Performance testing methods, devices, media and electronic equipment for aircraft engines

Abstract

The present invention relates to a performance test method, device, medium and electronic equipment for an aircraft engine. The test method comprises: obtaining a first performance parameter of an aircraft engine, wherein the first performance parameter is a flight parameter affecting the performance of the aircraft engine; dividing the first performance parameter into a plurality of flight states according to fuzzy reasoning; inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine; the extreme gradient model is trained by constructing a tree with a minimum loss function and a regular function for residual fitting; and detecting the performance of the aircraft engine according to the second performance parameter to obtain a performance detection result.

Description

Technical Field

[0001] The present invention relates to the field of aircraft engine performance testing, and in particular to an aircraft engine performance testing method, device, medium and electronic equipment. Background Art

[0002] Aircraft engines are the power core of aircraft. The health of aircraft engines is crucial to aviation safety. Therefore, the performance of aircraft engines must be tested frequently. Aircraft engine performance parameter testing refers to the monitoring of the main performance parameters that characterize the working conditions of aircraft engines, such as N2 (high-pressure rotor speed), N1 (fan speed), FF (fuel flow), and EGT (exhaust temperature) through certain technical means, and modeling and analysis based on actual monitoring data to obtain the changing trends of the working conditions of each component under different external environments. The changing trend of the performance parameters of aircraft engines can objectively reflect the performance degradation of aircraft engines. Due to the complex working process of aircraft engines, the changeable working environment, and the long-term high-speed rotation and high-temperature state of core components, their performance parameters show nonlinearity, non-stationarity, volatility and other characteristics, which brings great difficulties to the performance trend prediction of aircraft engines. Traditional prediction methods such as multivariate linear regression and autoregressive sliding average model have low prediction accuracy and weak robustness.

[0003] In traditional schemes, most of them adopt the post-flight diagnosis method, which mainly analyzes the gas path parameters, vibration parameters and oil parameters of the aircraft engine. The potential risk status data of the aircraft engine is found through abnormal parameters, and then it is judged whether there are abnormalities in the status of the units and subsystems related to the gas path in the aircraft engine as the basis for fault diagnosis. There is a delay in the fault diagnosis of the aircraft engine after the flight, and it is impossible to predict the over-limit state of the aircraft engine, which leads to untimely maintenance of the aircraft engine and great safety hazards. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present invention provides a performance testing method, device, medium and electronic equipment for an aircraft engine.

[0005] According to a first aspect of an embodiment of the present invention, a performance test method for an aircraft engine is provided, the test method comprising:

[0006] Acquiring a first performance parameter of an aircraft engine, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine;

[0007] dividing the first performance parameter into a plurality of flight states according to fuzzy reasoning;

[0008] Inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, wherein the extreme gradient model is trained by constructing a tree using a minimum loss function and a regularization function to perform residual fitting;

[0009] The performance of the aircraft engine is tested according to the second performance parameter to obtain a performance test result.

[0010] Optionally, the extreme gradient model is trained in the following manner:

[0011] Input the sample first parameter into the initial extreme gradient model;

[0012] According to the first parameter of the sample, an initial function is determined by minimizing a preset loss function and a regularization function;

[0013] According to preset decision tree parameters, transform the loss function in the initial function to obtain an objective function;

[0014] Determining a leaf weight function according to a sum of a first-order derivative and a sum of a second-order derivative of a first parameter of the sample;

[0015] Determining an optimal solution of the objective function according to the leaf weight function and the objective function;

[0016] The extreme gradient model is updated according to the optimal solution.

[0017] Optionally, transforming the loss function in the initial function according to preset decision tree parameters to obtain an objective function includes:

[0018] Performing a second-order Taylor expansion on the loss function in the initial function to obtain a simplified initial function;

[0019] Substituting the preset decision tree parameters into the simplified initial function, the objective function is obtained.

[0020] Optionally, dividing the first performance parameter into a plurality of flight states according to fuzzy reasoning includes:

[0021] Determining a target parameter according to the first performance parameter;

[0022] Based on the domain of the target parameter, the target parameter is fuzzified by a membership function to obtain a plurality of fuzzy sets;

[0023] Perform fuzzy reasoning on the multiple fuzzy sets according to preset fuzzy rules to obtain fuzzy conclusions;

[0024] The fuzzy conclusion is defuzzified by the maximum membership method to obtain a plurality of flight states.

