Aero-engine main bearing kinetic model correction method based on sensitive feature fusion

Abstract

The invention relates to the technical field of modeling, in particular to an aero-engine main bearing kinetic model correction method based on sensitive feature fusion, which comprises the following steps: constructing an aero-engine main bearing kinetic model; simulation data are generated according to the collected actual measurement data, time-frequency domain features of the actual measurement data and the simulation data are extracted, and an initial feature set is constructed; based on a binary improved horse swarm optimization algorithm, sensitive features are screened out from the initial feature set, a sensitive feature set is constructed, a component matrix coefficient of the sensitive feature set is calculated in combination with a principal component analysis method, and a fused sensitive feature expression is obtained; calculating fusion sensitive features of the measured data and the simulation data by using the fusion feature expression, and taking a difference calculation formula of the two as a target function for model correction; and performing optimization based on a horse group optimization algorithm and the target function, updating model parameters of the aeroengine main bearing dynamic model, and realizing correction of the aeroengine main bearing dynamic model.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of aircraft engine main bearing modeling, and in particular to a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion. Background Art

[0002] As a core component of aircraft engines, aircraft engine main bearings play a vital role in ensuring stable engine operation and power transmission. Widely used in engines of various aircraft, they are a crucial foundation for safe and efficient flight. Because aircraft engines operate in extremely complex and harsh environments, such as high temperatures, high pressures, high speeds, and intense vibrations, aircraft engine main bearings are highly susceptible to various types of failures, including pitting, spalling, and wear. These failures severely impact the normal operation of mechanical equipment and can result in significant safety incidents and economic losses.

[0003] To address this challenge, it is necessary to establish an accurate dynamic model of aero-engine main bearings and study their vibration characteristics when considering inner and outer race faults, thus providing an important theoretical basis for fault diagnosis of aero-engine main bearings. In recent years, numerous researchers have studied the dynamic response of bearings with single faults in the inner or outer races, elucidating the vibration mechanisms of faulty bearings from multiple perspectives and achieving fruitful results. Focusing on rolling bearings, some studies have established a four-degree-of-freedom dynamic model of a faulty ball bearing to explain the coupling mechanism of vibration signals during faults. Others have established a dynamic model of a faulty ball bearing based on Hertz contact theory and the law of conservation of energy, deriving energy equations for each bearing component and the impact excitation equation between the rolling element and the fault region, and simulating its vibration response. Furthermore, some have established a six-degree-of-freedom dynamic simulation model of a bearing with localized surface defects by considering high-speed effects, gyroscopic torque, and centrifugal forces.

[0004] However, the aforementioned dynamic models all overlook the impact of modeling parameters on the dynamic model. Specifically, the modeling parameters are derived from theoretical calculations and empirical estimates, without refinement using measured data. Consequently, they fail to fully reflect the real-world situation, leading to significant discrepancies between simulated and measured data and reducing their accuracy. To improve the consistency between simulated and measured data, finding, designing, and applying a correction method for the dynamic model of aircraft engine main bearings is crucial for aircraft engine health monitoring. Summary of the Invention

[0005] In view of this, an embodiment of the present application proposes a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion, which aims to scientifically correct the dynamic model of an aircraft engine main bearing so that the simulation data output by the corrected dynamic model of an aircraft engine main bearing is highly consistent with the measured data, providing an effective solution to the data imbalance and small sample problems in fault diagnosis applications.

[0006] In order to achieve the above-mentioned purpose, an embodiment of the present application proposes a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion, the method comprising the following steps: constructing an aircraft engine main bearing dynamic model; generating simulation data based on the collected measured data, and extracting the time-frequency domain features of the measured data and the simulation data to construct an initial feature set; based on the binary improved Horse Swarm Optimization algorithm, screening out sensitive features from the initial feature set, constructing a sensitive feature set, and combining the principal component analysis method to calculate the component matrix coefficients of the sensitive feature set, thereby obtaining a fused sensitive feature expression; using the fused feature expression to calculate the fused sensitive features of the measured data and the simulation data, and using the difference calculation formula between the two as the objective function of the model correction; optimizing based on the Horse Swarm Optimization algorithm and the objective function, updating the model parameters of the aircraft engine main bearing dynamic model, and realizing the correction of the aircraft engine main bearing dynamic model.

