Method, device and equipment for extracting degradation factors of aero-engine and medium

Abstract

The invention discloses an aero-engine degradation factor extraction method, device, equipment and medium, and relates to the technical field of aero-engines, the aero-engine degradation factor extraction method establishes an aero-engine linear guide system based on engine characteristics, adds guide consciousness to particles on the basis of a traditional particle swarm algorithm, and improves the performance of the aero-engine. A linear guide system participates in particle speed correction, the problem that degradation factor extraction is prone to falling into local optimum is solved, and accurate output of the degradation factor of the rotating part is achieved. Compared with the prior art, the method has the advantages that the multi-dimensional characteristic of the degradation problem of the aero-engine is fully utilized, the influence of the degradation factor on multi-dimensional output is simplified into a linear influence matrix, and the guide consciousness is constructed by directly multiplying the multi-dimensional deviation vector and the matrix, so that the extraction precision of the degradation factor of the aero-engine can be greatly improved; meanwhile, the overall calculation cost is low, complex operation and judgment are not needed, and the extraction rate and adaptability of the degradation factors can be guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aero-engines, in particular, to an aero-engine degradation factor extraction method, device, equipment and medium. BACKGROUND

[0002] High-precision extraction of aero-engine degradation factors is of great significance to flight safety, predictive maintenance and cost reduction. By quantifying the performance degradation of key components (such as compressor efficiency decline and turbine wear), potential failures can be warned in advance and maintenance strategies can be optimized, thereby avoiding sudden failures, prolonging component life and reducing unnecessary disassembly and inspection. In aero-engines, degradation factors are key indicators for quantifying the degree of aero-engine performance degradation, reflecting the irreversible decline of performance parameters (such as efficiency, thrust and aerodynamic stability) of key components (such as compressors, turbines, etc.) due to wear, corrosion and fouling.

[0003] Aero-engine degradation factor extraction is essentially a multi-objective optimization problem, aiming to infer the degree of performance degradation (performance decay or functional deterioration) of key components (such as compressors, turbines, combustion chambers, etc.) inside the engine through sensor data and engine models.

[0004] In existing aero-engine degradation factor extraction methods, traditional multi-objective optimization algorithms (such as genetic algorithms and particle swarm algorithms) have limitations such as high computational complexity, large parameter sensitivity, strong subjectivity of fuzzy weights, etc., making it difficult to adapt to the multi-dimensional dynamic nonlinear characteristics exhibited during engine degradation. At the same time, although optimization methods based on chaos theory improve global search ability, they are highly dependent on model accuracy (such as CFD models), and lack adaptability when dealing with dynamic operating conditions of the engine. These factors limit the real-time performance and accuracy of degradation factor extraction, affecting the effectiveness of engine fault prediction and maintenance optimization, and there is an urgent need to develop more efficient and adaptable degradation factor extraction technology. SUMMARY

[0005] The present application provides an aero-engine degradation factor extraction method to solve the technical problems of poor real-time performance, accuracy and adaptability of existing aero-engine degradation factor extraction technology.

[0006] The present application is achieved by the following scheme: An aero-engine degradation factor extraction method, comprising the steps of: S1, constructing an engine degradation factor linear influence matrix according to flight data and component-level models; S2, randomly initializing a particle population, initializing particle motion inertia, individual consciousness, group consciousness, and guide consciousness, and evaluating the speed and position of each particle to obtain an initial global optimum; S3, establishing a linear aero-engine guide based on engine characteristics, adding guide awareness calculation on the basis of traditional particle swarm algorithm particle motion inertia, individual awareness and group awareness, superimposing the speed and position of each particle to calculate the output deviation vector and evaluate the function fitness value of each particle; S4, updating the guide weight coefficient according to the current fitness of the particle, and then calculating the particle guide awareness according to the linear influence matrix, the output deviation vector and the guide awareness weight coefficient; S5, updating the historical optimal position of each particle based on the function fitness value of each particle, and then updating the global optimal position of the group; S6, verifying whether the global optimal position of the group meets the end condition, if not, continue to iterate until the end condition is met and exit.

[0007] Further, the step S1 specifically comprises the steps of: S11, respectively disturbing the flow and efficiency degradation factors of each rotating part of the engine model, and recording the changes of each state quantity after the disturbance of a single degradation factor; S12, constructing an engine degradation factor linear influence matrix α : In the formula, Δ η i is the disturbance amount of the first i engine degradation factor, Δ x ji is the first i degradation factor, and the first j engine state quantity change.

