A civil aviation engine fault diagnosis method and system based on big data drive

Abstract

The present invention relates to the technical field of engine fault test diagnosis, and in particular to a civil aviation engine fault diagnosis method and system based on big data drive. The system comprises an engine fault position positioning management module, an engine fault diagnosis management module, and an engine fault diagnosis result feedback module. Temperature feature recognition is performed by combining a body temperature distribution image parameter under a preset engine fault state with an intelligent recognition algorithm and collected engine body space temperature image parameters, so that the engine body temperature fault feature and position are scientifically and accurately identified. A physical three-dimensional model of the engine body is used to match the engine body temperature fault feature with the engine body temperature fault feature, so that the engine body temperature fault feature is accurately mapped to the physical model of the engine body, and the engine body fault position is intuitively and accurately located, thereby improving the positioning efficiency and accuracy of civil aviation engine fault diagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of engine fault test diagnosis, and in particular to a civil aviation engine fault diagnosis method and system driven by big data. Background Art

[0002] Aircraft engines are highly complex and sophisticated thermal machinery. Aircraft engines have developed into mature products with extremely high reliability. Aircraft engines in use include turbojet / turbofan engines, turboshaft / turboprop engines, ramjet engines and piston engines. Aircraft engines are further accelerating their development and will bring new major changes to the human aviation field. Traditional aircraft engines are developing towards gear transmission engines, variable cycle engines, multi-electric engines, intercooled regenerative engines and open rotor engines. Non-traditional pulse detonation engines, scramjet engines, turbine combination engines, as well as solar power and fuel cell power are also maturing. The development of these engines will make future aircraft faster, higher, farther, more economical and more reliable, and meet more stringent environmental protection requirements. It will also make hypersonic aircraft, trans-atmospheric aircraft and reusable round-trip transportation between the earth and the sky a reality; aircraft engines are the core components of aircraft, providing propulsion for aircraft and ensuring flight safety and stability. Aircraft engines also have the risk of failure and damage. Once a failure occurs, it will threaten the safety of flights. Therefore, the fault diagnosis and maintenance processing technology of aircraft engines is extremely important. Among them, the fault diagnosis of civil aviation engines is related to the safety of passengers' lives and property. Therefore, the safety of civil aviation engines is very important. The existing civil aviation engine fault diagnosis can only rely on manual and computer-assisted regular comprehensive maintenance and overhaul of the engine. The overhaul process requires the entire engine mechanical parts, electronic equipment, and control programs to be checked one by one. The workload is large and the engine fault diagnosis efficiency is low. The existing civil aviation engine fault diagnosis system cannot scientifically, accurately, and efficiently locate the engine fault point, as well as the intelligent engine's software and hardware fault point diagnosis and analysis, which reduces the efficiency and reliability of civil aviation engine fault diagnosis.

[0003] The Bat Algorithm (BA) algorithm is a random search algorithm that simulates the use of sonar by bats in nature to detect prey and avoid obstacles, that is, it simulates the most basic detection and positioning capabilities of bats using ultrasound to detect obstacles or prey and links it with the optimization objective function. The bionic principle of the BA algorithm maps the bat individuals of a population to NP feasible solutions in the D-dimensional problem space, simulates the optimization process and search into the movement process of individual bats in the population and the search for prey, and uses the fitness function value of the problem to measure the quality of the bat's position, and compares the individual survival of the fittest process to the iterative process of replacing poor feasible solutions with good feasible solutions in the optimization and search process.

[0004] The Chinese invention patent with the announcement number CN115326400B discloses a fault diagnosis method and electronic equipment for an aircraft engine surge detection system, which adopts the steps of 1, determining the fault status of the power system and the surge detection system; step 2, real-time acquisition of the in-machine fault status of the power system and the surge detection system, and when the power system and the surge detection system have a fault status, the corresponding fault status is stored in real time in the form of a fault code; step 3, connecting the ground fault diagnosis system with the data communication interface to realize the communication function between the ground fault diagnosis system and the airborne surge detection system; step 4, reading the fault code stored in the in-machine fault diagnosis system, locating the fault status of the power system and the surge detection system and displaying it; step 5, locating the sensor fault status and displaying it; step 6, judging the health status of the airborne surge detection system through the fault assessment module in the ground fault diagnosis system. However, the above technical solutions still need to check the mechanical parts, electronic equipment, and control program of the entire aircraft one by one, which has a large inspection workload and low engine fault diagnosis efficiency, and cannot scientifically, accurately, and efficiently locate the engine fault point position, as well as the intelligent aircraft software and hardware fault point diagnosis and analysis.

[0005] The Chinese patent with the announcement number CN113469995B discloses a method and system for diagnosing thermal faults of substation equipment; specifically discloses: the present invention relates to a method and system for diagnosing thermal faults of substation equipment. The method comprises: preprocessing an infrared image; determining a corresponding temperature matrix according to the preprocessed infrared image; using a K-means clustering algorithm to roughly segment the substation equipment target and background on the preprocessed infrared image; combining the differential information obtained by detecting the edge of the target with the Prewitt operator with the Chan-Vese model to determine an improved Chan-Vese model; using the roughly segmented infrared image as the initial segmentation condition of the above model, using the above model to finely segment the substation equipment target and background on the roughly segmented infrared image to determine the target position to be diagnosed; determining the temperature of the target position according to the target position and the temperature matrix; and diagnosing the thermal fault of the substation equipment according to the temperature of the target position. The present invention can improve the accuracy of thermal fault diagnosis of substation equipment.

[0006] The Chinese patent with publication number CN109163911A discloses an engine fuel system fault diagnosis method based on an improved bat algorithm to optimize ELM; specifically disclosed: The present invention relates to an engine fuel system fault diagnosis method based on an improved bat algorithm to optimize ELM, belonging to the field of mechanical engineering automation technology. The present invention uses sensors to collect vibration signals of the fuel system in real time and extract its feature vectors; then uses the bat algorithm to optimize the weights and thresholds of the extreme learning machine, and builds the optimized network structure of the extreme learning machine; then the extracted feature vectors are input into the optimized extreme learning machine for training to obtain an engine fuel system fault diagnosis model and perform engine fuel system fault diagnosis. The present invention has a fast classification speed and improves the accuracy of engine fuel system fault diagnosis. Summary of the invention

[0007] 1. Technical issues to be resolved

[0008] In order to solve the above-mentioned problem that the existing civil aviation engine fault diagnosis can only rely on manual combined with computer-assisted regular comprehensive maintenance and overhaul of the engine, the maintenance process requires the entire engine mechanical parts, electronic equipment, and control programs to be checked one by one, the workload is large and the engine fault diagnosis efficiency is low. The existing civil aviation engine fault diagnosis system cannot scientifically, accurately and efficiently locate the engine fault point, as well as the intelligent engine software and hardware fault point diagnosis and analysis, which reduces the efficiency and reliability of civil aviation engine fault diagnosis. The purpose of accurate and efficient positioning of the above engine fault points, intelligent engine hardware fault analysis, and intelligent engine software fault analysis is achieved.

