Method for evaluating control plan robustness of aero-engine

Abstract

The invention discloses an aero-engine control plan robustness evaluation method comprising the following steps: defining an aero-engine component mathematical model, and obtaining a robust performance factor from the mathematical model; the mathematical model comprises a fan mathematical model, a press mathematical model, a high-pressure turbine mathematical model and a low-pressure turbine mathematical model; an aero-engine control plan robustness evaluation function is defined and expressed as # imgabs0 #, # imgabs1 # to Exn are robustness measurement accumulation quantities of the key performance indexes under perturbation of different performance factors, and alpha 1, alpha 2,..., alpha n are proportions of robustness measurement indexes of the key performance indexes; and defining a robustness evaluation index, and according to the value Q key of the evaluation function, determining whether the robustness of the aero-engine control plan meets a design standard. According to the technical scheme, complete control plan multi-index robustness calculation can be formed so as to realize evaluation of engine control plan robustness.

Description

Technical Field

[0001] The present invention relates to the technical field of aircraft engine control, and in particular to a method for evaluating the robustness of an aircraft engine control plan. Background Art

[0002] Aircraft engines will face various complex flight conditions in actual flight. In order to ensure that aircraft engines can operate safely under different conditions, parameter control rules need to be formulated so that aircraft engines can achieve expected functions and performance in complex environments. These parameter control rules constitute the aircraft engine control plan. It can be seen that the aircraft engine control plan is the top-level design requirement that aircraft engines need to meet during operation. The steady-state and transitional performance of the engine operation are largely determined by the control plan. According to the environmental parameters of the engine flight, flight operation instructions and the current operating status of the engine, the engine control plan gives the expected value of the engine control parameters, that is, the expected aircraft engine performance. An excellent aircraft engine control plan should be able to give full play to the performance of the engine and ensure that the engine has high robustness throughout its life cycle.

[0003] Therefore, an objective technical solution that is closely related to the status of important components of aircraft engines is needed to evaluate the robustness of aircraft engine control plans. Summary of the invention

[0004] To achieve the above objectives, the present application provides a method for evaluating the robustness of an aircraft engine control plan, comprising the following steps:

[0005] Define the mathematical model of aircraft engine components and obtain the robust performance factor from the mathematical model; the mathematical model includes: fan mathematical model, press mathematical model, high-pressure turbine mathematical model and low-pressure turbine mathematical model; the robust performance factor includes: fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W a3 , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT ;

[0006] The robustness evaluation function of the aircraft engine control plan is defined as:

[0007] in, To E xnis the accumulated robustness measure of the key performance indicator under the perturbation of the robust performance factor, α1, α2, …, α n The proportion of robustness metrics of key performance indicators;

[0008] The robustness evaluation index is defined, and the robustness of the aircraft engine control plan is judged according to the key value Q of the evaluation function to determine whether it meets the design standards.

[0009] Among them, the input parameter of the fan mathematical model is the fan boost ratio π F , low pressure rotor physical speed n L , total temperature of fan component inlet T t2 , total pressure P t2 ;

[0010] According to the relative conversion speed of the fan and fan pressure ratio π F =P t21 / P t2 , obtain the fan flow health parameter W from the fan component characteristic data aF,cor , fan efficiency health parameter η F , expressed as: Among them, P t21 is the total pressure at the fan outlet, T t2d is the total temperature of the fan component inlet at the design point, n Ld is the physical speed of the low-pressure rotor at the design point.

[0011] The input parameters of the mathematical model of the press are: compressor pressure ratio π C , high pressure rotor physical speed n H , compressor component inlet total temperature T t22 , total pressure P t22 ; Among them, P t22 =P t21 , T t22 =T t21 , T t21 is: total temperature at fan outlet, P t21 is: total pressure at fan outlet;

[0012] Relative conversion speed according to compressor Compressor pressure ratio π C =P t3 / P t22 , the compressor flow health parameter W is obtained from the compressor component characteristic data a3 , compressor efficiency health parameter η C , expressed as:

[0013] Among them, P t3 is the total pressure at the compressor outlet.