[0008] Further, the step S3 has the steps of: S31, respectively calculating the particle motion inertia m , individual awareness k , group awareness , guide awareness of the first particle in the first iteration: In the formula, w is the inertia weight, controlling the retention proportion of the original speed, v is the particle speed; c 1 and c 2 are individual learning factor constants and group learning factor constants, respectively; r 1, r 2 are random numbers, which introduce randomness to avoid the algorithm from falling into local optimum; pmd and g d are the individual historical optimal position and the group historical optimal position, respectively; x is the current position of the particle; c 3 is the weight coefficient of the guide of the aero-engine, which is changed with the fitness value of the function and used to keep the order of magnitude of the guide consistent with the other three terms; γ is the adjustment amount of the degeneration factor, which is related to the linear influence matrix α and the deviation vector β ; S32, superimpose the calculated particle motion inertia, individual consciousness, group consciousness and guide consciousness to obtain the speed and position of each particle, respectively: wherein, is the speed of the m th particle obtained in the k th iteration, is the position of the m th particle obtained in the k th iteration; S33, substitute the position of each particle into the component-level model as the degeneration factor to perform calculation, compare with the sensor data to obtain the deviation vector m of the th particle: wherein, e j is the absolute error of the j th state variable; n is the sample number, and the absolute error average of n samples is taken as the absolute error in the case of equivalent degeneration level; S34, calculate the fitness value according to the deviation vector f m : wherein, a mj is the weight coefficient of the m th state variable of the j th particle, e mj is the absolute error of the m th state variable of the j th particle.

[0009] Further, the step S4 specifically comprises the step of: S41, calculate the order of magnitude adjustment amount according to the speed of the particle r m :​ In the formula, r m It is the first m The adjustment factor is on the order of magnitude of the number of particles, where maxAcc is the maximum particle velocity. α It is a linear influence matrix. β The deviation vector; S42. Adjust the amount according to the order of magnitude. r m Particle fitness value f m Update the guide weight coefficient of the aircraft engine c 3: In the formula, k 1. b , f 1. f 2 is a non-zero constant; when f m < f At 1 o'clock, cancel the aircraft engine guide; when f 1< f m < f 2. The guide weight coefficient changes with the fitness value; when f m > f At time 2, the guide weight coefficient is a non-zero constant; S43. Based on the linear influence matrix α Deviation vector β Updated Guide Consciousness Weighting Coefficient c 3. Recalculate the guide's awareness v 4: v 4= c 3× α × β .

[0010] This application also provides an apparatus for extracting degradation factors from aircraft engines, comprising: The linear influence matrix construction module is used to construct the linear influence matrix of engine degradation factors based on flight data and component-level models. The initial global optimum evaluation module is used to randomly initialize the particle population, initialize the particle motion inertia, individual consciousness, group consciousness, and guide consciousness, and evaluate the velocity and position of each particle to obtain the initial global optimum. The function fitness value calculation module is used to establish a linear aero-engine guide based on engine characteristics. On the basis of particle motion inertia, individual consciousness and group consciousness in the traditional particle swarm algorithm, the guide consciousness calculation is added. The velocity and position of each particle are obtained by superposition, and then the output deviation vector is calculated and the function fitness value of each particle is evaluated. The guide consciousness calculation module is used to update and calculate the guide weight coefficient based on the particle's current fitness, and then calculate the particle's guide consciousness based on the linear influence matrix, the output deviation vector, and the guide consciousness weight coefficient. The optimal position update module is used to update the historical optimal position of each particle based on the function fitness value of each particle, and then update the global optimal position of the population. The iterative verification module is used to verify whether the global optimal position of the population meets the termination condition. If it does not meet the condition, the iteration continues until the termination condition is met and the module exits.

[0011] Furthermore, the linear influence matrix construction module specifically includes: The state quantity recording module is used to perturb the flow rate and efficiency degradation factor of each rotating part of the engine model, and record the changes of each state quantity after a single degradation factor perturbation. The matrix construction module is used to construct the linear influence matrix of engine degradation factors. α : In the formula, Δ η i For the engine i The perturbation of each degradation factor, Δ x ji For the disturbance i The first degradation factor, the... j Changes in engine state.

[0012] Furthermore, the function fitness value calculation module includes: The integrated calculation module is used to calculate the first... m The first particle k Particle motion inertia in the next iteration Individual consciousness Group consciousness Guide awareness : In the formula, w It is the inertial weight, which controls the proportion of the original velocity that is retained. v It is the particle velocity; c 1 and c 2 represents the individual learning factor constant and the group learning factor constant, respectively; r 1.r 2 is a random number, which is introduced to avoid the algorithm falling into local optimum by randomness; p md and g d are the historical optimal position of the particle individual and the historical optimal position of the group respectively; x is the current position of the particle; c 3 is the weight coefficient of the guide of the aero-engine, which changes with the fitness value and is used to keep the order of magnitude of the guide consistent with the other three; γ is the degradation factor adjustment quantity, which is related to the linear influence matrix α and the bias vector β ; The velocity and position calculation module is used to superimpose the calculated particle motion inertia, individual consciousness, group consciousness and guide consciousness to obtain the velocity and position of each particle respectively: In the formula, is the velocity of the m th particle obtained in the k th iteration, is the position of the m th particle obtained in the k th iteration; The bias vector calculation module is used to take the position of each particle as the degradation factor, substitute it into the component-level model for calculation, and compare it with the sensor data to obtain the bias vector of the m th particle : In the formula, e j is the absolute error of the j th state variable; n is the sample number, and the absolute error average of n samples is taken as the absolute error in the case of equivalent degradation level; The fitness value calculation module is used to calculate the fitness value according to the bias vector f m : In the formula, a mj is the weight coefficient of the m th state variable of the j th particle, e mj is the absolute error of the m th state variable of the j th particle.