[0009] (II) Technical solution

[0010] The present invention is implemented by the following technical solution: a civil aviation engine fault diagnosis method based on big data drive, the method comprising the following steps:

[0011] S1, collecting temperature characteristic image data of the engine running body space;

[0012] S2, using a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, identifying the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generating engine running body space fault temperature distribution recognition result data;

[0013] S3. When the engine running body space fault temperature distribution recognition result data exists, collect the engine body physical space model feature data; match the engine running body space fault temperature distribution recognition result data with the engine body physical space model feature data according to the body space model structure, locate the position where the body space fault temperature feature is distributed on the surface of the engine body physical space model, and generate the engine body fault position location data;

[0014] S4, when the engine body fault position location data is completed, the engine fault position hardware image feature data is collected; the engine fault position hardware image feature data and the engine hardware fault image feature data are analyzed and identified by a data recognition algorithm (bat algorithm), and the engine body fault position hardware fault diagnosis result data is constructed;

[0015] S5. When the engine body fault position hardware fault diagnosis result data is completed, the engine fault position control software test result data is collected; the engine fault position control software test result data is compared with the engine control software test qualified result data to construct the engine fault position control software fault diagnosis result data;

[0016] S6, obtaining the engine running body spatial fault temperature distribution identification result data, the engine body fault location positioning data, the engine body fault location hardware fault diagnosis result data, the engine fault location control software fault diagnosis result data, and combining and processing them to generate engine fault comprehensive diagnosis result data;

[0017] S7: Push and feed back the engine fault comprehensive diagnosis result data to the engine fault diagnosis center.

[0018] Preferably, the operation steps of collecting the temperature characteristic image data of the engine running body space are as follows:

[0019] S11, using the thermal imaging infrared camera lens mounted on the drone to scan and capture the entire engine body space temperature characteristic image when the engine is running and generate the engine running body space temperature characteristic image data a jiti .

[0020] Preferably, the steps of using a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, identifying the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generating the engine running body space fault temperature distribution recognition result data are as follows:

[0021] S21, establish the engine running body space fault temperature characteristic image data set B = (b1, ..., b m ,…,b ε ), m = 1, 2, 3, ..., ε; where b m represents the mth engine running body space fault temperature characteristic image data, ε represents the maximum number of engine running body space fault temperature characteristic image data, and the engine running body space fault temperature characteristic image represents the body surface temperature distribution characteristic image when the engine is in a fault state;

[0022] S22, using a data recognition algorithm to identify the engine running body space temperature characteristic image data a jiti The engine running body space fault temperature characteristic image data set B has the engine running body space fault temperature characteristic image data b m Performing body space temperature characteristic image analysis, matching and identifying body space fault temperature characteristic image data, the operation steps of the data recognition algorithm to identify the fault temperature characteristic image data are as follows:

[0023] S221, parameter initialization: temperature feature identification bat population size N, maximum number of iterations T, objective function f(X), temperature feature identification bat individual i position Z i , i = 1, 2, ..., N; and temperature characteristics to identify the speed Φ of individual bat i i , the temperature characteristic identifies the sound wave frequency Γ of bat individual i i , temperature characteristics identify the sound wave loudness Λ of bat individual i i and temperature characteristics to identify the sound wave frequency Ψ of bat individual i i ;

[0024] S222. Find the optimal temperature feature in the current population to identify the bat position Z best , that is, the temperature feature recognition bat population searches for the engine running body space fault temperature feature image data set B in the search space of the engine running body space fault temperature feature image data a jiti The most matching engine running body space fault temperature characteristic image data b m The calculation formula for updating the parameters of bats using temperature characteristics is:

[0025]

[0026] in represents the speed of the temperature feature recognition bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the speed of the temperature feature recognition bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the position of the temperature feature recognition bat individual i in the search space of the engine running body spatial fault temperature feature image data set B in the t+1th iteration, represents the position of the temperature feature recognition bat individual i in the search space of the engine running body spatial fault temperature feature image data set B in the tth iteration, Z best It indicates the optimal position of the temperature feature to identify the individual bat in the search space of the engine running body space fault temperature feature image data set B, represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the frequency of sound waves of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the frequency of sound waves emitted by individual bat i in the search space of the temperature feature image data set B of the engine running body space fault by temperature feature identification; exp represents a function with values ​​(0, 1), α and β are constants and 0<α<1, β>0;

[0027] S223, generate a random number rand1, rand1 is a random number with a value of [0, 1]. When rand1>Ψ i , select the temperature feature of the optimal position from the search space of the engine running body space fault temperature feature image data set B to identify the bat individual, that is, search out the engine running body space temperature feature image data a jiti The most matching engine running body space fault temperature characteristic image data b m , the temperature characteristics of the optimal location are selected to identify individual bats through the formula exp→∞ generates a local temperature characteristic image data a of the engine running body space jiti Matched engine running body space fault temperature characteristic image data b m The solution of It represents the frequency of sound waves of the temperature feature image data set B in the engine running body space fault temperature feature image data set in the tth iteration. When rand1≤Ψ i When, according to the formula, Update temperature signatures to identify bat locations;

[0028] S224, generate another random number rand2, where rand2 is a random number with a value of [0, 1]; when rand2 < Λ i , and at this time the engine running body space fault temperature characteristic image data b m Corresponding objective function f(X) and engine running body space temperature characteristic image data a jiti If the fitness value of is better than the new solution in step S223, then accept the position and use the formula Z new =Z old +χ×Λ t , Adjust Λ i and ψ i ; where Z new represents the position of the individual bat after adjustment in the search space of the temperature feature image data set B of the engine running body space fault temperature feature, Z old represents the position of the temperature feature recognition bat individual in the search space of the engine running body space fault temperature feature image data set B before adjustment, Λ t represents the sound wave loudness of the individual bat identified by temperature feature in the engine running body space fault temperature feature image data set B in the t-th iteration, χ is an arbitrary number with a value of [-1, 1]; when rand2≥Λ i When, according to the formula, Update temperature signatures to identify bat locations;