[0025] Optionally, before inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, the method further comprises:

[0026] Decomposing the first performance parameter by wavelet transform to obtain a set of wavelet decomposition coefficients;

[0027] Determine the noise sub-coefficient and the signal sub-coefficient in the wavelet decomposition coefficient according to a preset threshold;

[0028] Processing the wavelet decomposition coefficients by setting the noise-like sub-coefficients to zero to obtain processed wavelet decomposition coefficients;

[0029] According to the processed wavelet decomposition coefficients, a first performance parameter after denoising is constructed.

[0030] Optionally, according to a preset threshold, determining the noise sub-coefficients and the signal sub-coefficients in the wavelet decomposition coefficients includes:

[0031] Determine the sub-coefficients in the wavelet decomposition coefficients that are greater than or equal to the preset threshold as the signal-type sub-coefficients;

[0032] The sub-coefficients of the wavelet decomposition coefficients that are smaller than the preset threshold are determined as the noise-type sub-coefficients.

[0033] Optionally, testing the performance of the aircraft engine according to the second performance parameter to obtain a performance test result includes:

[0034] constructing a performance change curve of the aircraft engine according to the second performance parameter;

[0035] According to the performance change curve, the performance of the aircraft engine is tested to obtain a performance test result.

[0036] According to a second aspect of an embodiment of the present invention, there is provided a performance testing device for an aircraft engine, comprising:

[0037] An acquisition module is configured to acquire a first performance parameter of the aircraft engine, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine;

[0038] a division module, configured to divide the first performance parameter into a plurality of flight states according to fuzzy reasoning;

[0039] A function module is configured to input the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, wherein the extreme gradient model is trained by constructing a tree through a minimum loss function and a regular function to perform residual fitting;

[0040] The determination module is configured to detect the performance of the aircraft engine according to the second performance parameter to obtain a performance detection result.

[0041] According to a third aspect of an embodiment of the present invention, there is provided a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the performance testing method for an aircraft engine provided by the first aspect of the present invention.

[0042] According to a fourth aspect of an embodiment of the present invention, there is provided an electronic device, including:

[0043] a memory having a computer program stored thereon;

[0044] A processor is used to execute the computer program in the memory to implement the steps of the method described in the first aspect of an embodiment of the present invention.

[0045] The technical solution provided by the embodiments of the present invention may have the following beneficial effects:

[0046] By confirming the flight status through fuzzy reasoning, the impact of misjudgment of the flight status on the performance analysis of the aircraft is reduced. Furthermore, by analyzing the first performance parameters under various flight states based on the extreme gradient model, the stability and accuracy of the performance test results are effectively improved. The performance changes of the aircraft engine are predicted based on the performance test results, which improves the safety of the aircraft, plays a key role in the health management of the engine, and effectively reduces the maintenance cost of the aircraft engine.

[0047] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0049] Figure 1 The present invention is a flow chart of a method for testing performance of an aircraft engine according to an exemplary embodiment.

[0050] Figure 2 The figure is a flowchart of a method for training an extreme gradient model according to an exemplary embodiment.

[0051] Figure 3 An implementation according to an exemplary embodiment is shown Figure 2 Flowchart of step S23 in FIG.

[0052] Figure 4 An implementation according to an exemplary embodiment is shown Figure 1 Flowchart of step S12 in FIG.

[0053] Figure 5 An implementation according to an exemplary embodiment is shown Figure 1 Flowchart of step S14 in FIG.

[0054] Figure 6 The present invention is a block diagram of a performance testing device for an aircraft engine according to an exemplary embodiment.

[0055] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Instead, they are merely examples of devices and methods consistent with some aspects of the present invention as detailed in the appended claims.

[0057] It should be noted that all actions of acquiring signals, information or data in the present invention are performed in compliance with the corresponding data protection laws and policies of the country where the device is located and with the authorization given by the owner of the corresponding device.

[0058] Figure 1 is a flow chart of a method for testing performance of an aircraft engine according to an exemplary embodiment. Figure 1 As shown, the following steps are included.

[0059] In step S11, a first performance parameter of the aircraft engine is obtained, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine.

[0060] In a specific implementation, a first performance parameter of an aircraft engine is obtained through a QAR (Quick Access Recorder), wherein the frequency of QAR acquiring data is 1 Hz, and the first performance parameter is the value of flight parameters of multiple aircraft at different times.