[0007] In order to achieve the above-mentioned purpose, an embodiment of the present application also proposes an aircraft engine main bearing dynamics model correction system based on sensitive feature fusion, the system comprising: a model construction module for constructing an aircraft engine main bearing dynamics model; an initial feature set construction module for generating simulation data based on the collected measured data, and extracting the time-frequency domain features of the measured data and the simulation data to construct an initial feature set; a sensitive feature set construction module for screening out sensitive features from the initial feature set based on the binary improved Horseshoe Optimization algorithm, constructing a sensitive feature set, and combining the principal component analysis method to calculate the component matrix coefficients of the sensitive feature set, thereby obtaining a fused sensitive feature expression; an objective function establishment module for calculating the fused sensitive features of the measured data and the simulation data using the fused feature expression, and using the difference calculation formula between the two as the objective function of the model correction; an optimization module for optimizing based on the Horseshoe Optimization algorithm and the objective function, updating the model parameters of the aircraft engine main bearing dynamics model, and realizing the correction of the aircraft engine main bearing dynamics model.

[0008] In order to achieve the above-mentioned purpose, an embodiment of the present application also proposes an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion.

[0009] In order to achieve the above-mentioned purpose, an embodiment of the present application further proposes a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement the above-mentioned method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion.

[0010] This application proposes a method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion. First, the dynamic model of the aero-engine main bearing is constructed. Then, simulation data is generated based on the collected measured data, and the time-frequency domain features of the measured data and the simulation data are extracted to construct an initial feature set. Then, based on the binary improved horse swarm optimization algorithm, sensitive features are screened out from the initial feature set to construct a sensitive feature set. The binary improved horse swarm optimization algorithm can effectively screen out sensitive features in the initial feature set, which lays a solid foundation for subsequent model correction. Next, combined with the principal component analysis method, the component matrix coefficients of the sensitive feature set are calculated to obtain a fused sensitive feature expression. The fused feature expression is then used to calculate the fused sensitive features of the measured data and the simulation data, and the difference calculation formula between the two is used as the objective function of the model correction. Finally, based on the horse swarm optimization algorithm and the objective function, optimization is performed to update the model parameters of the aero-engine main bearing dynamic model to achieve correction of the aero-engine main bearing dynamic model. The simulation data output by the corrected aero-engine main bearing dynamic model is highly consistent with the measured data, which provides an effective solution to the data imbalance and small sample problems in fault diagnosis applications.

[0011] Optionally, constructing a dynamic model of an aircraft engine main bearing includes: The aircraft engine main bearing is considered as a faulty bearing, and a dynamic model is established that takes into account the fault size, speed, load, and fault type of the faulty bearing. The dynamic model mainly consists of an inner ring, rolling elements, outer ring, and cage. The inner ring is assumed to rotate with the rotating shaft, and the outer ring is fixed to the bearing seat. The dynamic model captures the axial and radial motion of the inner and outer rings. During the simulation process, the stiffness and damping of each component are assumed to be fixed values. The operating parameters of the dynamic model are consistent with those of the real bearing. According to Newton's second law, the centrifugal force caused by installation and manufacturing deviation is considered and the dynamic model is established. The four differential equations of motion related to the inner and outer rings in the plane are: ; in, Indicates the mass of the bearing inner ring, represents the damping coefficient of the bearing inner ring, Indicates the stiffness of the bearing inner ring, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, Indicates that the external load is The directional component of force, is the eccentricity, is the shaft speed, For the moment, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, represents the acceleration due to gravity, Indicates that the external load is The directional component of force, Indicates the mass of the bearing outer ring, represents the damping coefficient of the bearing outer ring, Indicates the stiffness of the bearing outer ring, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction.

[0012] Optionally, after establishing and obtaining the four motion differential equations of the dynamic model related to the inner ring and the outer ring in the plane, the method further includes: The fourth-order Runge-Kutta method is used to solve the four motion differential equations related to the inner and outer rings in the plane of the dynamic model to obtain the vibration responses of the rolling elements, inner and outer rings. The ode45 solver is then used to calculate the dynamic response of the faulty bearing. When the rolling element passes through a defect, it will produce a certain contact deformation. The contact deformation formula of the rolling element is as follows: ; in, Indicates the The angular position of the rolling element, Indicates the radial clearance of the bearing, Indicates the deformation release caused by local defects, Indicates the The contact deformation of the rolling elements.