[0029] S225, sorting the fitness values ​​of all individuals in the temperature feature recognition bat population, that is, searching for all engine running body space fault temperature feature image data b m The engine running body space temperature characteristic image data a jiti Sort the fitness values ​​and find the current best Z best , that is, find the characteristic image data a of the engine running body space temperature jiti The most matching engine running body space fault temperature characteristic image data b m ;

[0030] S226, repeating steps S222 to S225; when the maximum number of iterations T is met, outputting the temperature characteristic image data a of the engine running body space jiti The most matching engine running body space fault temperature characteristic image data b m ;

[0031] S23, the engine running body space fault temperature characteristic image data b outputted in S226 m The engine running body space temperature characteristic image data a jitiPerform body space temperature feature image matching and overlap, and mark the fault temperature feature by color in the engine running body space temperature feature image data a jiti The distribution characteristics and position characteristics in the engine running body are used to generate the engine running body space fault temperature distribution recognition result data a guzhang .

[0032] Preferably, when the engine running body space fault temperature distribution recognition result data exists, the engine body physical space model feature data is collected; the engine running body space fault temperature distribution recognition result data and the engine body physical space model feature data are matched according to the body space model structure, the position of the body space fault temperature feature distributed on the surface of the engine body physical space model is located, and the engine body fault position location data is generated. The operation steps are as follows:

[0033] S31, when the engine is running, the body space fault temperature distribution identification result data a guzhang When it exists, the UAV is equipped with a laser 3D scanner to perform 3D modeling on the engine body to generate the engine body physical space model feature data c shiwu ;

[0034] S32, the engine running body space fault temperature distribution identification result data a guzhang The engine body physical space model characteristic data c shiwu According to the engine body structure, the engine body spatial fault temperature characteristics and position characteristics are mapped to the corresponding engine body physical space model feature surface, and the engine body fault location positioning data c is generated. dingwei .

[0035] Preferably, when the engine body fault position positioning data is completed, the engine fault position hardware image feature data is collected; the engine fault position hardware image feature data and the engine hardware fault image feature data are analyzed and identified by a data recognition algorithm to construct the engine body fault position hardware fault diagnosis result data in the following steps:

[0036] S41, when the engine body fault location positioning data c is completed dingwei When the engine body fault location positioning data c dingwei Control the drone equipped with a camera to collect the image features of mechanical parts and electronic equipment at the fault location of the engine body when it is not running, and generate a hardware image feature data set D = (d1,…,d n ,…,d φ ), n=1,2,3,…,φ; where d nrepresents the hardware image feature data of the engine fault position corresponding to the nth engine hardware collected, and φ represents the maximum value of the number of engine hardware;

[0037] S42. Establishing an engine hardware fault image feature data set where d' j represents the j-th engine hardware fault image feature data, Indicates the maximum value of the number of engine hardware fault types, where the engine fault location hardware fault includes one or more of engine structure failure, component wear, rotor failure, thrust component failure, turbine failure, cylinder failure, fuel system failure, and oxygen supply system failure;

[0038] S43, using the data recognition algorithm in step S22 to identify the engine fault location hardware image feature data set D. n and the engine hardware fault image feature data set D' j The hardware image feature data of the engine hardware fault image is matched, and the hardware fault diagnosis result data D″ of the engine body fault position is constructed through analysis;

[0039] When n With d' j If the match is successful, the output engine body fault location hardware fault diagnosis result data D" is d' 硬件 ;

[0040] When n With d' j If the match is not successful, the output engine body fault location hardware fault diagnosis result data D″ is none.

[0041] Preferably, when the engine body fault position hardware fault diagnosis result data is completed, the engine fault position control software test result data is collected; the engine fault position control software test result data is compared with the engine control software test qualified result data, and the operation steps of constructing the engine fault position control software fault diagnosis result data are as follows:

[0042] S51, when the hardware fault diagnosis result data of the engine body fault position is completed, based on the engine body fault position positioning data c dingwei , through the engine software test platform, the control software of the control hardware action corresponding to the fault position of the engine body in the stopped state is tested and the ratio of the normal test running times to the total test running times is measured, and the engine fault position control software test result data e is generated;

[0043] S52. Establish engine control software test qualified result data hege , the control software test qualified result represents the minimum value of the control software test result data of the normal operation of the engine body hardware control software;

[0044] S53, comparing the engine fault location control software test result data e with the engine control software test qualified result data e hege Perform numerical comparison and construct engine fault location control software fault diagnosis result data E;

[0045] When e≥e hege , the engine fault location control software fault diagnosis result data E is that there is no fault in the hardware control software;

[0046] when e<e hege , the engine fault location control software fault diagnosis result data E indicates that there is a fault in the hardware control software.

[0047] Preferably, the operation steps of obtaining the engine running body spatial fault temperature distribution identification result data, the engine body fault position location data, the engine body fault position hardware fault diagnosis result data, and the engine fault position control software fault diagnosis result data and combining and processing to generate engine fault comprehensive diagnosis result data are as follows:

[0048] S61, obtaining the engine running body space fault temperature distribution identification result data a guzhang , the engine body fault location positioning data c dingwei , the engine body fault position hardware fault diagnosis result data D″, the engine fault position control software fault diagnosis result data E;

[0049] S62, the engine running body space fault temperature distribution identification result data a guzhang , the engine body fault location positioning data c dingwei , the engine body fault position hardware fault diagnosis result data D″, the engine fault position control software fault diagnosis result data E are combined and processed to generate the engine fault comprehensive diagnosis result data set G=(a guzhang ,c dingwei ,D”,E).

[0050] Preferably, the operation steps of pushing and feeding back the engine fault comprehensive diagnosis result data to the engine fault diagnosis center are as follows:

[0051] S71. Push and feed back the engine fault comprehensive diagnosis result data set G to the engine fault diagnosis center via mobile network communication, and maintenance personnel perform targeted maintenance and repair operations on the engine based on the engine fault comprehensive diagnosis result data set G.