[0061] It is worth noting that the aircraft engine is a nonlinear system, and the first performance parameter includes at least one of the following: throttle lever position (TLA), Mach number (Mach), aircraft total air temperature (TAT), aircraft weight (GW), engine speed, and flight altitude (ALT).

[0062] Among them, the throttle lever position is the control input of the aircraft, and the Mach number, the total temperature of the atmosphere outside the aircraft, the aircraft weight and the flight altitude are the state parameters of the aircraft.

[0063] In step S12, the first performance parameter is divided into a plurality of flight states according to fuzzy reasoning.

[0064] Optionally, in order to meet the propulsion requirements of the aircraft in different flight states, the automatic thrust system of the aircraft will adopt different thrust control modes, wherein the change law of thrust is different in different thrust control modes. For example, civil aviation engines include the following constant thrust modes: takeoff thrust, maximum continuous thrust, flexible takeoff thrust, maximum climb thrust and slow-speed thrust. Similarly, the aircraft can also control the thrust through the throttle system. In the constant thrust mode, according to different flight states, the aircraft engine has different engine speeds and loads, so that the first performance parameter has different change laws and functional relationships. In view of this, the first performance parameter can be divided into multiple flight states before modeling and analysis to avoid the influence of different flight states on engine performance detection.

[0065] The flight state of the aircraft includes at least one of the following: a taxiing (T) state, a stop (S) state, a climb-and-takeoff (CLB) state, a step-and-level (LEV) state, a cruise (CRU) state, and a descending (DES) state.

[0066] It is worth noting that the aircraft uses a gradient crawling method in the ascent and descent stages, that is, the process of the aircraft's ascent and descent is not continuous. The step-level flight stage in the aircraft's gradient ascent or gradient descent can be included in the flight state. When the aircraft's flight state is changing, since there is no specific boundary judgment in the flight state, if the changing flight state is regarded as a flight state, serious errors will occur. In this case, judging the aircraft's flight state through fuzzy reasoning can effectively reduce the error in judging the flight state. For example, when the aircraft is in cruise mode, the aircraft will frequently change its altitude. The aircraft's climbing and descending correspond to different flight states. Fuzzy reasoning can be used to identify the flight stage to improve the accuracy of the detection results.

[0067] In step S13, the first performance parameter of the same flight state is input into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine. The extreme gradient model is trained by constructing a tree using a minimum loss function and a regularization function for residual fitting.

[0068] It is worth noting that the extreme gradient model fits the residual of the previous tree by continuously generating trees, thereby determining the law of how the second performance parameter used to characterize the performance of the aircraft engine changes with the first performance parameter under different flight conditions of the aircraft.

[0069] The second performance parameter includes at least one of the following: N2 (high-pressure rotor speed), N1 (fan speed), FF (fuel flow), and EGT (exhaust gas temperature).

[0070] In step S14, the performance of the aircraft engine is tested according to the second performance parameter to obtain a performance test result.

[0071] The changing pattern of the aircraft engine performance is determined based on the second performance parameter. The changing pattern of the engine performance can objectively reflect the performance degradation of the aircraft engine. It is easy to understand that by detecting whether the performance of the aircraft engine is degraded, it can be determined whether the performance test results of the aircraft engine are normal.

[0072] Based on the above scheme, the flight status is confirmed through fuzzy reasoning, which reduces the impact of misjudgment of the flight status on the performance analysis of the aircraft. Furthermore, by analyzing the first performance parameters under various flight states based on the extreme gradient model, the stability and accuracy of the performance test results are effectively improved. The performance changes of the aircraft engine are predicted according to the performance test results, which improves the safety of the aircraft, plays a key role in the health management of the engine, and effectively reduces the maintenance cost of the aircraft engine.

[0073] Figure 2 is a flowchart of a training method for an extreme gradient model according to an exemplary embodiment. Figure 2 As shown, the extreme gradient model is trained in the following way:

[0074] S21, inputting the first parameter of the sample into the initial extreme gradient model.