[0013] Further, the guide awareness calculation module specifically comprises: a magnitude adjustment amount calculation module, configured to calculate a magnitude adjustment amount according to the velocity of the particle r m : wherein, r m is the magnitude adjustment amount of the i th particle, maxAcc is the maximum velocity of the particle, m is a linear influence matrix, α is a bias vector; β a guide weight coefficient updating module, configured to update the guide weight coefficient according to the magnitude adjustment amount r m , the particle fitness value f m update the guide weight coefficient of the aero-engine c 3: wherein, k 1, b , f 1, f 2 is a constant not equal to 0; when f m < f 1, the guide of the aero-engine is cancelled; when f 1 f m < f 2, the guide weight coefficient changes with the fitness value; when f m > f 2, the guide weight coefficient is a constant not equal to 0; a guide awareness updating module, configured to update the guide awareness according to the linear influence matrix α , the bias vector β , and the updated guide awareness weight coefficient c 3 v 4: v 4= c 3× α × β .

[0014] Another aspect of the present application also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the aero-engine degradation factor extraction method when running the computer program.

[0015] ​The application also provides a storage medium including a stored program which, when executed, controls a device in which the storage medium is located to perform the steps of the aero-engine degradation factor extraction method.

[0016] Compared with the prior art, the application has the following beneficial effects: The application provides an aero-engine degradation factor extraction method. The method establishes a linear aero-engine guide based on engine characteristics. On the basis of particle motion inertia, individual consciousness and group consciousness of a traditional particle swarm algorithm, a guide consciousness established based on engine characteristics is added. The linear guide system is used to change the particle speed, so as to solve the problem that the degradation factor extraction easily falls into local optimization, and to realize accurate output of the degradation factor of the rotating part. Since the linear aero-engine guide proposed in the application is linear, the linear guide has a small calculation cost and does not need complex operation and judgment, and the degradation factor extraction rate can be ensured. Compared with the traditional method, the degradation factor extraction method of the application not only greatly improves the degradation factor extraction accuracy and calculation efficiency, but also has excellent stability and adaptability, and can be widely applied to performance degradation evaluation and life prediction of rotating parts in aero-engines, gas turbines and other systems.

[0017] In addition to the purposes, features and advantages described above, the application has other purposes, features and advantages. The application will be described in further detail below with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the application and, together with the specification, serve to explain the principles of the application.

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, for those skilled in the art, other drawings can also be obtained based on these drawings without creative labor. Figure 1 is a flowchart of the aero-engine degradation factor extraction method of the preferred embodiment of the application; Figure 2 is a flowchart of the aero-engine degradation factor extraction method of another preferred embodiment of the application; Figure 3 is a schematic diagram of the aero-engine degradation factor extraction device module of the preferred embodiment of the application; Figure 4 is a schematic diagram of the sub-module of the linear influence matrix construction module of the preferred embodiment of the application; Figure 5is a submodule schematic diagram of a function fitness value calculation module of a preferred embodiment of the present application; Figure 6 is a submodule schematic diagram of a guide awareness calculation module of a preferred embodiment of the present application; Figure 7 is an electronic device entity schematic block diagram of a preferred embodiment of the present application; Figure 8 is an internal structure diagram of a computer device of a preferred embodiment of the present application. DETAILED DESCRIPTION

[0020] It should be understood that the specific embodiments described herein are merely intended to explain the technical solutions of the present application, and are not intended to limit the present application.

[0021] In order to better understand the technical solutions of the present application, the following will be described in detail in combination with the drawings of the specification and specific embodiments.

[0022] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an aero-engine degradation factor extraction device capable of realizing the above functions. The following takes the aero-engine degradation factor extraction device as an example to describe the present embodiment and the following embodiments.