[0052] A system for implementing the civil aviation engine fault diagnosis method based on big data drive, the system comprising an engine fault location location management module, an engine fault diagnosis management module, and an engine fault diagnosis result feedback module;

[0053] The engine fault location management module includes an engine running body space temperature feature acquisition unit, an engine running body space fault temperature feature storage unit, an engine running body space fault temperature feature recognition unit, an engine body physical space model feature acquisition unit, and an engine body fault location analysis unit;

[0054] The engine running body space temperature feature acquisition unit collects the engine running body space temperature feature image data through the thermal imaging infrared shooting lens carried by the unmanned aerial vehicle; the engine running body space fault temperature feature storage unit is used to store the engine running body space fault temperature feature image data; the engine running body space fault temperature feature recognition unit uses a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, recognizes the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generates the engine running body space fault temperature distribution recognition result data; the engine body physical space model feature acquisition unit collects the engine body physical space model feature data through the unmanned aerial vehicle equipped with a laser three-dimensional scanner; the engine body fault position positioning analysis unit matches the engine running body space fault temperature distribution recognition result data with the engine body physical space model feature data according to the body space model structure, locates the position where the body space fault temperature feature is distributed on the surface of the engine body physical space model, and generates the engine body fault position positioning data;

[0055] The engine fault diagnosis management module includes an engine fault location hardware image acquisition unit, an engine hardware fault image storage unit, an engine fault location hardware fault diagnosis unit, an engine fault location software test result acquisition unit, an engine software test qualified result storage unit, and an engine fault location software fault diagnosis unit;

[0056] The engine fault position hardware image acquisition unit acquires engine fault position hardware image feature data through a drone-mounted shooting device; the engine hardware fault image storage unit is used to store engine hardware fault image feature data; the engine fault position hardware fault diagnosis unit uses a data recognition algorithm to perform hardware image feature analysis and recognition on the engine fault position hardware image feature data and the engine hardware fault image feature data, and constructs engine body fault position hardware fault diagnosis result data; the engine fault position software test result acquisition unit acquires engine fault position control software test result data through an engine software test platform; the engine software test qualified result storage unit is used to store engine control software test qualified result data; the engine fault position software fault diagnosis unit compares the engine fault position control software test result data with the engine control software test qualified result data, and constructs engine fault position control software fault diagnosis result data;

[0057] The engine fault diagnosis result feedback module includes an engine fault diagnosis result data acquisition unit, an engine fault diagnosis result data processing unit, and an engine fault diagnosis result feedback unit;

[0058] The engine fault diagnosis result data acquisition unit is used to obtain the engine running body space fault temperature distribution identification result data, the engine body fault position positioning data, the engine body fault position hardware fault diagnosis result data, and the engine fault position control software fault diagnosis result data; the engine fault diagnosis result data processing unit combines the engine running body space fault temperature distribution identification result data, the engine body fault position positioning data, the engine body fault position hardware fault diagnosis result data, and the engine fault position control software fault diagnosis result data to generate engine fault comprehensive diagnosis result data; the engine fault diagnosis result feedback unit pushes and feeds back the engine fault comprehensive diagnosis result data to the engine fault diagnosis center.

[0059] (III) Beneficial effects

[0060] The present invention provides a civil aviation engine fault diagnosis method and system based on big data drive, which has the following beneficial effects:

[0061] 1. The engine running body space fault temperature feature storage unit and the engine running body space fault temperature feature recognition unit cooperate with each other to adopt the body temperature distribution image parameters under the preset engine fault state combined with the intelligent recognition algorithm and the collected engine body space temperature image parameters to perform temperature feature recognition, so as to scientifically and accurately identify the engine body temperature fault characteristics and positions. The engine body physical space model feature acquisition unit and the engine body fault position positioning analysis unit cooperate with each other to use the engine body physical three-dimensional model and the engine body temperature fault characteristics to match the body model structure, so as to realize the accurate mapping of the engine body temperature fault characteristics to the engine body physical model, intuitively and accurately locate the engine body fault position, and improve the positioning efficiency and accuracy of civil aviation engine fault diagnosis.

[0062] 2. The engine fault location hardware image acquisition unit and the engine fault location hardware fault diagnosis unit cooperate with each other to use the engine fault location hardware image in combination with the intelligent recognition algorithm to match the engine hardware fault image, thereby realizing intelligent and efficient hardware analysis of the engine fault location. The engine fault location software test result acquisition unit and the engine fault location software fault diagnosis unit cooperate with each other to use the engine fault location control software test result parameters and the preset engine control software test qualified result parameters to perform numerical comparison, thereby realizing digital analysis and management of engine software operation failures, improving the efficiency and reliability of civil aviation engine fault diagnosis, and ensuring the service life and safety of civil aviation engines.

[0063] 3. Through the engine fault diagnosis result data acquisition unit and mutual cooperation, accurate information collection and processing of engine fault location information, engine hardware fault characteristics, and engine software fault characteristics are achieved. At the same time, the engine-generated fault information is combined to generate comprehensive engine fault analysis data, which is convenient for maintenance personnel to browse and manage engine fault information, and realize scientific and standardized engine fault diagnosis management. The engine fault diagnosis result feedback unit feeds back and pushes engine fault diagnosis information, realizing efficient and accurate feedback of engine fault diagnosis information. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic diagram of a module of a civil aviation engine fault diagnosis system driven by big data provided by the present invention;

[0065] Figure 2 A flowchart of a civil aviation engine fault diagnosis method driven by big data provided by the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] The embodiments of the civil aviation engine fault diagnosis method and system based on big data drive are as follows:

[0068] See also Figure 1 - Figure 2 A civil aviation engine fault diagnosis method based on big data drive includes the following steps:

[0069] S1, collecting temperature characteristic image data of the engine running body space;

[0070] S2, using a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, identifying the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generating engine running body space fault temperature distribution recognition result data;

[0071] S3. When the engine running body space fault temperature distribution recognition result data exists, collect the engine body physical space model feature data; match the engine running body space fault temperature distribution recognition result data with the engine body physical space model feature data according to the body space model structure, locate the position where the body space fault temperature feature is distributed on the surface of the engine body physical space model, and generate the engine body fault position location data;

[0072] S4, when the engine body fault position location data is completed, the engine fault position hardware image feature data is collected; the engine fault position hardware image feature data and the engine hardware fault image feature data are analyzed and identified by a data recognition algorithm to construct the engine body fault position hardware fault diagnosis result data;

[0073] S5. When the engine body fault position hardware fault diagnosis result data is completed, the engine fault position control software test result data is collected; the engine fault position control software test result data is compared with the engine control software test qualified result data to construct the engine fault position control software fault diagnosis result data;

[0074] S6, obtaining engine running body spatial fault temperature distribution identification result data, engine body fault location positioning data, engine body fault location hardware fault diagnosis result data, engine fault location control software fault diagnosis result data, and combining and processing them to generate engine fault comprehensive diagnosis result data;

[0075] S7. Push and feed back the engine fault comprehensive diagnosis result data to the engine fault diagnosis center.