[0013] Optionally, extract the time-frequency domain features of the measured data and the simulated data to construct an initial feature set, including: Extract 25 time-frequency domain features of measured data and simulation data to construct the initial feature set; Among them, the 25 time-frequency domain features extracted include: maximum value, minimum value, mean value, median, peak value, peak-to-peak value, rectified mean value, variance, standard deviation, kurtosis, skewness, root mean square, form factor, peak factor, impulse factor, root mean square amplitude, margin factor, kurtosis factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance, frequency standard deviation, spectral entropy and signal complexity.

[0014] Optionally, based on the binary improved horse swarm optimization algorithm, sensitive features are screened out from the initial feature set, including: Based on the binary improved horse herd optimization algorithm, a transfer function is used to convert the velocity value into a probability value to specify the probability of updating the binary solution, thereby screening out sensitive features from the initial feature set; The transfer function used is the V-type transfer function, and its calculation formula is: ; in, Indicates in In the iteration The horse in the The speed of the dimension, represents a V-type transfer function; The horse's position based on its speed is updated using the following formula: ; in, Indicates in In the iteration The horse in the The position in the dimension, Indicates the updated The horse in the Position in the dimension.

[0015] Optionally, the principal component analysis method is combined to calculate the component matrix coefficients of the sensitive feature set, thereby obtaining the fused sensitive feature expression, including: The sensitive feature set is normalized using the following formula: ; in, represents the sensitive feature set, represents the mean of the sensitive feature set, represents the standard deviation of the sensitive feature set, represents the normalized sensitive feature set; The covariance matrix is ​​calculated using the following formula: ; in, represents the covariance matrix, Indicates the total number of vibration signals; The eigenvalues ​​and eigenvectors are calculated using the following formula: ; ; in, Indicates the eigenvalue matrix, Indicates the eigenvector matrix; The largest eigenvalue matrix Convert to a diagonal matrix and combine the largest eigenvector matrix , the factor loading coefficient is calculated by the following formula: ; in, represents the calculated factor loading coefficient, Indicates finding the diagonal matrix; Based on the following formula, and , calculate the component matrix coefficients: ; in, Represents the calculated component matrix coefficients.

[0016] Optionally, the fusion sensitive features of the measured data and the simulation data are calculated using the fusion feature expression, and the difference calculation formula between the two is used as the objective function of the model correction, including: Filter out the component matrix coefficients greater than 0, perform weighted calculation with the sensitive features of the measured data and simulation data, and obtain the fused sensitive features of the measured data and simulation data; The difference between the fusion sensitive features of the measured data and the simulation data is calculated as the objective function of the model correction through the following formula: ; in, represents the objective function of model modification, represents the fusion sensitive features of simulation data, Represents the fused sensitive features of the measured data. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the related technologies, the following is a brief introduction to the drawings required for use in the embodiments of the present application or the description of the related technologies. Obviously, the following drawings are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. The drawings described here are only used to explain the present application and are not used to limit the present application.

[0018] Figure 1 This is a flow chart of a method for correcting a dynamic model of an aircraft engine main bearing based on sensitive feature fusion, provided in one embodiment of the present application; Figure 2 This is a visual schematic diagram of a process for correcting the dynamics model of an aircraft engine main bearing, provided in one embodiment of the present application; Figure 3 A vibration signal comparison diagram is provided in one embodiment of the present application; Figure 4 An envelope signal comparison diagram is provided in one embodiment of the present application; Figure 5 This is a schematic structural diagram of a system for correcting a dynamic model of an aircraft engine main bearing based on sensitive feature fusion, provided in another embodiment of the present application; Figure 6 It is a structural diagram of an electronic device provided in another embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. Those skilled in the art will appreciate that in the embodiments of the present application, many technical details are proposed to enable the reader to better understand. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The following embodiments can be combined with each other and referenced to each other under the premise of no contradiction.