[0023] As shown in Figure 1 The preferred embodiment of the present application provides an aero-engine degradation factor extraction method, comprising the steps of: S1, constructing an engine degradation factor linear influence matrix according to flight data and component level model; S2, randomly initializing a particle population, initializing particle motion inertia, individual awareness, group awareness, and guide awareness, and evaluating the speed and position of each particle to obtain an initial global optimum; S3, establishing a linear aero-engine guide based on engine characteristics, adding guide awareness calculation on the basis of traditional particle swarm algorithm particle motion inertia, individual awareness, and group awareness, superimposing the speed and position of each particle, then calculating the output deviation vector and evaluating the function fitness value of each particle; S4, updating the guide weight coefficient according to the current fitness of the particle, and then calculating the guide awareness of the particle according to the linear influence matrix, the output deviation vector, and the guide awareness weight coefficient; S5, updating the historical optimal position of each particle based on the function fitness value of each particle, and then updating to obtain the global optimal position of the group; S6, verifying whether the global optimal position of the group meets the end condition, if not, continue to iterate until the end condition is met and exit.

[0024] The embodiment proposes an aero-engine degradation factor extraction method, which establishes a linear aero-engine guide based on engine characteristics. Based on the inertia of particle motion, individual consciousness and group consciousness of the traditional particle swarm algorithm, a guide consciousness based on the establishment of engine characteristics is added. The linear guide system is used to change the particle speed, which innovatively solves the problem of easy falling into local optimum in degradation factor extraction, realizes the accurate output of the degradation factor of rotating parts, and ensures the degradation factor extraction rate. Compared with the traditional method, the degradation factor extraction method of the embodiment not only greatly improves the degradation factor extraction accuracy and calculation efficiency, but also has excellent stability and adaptability, and can be widely applied to the performance degradation evaluation and life prediction of rotating parts in aero-engine, gas turbine and other systems.

[0025] Preferably, the step S1 specifically comprises the steps of: S11, respectively perturbing the flow and efficiency degradation factors of each rotating part of the engine model, and recording the changes of each state quantity after the single degradation factor is perturbed; S12, constructing an engine degradation factor linear influence matrix α : In the formula, Δ η i is the perturbation amount of the first i engine degradation factor, Δ x ji is the first i degradation factor, and the first j engine state quantity change.

[0026] The steps S11-S12 of the embodiment realize the construction of the engine degradation factor linear influence matrix by respectively perturbing the flow and efficiency degradation factors of each rotating part of the engine model, and the changes of each state quantity after the single degradation factor is perturbed. The advantages include directly quantifying the sensitivity of unit degradation factor to a specific state quantity, establishing a direct mapping relationship between the degradation factor and the state quantity, providing a physically interpretable mathematical basis for subsequent reverse extraction of the degradation factor, and realizing linear dimension reduction of a complex nonlinear system.

[0027] Preferably, the step S3 has the steps of: S31, respectively calculating the inertia of particle motion m , individual consciousness k , group consciousness , guide consciousness of the first particle in the first iteration. wherein, w is the inertia weight, controlling the preservation proportion of the original velocity, v is the particle velocity; c 1 and c 2 are respectively the individual learning factor constant and the group learning factor constant; r 1, r 2 is a random number, which is introduced to avoid the algorithm from falling into local optimum; p md and g d are respectively the individual historical optimal position and the group historical optimal position; x is the current position of the particle; c 3 is the weight coefficient of the guide of the aero-engine, which is changed with the fitness value and is used to make the order of magnitude of the guide consciousness consistent with the other three; γ is the degeneration factor adjustment amount, which is related to the linear influence matrix α and the bias vector β ; S32, the calculated particle motion inertia, individual consciousness, group consciousness and guide consciousness are superimposed to obtain the velocity and position of each particle respectively: wherein, is the velocity obtained by the m th particle in the k th iteration, is the position obtained by the m th particle in the k th iteration; S33, the position of each particle is taken as the degeneration factor, which is substituted into the component-level model to be calculated, and the deviation vector m of the th particle is obtained by comparing with the sensor data: wherein, e j is the absolute error of the j th state variable; n is the sample number, and the absolute error average of n samples is taken as the absolute error in the case of equivalent degeneration level; S34, the fitness value is calculated according to the deviation vector f m : wherein, a mj is the​m The first particle j The weighting coefficients of each state variable e mj It is the first m The first particle j The absolute error of each state variable.

[0028] This embodiment calculates the particle motion inertia, individual consciousness, group consciousness, and guide consciousness by superimposing them in steps S31-S34. After obtaining the velocity and position of each particle, the position is used as a degradation factor and substituted into the component-level model to calculate the deviation vector. Finally, the fitness value is calculated based on the deviation vector. Its advantages include: directly mapping the particle position to the candidate degradation factor and comparing it with the actual sensor data in the component-level model, ensuring that each particle optimization is based on the actual physical laws of engine operation, solving the problem that the optimization direction of traditional algorithms is divorced from engineering reality; and by guiding the particles to focus in the direction of minimizing deviation through guide consciousness, and combining the deviation vector for comprehensive evaluation of multiple state quantities, the number of algorithm iterations can be reduced.