[0076] For further information, see Figure 1 - Figure 2 The operation steps for collecting the temperature characteristic image data of the engine running body space are as follows:

[0077] S11, using the thermal imaging infrared camera lens mounted on the drone to scan and capture the entire engine body space temperature characteristic image when the engine is running and generate the engine running body space temperature characteristic image data a jiti .

[0078] The steps of using a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, identifying the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generating the engine running body space fault temperature distribution recognition result data are as follows:

[0079] S21, establish the engine running body space fault temperature characteristic image data set B = (b1, ..., b m ,…,bε), m=1,2,3,…,ε; where b m represents the mth engine running body space fault temperature characteristic image data, ε represents the maximum number of engine running body space fault temperature characteristic image data, and the engine running body space fault temperature characteristic image represents the body surface temperature distribution characteristic image when the engine is running in a fault state;

[0080] S22, using a data recognition algorithm to convert the engine running body space temperature characteristic image data a jiti The engine running body space fault temperature characteristic image data set B has the engine running body space fault temperature characteristic image data b m Perform body space temperature characteristic image analysis, match and identify body space fault temperature characteristic image data, and the operation steps of the data recognition algorithm to identify the fault temperature characteristic image data are as follows:

[0081] S221, parameter initialization: temperature feature identification bat population size N, maximum number of iterations T, objective function f(X), temperature feature identification bat individual i position Zi , i = 1, 2, ..., N; and temperature characteristics to identify the speed Φ of individual bat i i , the temperature characteristic identifies the sound wave frequency Γ of bat individual i i , temperature characteristics identify the sound wave loudness Λ of bat individual i i and temperature characteristics to identify the sound wave frequency Ψ of bat individual i i ;

[0082] S222. Find the optimal temperature feature in the current population to identify the bat position Z best , that is, the temperature feature recognition bat population searches for the engine running body space fault temperature feature image data set B in the search space of the engine running body space fault temperature feature image data a jiti The most matching engine running body space fault temperature characteristic image data b m The calculation formula for updating the parameters of bats using temperature characteristics is:

[0083] in represents the speed of the temperature feature recognition bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the speed of the temperature feature recognition bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the position of the temperature feature recognition bat individual i in the search space of the engine running body spatial fault temperature feature image data set B in the t+1th iteration, represents the position of the temperature feature recognition bat individual i in the search space of the engine running body spatial fault temperature feature image data set B in the tth iteration, Z best It indicates the optimal position of the temperature feature to identify the individual bat in the search space of the engine running body space fault temperature feature image data set B, represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the frequency of sound waves of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the frequency of sound waves emitted by individual bat i in the search space of the temperature feature image data set B of the engine running body space fault by temperature feature identification; exp represents a function with values ​​(0, 1), α and β are constants and 0<α<1, β>0;

[0084] S223, generate a random number rand1, rand1 is a random number with a value of [0, 1]. When rand1>Ψ i , select the temperature feature of the optimal position from the search space of the engine running body space fault temperature feature image data set B to identify the bat individual, that is, search out the engine running body space temperature feature image data a jiti The most matching engine running body space fault temperature characteristic image data b m , the temperature characteristics of the optimal location are selected to identify individual bats through the formula exp→∞ generates a local temperature characteristic image data a of the engine running body space jiti Matched engine running body space fault temperature characteristic image data b m The solution of It represents the frequency of sound waves of the temperature feature image data set B in the engine running body space fault temperature feature image data set in the tth iteration. When rand1≤Ψ i When, according to the formula, Update temperature signatures to identify bat locations;

[0085] S224, generate another random number rand2, where rand2 is a random number with a value of [0, 1]; when rand2 < Λ i , and at this time the engine running body space fault temperature characteristic image data b m Corresponding objective function f(X) and engine running body space temperature characteristic image data a jiti If the fitness value of is better than the new solution in step S223, then accept the position and use the formula Z new =Z old +χ×Λ t , Adjust Λ i and ψ i ; where Z new represents the position of the individual bat after adjustment in the search space of the temperature feature image data set B of the engine running body space fault temperature feature, Z old represents the position of the temperature feature recognition bat individual in the search space of the engine running body space fault temperature feature image data set B before adjustment, Λ trepresents the sound wave loudness of the individual bat identified by temperature feature in the engine running body space fault temperature feature image data set B in the t-th iteration, χ is an arbitrary number with a value of [-1, 1]; when rand2≥Λ i When, according to the formula, Update temperature signatures to identify bat locations;

[0086] S225, sorting the fitness values ​​of all individuals in the temperature feature recognition bat population, that is, searching for all engine running body space fault temperature feature image data b m The engine running body space temperature characteristic image data a jiti Sort the fitness values ​​and find the current best Z best , that is, find the characteristic image data a of the engine running body space temperature jiti The most matching engine running body space fault temperature characteristic image data b m ;

[0087] S226, repeating steps S222 to S225; when the maximum number of iterations T is met, outputting the temperature characteristic image data a of the engine running body space jiti The most matching engine running body space fault temperature characteristic image data b m ;

[0088] S23, the engine running body space fault temperature characteristic image data b outputted from S226 m The engine running body space temperature characteristic image data a jiti Perform body space temperature feature image matching and overlap, and identify the fault temperature feature by color in the engine running body space temperature feature image data a jiti The distribution characteristics and position characteristics in the engine running body are used to generate the engine running body space fault temperature distribution recognition result data a guzhang .

[0089] When the engine running body space fault temperature distribution recognition result data exists, the engine body physical space model feature data is collected; the engine running body space fault temperature distribution recognition result data and the engine body physical space model feature data are matched according to the body space model structure, the position of the body space fault temperature feature distributed on the surface of the engine body physical space model is located, and the engine body fault position location data is generated. The operation steps are as follows:

[0090] S31, when the engine is running, the body space fault temperature distribution identification result data a guzhang When it exists, the UAV is equipped with a laser 3D scanner to perform 3D modeling on the engine body to generate the engine body physical space model feature data cshiwu ;

[0091] S32, the engine running body space fault temperature distribution identification result data a guzhang The engine body physical space model characteristic data c shiwu According to the engine body structure, the engine body spatial fault temperature characteristics and position characteristics are mapped to the corresponding engine body physical space model feature surface, and the engine body fault location positioning data c is generated. dingwei .