[0020] One embodiment of the present application proposes a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion, which is applied to an electronic device, wherein the electronic device can be a terminal or a server. This embodiment and the following embodiments are all described using a server as an example. The following describes in detail the implementation details of the method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion proposed in this embodiment. The following content is only provided for ease of understanding and is not required for the implementation of this solution.

[0021] The specific process of the method for correcting the dynamic model of the main bearing of an aero-engine based on sensitive feature fusion proposed in this embodiment can be as follows: Figure 1 As shown, the visual details are as follows Figure 2 As shown, the method includes: Step 101: construct a dynamic model of an aircraft engine main bearing.

[0022] In practice, the prerequisite for modifying the dynamic model of an aircraft engine main bearing is to first construct a dynamic model of the aircraft engine main bearing. The purpose of modifying the dynamic model is to monitor the health of the aircraft engine. Therefore, the aircraft engine main bearing must be treated as a faulty bearing during modeling. Leveraging Hertz contact theory, Newton's second law, and Lagrange's equations, a dynamic model is established that takes into account the fault size, speed, load, and fault type of the faulty bearing. The established dynamic model primarily consists of an inner ring, rolling elements, an outer ring, and a cage. The inner ring is assumed to rotate with the axis of rotation, while the outer ring is fixed to the bearing seat. The dynamic model captures the axial and radial motion of the inner and outer rings. Furthermore, during the simulation, the stiffness and damping of each component are assumed to be fixed, ensuring that the operating parameters of the dynamic model remain consistent with those of a real bearing.

[0023] According to Newton's second law, considering the centrifugal force caused by installation and manufacturing deviation, the server needs to establish a dynamic model. The four differential equations of motion related to the inner and outer rings in the plane are: ; in, Indicates the mass of the bearing inner ring, represents the damping coefficient of the bearing inner ring, Indicates the stiffness of the bearing inner ring, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, Indicates that the external load is The directional component of force, is the eccentricity, is the shaft speed, For the moment, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, represents the acceleration due to gravity, Indicates that the external load is The directional component of force, Indicates the mass of the bearing outer ring, represents the damping coefficient of the bearing outer ring, Indicates the stiffness of the bearing outer ring, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction.

[0024] In one example, after completing the construction of the dynamic model of the aircraft engine main bearing, the server also needs to use the fourth-order Runge-Kutta method to solve the four motion differential equations related to the inner and outer rings in the plane of the established dynamic model, obtain the vibration response of the rolling elements, inner and outer rings, and use the ode45 solver to calculate the dynamic response of the faulty bearing.

[0025] In the established dynamic model, a single-point fault of the inner or outer ring of the bearing is introduced, which simplifies the localized fault into a small rectangular spalling and reasonably ignores the complex mechanism of the failure mode difference.

[0026] When the rolling element passes through a defect, it will produce a certain contact deformation. The contact deformation formula of the rolling element is as follows: ; in, Indicates the The angular position of the rolling element, Indicates the radial clearance of the bearing, Indicates the deformation release caused by local defects, Indicates the The contact deformation of the rolling elements.

[0027] Step 102 : Generate simulation data based on the collected measured data, extract time-frequency domain features of the measured data and the simulation data, and construct an initial feature set.

[0028] In the specific implementation, after constructing the dynamic model of the aircraft engine main bearing, the server needs to collect measured data and generate simulation data based on the fault size, speed, load, and fault type of the measured data collected, and extract the time-frequency domain features of the measured data and simulation data to construct an initial feature set.

[0029] In one example, the server needs to extract 25 time-frequency domain features of measured data and simulation data to construct an initial feature set.

[0030] Among them, the 25 time-frequency domain features extracted include: maximum value, minimum value, mean value, median, peak value, peak-to-peak value, rectified mean value, variance, standard deviation, kurtosis, skewness, root mean square, form factor, peak factor, impulse factor, root mean square amplitude, margin factor, kurtosis factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance, frequency standard deviation, spectral entropy and signal complexity.

[0031] Taking kurtosis as an example, kurtosis indicates the sharpness of the vibration signal distribution over a period of time. It is the fourth-order normalized moment of the vibration signal. Its calculation formula is as follows: ; in, Represents the total number of vibration signals, Indicates the A vibration signal, Indicates the The mean value of the vibration signal, Indicates the The standard deviation of the vibration signal, Indicates the The kurtosis of a vibration signal.