[0029] Preferably, step S4 specifically includes the following steps: S41. Calculate the order-of-magnitude adjustment based on the particle velocity. r m : In the formula, r m It is the first m The adjustment factor is on the order of magnitude of the number of particles, where maxAcc is the maximum particle velocity. α It is a linear influence matrix. β The deviation vector; S42. Adjust the amount according to the order of magnitude. r m Particle fitness value f m Update the guide weight coefficient of the aircraft engine c 3: In the formula, k 1. b , f 1. f 2 is a non-zero constant; when f m < f At 1 o'clock, cancel the aircraft engine guide; when f 1< f m < f 2. The guide weight coefficient changes with the fitness value; when f m >f 2 the guide weight coefficient is a constant not equal to 0; S43, according to the linear influence matrix α , the deviation vector β , the updated guide awareness weight coefficient c 3 Recalculate the guide awareness v 4: v 4= c 3× α × β .

[0030] The embodiment first calculates the order of magnitude adjustment quantity through steps S41-S43, and then adjusts the guide weight coefficient according to the order of magnitude adjustment quantity r m , the particle fitness value f m The guide weight coefficient of the aero-engine is updated, and finally the linear influence matrix α , the deviation vector β , the updated guide awareness weight coefficient c 3 Recalculate the guide awareness for iterative update of the guide awareness, which has the advantages of: through the order of magnitude adjustment quantity, the order of magnitude of the guide awareness, the inertia awareness, the individual awareness and the group awareness can be ensured to be consistent, avoiding that a certain awareness item excessively dominates the optimization; and adaptive adjustment of the guide weight coefficient can intelligently switch strategies at different search stages; at the same time, through matrix multiplication, the multi-state quantity deviation is compressed into a degradation factor correction direction vector, which can ensure that the guide awareness accurately points to the real value of the degradation factor.

[0031] Preferably, as Figure 2 indicated, another preferred embodiment of the present application provides an aero-engine degradation factor extraction method, which establishes an aero-engine linear guide matrix, and uses the aero-engine guide matrix to assist the particle swarm optimization algorithm to solve the problem that the traditional particle swarm optimization algorithm is easy to fall into local optimum when extracting the degradation factor, realizes accurate output of the degradation factor of the rotating part, and obtains the change rule of the degradation factor.

[0032] Preferably, as Figure 3 indicated, the present application further provides an aero-engine degradation factor extraction device, which comprises: A linear influence matrix construction module is configured to construct an engine degradation factor linear influence matrix according to flight data and component-level models; An initial global optimum evaluation module is configured to randomly initialize a particle population, initialize the particle motion inertia, individual awareness, group awareness, and guide awareness, and evaluate the speed and position of each particle to obtain an initial global optimum; The function fitness value calculation module is used to establish a linear aero-engine guide based on engine characteristics. On the basis of particle motion inertia, individual consciousness and group consciousness in the traditional particle swarm algorithm, the guide consciousness calculation is added. The velocity and position of each particle are obtained by superposition, and then the output deviation vector is calculated and the function fitness value of each particle is evaluated. The guide consciousness calculation module is used to update and calculate the guide weight coefficient based on the particle's current fitness, and then calculate the particle's guide consciousness based on the linear influence matrix, the output deviation vector, and the guide consciousness weight coefficient. The optimal position update module is used to update the historical optimal position of each particle based on the function fitness value of each particle, and then update the global optimal position of the population. The iterative verification module is used to verify whether the global optimal position of the population meets the termination condition. If it does not meet the condition, the iteration continues until the termination condition is met and the module exits.

[0033] The aero-engine degradation factor extraction device provided in this embodiment adopts the aero-engine degradation factor extraction method in the above embodiments, solving the technical problems of poor real-time performance, accuracy, and adaptability of existing aero-engine degradation factor extraction technologies. Compared with the prior art, the beneficial effects of the aero-engine degradation factor extraction device provided in this application are the same as those of the aero-engine degradation factor extraction method provided in the above embodiments, and other technical features in the aero-engine degradation factor extraction device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0034] Preferably, such as Figure 4 As shown, the linear influence matrix construction module specifically includes: The state quantity recording module is used to perturb the flow rate and efficiency degradation factor of each rotating part of the engine model, and record the changes of each state quantity after a single degradation factor perturbation. The matrix construction module is used to construct the linear influence matrix of engine degradation factors. α : In the formula, Δ η i For the engine i The perturbation of each degradation factor, Δ x ji For the disturbance i The first degradation factor, the... j Changes in engine state.