[0092] The engine running body spatial fault temperature feature storage unit and the engine running body spatial fault temperature feature recognition unit cooperate with each other to adopt the body temperature distribution image parameters under the preset engine fault state combined with the intelligent recognition algorithm and the collected engine body spatial temperature image parameters to perform temperature feature recognition, so as to scientifically and accurately identify the engine body temperature fault characteristics and positions. The engine body physical space model feature acquisition unit and the engine body fault position positioning analysis unit cooperate with each other to use the engine body physical three-dimensional model and the engine body temperature fault characteristics to match the body model structure, so as to achieve accurate mapping of the engine body temperature fault characteristics to the engine body physical model, intuitively and accurately locate the engine body fault position, and improve the positioning efficiency and accuracy of civil aviation engine fault diagnosis.

[0093] For further information, see Figure 1 - Figure 2 When the engine body fault position location data is completed, the engine fault position hardware image feature data is collected; the engine fault position hardware image feature data and the engine hardware fault image feature data are analyzed and identified by using a data recognition algorithm, and the operation steps for constructing the engine body fault position hardware fault diagnosis result data are as follows:

[0094] S41, when the engine body fault location positioning data c is completed dingwei When the engine body fault location data c is used, dingwei Control the drone equipped with a camera to collect the image features of mechanical parts and electronic equipment at the fault location of the engine body when it is not running, and generate a hardware image feature data set D = (d1,…,d n ,…,d φ ), n=1,2,3,…,φ; where d n represents the hardware image feature data of the engine fault position corresponding to the nth engine hardware collected, and φ represents the maximum value of the number of engine hardware;

[0095] S42. Establishing an engine hardware fault image feature data set where d' j represents the j-th engine hardware fault image feature data, Indicates the maximum number of engine hardware fault types. Engine fault location hardware faults include one or more of engine structure failure, component wear, rotor failure, thrust component failure, turbine failure, cylinder failure, fuel system failure, and oxygen supply system failure.

[0096] S43, using the data recognition algorithm in step S22 to identify the engine fault location hardware image feature data set D of the engine fault location hardware image feature data set D. n and the engine hardware fault image feature data set D' j The hardware image feature data of the engine hardware fault image is matched, and the hardware fault diagnosis result data D″ of the engine body fault position is constructed through analysis;

[0097] When n With d' j If the match is successful, the output engine body fault location hardware fault diagnosis result data D" is d' 硬件 ;

[0098] When n With d' j If the match is not successful, the output engine body fault location hardware fault diagnosis result data D″ is none.

[0099] When the hardware fault diagnosis result data of the engine body fault position is completed, the engine fault position control software test result data is collected; the engine fault position control software test result data is compared with the engine control software test qualified result data, and the operation steps for constructing the engine fault position control software fault diagnosis result data are as follows:

[0100] S51, when the hardware fault diagnosis result data of the engine body fault position is completed, according to the engine body fault position positioning data c dingwei , through the engine software test platform, the control software of the control hardware action corresponding to the fault position of the engine body in the stopped state is tested and the ratio of the normal test running times to the total test running times is measured, and the engine fault position control software test result data e is generated;

[0101] S52. Establish engine control software test qualified result data hege , the qualified result of the control software test indicates the minimum value of the control software test result data of the normal operation of the engine body hardware control software;

[0102] S53, compare the engine fault position control software test result data e and the engine control software test qualified result data e hege Perform numerical comparison and construct engine fault location control software fault diagnosis result data E;

[0103] When e≥e hege , the engine fault location control software fault diagnosis result data E is that there is no fault in the hardware control software;

[0104] when e<e hege , the engine fault location control software fault diagnosis result data E indicates that there is a fault in the hardware control software.

[0105] The engine fault location hardware image acquisition unit and the engine fault location hardware fault diagnosis unit cooperate with each other to use the engine fault location hardware image in combination with an intelligent recognition algorithm to match the engine hardware fault image, thereby realizing intelligent and efficient hardware analysis of the engine fault location. The engine fault location software test result acquisition unit and the engine fault location software fault diagnosis unit cooperate with each other to perform numerical comparison between the engine fault location control software test result parameters and the preset engine control software test qualified result parameters, thereby realizing digital analysis and management of engine software operation failures, thereby improving the efficiency and reliability of civil aviation engine fault diagnosis and ensuring the service life and safety of civil aviation engines.

[0106] For further information, see Figure 1 - Figure 2 The steps for obtaining the engine running body spatial fault temperature distribution identification result data, the engine body fault location positioning data, the engine body fault location hardware fault diagnosis result data, and the engine fault location control software fault diagnosis result data and combining and processing them to generate the engine fault comprehensive diagnosis result data are as follows:

[0107] S61, obtaining the engine running body space fault temperature distribution identification result data a guzhang 、Engine body fault location location data c dingwei , engine body fault location hardware fault diagnosis result data D″, engine fault location control software fault diagnosis result data E;

[0108] S62, the engine running body space fault temperature distribution identification result data a guzhang , Engine body fault location location data c dingwei , engine body fault location hardware fault diagnosis result data D″, engine fault location control software fault diagnosis result data E are combined and processed to generate engine fault comprehensive diagnosis result data set G=(a guzhang ,c dingwei,D”,E).

[0109] The steps to push the engine fault comprehensive diagnosis result data back to the engine fault diagnosis center are as follows:

[0110] S71. Push and feed back the engine fault comprehensive diagnosis result data set G to the engine fault diagnosis center via mobile network communication, and maintenance personnel perform targeted maintenance and repair work on the engine based on the engine fault comprehensive diagnosis result data set G.

[0111] Through the engine fault diagnosis result data acquisition unit and their mutual cooperation, accurate information collection and processing of engine fault location information, engine hardware fault characteristics, and engine software fault characteristics are achieved. At the same time, the engine-generated fault information is combined to generate comprehensive engine fault analysis data, which is convenient for maintenance personnel to browse and manage engine fault information, thereby realizing scientific and standardized engine fault diagnosis management. The engine fault diagnosis result feedback unit feeds back and pushes engine fault diagnosis information, thereby realizing efficient and accurate feedback of engine fault diagnosis information.