[0075] It is worth noting that the first performance parameter of the same flight state is divided into a training set, a validation set and a test set according to a preset ratio. The initial extreme gradient model is trained using the training set. The initial extreme gradient model is evaluated using the validation set through cross-validation to obtain the target extreme gradient model. The test set is then input into the target extreme gradient model to verify the fitting accuracy. If the accuracy target meets the preset error requirement, the extreme gradient model training is completed. If the accuracy target does not meet the preset error requirement, the first performance parameter is re-divided according to the predicted ratio for training. In one embodiment, the preset ratio is 7 to 3 to 2.

[0076] In one embodiment, the basis of the initial gradient model is the following weak classifier combination function:

[0077] in: is the model prediction value of the i-th sample, K is the number of trees, and f k is the kth tree, x i is the i-th sample of the input.

[0078] S22, determining an initial function by minimizing a preset loss function and a regularization function according to the first parameter of the sample;

[0079] Furthermore, the extreme gradient model minimizes the loss function and the regular function Ω(f k ) to find the best initial function, the initial function is as follows:

[0080]

[0081] Among them, the loss function For the predicted value and the true value y i The error, regular function Ω(f k ) represents the complexity of each tree, T is the number of leaf nodes, w is the weight of the leaf node, γ is the complexity cost of the new leaf node, and λ can control the weight of the leaf node not to be too large to prevent overfitting.

[0082] S23, transforming the loss function in the initial function according to preset decision tree parameters to obtain an objective function;

[0083] S24, determining a leaf weight function according to the sum of the first-order derivatives and the sum of the second-order derivatives of the first parameter of the sample;

[0084] In one embodiment, the sum of the first-order derivatives is The sum of the second-order derivatives is Leaf weight function As shown below:

[0085] S25, determining an optimal solution of the objective function according to the leaf weight function and the objective function;

[0086] In one implementation, the leaf weight function is substituted into the objective function to obtain the minimum value of the objective function. The smaller the minimum value of the objective function, the better the tree structure and the better the optimal solution of the objective function. As shown below:

[0087] S26. Update the extreme gradient model according to the optimal solution.

[0088] Figure 3 An implementation according to an exemplary embodiment is shown Figure 2 Flow chart of step S23 in FIG. Figure 3 As shown, the following steps are included:

[0089] S231. Perform a second-order Taylor expansion on the loss function in the initial function to obtain a simplified initial function.

[0090] Specifically, it can be seen from the initial function that the prediction result of the tth round is the sum of the prediction result of the t-1th round and the decision tree function generated by the tth iteration. The simplified initial function is as follows: (t) =

[0091] in, is the first-order derivative of the loss function, is the second-order derivative of the loss function. is a constant term and does not affect the optimal solution.

[0092] S232: Substitute the preset decision tree parameters into the simplified initial function to obtain the objective function.

[0093] In one implementation, the sample set of the jth leaf is defined as follows: j ={i|q(x t )=j}, the objective function becomes:

[0094] Figure 4 An implementation according to an exemplary embodiment is shown Figure 1 The flowchart of step S12 in FIG. 1 comprises the following steps:

[0095] S121. Determine a target parameter according to the first performance parameter.

[0096] In one embodiment, based on the first performance parameter, target parameters are obtained from the onboard recorder through data decoding software, wherein the target parameters include at least one of the following: flight altitude, aircraft engine temperature, pressure altitude, vertical speed, airspeed and Mach number.

[0097] S122. Based on the domain of the target parameter, the target parameter is fuzzified by a membership function to obtain a plurality of fuzzy sets.

[0098] In one embodiment, the vertical velocity is fuzzified using a Z-shaped membership function to obtain three fuzzy sets of negative (NV), zero (ZV), and positive (PV). The halo value range corresponding to the vertical velocity is [-5000, 5000] ft / min, and the membership value range is [0, 1].

[0099] Optionally, the membership function may also be an S-shaped membership function or a triangular membership function.

[0100] In another example, the pressure altitude is fuzzified using an S-shaped membership function, and the pressure altitude is divided into three fuzzy sets: low (LH), medium (MH), and high (HH), and the domain value range is [-2000,40000]ft;

[0101] As another example, the airspeed is fuzzified using a triangular membership function, and the airspeed is divided into three fuzzy sets: low (LV), medium (MV), and high (HV), and the domain value range is [0,400] knots.

[0102] S123. Perform fuzzy reasoning on the multiple fuzzy sets according to preset fuzzy rules to obtain fuzzy conclusions.