[0014] The input parameter of the high-pressure turbine mathematical model is the high-pressure turbine pressure ratio π C , high pressure rotor physical speed n H , High-pressure turbine component inlet total temperature T t4 , total pressure P t4 ;

[0015] Relative conversion speed according to high pressure turbine and high pressure turbine pressure drop ratio π C =P t41 / P t4 , obtain the high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , expressed as:

[0016] Where P t41 is the total pressure at the outlet of the high-pressure turbine component, T t41 is the total temperature at the outlet of the high-pressure turbine components, T t41d is the total temperature at the inlet of the high-pressure turbine components at the design point, n Hd is the physical speed of the high pressure rotor at the design point.

[0017] The input parameters of the low-pressure turbine mathematical model include: low-pressure turbine pressure ratio π LT ;

[0018] Low-pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT , expressed as:

[0019]

[0020] The input parameters of the mathematical model of the low-pressure turbine include the physical speed n L , total pressure P t45 , inlet total temperature T t45 ;

[0021] The output parameters of the low-pressure turbine mathematical model are: low-pressure turbine outlet flow rate W 55 , total pressure P t5 , total temperature T t5 , Power N LT ;

[0022] There is a transition section at the outlet of the low-pressure turbine. When the airflow passes through the transition section, a total pressure loss is generated. The transition section can be expressed as: W 55 =W5,T 55 =T5,P t55 =P t5 σ LT ; In the formula, σ LT is the total pressure loss coefficient of the transition section, Pt55 is the total pressure at the low-pressure turbine inlet, T 55 is the total temperature at the high-pressure turbine inlet, and W5 is the low-pressure turbine inlet flow rate.

[0023] Furthermore, before defining the robustness evaluation function of the aircraft engine control plan, the robustness metric accumulation of the key performance indicators under the perturbation of the robust performance factor is obtained, including:

[0024] Performing data sampling to obtain random perturbation combination samples of the key performance indicators;

[0025] According to the random perturbation combination sample, the dimensionless deviation caused by the fluctuation of the key performance indicators is calculated, which is expressed as: Wherein, n is the number of selected robust performance factors, i is the robust performance factor number, and i<=n; per is the simulation output when the robust performance factor produces uncertainty fluctuations; norm is the simulation output when the robust performance factor does not produce uncertainty fluctuations;

[0026] The calculation method of the robustness metric accumulation is defined as: in, j=1,2,...,N, N is the number of random combinations of the robust performance factor perturbation Δ, j is the combination sequence number, and j<=N.

[0027] Wherein, obtaining the random perturbation combination sample of the key performance indicator includes:

[0028] Determine the state of the Markov chain, expressed as: Δ: = {Δ1, Δ2, ..., Δ8} T , where Δ1, Δ2, ..., Δ8 are respectively: fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W a3 , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT perturbation;

[0029] The Markov chain at the i-th moment is expressed as: Δ (i) :={Δ1 (i) ,Δ2 (i) ,...,Δ8 (i)} T ;

[0030] Define the random variable Δ j The probability density function of is: in, is the random variable Δ j The probability density function of

[0031] Data sampling is performed to obtain random perturbation combination samples of the key performance indicators, which are expressed as:

[0032] According to the present invention, a robust performance factor can be defined based on the changes in indicators when important components of an aircraft engine, such as a fan, a compressor, and a turbine, are degraded. By analyzing each robust performance factor, the degree of degradation of the important components is combined with the engine control plan to define the degree of influence. On this basis, an engine robustness evaluation function is established to form a complete control plan robustness calculation to achieve an evaluation of the robustness of the engine control plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 is a step diagram of a method for evaluating robustness of an aircraft engine control plan provided in accordance with an embodiment of the present invention;

[0034] Figure 2 It is a data logic diagram of an evaluation method for robustness of an aircraft engine control plan provided in an embodiment of the present invention;

[0035] Figure 3 is a flow chart of the MH sampling algorithm provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The aircraft engine control plan is the established planning target that the key performance indicators of the aircraft engine should show during the entire flight process. In other words, the key performance indicators of the engine when working according to the corresponding control plan can evaluate the rationality of the plan. Due to the influence of performance factors such as incoming flow conditions, individual differences, and performance degradation, even if the aircraft engine works along a specific control plan, the key performance indicators will deviate. In the present invention, the deviations of these key performance indicators are used as engine robust performance factors, and the key performance indicators are used as quantitative measurement indicators of the robustness of the aircraft engine control plan.