[0112] A system for civil aviation engine fault diagnosis method based on big data drive is implemented, the system includes an engine fault location location management module, an engine fault diagnosis management module, and an engine fault diagnosis result feedback module;

[0113] The engine fault location management module includes an engine running body space temperature feature acquisition unit, an engine running body space fault temperature feature storage unit, an engine running body space fault temperature feature recognition unit, an engine body physical space model feature acquisition unit, and an engine body fault location analysis unit;

[0114] An engine running body space temperature feature acquisition unit, which collects engine running body space temperature feature image data through a thermal imaging infrared camera lens mounted on a drone; an engine running body space fault temperature feature storage unit, which is used to store engine running body space fault temperature feature image data; an engine running body space fault temperature feature recognition unit, which uses a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, recognizes the body space fault temperature feature and the position data of the fault temperature feature in the body space temperature feature image, and generates engine running body space fault temperature distribution recognition result data; an engine body physical space model feature acquisition unit, which collects engine body physical space model feature data through a laser three-dimensional scanner mounted on a drone; an engine body fault position positioning analysis unit, which matches the engine running body space fault temperature distribution recognition result data with the engine body physical space model feature data according to the body space model structure, locates the position where the body space fault temperature feature is distributed on the surface of the engine body physical space model, and generates engine body fault position positioning data;

[0115] The engine fault diagnosis management module includes an engine fault location hardware image acquisition unit, an engine hardware fault image storage unit, an engine fault location hardware fault diagnosis unit, an engine fault location software test result acquisition unit, an engine software test qualified result storage unit, and an engine fault location software fault diagnosis unit;

[0116] The engine fault location hardware image acquisition unit collects the engine fault location hardware image feature data through the drone-mounted shooting equipment; the engine hardware fault image storage unit is used to store the engine hardware fault image feature data; the engine fault location hardware fault diagnosis unit uses a data recognition algorithm to perform hardware image feature analysis and recognition on the engine fault location hardware image feature data and the engine hardware fault image feature data, and constructs the engine body fault location hardware fault diagnosis result data; the engine fault location software test result acquisition unit collects the engine fault location control software test result data through the engine software test platform; the engine software test qualified result storage unit is used to store the engine control software test qualified result data; the engine fault location software fault diagnosis unit compares the engine fault location control software test result data with the engine control software test qualified result data, and constructs the engine fault location control software fault diagnosis result data;

[0117] The engine fault diagnosis result feedback module includes an engine fault diagnosis result data acquisition unit, an engine fault diagnosis result data processing unit, and an engine fault diagnosis result feedback unit;

[0118] The engine fault diagnosis result data acquisition unit is used to obtain the engine running body space fault temperature distribution identification result data, the engine body fault position positioning data, the engine body fault position hardware fault diagnosis result data, and the engine fault position control software fault diagnosis result data; the engine fault diagnosis result data processing unit combines the engine running body space fault temperature distribution identification result data, the engine body fault position positioning data, the engine body fault position hardware fault diagnosis result data, and the engine fault position control software fault diagnosis result data to generate the engine fault comprehensive diagnosis result data; the engine fault diagnosis result feedback unit pushes the engine fault comprehensive diagnosis result data to the engine fault diagnosis center.

[0119] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A civil aviation engine fault diagnosis method based on big data, characterized in that: The method comprises the following steps: S1, collecting temperature characteristic image data of the engine running body space, said S1 comprises the following steps: S11, using the thermal imaging infrared camera lens mounted on the drone to scan and capture the entire engine body space temperature characteristic image when the engine is running and generate the engine running body space temperature characteristic image data a jiti ; S2, using a data recognition algorithm to perform body space temperature feature image analysis on the engine running body space temperature feature image data and the engine running body space fault temperature feature image data, identifying the body space fault temperature feature and the fault temperature feature and the position data of their distribution in the body space temperature feature image, and generating engine running body space fault temperature distribution recognition result data, wherein S2 includes the following steps: S21, establish the engine running body space fault temperature characteristic image data set B = (b1, ..., b m ,…,b ε ), m = 1, 2, 3, ..., ε; where b m represents the mth engine running body space fault temperature characteristic image data, ε represents the maximum number of engine running body space fault temperature characteristic image data, and the engine running body space fault temperature characteristic image represents the body surface temperature distribution characteristic image when the engine is running in a fault state; S22, using a data recognition algorithm to identify the engine running body space temperature characteristic image data a jiti The engine running body space fault temperature characteristic image data set B has the engine running body space fault temperature characteristic image data b m Performing body space temperature characteristic image analysis, matching and identifying body space fault temperature characteristic image data, the operation steps of the data recognition algorithm to identify the fault temperature characteristic image data are as follows: S221, parameter initialization: temperature feature identification bat population size N, maximum number of iterations T, objective function f(X), temperature feature identification bat individual i position Z i , i = 1, 2, ..., N; and temperature characteristics to identify the speed Φ of individual bat i i , the temperature characteristic identifies the sound wave frequency Γ of bat individual i i , temperature characteristics identify the sound wave loudness Λ of bat individual i i and temperature characteristics to identify the sound wave frequency ψ of bat individual i i ; S222. Find the optimal temperature feature in the current population to identify the bat position Z best , that is, the temperature feature recognition bat population searches for the engine running body space fault temperature feature image data set B in the search space of the engine running body space fault temperature feature image data a jiti The most matching engine running body space fault temperature characteristic image data b m ; S223, generate a random number rand1, rand1 is a random number with a value of [0, 1]. When rand1>ψ i , select the temperature feature of the optimal position from the engine running body space fault temperature feature image data set B search space to identify the bat individual, through the formula exp→∞ generates a local temperature characteristic image data a of the engine running body space jiti Matched engine running body space fault temperature characteristic image data b m The solution; when rand1≤ψ i When, according to the formula, Update temperature signatures to identify bat locations; represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the sound wave loudness of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the frequency of sound waves of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t+1th iteration, represents the frequency of sound waves of the temperature feature-recognized bat individual i in the search space of the engine running body space fault temperature feature image data set B in the t-th iteration, represents the frequency of sound waves emitted by individual bat i in the search space of the temperature feature image data set B of the engine running body space fault by temperature feature identification; exp represents a function with values ​​(0, 1), α and β are constants and 0<α<1, β>0; S224, generate another random number rand2, where rand2 is a random number with a value of [0, 1]; when rand2 < Λ i , and at this time the engine running body space fault temperature characteristic image data b m Corresponding objective function f(X) and engine running body space temperature characteristic image data a jiti If the fitness value of is better than the new solution in step S223, the current position is accepted and the solution is calculated according to the formula Z. new =Z old +χ×Λ t , Adjust Λ i and ψ i ; where Z new represents the position of the individual bat after adjustment in the search space of the temperature feature image data set B of the engine running body space fault temperature feature, Z old represents the position of the temperature feature recognition bat individual in the search space of the engine running body space fault temperature feature image data set B before adjustment, Λ t represents the sound wave loudness of the individual bat identified by temperature feature in the engine running body space fault temperature feature image data set B in the t-th iteration, χ is an arbitrary number with a value of [-1, 1]; when rand2≥Λ i When, according to the formula, Update temperature signatures to identify bat locations; S225, sorting the fitness values ​​of all individuals in the current temperature feature recognition bat population, that is, searching for all engine running body space fault temperature feature image data b m The engine running body space temperature characteristic image data a jiti Sort the fitness values ​​and find the current best Z best , that is, find the characteristic image data a of the engine running body space temperature jiti The most matching engine running body space fault temperature characteristic image data b m ; S226, repeating steps S222 to S225; when the maximum number of iterations T is met, outputting the temperature characteristic image data a of the engine running body space jiti The most matching engine running body space fault temperature characteristic image data b m ; S23, the engine running body space fault temperature characteristic image data b output in step S226 m The engine running body space temperature characteristic image data a jiti Perform body space temperature feature image matching and overlap, and mark the fault temperature feature by color in the engine running body space temperature feature image data a jiti The distribution characteristics and position characteristics in the engine running body are used to generate the engine running body space fault temperature distribution recognition result data a guzhang ; S3. When the engine running body space fault temperature distribution recognition result data exists, collect the engine body physical space model feature data; match the engine running body space fault temperature distribution recognition result data with the engine body physical space model feature data according to the body space model structure, locate the position where the body space fault temperature feature is distributed on the surface of the engine body physical space model, and generate the engine body fault position location data; S4, when the engine body fault position location data is completed, the engine fault position hardware image feature data is collected; the engine fault position hardware image feature data and the engine hardware fault image feature data are analyzed and identified by a data recognition algorithm to construct the engine body fault position hardware fault diagnosis result data; S5. When the engine body fault position hardware fault diagnosis result data is completed, the engine fault position control software test result data is collected; the engine fault position control software test result data is compared with the engine control software test qualified result data to construct the engine fault position control software fault diagnosis result data; S6, obtaining the engine running body spatial fault temperature distribution identification result data, the engine body fault location positioning data, the engine body fault location hardware fault diagnosis result data, the engine fault location control software fault diagnosis result data, and combining and processing them to generate engine fault comprehensive diagnosis result data; S7: Push and feed back the engine fault comprehensive diagnosis result data to the engine fault diagnosis center.