[0103] It is worth noting that, according to the characteristics of the target parameters, the fuzzy rules are braked in the form of if-then and stored in the fuzzy rule base in advance. According to the fuzzy rule base, the Mamdani fuzzy reasoning method is used for reasoning to obtain the fuzzy conclusion.

[0104] In a specific implementation, the fuzzy rules are as follows:

[0105] Rule 1: When the pressure altitude is LH, the airspeed is LV, and the vertical speed is ZV, the flight status is T or S.

[0106] Rule 2: When the pressure altitude is MH, the airspeed is MV, and the vertical speed is PV, the flight state is CLB.

[0107] Rule 3: When the pressure altitude is HH, the airspeed is HV, and the vertical speed is ZV, the flight status is CRU.

[0108] Rule 4: When the pressure altitude is MH, the airspeed is MV, and the vertical speed is NV, the flight state is DES.

[0109] Rule 5: When the pressure altitude is MH, the airspeed is MV, and the vertical speed is ZV, the flight state is LEV.

[0110] Rule 6: When the pressure altitude is HH, the airspeed is HV, and the vertical speed is PV, the flight state is CLB.

[0111] Rule 7: When the pressure altitude is HH, the airspeed is MV, and the vertical speed is NV, the flight state is DES.

[0112] S124. Defuzzify the fuzzy conclusion by using the maximum membership method to obtain a plurality of flight states.

[0113] It is worth mentioning that the defuzzification process is to convert the fuzzy conclusion into a specific output through the membership method. Since there may be one or more fuzzy subsets in a fuzzy set, there may be one or more points on the membership curve that take the maximum value. In view of this, the maximum membership method is used to defuzzify the fuzzy conclusion and obtain a collection of multiple maximum values.

[0114] Optionally, before inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, the method further comprises:

[0115] The first performance parameter is decomposed by wavelet transform to obtain a set of wavelet decomposition coefficients.

[0116] According to a preset threshold, the noise sub-coefficient and the signal sub-coefficient in the wavelet decomposition coefficient are determined.

[0117] The wavelet decomposition coefficients are processed by setting the noise-like sub-coefficients to zero, thereby obtaining processed wavelet decomposition coefficients.

[0118] According to the processed wavelet decomposition coefficients, a first performance parameter after denoising is constructed.

[0119] It is worth noting that the first performance parameters collected by QAR have many wild values ​​and jitters due to the complex and changeable flight environment. Direct use will affect the accuracy of the data. Therefore, before the first performance parameters are input into the extreme gradient model, the first performance parameters are processed by the wavelet threshold method to improve the accuracy of the data.

[0120] Optionally, according to a preset threshold, determining the noise sub-coefficients and the signal sub-coefficients in the wavelet decomposition coefficients includes:

[0121] The sub-coefficients in the wavelet decomposition coefficients that are greater than or equal to the preset threshold are determined as the signal-type sub-coefficients.

[0122] The sub-coefficients of the wavelet decomposition coefficients that are smaller than the preset threshold are determined as the noise-type sub-coefficients.

[0123] Figure 5 An implementation according to an exemplary embodiment is shown Figure 1 The flowchart of step S14 in FIG. 1 comprises the following steps:

[0124] S141. Construct a performance change curve of the aircraft engine according to the second performance parameter.

[0125] S142. According to the performance change curve, the performance of the aircraft engine is tested to obtain a performance test result.

[0126] For example, in the process of an aircraft climbing from an altitude of 10,000 feet to an altitude of 25,000 feet, the predicted performance change data of the aircraft engine during the climb can be obtained through the performance change curve, and the actual aircraft engine performance change data is collected. According to the actual performance change data and the predicted performance change data, the detection results are obtained. Through the above monitoring results, the failure of the aircraft engine is predicted, so as to achieve early maintenance or take corresponding emergency measures.

[0127] Figure 6 FIG. 1 is a block diagram of a performance testing device 1200 for an aircraft engine according to an exemplary embodiment. Figure 6 The device includes an acquisition module 1201, a division module 1202, a function module 1203 and a determination module 1204.

[0128] The acquisition module 1201 is configured to acquire a first performance parameter of an aircraft engine, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine.

[0129] The division module 1202 is configured to divide the first performance parameter into multiple flight states based on fuzzy reasoning.

[0130] Function module 1203 is configured to input the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, wherein the extreme gradient model is trained by constructing a tree using a minimum loss function and a regular function for residual fitting.