[0037] In actual operation, aircraft engines may face uncertain factors of engine internal degradation, and there are differences between these parameters, which are interconnected and mutually restricted. These uncertainties lead to the uncertainty of aircraft engine control plans. The traditional robustness analysis method is to directly add uncertainty factors into the system for simulation. However, due to the excessive number of uncertain performance factors faced by the transition state control plan during operation, there are also many classification combinations of various factors, which are difficult to exhaust. Therefore, in the present invention, the Markov Chain Monte Carlo (MCMC) method is adopted to simulate the perturbation of uncertain performance factors faced by the engine during flight, so as to evaluate the influence of performance factor perturbations such as engine individual differences and performance degradation on the key indicators of the engine control plan: by analyzing and calculating the robustness quantitative measurement indicators of the key performance indicators of the control plan, the formulation of the robustness evaluation function of the control plan is completed. On the basis of establishing a complete robustness analysis method for the aircraft engine control plan, a robustness evaluation standard is provided for the engine control plan to evaluate the robust performance of the engine control plan.

[0038] The Markov Chain Monte Carlo (MCMC) method is a random simulation method. The MCMC method theory is based on the following Bayesian theory:

[0039]

[0040] Among them, P(θ) represents the prior probability distribution, f(θ|X) represents the posterior distribution of the parameter, and f(X|θ) represents the likelihood function, which can represent the similarity between the assumed prior probability distribution and the posterior distribution.

[0041] When the prior distribution and likelihood function are known, the posterior distribution of the parameter can be calculated. However, in general, the distribution of the prior distribution and the likelihood function are relatively complex, which makes the posterior distribution very complex and difficult to calculate directly. Therefore, the MCMC method is used to solve the situation where the posterior distribution is difficult to calculate or difficult to obtain through simple random sampling.

[0042] Markov chain Monte Carlo method is widely used in sampling of higher-dimensional distribution functions. The main idea of ​​using Markov chain Monte Carlo method for sampling is to construct a Markov chain that satisfies the stationary distribution condition. The starting state of the Markov chain can be arbitrary, but the stationary distribution of the Markov chain is required to be the target distribution p(x) of sampling. By using the Monte Carlo method to complete sample sampling, sample data that satisfies the target distribution p(x) can be obtained. When the Markov chain reaches a stable convergence state, it can be considered that the subsequent sampling results all meet the target distribution conditions.

[0043] The transfer matrix affects the convergence properties of the Markov chain. If we want to obtain the target stationary distribution π, we only need to find the corresponding transfer matrix P. However, generally speaking, for a target stationary distribution π, if we randomly obtain a transfer matrix Q, the stationary and detailed conditions are difficult to meet, that is,

[0044] π i Q ij ≠π j Q ji

[0045] To solve this problem, we introduce the acceptance rate and get

[0046] π i Q ij α(i,j)≠π j Q ji α(j,i)

[0047] If α satisfies

[0048] α(i,j)=π j Q ji

[0049] α(j,i)=π i Q ij

[0050] At this time there is

[0051] P ij =α(i,j)Q ij

[0052] The above formula shows that the solution process changes from finding the target transfer matrix P to constructing a random transfer matrix Q and acceptance rate α.

[0053] The specific implementation of the present invention is described in detail below with reference to the accompanying drawings.

[0054] Figure 1 A method step diagram for evaluating the robustness of an aircraft engine control plan by the present invention is provided, comprising:

[0055] S100: Define mathematical models of aircraft engine components and obtain robust performance factors from the mathematical models;

[0056] The present invention aims at the impact on the engine transient control plan when the fan, compressor and turbine of the aircraft engine are degraded, so the mathematical model constructed includes: a fan mathematical model, a compressor mathematical model, a high-pressure turbine mathematical model and a low-pressure turbine mathematical model;

[0057] 1) Fan mathematical model:

[0058] In actual applications, the fan component is located on the low-pressure rotor of the engine. The fan compresses the air coming in from the inlet duct. The compressed air is divided into two streams. The airflow at the blade root enters the compressor component for further compression, and the airflow at the blade tip enters the outer duct.