[0032] Step 103 , based on the binary improved horse group optimization algorithm, sensitive features are screened out from the initial feature set to construct a sensitive feature set, and the component matrix coefficients of the sensitive feature set are calculated in combination with the principal component analysis method, thereby obtaining a fused sensitive feature expression.

[0033] In the specific implementation, after constructing the initial feature set, the server needs to screen out sensitive features from the initial feature set based on the binary improved Horseshoe Optimization algorithm (BHOA algorithm), construct a sensitive feature set, and combine it with the principal component analysis method (PCA method) to calculate the component matrix coefficients of the sensitive feature set to obtain the fused sensitive feature expression.

[0034] In one example, the traditional HOA algorithm uses a continuous search space during optimization. However, feature selection is a binary optimization problem, where decision variables have different formats and can only take on the values ​​"0" and "1." Therefore, the server uses a binary modified horse herd optimization algorithm, employing a transfer function to convert velocity values ​​into probability values. This specifies the probability of updating the binary solution, thereby filtering out sensitive features from the initial feature set.

[0035] The transfer function used by the server is a V-type transfer function, and its calculation formula is: ; in, Indicates in In the iteration The horse in the The speed of the dimension, represents a V-type transfer function.

[0036] The horse's position based on its speed is updated using the following formula: ; in, Indicates in In the iteration The horse in the The position in the dimension, Indicates the updated The horse in the Position in the dimension.

[0037] In one example, the principal component analysis process is implemented by the following steps.

[0038] First, the sensitive feature set is normalized using the following formula: ; in, represents the sensitive feature set, represents the mean of the sensitive feature set, represents the standard deviation of the sensitive feature set, Represents the normalized sensitive feature set.

[0039] Next, calculate the covariance matrix using the following formula: ; in, represents the covariance matrix, Indicates the total number of vibration signals.

[0040] Then, the eigenvalues ​​and eigenvectors are calculated using the following formula: ; ; in, Indicates the eigenvalue matrix, Indicates the eigenvector matrix.

[0041] Then, the largest eigenvalue matrix Convert to a diagonal matrix and combine the largest eigenvector matrix , the factor loading coefficient is calculated by the following formula: ; in, represents the calculated factor loading coefficient, It means to find the diagonal matrix.

[0042] Finally, through the following formula, based on and , calculate the component matrix coefficients: ; in, Represents the calculated component matrix coefficients.

[0043] Step 104: Calculate the fusion sensitive features of the measured data and the simulation data using the fusion feature expression, and use the difference calculation formula between the two as the objective function of the model correction.

[0044] In a specific implementation, after obtaining the fusion feature expression, the server can use the fusion feature expression to calculate the fusion sensitive features of the measured data and the simulation data, and use the difference calculation formula between the two as the objective function of the model correction.

[0045] In one example, after calculating the component matrix coefficients, the server needs to screen out the component matrix coefficients greater than 0, perform weighted calculations on the sensitive features of the measured data and the simulation data, and obtain the fused sensitive features of the measured data and the simulation data.

[0046] In one example, the server uses the following formula to calculate the difference between the fused sensitive features of the measured data and the simulated data as the objective function for model correction: ; in, represents the objective function of model modification, represents the fusion sensitive features of simulation data, Represents the fused sensitive features of the measured data.

[0047] Step 105 : Optimize based on the Horse Swarm Optimization algorithm and the objective function, update the model parameters of the aircraft engine main bearing dynamics model, and implement the correction of the aircraft engine main bearing dynamics model.

[0048] In the specific implementation, after establishing the objective function, the server can perform optimization based on the Horseshoe Optimization algorithm and the objective function, thereby updating the model parameters of the aircraft engine main bearing dynamics model, and setting the upper and lower limits of the parameters to ensure that the updated parameters are reasonable in the physical sense, and ultimately realize the correction of the aircraft engine main bearing dynamics model.