[0035] Preferably, such as Figure 5 As shown, the function fitness value calculation module includes: The integrated calculation module is used to calculate the first... m The first particlek Particle motion inertia of sub-iteration Individual consciousness Group consciousness Guidance consciousness : wherein, w is the inertia weight, controlling the retention proportion of original velocity, v is the particle velocity; c 1 and c 2 are respectively the individual learning factor constant and the group learning factor constant; r 1, r 2 are random numbers, introducing randomness to avoid the algorithm falling into local optimum; p md and g d are respectively the particle individual historical optimal position and the group historical optimal position; x is the current position of the particle; c 3 is the guide weight coefficient of the aero-engine, which changes with the fitness value and is used to keep the order of magnitude of the guidance consciousness consistent with the other three; γ is the degradation factor adjustment amount, related to the linear influence matrix α and the bias vector β ; The velocity and position calculation module is used to superimpose the calculated particle motion inertia, individual consciousness, group consciousness and guidance consciousness to obtain the velocity and position of each particle respectively: wherein, is the velocity obtained by the m th particle in the k th iteration, is the position obtained by the m th particle in the k th iteration; The bias vector calculation module is used to take the position of each particle as the degradation factor, substitute it into the component-level model for calculation, and compare it with the sensor data to obtain the bias vector of the m th particle: wherein, e j is the absolute error of the j th state variable; n is the sample number, and the absolute error average value of n samples is taken as the absolute error in the case of equivalent degradation level; ​The fitness value calculation module is used to calculate the fitness value based on the bias vector. f m : In the formula, a mj It is the first m The first particle j The weighting coefficients of each state variable e mj It is the first m The first particle j The absolute error of each state variable.

[0036] Preferably, such as Figure 6 As shown, the guide consciousness calculation module specifically includes: The order-of-magnitude adjustment calculation module is used to calculate the order-of-magnitude adjustment based on the particle velocity. r m : In the formula, r m It is the first m The adjustment factor is on the order of magnitude of the number of particles, where maxAcc is the maximum particle velocity. α It is a linear influence matrix. β The deviation vector; The wizard-driven weight coefficient update module is used to adjust the amount based on the order of magnitude. r m Particle fitness value f m Update the guide weight coefficient of the aircraft engine c 3: In the formula, k 1. b , f 1. f 2 is a non-zero constant; when f m < f At 1 o'clock, cancel the aircraft engine guide; when f 1< f m < f 2. The guide weight coefficient changes with the fitness value; when f m > f At time 2, the guide weight coefficient is a non-zero constant; The guide awareness update module is used to update the linear influence matrix. α Deviation vector β Updated Guide Consciousness Weighting Coefficientc 3 Recalculate the guide awareness v 4: v 4= c 3x α x β .

[0037] As Figure 7 shown, the preferred embodiment of the present application also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the aero-engine degradation factor extraction method in the above embodiment when executing the computer program.

[0038] The electronic device provided by the present application adopts the aero-engine degradation factor extraction method in the above embodiment to solve the technical problems of poor real-time performance, accuracy and adaptability of existing aero-engine degradation factor extraction technology. Compared with the prior art, the electronic device provided by the present application has the same beneficial effects as the aero-engine degradation factor extraction method provided by the above embodiment, and other technical features in the electronic device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0039] As Figure 8 shown, the preferred embodiment of the present application also provides a computer device, which can be a terminal or a living body detection server, and its internal structure diagram can be as shown in Figure 8 The computer device comprises a processor, a memory and a network interface connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is used to communicate with other computer devices outside through network connection. The computer program is executed by the processor to implement the steps of the above aero-engine degradation factor extraction method.

[0040] Those skilled in the art can understand Figure 8 that the structure shown in the above embodiment is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0041] The computer device provided in the application adopts the aero-engine degradation factor extraction method in the above embodiment, and solves the technical problems of poor real-time performance, accuracy and adaptability of the existing aero-engine degradation factor extraction technology. Compared with the prior art, the computer device provided in the application has the same beneficial effects as the aero-engine degradation factor extraction method provided in the above embodiment, and other technical features in the electronic device are the same as the features disclosed in the above embodiment method, which will not be described here.

[0042] The preferred embodiment of the application also provides a storage medium including a stored program, which controls the device where the storage medium is located to perform the steps of the aero-engine degradation factor extraction method in the above embodiment when the program is running.

[0043] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0044] If the functions of the method of the embodiment are realized in the form of software function units and sold or used as independent products, they can be stored in one or more computer readable storage media. Based on this understanding, the part of the prior art or the part of the technical solution of the embodiments of the application can be embodied in the form of a software product stored in a storage medium, including a plurality of instructions for causing a computing device (which can be a personal computer, a server, a mobile computing device or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0045] Those skilled in the art should understand that the embodiments of the application can be provided as a method, a system or a computer program product. Therefore, the application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes. The solutions in the embodiments of the application can be implemented in various computer languages, such as object-oriented programming language C++ and embedded programming language C.