[0059] The input parameter of the fan mathematical model is the fan pressure ratio π F , low pressure rotor physical speed n L , total temperature of fan component inlet T t2 , total pressure P t2 ;

[0060] In the fan mathematical model, according to the relative conversion speed of the fan and fan pressure ratio π F =P t21 / P t2 , obtain the fan flow health parameter W from the fan component characteristic data aF,cor , fan efficiency health parameter η F , expressed as: Among them, P t21 is the total pressure at the fan outlet, T t2d is the total temperature of the fan component inlet at the design point, n Ld is the physical speed of the low-pressure rotor at the design point;

[0061] Among them, the fan outlet flow rate W a21 and fan inlet flow W a2 Equal, that is:

[0062] W a21 =W a2 ,and C mf is the correction factor;

[0063] Then the fan efficiency health parameter η F =C ηF η LF,cor , where C ηF To correct the parameters.

[0064] 2) Press mathematical model:

[0065] In actual applications, the compressor components are driven by the high-pressure turbine components, and their function is to further compress the air so that the high-temperature, high-pressure gas flows into the combustion chamber.

[0066] The input parameters of the mathematical model of the press are: compressor pressure ratio π C , high pressure rotor physical speed n H , compressor component inlet total temperature T t22 , total pressure P t22 ;

[0067] Among them, P t22 =P t21 , T t22 =T t21 , T t21 is the total temperature at the fan outlet, P t21 is the total pressure at the fan outlet;

[0068] In the mathematical model of the press, the relative conversion speed of the compressor is calculated based on Compressor pressure ratio π C =P t3 / P t22 , the compressor flow health parameter W is obtained from the compressor component characteristic data a3 , compressor efficiency health parameter η C , expressed as:

[0069] Among them, P t3 is the total pressure at the compressor outlet;

[0070] And the compressor inlet flow W a22 and compressor flow health parameter W a3 Equal, that is: W a3 =W a22 ;

[0071] but:

[0072] The compressor efficiency health parameter is the actual compressor efficiency, expressed as: η C =C ηC n HC,cor . Among them, C ηC To correct the parameters.

[0073] 3) High-pressure turbine mathematical model

[0074] In aircraft engines, the main function of the high-pressure turbine components is to convert the internal energy of the gas into mechanical energy. The high-temperature and high-pressure gas after the combustion chamber expands and does work in the turbine components, driving the compression components in front to rotate.

[0075] The input parameter of the high-pressure turbine mathematical model is the high-pressure turbine pressure ratio π C , high pressure rotor physical speed n H , High-pressure turbine component inlet total temperature T t4 , total pressure P t4 ;

[0076] In the high-pressure turbine mathematical model, the high-pressure turbine relative conversion speed is calculated based on and high pressure turbine pressure drop ratio π C =P t41 / P t4, obtain the high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , expressed as: Where P t41 is the total pressure at the outlet of the high-pressure turbine component, T t41 is the total temperature at the outlet of the high-pressure turbine components, T t41d is the total temperature at the inlet of the high-pressure turbine components at the design point, n Hd is the physical speed of the high-pressure rotor at the design point;

[0077] High pressure turbine inlet flow rate W g4 The relationship with the high-pressure turbine flow health parameter is:

[0078]

[0079] The high-pressure turbine inlet flow is equal to the high-pressure turbine outlet flow, that is, W a44 =W a41 , W a41 is the high-pressure turbine inlet flow rate, W a44 is the high-pressure turbine inlet flow rate; then: high-pressure turbine efficiency health parameter η HT The algorithm is: HT =C HT n HT,cor , where: C HT To correct the parameters.

[0080] 4) Low-pressure turbine mathematical model:

[0081] The low-pressure turbine components adopt a modeling method similar to that of the high-pressure turbine components. The input parameters of the low-pressure turbine mathematical model include: LT , physical speed n L , total pressure P t45 , inlet total temperature T t45 ;

[0082] The output parameters of the low-pressure turbine mathematical model include: low-pressure turbine outlet flow rate W 55 , total pressure P t5 , total temperature T t5 , Power N LT ;

[0083] The difference from the high-pressure turbine is that there is a transition section at the outlet of the low-pressure turbine, and the total pressure loss will be generated when the airflow passes through. The transition section can be expressed as: W 55 =W5,T 55 =T5,

[0084] P t55 =P t5 σ LT ; In the formula, σ LTis the total pressure loss coefficient of the transition section, P t55 is the total pressure at the low-pressure turbine inlet, T 55 is the total temperature at the high-pressure turbine inlet, and W5 is the low-pressure turbine inlet flow rate.