[0131] The determination module 1204 is configured to detect the performance of the aircraft engine according to the second performance parameter to obtain a performance detection result.

[0132] Optionally, the model submodule in the function module is configured as follows:

[0133] According to the first parameter of the sample, an initial function is determined by minimizing a preset loss function and a regularization function.

[0134] According to preset decision tree parameters, the loss function in the initial function is transformed to obtain the objective function.

[0135] A leaf weight function is determined according to the sum of the first-order derivatives and the sum of the second-order derivatives of the first parameter of the sample.

[0136] Determining an optimal solution of the objective function according to the leaf weight function and the objective function;

[0137] The extreme gradient model is updated according to the optimal solution.

[0138] Optionally, the units in the model submodule are configured as:

[0139] Perform a second-order Taylor expansion on the loss function in the initial function to obtain a simplified initial function.

[0140] Substituting the preset decision tree parameters into the simplified initial function, the objective function is obtained.

[0141] Optionally, the division module 1202 includes:

[0142] The first determination submodule is configured to determine a target parameter according to the first performance parameter.

[0143] The fuzzification submodule is configured to fuzzify the target parameter through a membership function based on the domain of the target parameter to obtain multiple fuzzy sets.

[0144] The reasoning submodule is configured to perform fuzzy reasoning on the multiple fuzzy sets according to preset fuzzy rules to obtain fuzzy conclusions.

[0145] The defuzzification submodule is configured to defuzzify the fuzzy conclusion by using the maximum membership method to obtain a plurality of the flight states.

[0146] Optionally, the acquisition module 1201 further includes:

[0147] The decomposition submodule is configured to decompose the first performance parameter through wavelet transform to obtain a set of wavelet decomposition coefficients.

[0148] The second determination submodule is configured to determine the noise-like sub-coefficients and the signal-like sub-coefficients in the wavelet decomposition coefficients according to a preset threshold.

[0149] The processing submodule is configured to process the wavelet decomposition coefficients by setting the noise-like sub-coefficients to zero, so as to obtain processed wavelet decomposition coefficients.

[0150] The first construction submodule is configured to construct a first performance parameter after denoising according to the processed wavelet decomposition coefficients.

[0151] Optionally, the second determining submodule includes:

[0152] A first determining unit is configured to determine a sub-coefficient greater than or equal to the preset threshold in the wavelet decomposition coefficient as the signal-type sub-coefficient;

[0153] The second determining unit is configured to determine the sub-coefficients smaller than the preset threshold in the wavelet decomposition coefficients as the noise-type sub-coefficients.

[0154] Optionally, the determining module 1204 includes:

[0155] The second construction submodule is configured to construct a performance change curve of the aircraft engine according to the second performance parameter.

[0156] The detection submodule is configured to detect the performance of the aircraft engine according to the performance change curve to obtain a performance detection result.

[0157] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0158] The present invention also provides a computer-readable storage medium on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the performance testing method of the aircraft engine provided by the present invention are implemented.

[0159] The present invention also provides an electronic device, comprising:

[0160] a memory having a computer program stored thereon;

[0161] A processor is used to execute the computer program in the memory to implement the steps of the performance testing method of the aircraft engine.

[0162] Figure 7 FIG. 7 is a block diagram of an electronic device 700 according to an exemplary embodiment. Figure 7 As shown, the electronic device 700 may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0163] The processor 701 is used to control the overall operation of the electronic device 700 to complete all or part of the steps in the above-mentioned performance test method of the aircraft engine. The memory 702 is used to store various types of data to support the operation of the electronic device 700, and these data may include, for example, instructions for any application or method used to operate on the electronic device 700, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 702 or sent through the communication component 705. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 704 provides an interface between the processor 701 and other interface modules, and the above-mentioned other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 705 may include: Wi-Fi module, Bluetooth module, NFC module, etc.

[0164] In an exemplary embodiment, the electronic device 700 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned aircraft engine performance test method.

[0165] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned performance test method of the aircraft engine are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 702 including program instructions, and the above-mentioned program instructions can be executed by the processor 701 of the electronic device 700 to complete the above-mentioned performance test method of the aircraft engine.

[0166] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the present invention. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present invention is indicated by the following claims.