[0049] This embodiment proposes a method for correcting an aeroengine main bearing dynamics model based on sensitive feature fusion. First, the aeroengine main bearing dynamics model is constructed. Simulation data is then generated based on the collected measured data. Time-frequency domain features of the measured and simulated data are extracted to construct an initial feature set. Sensitive features are then filtered from the initial feature set using a binary improved Horse Swarm Optimization algorithm to construct a sensitive feature set. The binary improved Horse Swarm Optimization algorithm can effectively filter out sensitive features from the initial feature set, laying a solid foundation for subsequent model correction. Next, principal component analysis is combined to calculate the component matrix coefficients of the sensitive feature set to obtain a fused sensitive feature expression. The fused feature expression is then used to calculate the fused sensitive features of the measured and simulated data. The difference between the two is calculated as the objective function for model correction. Finally, optimization is performed based on the Horse Swarm Optimization algorithm and the objective function to update the model parameters of the aeroengine main bearing dynamics model, thereby correcting the aeroengine main bearing dynamics model. The simulated data output by the corrected aeroengine main bearing dynamics model is highly consistent with the measured data, providing an effective solution to data imbalance and small sample size issues in fault diagnosis applications.

[0050] The steps of the various methods above are divided for clarity of description only. During implementation, they can be combined into a single step, or some steps can be broken down into multiple steps. As long as they contain the same logical relationships, they are all within the scope of protection of this application. Adding minor modifications or introducing minor design changes to the algorithm or process, but not changing the core design of the algorithm or process, are also within the scope of protection of this application.

[0051] Another embodiment of the present application proposes an aircraft engine main bearing dynamics model correction system based on sensitive feature fusion. The details of the aircraft engine main bearing dynamics model correction system based on sensitive feature fusion proposed in this embodiment are described in detail below. The following content is only the implementation details provided for easy understanding and is not necessary for the implementation of this example.

[0052] In one embodiment, we used the Western Reserve Bearing Dataset with a speed of 1797 r / min, including healthy, inner ring fault, and outer ring fault (fault size: 0.007 inches and 0.014 inches), a total of 5 types of measured data (1 type of healthy, 2 types of inner ring fault, 2 types of outer ring fault) to verify the effectiveness of the method. The fault size, speed, load, and fault type of the 5 types of measured data were input into the dynamic model to generate 5 types of simulation data and perform sliding window sampling. The above method was then used to calculate the fused sensitive features of the measured data and the fused sensitive features of the simulation data, and then the dynamic model was corrected. The comparison diagram of the optimized vibration signal is shown in the figure below. Figure 3 As shown, the envelope spectrum of the inner race fault is as follows Figure 4 It can be seen that the method for correcting the dynamic model of the main bearing of an aero-engine based on sensitive feature fusion proposed in this application is scientific and effective.

[0053] The structure of the aircraft engine main bearing dynamic model correction system based on sensitive feature fusion proposed in this embodiment is as follows: Figure 5 As shown, it includes: a model construction module 201, an initial feature set construction module 202, a sensitive feature set construction module 203, an objective function establishment module 204 and an optimization module 205.

[0054] The model building module 201 is used to build a dynamic model of the aircraft engine main bearing.

[0055] The initial feature set construction module 202 is used to generate simulation data based on the collected measured data, and extract time-frequency domain features of the measured data and the simulation data to construct an initial feature set.

[0056] The sensitive feature set construction module 203 is used to screen out sensitive features from the initial feature set based on the binary improved horse group optimization algorithm, construct a sensitive feature set, and calculate the component matrix coefficients of the sensitive feature set in combination with the principal component analysis method, thereby obtaining a fused sensitive feature expression.

[0057] The objective function establishment module 204 is used to calculate the fusion sensitive features of the measured data and the simulation data using the fusion feature expression, and use the difference calculation formula between the two as the objective function of the model correction.

[0058] The optimization module 205 is used to perform optimization based on the Horse Swarm Optimization algorithm and the objective function, update the model parameters of the aircraft engine main bearing dynamics model, and realize the correction of the aircraft engine main bearing dynamics model.

[0059] It is not difficult to find that this embodiment is a system embodiment corresponding to the above-mentioned method embodiment, and this embodiment can be implemented in conjunction with the above-mentioned method embodiment. The relevant technical details and technical effects mentioned in the above-mentioned method embodiment are still valid in this embodiment, and to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the above-mentioned method embodiment.

[0060] It is worth mentioning that all modules and modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovation of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that other units do not exist in this embodiment.