[0085] Low-pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT , expressed as:

[0086]

[0087] Among them, the relationship between the low-pressure turbine inlet flow rate W5 and the low-pressure turbine flow rate health parameter is:

[0088]

[0089] Low pressure turbine outlet flow W a55 Equal to the low-pressure turbine inlet flow rate W5, that is: W a55 =W5;

[0090] Low-pressure turbine efficiency health parameter η LT The algorithm is: LT =C LT n LT,cor , where: C LT To correct the parameters.

[0091] Through the above mathematical model, the robust performance factor is constructed, including: fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W aC,cor , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT .

[0092] S110: Define the robustness evaluation function of the aircraft engine control plan:

[0093] In this step, first, the robustness metric accumulation of the key performance indicator under different robust performance factor perturbations is obtained, including:

[0094] Step S111: performing data sampling to obtain random perturbation combination samples of the key performance indicators;

[0095] The degradation of the fan, compressor and turbine of an aircraft engine affects the engine transition state control plan. The deviation of the key performance indicator means that the aircraft engine control plan is affected by the robust performance factor of the engine. In the present invention, the state of the Markov chain is defined as represented by an 8-dimensional random variable, that is, it has 8 robust performance factors.

[0096] Specifically, before data sampling, the process of obtaining the random perturbation combination sample of the key performance indicator includes:

[0097] Determine the state of the Markov chain, expressed as: Δ: = {Δ1, Δ2, ..., Δ8} T , where Δ1, Δ2, ..., Δ8 are respectively: fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W a3 , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT perturbation;

[0098] The Markov chain at the i-th moment is the distribution of the performance index at the i-th moment, expressed as:

[0099] Δ (i) :={Δ1 (i) ,Δ2 (i) ,...,Δ8 (i)} T , where Δ j (i) is the jth component of the robust performance factor at the i-th moment;

[0100] Define the random variable Δ j The probability density function of is: in, is the random variable Δ j The probability density function of

[0101] During initialization, select the initial state point Δ 0 =0, which means that the engine is not degraded at this time and all components are in normal operation. By sampling with the MCMC method, we can obtain random perturbation combination samples of multiple groups of performance factors, expressed as: Sampling methods such as Figure 3 shown.

[0102] The key performance index of the aircraft engine transient control plan is X: = {x1, x2, ... x n}, such as thrust linearity, transition state adjustment time, rotor speed, turbine inlet temperature, compressor inlet pressure, compressor surge margin and other parameters, as proposed in the present invention, can be used to reflect the performance output of the engine. When the aircraft engine fluctuates due to the performance factor Δ:={Δ1, Δ2, ..., Δ8}, the fluctuation of the key performance indicators of the aircraft engine control plan in the simulation can be used to calculate the robustness quantitative measurement index of a certain key performance indicator X, as shown in step S112.

[0103] Step S112: Calculate the dimensionless deviation caused by the fluctuation of the key performance indicator according to the random perturbation combination sample, expressed as: Wherein, n is the number of selected robust performance factors, i is the robust performance factor number, and i<=n; per is the simulation output when the performance factor produces uncertainty fluctuations; norm is the simulation output when the performance factor does not produce uncertainty fluctuations;

[0104] According to the calculation method of dimensionless deviation, each performance index of the aircraft engine control plan can be calculated to obtain the corresponding robustness quantitative measurement result, that is, dimensionless deviation. The calculation needs to consider the perturbation combination of performance factors that conform to their own distribution, and analyze the perturbation of the key performance indicators of the engine control plan under different perturbation combinations, and calculate the accumulation of robustness measurement indicators of advanced control algorithms. This robustness metric can be used to analyze the impact of performance factors.

[0105] Step S113: define the calculation method of the robustness metric accumulation amount as:

[0106] in, j=1,2,...,N, N is the number of random combinations of the robust performance factor perturbation Δ, j is the combination sequence number, and j<=N.

[0107] At this point, the robustness evaluation of the aircraft engine control plan can be performed.