[0046] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0047] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0048] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0049] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0050] The computer program product provided by the present application solves the technical problem of poor real-time performance, accuracy and adaptability of the existing aero-engine degradation factor extraction technology. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the aero-engine degradation factor extraction method provided by the above-mentioned embodiments, and are not described here.

[0051] Although the preferred embodiments of the present application have been described, those skilled in the art, once they know the basic creative concept, can make additional changes and modifications to these embodiments. Therefore, the appended claims are intended to include the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0052] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for extracting degradation factors from aero-engines, characterized in that, Including the following steps: S1. Based on flight data and component-level models, construct a linear influence matrix of engine degradation factors; S2. Randomly initialize the particle population, initialize the particle motion inertia, individual consciousness, group consciousness, and guide consciousness, and evaluate the speed and position of each particle to obtain the initial global optimum; S3. Based on engine characteristics, establish a linear aero-engine guide. On the basis of particle motion inertia, individual consciousness and group consciousness in the traditional particle swarm algorithm, add guide consciousness calculation, superimpose to obtain the velocity and position of each particle, then calculate the output deviation vector and evaluate the function fitness value of each particle. S4. Calculate the guide weight coefficient based on the particle's current fitness, and then calculate the particle's guide consciousness based on the linear influence matrix, output deviation vector, and guide consciousness weight coefficient. S5. Based on the function fitness value of each particle, update the historical best position of each particle, and then update the global best position of the population. S6. Verify whether the global optimal position of the population satisfies the termination condition. If not, continue iterating until the termination condition is met and exit.

2. The method for extracting degradation factors from aero-engines according to claim 1, characterized in that, Step S1 specifically includes the following steps: S11. Perturb the flow rate and efficiency degradation factor of each rotating component of the engine model respectively, and record the changes of each state quantity after the single degradation factor perturbation. S12. Construct the linear influence matrix of engine degradation factors. α : In the formula, Δ η i For the engine i The perturbation of each degradation factor, Δ x ji For the disturbance i The first degradation factor, the... j Changes in engine state.

3. The method for extracting degradation factors from aero-engines according to claim 2, characterized in that, Step S3 includes the following steps: S31, calculate the first one respectively m The first particle k Particle motion inertia in the next iteration Individual consciousness Group consciousness Guide awareness : In the formula, w It is the inertial weight, which controls the proportion of the original velocity that is retained. v It is the particle velocity; c 1 and c 2 represents the individual learning factor constant and the group learning factor constant, respectively; r 1. r 2 is a random number, introducing randomness to prevent the algorithm from getting trapped in local optima; p md and g d These are the individual particle's historical best position and the group's historical best position, respectively. x It is the particle's current position; c 3 is the guide weight coefficient for aero-engines, which changes with the fitness value of the function and is used to keep the order of magnitude of the guide consciousness consistent with the other three items; γ It is the degradation factor moderating factor, and the linear influence matrix. α Sum of deviation vectors β related; S32. Superimpose the calculated particle inertia, individual consciousness, group consciousness, and guide consciousness to obtain the velocity and position of each particle: In the formula, It is the first m Particle iteration k The speed obtained this time, It is the first m Particle iteration k The position obtained next; S33. Using the position of each particle as a degradation factor, substituting it into the component-level model for calculation, and comparing it with the sensor data to obtain the first... m The deviation vector of each particle : In the formula, e j It is the first j The absolute error of each state quantity; n It is the sample size, taken under the condition of comparable degradation levels. n The average absolute error of each sample is taken as the absolute error; S34. Calculate the fitness value based on the deviation vector. f m : In the formula, a mj It is the first m The first particle j The weighting coefficients of each state variable e mj It is the first m The first particle j The absolute error of each state variable.