[0061] Another embodiment of the present application provides an electronic device, such as Figure 6 As shown, it includes: at least one processor 301; and a memory 302 that is communicatively connected to the at least one processor 301; wherein the memory 302 stores instructions that can be executed by the at least one processor 301, and the instructions are executed by the at least one processor 301 to enable the at least one processor 301 to execute a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion as described in the above method embodiment.

[0062] The memory and processor are connected using a bus, which includes any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and therefore will not be described further in this article. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. Data processed by the processor is transmitted on a wireless medium via an antenna. Furthermore, the antenna also receives data and transmits it to the processor.

[0063] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0064] Another embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it can implement a method for correcting the dynamic model of an aircraft engine main bearing based on sensitive feature fusion as described in the above method embodiment.

[0065] That is, those skilled in the art will understand that all or part of the steps in the above-described method embodiments can be implemented by instructing the relevant hardware through a program, which is stored in a storage medium and includes a number of instructions for causing a device (such as a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps in the method embodiments of the present application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory, a random access memory, a magnetic disk, or an optical disk.

[0066] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application. In actual applications, various modifications may be made to the embodiments in form and detail without departing from the spirit and scope of the present application. Those skilled in the art will appreciate that improvements and modifications may be made without departing from the principles of the present application, and such improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion, characterized in that: The method comprises: Construct a dynamic model of aero-engine main bearings; Generate simulation data based on the collected measured data, extract the time-frequency domain features of the measured data and simulation data, and construct an initial feature set; Based on the binary improved horse group optimization algorithm, sensitive features are screened from the initial feature set to construct a sensitive feature set. Combined with the principal component analysis method, the component matrix coefficients of the sensitive feature set are calculated to obtain the fused sensitive feature expression. The fusion feature expression is used to calculate the fusion sensitive features of the measured data and the simulation data, and the difference calculation formula between the two is used as the objective function of the model correction; Based on the Horse Swarm Optimization algorithm and the objective function, the model parameters of the aero-engine main bearing dynamics model are updated to realize the correction of the aero-engine main bearing dynamics model.

2. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 1, characterized in that: Construct a dynamic model of an aircraft engine main bearing, including: The aircraft engine main bearing is considered as a faulty bearing, and a dynamic model is established that takes into account the fault size, speed, load, and fault type of the faulty bearing. The dynamic model mainly consists of an inner ring, rolling elements, outer ring, and cage. The inner ring is assumed to rotate with the rotating shaft, and the outer ring is fixed to the bearing seat. The dynamic model captures the axial and radial motion of the inner and outer rings. During the simulation process, the stiffness and damping of each component are assumed to be fixed values. The operating parameters of the dynamic model are consistent with those of the real bearing. According to Newton's second law, the centrifugal force caused by installation and manufacturing deviation is considered and the dynamic model is established. The four differential equations of motion related to the inner and outer rings in the plane are: ； in, Indicates the mass of the bearing inner ring, represents the damping coefficient of the bearing inner ring, Indicates the stiffness of the bearing inner ring, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, Indicates that the external load is The directional component of force, is the eccentricity, is the shaft speed, For the moment, Indicates that the inner ring of the bearing is The acceleration in the direction, Indicates that the inner ring of the bearing is Speed ​​in direction, Indicates that the inner ring of the bearing is Displacement in direction, represents the acceleration due to gravity, Indicates that the external load is The directional component of force, Indicates the mass of the bearing outer ring, represents the damping coefficient of the bearing outer ring, Indicates the stiffness of the bearing outer ring, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction, Indicates that the outer ring of the bearing is The acceleration in the direction, Indicates that the outer ring of the bearing is Speed ​​in direction, Indicates that the outer ring of the bearing is Displacement in direction.

3. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 2, characterized in that: After establishing and obtaining the four motion differential equations of the dynamic model related to the inner ring and the outer ring in the plane, the method further includes: The fourth-order Runge-Kutta method is used to solve the four motion differential equations related to the inner and outer rings in the plane of the dynamic model to obtain the vibration responses of the rolling elements, inner and outer rings. The ode45 solver is then used to calculate the dynamic response of the faulty bearing. When the rolling element passes through a defect, it will produce a certain contact deformation. The contact deformation formula of the rolling element is as follows: ； in, Indicates the The angular position of the rolling element, Indicates the radial clearance of the bearing, Indicates the deformation release caused by local defects, Indicates the The contact deformation of the rolling elements.

4. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 1, characterized in that: Extract the time-frequency domain features of measured data and simulation data and construct an initial feature set, including: Extract 25 time-frequency domain features of measured data and simulation data to construct the initial feature set; Among them, the 25 time-frequency domain features extracted include: maximum value, minimum value, mean value, median, peak value, peak-to-peak value, rectified mean value, variance, standard deviation, kurtosis, skewness, root mean square, form factor, peak factor, impulse factor, root mean square amplitude, margin factor, kurtosis factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance, frequency standard deviation, spectral entropy and signal complexity.

5. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 1, characterized in that: Based on the binary improved horse swarm optimization algorithm, sensitive features are screened from the initial feature set, including: Based on the binary improved horse herd optimization algorithm, a transfer function is used to convert the velocity value into a probability value to specify the probability of updating the binary solution, thereby screening out sensitive features from the initial feature set; The transfer function used is the V-type transfer function, and its calculation formula is: ； in, Indicates in In the iteration The horse in the The speed of the dimension, represents a V-type transfer function; The horse's position based on its speed is updated using the following formula: ； in, Indicates in In the iteration The horse in the The position in the dimension, Indicates the updated The horse in the Position in the dimension.

6. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 5, characterized in that: Combined with the principal component analysis method, the component matrix coefficients of the sensitive feature set are calculated to obtain the fused sensitive feature expression, including: The sensitive feature set is normalized using the following formula: ； in, represents the sensitive feature set, represents the mean of the sensitive feature set, represents the standard deviation of the sensitive feature set, represents the normalized sensitive feature set; The covariance matrix is ​​calculated using the following formula: ； in, represents the covariance matrix, Indicates the total number of vibration signals; The eigenvalues ​​and eigenvectors are calculated using the following formula: ； ； in, Indicates the eigenvalue matrix, Indicates the eigenvector matrix; The largest eigenvalue matrix Convert to a diagonal matrix and combine the largest eigenvector matrix , the factor loading coefficient is calculated by the following formula: ； in, represents the calculated factor loading coefficient, Indicates finding the diagonal matrix; Based on the following formula, and , calculate the component matrix coefficients: ； in, Represents the calculated component matrix coefficients.

7. The method for correcting the dynamic model of an aero-engine main bearing based on sensitive feature fusion according to claim 6, characterized in that: The fusion feature expression is used to calculate the fusion sensitive features of the measured data and the simulation data, and the difference calculation formula between the two is used as the objective function of the model correction, including: Filter out the component matrix coefficients greater than 0, perform weighted calculation with the sensitive features of the measured data and simulation data, and obtain the fused sensitive features of the measured data and simulation data; The difference between the fusion sensitive features of the measured data and the simulation data is calculated as the objective function of the model correction through the following formula: ； in, represents the objective function of model modification, represents the fusion sensitive features of simulation data, Represents the fused sensitive features of the measured data.

8. An aircraft engine main bearing dynamics model correction system based on sensitive feature fusion, characterized in that: The system comprises: Model building module, used to build the dynamic model of the aircraft engine main bearing; An initial feature set construction module is used to generate simulation data based on the collected measured data, extract the time-frequency domain features of the measured data and the simulation data, and construct an initial feature set; The sensitive feature set construction module is used to screen out sensitive features from the initial feature set based on the binary improved horse group optimization algorithm, construct the sensitive feature set, and calculate the component matrix coefficients of the sensitive feature set in combination with the principal component analysis method, thereby obtaining the fused sensitive feature expression; The objective function establishment module is used to calculate the fusion sensitive features of the measured data and the simulation data using the fusion feature expression, and use the difference calculation formula between the two as the objective function of the model correction; The optimization module is used to optimize based on the Horse Swarm Optimization algorithm and the objective function, update the model parameters of the aircraft engine main bearing dynamics model, and realize the correction of the aircraft engine main bearing dynamics model.

9. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the aircraft engine main bearing dynamic model correction method based on sensitive feature fusion as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it can implement an aircraft engine main bearing dynamics model correction method based on sensitive feature fusion as described in any one of claims 1 to 7.

Images