[0108] According to the combination form of uncertainty parameter perturbation that satisfies the probability distribution, a quantitative measurement method for the robustness of the control plan based on key performance indicators is given. However, the robustness quantitative measurement result at this time is only a single performance indicator for the transition state control plan, and cannot be directly used as a quantitative measurement indicator for the overall robustness of the control plan. Therefore, in this step, a unified evaluation function is used to evaluate and analyze the robust performance of the control plan.

[0109] Based on this, the robustness evaluation function of the aircraft engine control plan is defined as:

[0110] in, To E xn is the robustness metric accumulation of the key performance indicators under different robust performance factor perturbations, α1, α2, …, α n It is the proportion of the robustness measurement index of the key performance indicator, that is, the weight.

[0111] The selection of weights is affected by the control plan objectives. The influence of each key performance indicator of the control plan is comprehensively considered and can be incorporated into the evaluation function to build a multi-indicator and multi-dimensional robustness comprehensive evaluation system. It should be noted that the weights should conform to: α1+α2+…+α n =1.

[0112] Among them, the weight of a key performance indicator reflects the importance of the indicator in the entire control plan. For example: for the thrust control plan, thrust linearity is an important robust performance indicator, and its corresponding robustness measurement weight accounts for the main part in the control plan evaluation function; and for the transition state control plan, the key parameter of the transition state adjustment time needs to be paid special attention to, and the corresponding weight coefficient also needs to be adjusted upward. Of course, for different control plan requirements, or different specific aircraft engine objects, it is necessary to comprehensively consider the selection of weights for the control plan robustness evaluation function.

[0113] S120: Define a robustness evaluation index, and determine whether the robustness of the aircraft engine control plan meets the design standards based on the value Q of the evaluation function.

[0114] After the quantitative index system of control plan robustness is established, the robustness of aircraft engine control can be analyzed: that is, the robust performance of the control plan can be evaluated by the size of the evaluation function Q.

[0115] For different control plans, the robustness evaluation results obtained according to the multi-index robustness analysis method provided by the present invention are different, which is reflected in the robustness evaluation function Q, that is, the better the robustness performance, the smaller the value of Q; the worse the robustness, the larger the value of Q. Therefore, the robustness evaluation index Q is defined as std , when Q is less than Q std When , the robustness of the aircraft engine control plan meets the design standards.

[0116] In general, the robustness evaluation process of the aircraft engine control plan proposed by the present invention is as follows: Figure 2 As shown: Combined with the mathematical model of aircraft engine components, data sampling is carried out on the aircraft engine, that is, multiple groups of random perturbation combinations of performance factors based on the Markov Chain Monte Carlo method are collected, and on this basis, quantitative robustness measurement is carried out to generate a robustness evaluation function.

[0117] In the present invention, the changes in indicators when important components such as the fan, compressor, and turbine of an aircraft engine degrade affect the engine's transitional control plan. The Markov chain Monte Carlo method is introduced to randomly sample the engine's performance factors, and obtain an uncertainty random combination sequence that satisfies the probability distribution to simulate the engine's operating conditions. The robustness measurement index of the control plan is obtained through simulation calculation, and the degree of influence of each performance factor on the engine control plan is analyzed. On this basis, the robustness evaluation function of the engine is established, forming a complete multi-index robustness analysis method for the control plan, so as to achieve the purpose of evaluating the robustness of the complex nonlinear object of the engine control plan.

[0118] The above disclosures are only several specific embodiments of the present invention; however, the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A method for evaluating the robustness of an aircraft engine control plan, characterized in that: The following steps are involved: defining a mathematical model of an aircraft engine component and obtaining a robust performance factor from the mathematical model; The mathematical model includes: a fan mathematical model, a press mathematical model, a high-pressure turbine mathematical model and a low-pressure turbine mathematical model; the robust performance factor includes: a fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W a3 , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT ; The robustness evaluation function of the aircraft engine control plan is defined as: in, To E xn is the accumulated robustness measure of the key performance indicator under the perturbation of the robust performance factor, α1, α2, …, α n The proportion of robustness metrics of key performance indicators; A robustness evaluation index is defined, and whether the robustness of the aircraft engine control plan meets the design standard is judged according to the key value Q of the evaluation function.