4. The method for extracting degradation factors from aero-engines according to claim 3, characterized in that, Step S4 specifically includes the following steps: S41. Calculate the order-of-magnitude adjustment based on the particle velocity. r m : In the formula, r m It is the first m The adjustment factor is on the order of magnitude of the number of particles, where maxAcc is the maximum particle velocity. α It is a linear influence matrix. β The deviation vector; S42. Adjust the amount according to the order of magnitude. r m Particle fitness value f m Update the guide weight coefficient of the aircraft engine c 3: In the formula, k 1. b , f 1. f 2 is a non-zero constant; when f m < f At 1 o'clock, cancel the aircraft engine guide; when f 1< f m < f 2. The guide weight coefficient changes with the fitness value; when f m > f At time 2, the guide weight coefficient is a non-zero constant; S43. Based on the linear influence matrix α Deviation vector β Updated Guide Consciousness Weighting Coefficient c 3. Recalculate the guide's awareness v 4: v 4= c 3× α × β 。 5. A device for extracting degradation factors from an aero-engine, characterized in that, include: The linear influence matrix construction module is used to construct the linear influence matrix of engine degradation factors based on flight data and component-level models. The initial global optimum evaluation module is used to randomly initialize the particle population, initialize the particle motion inertia, individual consciousness, group consciousness, and guide consciousness, and evaluate the velocity and position of each particle to obtain the initial global optimum. The function fitness value calculation module is used to establish a linear aero-engine guide based on engine characteristics. On the basis of particle motion inertia, individual consciousness and group consciousness in the traditional particle swarm algorithm, the guide consciousness calculation is added. The velocity and position of each particle are obtained by superposition, and then the output deviation vector is calculated and the function fitness value of each particle is evaluated. The guide consciousness calculation module is used to update and calculate the guide weight coefficient based on the particle's current fitness, and then calculate the particle's guide consciousness based on the linear influence matrix, the output deviation vector, and the guide consciousness weight coefficient. The optimal position update module is used to update the historical optimal position of each particle based on the function fitness value of each particle, and then update the global optimal position of the population. The iterative verification module is used to verify whether the global optimal position of the population meets the termination condition. If it does not meet the condition, the iteration continues until the termination condition is met and the module exits.

6. The aero-engine degradation factor extraction device according to claim 5, characterized in that, The linear influence matrix construction module specifically includes: The state quantity recording module is used to perturb the flow rate and efficiency degradation factor of each rotating part of the engine model, and record the changes of each state quantity after a single degradation factor perturbation. The matrix construction module is used to construct the linear influence matrix of engine degradation factors. α : In the formula, Δ η i For the engine i The perturbation of each degradation factor, Δ x ji For the disturbance i The first degradation factor, the... j Changes in engine state.

7. The aero-engine degradation factor extraction device according to claim 6, characterized in that, The function fitness value calculation module includes: The integrated calculation module is used to calculate the first... m The first particle k Particle motion inertia in the next iteration Individual consciousness Group consciousness Guide awareness : In the formula, w It is the inertial weight, which controls the proportion of the original velocity that is retained. v It is the particle velocity; c 1 and c 2 represents the individual learning factor constant and the group learning factor constant, respectively; r 1. r 2 is a random number, introducing randomness to prevent the algorithm from getting trapped in local optima; p md and g d These are the individual particle's historical best position and the group's historical best position, respectively. x It is the particle's current position; c 3 is the guide weight coefficient for aero-engines, which changes with the fitness value of the function and is used to keep the order of magnitude of the guide consciousness consistent with the other three items; γ It is the degradation factor moderating factor, and the linear influence matrix. α Sum of deviation vectors β related; The velocity and position calculation module is used to superimpose the calculated particle motion inertia, individual consciousness, group consciousness, and guide consciousness to obtain the velocity and position of each particle. In the formula, It is the first m Particle iteration k The speed obtained this time, It is the first m Particle iteration k The position obtained next; The deviation vector calculation module is used to take the position of each particle as a degradation factor, substitute it into the component-level model for calculation, and compare it with the sensor data to obtain the first deviation vector. m The deviation vector of each particle : In the formula, e j It is the first j The absolute error of each state quantity; n It is the sample size, taken under the condition of comparable degradation levels. n The average absolute error of each sample is taken as the absolute error; The fitness value calculation module is used to calculate the fitness value based on the bias vector. f m : In the formula, a mj It is the first m The first particle j The weighting coefficients of each state variable e mj It is the first m The first particle j The absolute error of each state variable.

8. The aero-engine degradation factor extraction device according to claim 7, characterized in that, The guide awareness calculation module specifically includes: The order-of-magnitude adjustment calculation module is used to calculate the order-of-magnitude adjustment based on the particle velocity. r m : In the formula, r m It is the first m The adjustment factor is on the order of magnitude of the number of particles, where maxAcc is the maximum particle velocity. α It is a linear influence matrix. β The deviation vector; The wizard-driven weight coefficient update module is used to adjust the amount based on the order of magnitude. r m Particle fitness value f m Update the guide weight coefficient of the aircraft engine c 3: In the formula, k 1. b , f 1. f 2 is a non-zero constant; when f m < f At 1 o'clock, cancel the aircraft engine guide; when f 1< f m < f 2. The guide weight coefficient changes with the fitness value; when f m > f At time 2, the guide weight coefficient is a non-zero constant; The guide awareness update module is used to update the linear influence matrix. α Deviation vector β Updated Guide Consciousness Weighting Coefficient c 3. Recalculate the guide's awareness v 4: v 4= c 3× α × β 。 9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the aero-engine degradation factor extraction method as described in any one of claims 1 to 4.

10. A storage medium comprising a stored program, characterized in that, When the program is running, it controls the device containing the storage medium to perform the steps of the aero-engine degradation factor extraction method as described in any one of claims 1 to 4.

Images