[0167] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for testing the performance of an aircraft engine, characterized in that: The test method includes: Acquiring a first performance parameter of an aircraft engine, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine; dividing the first performance parameter into a plurality of flight states according to fuzzy reasoning; Inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, wherein the extreme gradient model is trained by constructing a tree using a minimum loss function and a regularization function to perform residual fitting; Testing the performance of the aircraft engine according to the second performance parameter to obtain a performance test result; The extreme gradient model is trained in the following way: Input the sample first parameter into the initial extreme gradient model; According to the first parameter of the sample, an initial function is determined by minimizing a preset loss function and a regularization function; According to preset decision tree parameters, the loss function in the initial function is transformed to obtain an objective function; Determining a leaf weight function according to a sum of a first-order derivative and a sum of a second-order derivative of a first parameter of the sample; Determining an optimal solution of the objective function according to the leaf weight function and the objective function; updating the extreme gradient model according to the optimal solution; The step of transforming the loss function in the initial function according to the preset decision tree parameters to obtain the objective function includes: Performing a second-order Taylor expansion on the loss function in the initial function to obtain a simplified initial function; Substituting the preset decision tree parameters into the simplified initial function, the objective function is obtained.

2. The testing method according to claim 1, characterized in that: The dividing the first performance parameter into a plurality of flight states according to fuzzy reasoning comprises: Determining a target parameter according to the first performance parameter; Based on the domain of the target parameter, the target parameter is fuzzified by a membership function to obtain a plurality of fuzzy sets; Perform fuzzy reasoning on the multiple fuzzy sets according to preset fuzzy rules to obtain fuzzy conclusions; The fuzzy conclusion is defuzzified by the maximum membership method to obtain a plurality of flight states.

3. The testing method according to claim 1, characterized in that: Before inputting the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, the method includes: Decomposing the first performance parameter by wavelet transform to obtain a set of wavelet decomposition coefficients; Determine the noise sub-coefficient and the signal sub-coefficient in the wavelet decomposition coefficient according to a preset threshold; Processing the wavelet decomposition coefficients by setting the noise-like sub-coefficients to zero to obtain processed wavelet decomposition coefficients; According to the processed wavelet decomposition coefficients, a first performance parameter after denoising is constructed.

4. The testing method according to claim 3, characterized in that: According to a preset threshold, determining the noise sub-coefficients and the signal sub-coefficients in the wavelet decomposition coefficients includes: Determine the sub-coefficients in the wavelet decomposition coefficients that are greater than or equal to the preset threshold as the signal-type sub-coefficients; The sub-coefficients of the wavelet decomposition coefficients that are smaller than the preset threshold are determined as the noise-type sub-coefficients.

5. The testing method according to any one of claims 1 to 4, characterized in that: The step of testing the performance of the aircraft engine according to the second performance parameter to obtain a performance test result includes: constructing a performance change curve of the aircraft engine according to the second performance parameter; According to the performance change curve, the performance of the aircraft engine is tested to obtain a performance test result.

6. A performance test device for an aircraft engine, characterized in that: include: An acquisition module is configured to acquire a first performance parameter of the aircraft engine, where the first performance parameter is a flight parameter that affects the performance of the aircraft engine; a division module, configured to divide the first performance parameter into a plurality of flight states according to fuzzy reasoning; A function module is configured to input the first performance parameter of the same flight state into a pre-trained extreme gradient model to obtain a second performance parameter for characterizing the performance of the aircraft engine, wherein the extreme gradient model is trained by constructing a tree through a minimum loss function and a regular function to perform residual fitting; a determination module configured to detect the performance of the aircraft engine according to the second performance parameter to obtain a performance detection result; The extreme gradient model is trained in the following way: Input the sample first parameter into the initial extreme gradient model; According to the first parameter of the sample, an initial function is determined by minimizing a preset loss function and a regularization function; According to preset decision tree parameters, the loss function in the initial function is transformed to obtain an objective function; Determining a leaf weight function according to a sum of a first-order derivative and a sum of a second-order derivative of a first parameter of the sample; Determining an optimal solution of the objective function according to the leaf weight function and the objective function; updating the extreme gradient model according to the optimal solution; The step of transforming the loss function in the initial function according to the preset decision tree parameters to obtain the objective function includes: Performing a second-order Taylor expansion on the loss function in the initial function to obtain a simplified initial function; Substituting the preset decision tree parameters into the simplified initial function, the objective function is obtained.

7. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 5 are implemented.

8. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 5.

Images