be automatically retrieved and filled into the failure analysis report template.

[0067] Step 4.3: Call the data stored in the mechanical processing database, welding processing database and special processing database, and use the genetic algorithm to predict the surface integrity data and fatigue life data of the part based on the part name, part damage feature name, processing procedure name, processing parameters and the called data.

[0068] The method for predicting the surface integrity data and fatigue life data of the part is as follows: randomly generating an initial population based on material processing basic data, process basic data and process data called from a mechanical processing database, a welding processing database and a special processing database, wherein each individual represents a set of surface integrity data and fatigue life data of the part; selecting a fitness function, calculating the fitness of each individual, performing selection, crossover and mutation operations on the individuals according to the fitness of the individuals to obtain a new population, then calculating the fitness of each individual in the new population, and repeating the selection, crossover and mutation operations of the individuals until the maximum number of iterations is reached or the fitness reaches a preset threshold, and the iteration is stopped, and the individual with the highest fitness in the population in the current iteration round is obtained as the surface integrity data and fatigue life data of the part.

[0069] In this embodiment, a related enabling tool is developed, which can automatically predict the surface integrity and fatigue life of the failure part from the material processing basic database, process basic database, and process database in the mechanical processing database, welding processing database, and special processing database according to the genetic algorithm according to the input part name, part damage feature name, processing procedure name, and processing parameters, and fill them into the failure analysis report template. Specifically, the initial population is randomly generated according to the material processing basic data, process basic data, and process data called from the mechanical processing database, welding processing database, and special processing database, wherein each individual represents a set of surface integrity data and fatigue life data of the part; a suitable fitness function is selected, the fitness of each individual is calculated, the individuals are sorted in order from high to low fitness, the individual with the highest fitness is selected from the sorting results for genetic operation, the selected individual is cross-operated, a new individual is generated, and the new individual is mutated to obtain a new population, and the fitness of each individual in the new population is calculated, and the above genetic operation is repeated until the maximum number of iterations is reached or the fitness reaches a preset threshold, and the iteration is stopped, and the individual with the highest fitness in the population under the current iteration round is obtained as the optimal solution.

[0070] Step 4.4: Fill in the failure mechanism, failure cause, recommended measures, analysis conclusions, surface integrity data and fatigue life data of the part into the failure analysis report template to obtain a failure analysis report.

[0071] In this embodiment, if Figure 3 As shown, the technician inputs the name of the latest failed part, the name of the part damage feature, the name of the processing procedure, the processing parameters, the failure phenomenon description, the metallographic picture and other contents into the failure analysis knowledge base input interface, and uses the failure analysis enabling tool to automatically generate a report containing the failure mechanism, failure cause, recommended measures, analysis conclusions, surface integrity of the failed part, and fatigue life prediction of the failed part.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope defined by the claims of the present invention.

Claims

1. A method for constructing and using an aircraft engine failure analysis knowledge base, characterized in that: The method comprises the following steps: Step 1: Build the knowledge base framework of aircraft engine failure analysis by analyzing the processing types and processes of different parts in aircraft engines; Step 2: Collect historical processing data of different parts in aircraft engines; Step 3: storing the historical processing data of the aircraft engine into the aircraft engine failure analysis knowledge base framework to obtain the aircraft engine failure analysis knowledge base; Step 4: Obtain the failure parameters of any part in the aircraft engine, input them into the aircraft engine failure analysis knowledge base for failure analysis, and obtain a failure analysis report.

2. The method for constructing and using an aero-engine failure analysis knowledge base according to claim 1, characterized in that: The aero-engine failure analysis knowledge base framework described in step 1 includes: a failure analysis enabling tool, a machining database, a welding processing database, a special processing database, and a failure database; The failure analysis enabling tool is used to call data in a mechanical processing database, a welding processing database, a special processing database, and a failure database, and use a graph neural network algorithm and a genetic algorithm to perform failure analysis on input failure phenomena, processing parameters, and failure metallographic images, and generate a failure analysis report; The machining database is used to store machining data of different parts in the aircraft engine; The welding processing database is used to store welding processing data of different parts in the aircraft engine; The special processing database is used to store special processing data of different parts in the aircraft engine; The failure database is used to store historical failure data of different parts in the aircraft engine; The mechanical processing database, welding processing database and special processing database all include: a material processing basic database, a process basic database and a process database.