2. The method for diagnosing civil aviation engine faults based on big data drive according to claim 1, characterized in that: The S3 comprises the following steps: S31, when the engine is running, the body space fault temperature distribution identification result data a guzhang When it exists, the UAV is equipped with a laser 3D scanner to perform 3D modeling on the engine body to generate the engine body physical space model feature data c shiwu ; S32, the engine running body space fault temperature distribution identification result data a guzhang The engine body physical space model characteristic data c shiwu According to the engine body structure, the engine body spatial fault temperature characteristics and position characteristics are mapped to the corresponding engine body physical space model feature surface, and the engine body fault location positioning data c is generated. dingwei .

3. The method for diagnosing civil aviation engine faults based on big data drive according to claim 2 is characterized by: The S4 comprises the following steps: S41, when the engine body fault location positioning data c is completed dingwei When the engine body fault location positioning data c dingwei Control the drone equipped with a camera to collect the image features of mechanical parts and electronic equipment at the fault location of the engine body when it is not running, and generate a hardware image feature data set D = (d1,…,d n ,…,d φ ), n=1,2,3,…,φ; where d n represents the hardware image feature data of the engine fault position corresponding to the nth engine hardware collected, and φ represents the maximum value of the number of engine hardware; S42. Establishing an engine hardware fault image feature data set where d' j represents the j-th engine hardware fault image feature data, Indicates the maximum value of the number of engine hardware fault types, where the engine fault location hardware fault includes one or more of engine structure failure, component wear, rotor failure, thrust component failure, turbine failure, cylinder failure, fuel system failure, and oxygen supply system failure; S43, using the data recognition algorithm in step S22 to identify the engine fault location hardware image feature data set D. n and the engine hardware fault image feature data set D' j The hardware image feature data of the engine hardware fault image is matched, and the hardware fault diagnosis result data D″ of the engine body fault position is constructed through analysis; When n With d' j If the match is successful, the output engine body fault location hardware fault diagnosis result data D" is d' 硬件 ; When n With d' j If the match is not successful, the output engine body fault location hardware fault diagnosis result data D″ is none.

4. The method for diagnosing civil aviation engine faults based on big data drive according to claim 3 is characterized by: The S5 comprises the following steps: S51, when the hardware fault diagnosis result data of the engine body fault position is completed, based on the engine body fault position positioning data c dingwei , through the engine software test platform, the control software of the control hardware action corresponding to the fault position of the engine body in the stopped state is tested and the ratio of the normal test running times to the total test running times is measured, and the engine fault position control software test result data e is generated; S52. Establish engine control software test qualified result data hege , the control software test qualified result represents the minimum value of the control software test result data of the normal operation of the engine body hardware control software; S53, comparing the engine fault location control software test result data e with the engine control software test qualified result data e hege Perform numerical comparison and construct engine fault location control software fault diagnosis result data E; When e≥e hege , the engine fault location control software fault diagnosis result data E is that there is no fault in the hardware control software; when e<e hege , the engine fault location control software fault diagnosis result data E indicates that there is a fault in the hardware control software.

5. The method for diagnosing civil aviation engine faults based on big data drive according to claim 4 is characterized by: The S6 comprises the following steps: S61, obtaining the engine running body space fault temperature distribution identification result data a guzhang , the engine body fault location positioning data c dingwei , the engine body fault position hardware fault diagnosis result data D″, the engine fault position control software fault diagnosis result data E; S62, the engine running body space fault temperature distribution identification result data a guzhang , the engine body fault location positioning data c dingwei , the engine body fault position hardware fault diagnosis result data D″, the engine fault position control software fault diagnosis result data E are combined and processed to generate the engine fault comprehensive diagnosis result data set G=(a guzhang ,c dingwei ,D”,E).

6. The method for diagnosing civil aviation engine faults based on big data drive according to claim 5 is characterized by: The S7 comprises the following steps: S71. Push and feed back the engine fault comprehensive diagnosis result data set G to the engine fault diagnosis center via mobile network communication, and maintenance personnel perform targeted maintenance and repair operations on the engine based on the engine fault comprehensive diagnosis result data set G.

7. A system for implementing a civil aviation engine fault diagnosis method based on big data drive as described in any one of claims 1 to 6.

Images