3. The method for constructing and using an aero-engine failure analysis knowledge base according to claim 2, characterized in that: The historical processing data in step 2 includes: mechanical processing data, welding processing data, special processing data and failure data; The mechanical processing data, welding processing data, and special processing data all include: basic material processing data, basic process data, and process data; The failure data includes: failed products, failure phenomena, processing parameters, metallographic images, failure mechanisms, failure causes, recommended measures, analysis conclusions and failure analysis report templates; The method for collecting historical processing data of different parts in the aircraft engine is as follows: for any part that uses any processing method, different materials and processing tools are selected for the part and process parameters are set, and the surface integrity and fatigue life of the part are obtained by conducting simulated processing tests, and the basic material processing data of the part is recorded; different processing tools are selected for the part and the characteristics and process parameters of the part are set, and the dimension detection data, surface integrity data and fatigue life data of the part are obtained by conducting simulated processing tests, and the basic process data and process data of the part are recorded; for the failure results of different parts in the production site, the failed product, failure phenomenon, processing parameters, metallographic pictures, failure mechanism, failure cause, recommended measures, analysis conclusion and failure analysis report template corresponding to the failure result are obtained through analysis.

4. The method for constructing and using an aero-engine failure analysis knowledge base according to claim 3 is characterized in that: The machining data in step 3 include: material type, material model, heat treatment status, specimen length, specimen width, specimen thickness, cooling method, cutting / grinding speed, feed rate, cutting depth, machining noise, machining force, machining temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, tool wear, specimen fatigue, feature name, feature size, process method, machining equipment model, tool brand, spindle power, vibration, machining temperature, surface roughness, part name, tool number, gauge number, clamping method, machining sequence and dimensional accuracy; The welding processing data includes: material type, material model, heat treatment state, specimen length, specimen width, specimen thickness, welding wire type, cooling method, welding current, feed speed, weld width, processing noise, processing temperature, surface roughness, residual stress, surface hardness, plastic deformation layer, specimen fatigue, electrode type, current pulse interval, electrode loss, feature name, feature size, process method, processing equipment model and welding wire brand; The special processing data include: feature name, feature size, process method, material model, heat treatment status, processing equipment model, electrode material brand, cooling method, processing current, current pulse interval, feed speed, processing noise, processing temperature, surface roughness, residual stress, surface hardness, electrode loss, plastic deformation layer, specimen fatigue, part name, tool brand, tool number, gauge number, clamping method, processing sequence, welding current, weld width, dimensional accuracy, tool and grinder wear and dimensional accuracy.

5. The method for constructing and using an aero-engine failure analysis knowledge base according to claim 4 is characterized in that: The method for storing the historical machining data of the aircraft engine into the aircraft engine failure analysis knowledge base framework in step 3 is: storing the material machining basic data, process basic data and process data of the machining data into the material machining basic database, process basic database and process database of the machining database respectively; The material processing basic data, process basic data and process data of the welding processing data are stored in the material processing basic database, process basic database and process database in the welding processing database respectively; the material processing basic data, process basic data and process data of the special processing data are stored in the material processing basic database, process basic database and process database in the special processing database respectively; The failure data is stored in the failure database.

6. The method for constructing and using an aero-engine failure analysis knowledge base according to claim 5, characterized in that: The step 4 further comprises: Step 4.1: Obtain failure parameters of any part in the aircraft engine, including: part name, part damage feature name, processing procedure name, processing parameters, failure phenomenon and metallographic images; Step 4.2: Use the graph neural network algorithm to extract features from the failure phenomenon and metallographic images of the part, and match the extracted features with the failure data stored in the failure database to obtain the failure mechanism, failure cause, recommended measures and analysis conclusions of the part; Step 4.3: Call the data stored in the mechanical processing database, the welding processing database and the special processing database, and use the genetic algorithm to predict the surface integrity data and fatigue life data of the part based on the part name, part damage feature name, processing procedure name, processing parameters and the called data; Step 4.4: Fill in the failure mechanism, failure cause, recommended measures, analysis conclusions, surface integrity data and fatigue life data of the part into the failure analysis report template to obtain a failure analysis report.

7. A method for constructing and using an aero-engine failure analysis knowledge base according to claim 6, characterized in that: The method for predicting the surface integrity data and fatigue life data of the part described in step 4.3 is: randomly generate an initial population based on the material processing basic data, process basic data and process data called from the mechanical processing database, welding processing database and special processing database, where each individual represents a set of surface integrity data and fatigue life data of the part; select a fitness function, calculate the fitness of each individual, select, crossover and mutate the individuals according to their fitness, obtain a new population, calculate the fitness of each individual in the new population, and repeat the individual selection, crossover and mutation operations until the maximum number of iterations is reached or the fitness reaches a preset threshold and stops the iteration, and obtains the individual with the highest fitness in the population in the current iteration round as the surface integrity data and fatigue life data of the part.

Images