2. The evaluation method according to claim 1, characterized in that: The input parameter of the fan mathematical model is the fan pressure ratio π F , low pressure rotor physical speed n L , total temperature at the fan component inlet T t2 , total pressure P t2 ; According to the relative conversion speed of the fan and fan pressure ratio π F =P t21 / P t2 , obtain the fan flow health parameter W from the fan component characteristic data aF,cor , fan efficiency health parameter η F , expressed as: Among them, P t21 is the total pressure at the fan outlet, T t2d is the total temperature of the fan component inlet at the design point, n Ld is the physical speed of the low-pressure rotor at the design point.

3. The evaluation method according to claim 1, characterized in that: The input parameters of the mathematical model of the press are: compressor pressure ratio π C , high pressure rotor physical speed n H , compressor component inlet total temperature T t22 , total pressure P t22 ; Among them, P t22 =P t21 , T t22 =T t21 , T t21 is: total temperature at fan outlet, P t21 is: total pressure at fan outlet; Relative conversion speed according to compressor Compressor pressure ratio π C =P t3 / P t22 , the compressor flow health parameter W is obtained from the compressor component characteristic data a3 , compressor efficiency health parameter η C , expressed as: Among them, P t3 is the total pressure at the compressor outlet.

4. The evaluation method according to claim 1, characterized in that: The input parameter of the high-pressure turbine mathematical model is the high-pressure turbine pressure drop ratio π C , high pressure rotor physical speed n H , High-pressure turbine component inlet total temperature T t4 , total pressure P t4 ; Relative conversion speed according to high pressure turbine and high pressure turbine pressure drop ratio π C =P t41 / P t4 , obtain the high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , expressed as: Where P t41 is the total pressure at the outlet of the high-pressure turbine component, T t41 is the total temperature at the outlet of the high-pressure turbine components, T t41d is the total temperature at the inlet of the high-pressure turbine components at the design point, n Hd is the physical speed of the high pressure rotor at the design point.

5. The evaluation method according to claim 4, characterized in that: The input parameters of the low-pressure turbine mathematical model also include: low-pressure turbine drop pressure ratio π LT ; The low-pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT , expressed as:

6. The evaluation method according to claim 5, characterized in that: The input parameters of the low-pressure turbine mathematical model are: L , total pressure P t45 , inlet total temperature T t45 ; The output parameters of the low-pressure turbine mathematical model are: low-pressure turbine outlet flow rate W 55 , total pressure P t5 , total temperature T t5 , Power N LT ; There is a transition section at the outlet of the low-pressure turbine. When the airflow passes through the transition section, a total pressure loss is generated. The transition section can be expressed as: W 55 =W5,T 55 =T5,P t55 =P t5 σ LT ; In the formula, σ LT is the total pressure loss coefficient of the transition section, P t55 is the total pressure at the low-pressure turbine inlet, T 55 is the total temperature at the high-pressure turbine inlet, and W5 is the low-pressure turbine inlet flow rate.

7. The evaluation method according to claim 1, characterized in that: Before defining the robustness evaluation function of the aircraft engine control plan, the robustness metric accumulation of the key performance indicator under the perturbation of the robust performance factor is obtained, including: Performing data sampling to obtain random perturbation combination samples of the key performance indicators; According to the random perturbation combination sample, the dimensionless deviation caused by the fluctuation of the key performance indicators is calculated, which is expressed as: Wherein, n is the number of selected robust performance factors, i is the robust performance factor number, and i<=n; per is the simulation output when the robust performance factor produces uncertainty fluctuations; norm is the simulation output when the robust performance factor does not produce uncertainty fluctuations; The calculation method of the robustness metric accumulation is defined as: in, N is the number of random combinations of the robust performance factor perturbation Δ, j is the combination sequence number, and j<=N.

8. The evaluation method according to claim 7, characterized in that: The step of obtaining the random perturbation combination sample of the key performance indicator comprises: Determine the state of the Markov chain, expressed as: Δ: = {Δ1, Δ2, ..., Δ8} T , where Δ1, Δ2, ..., Δ8 are respectively: fan flow health parameter W aF,cor , Fan efficiency health parameter η F , compressor flow health parameter W a3 , compressor efficiency health parameter η C , high pressure turbine flow health parameter W gHT,cor , high pressure turbine efficiency health parameter η HT , low pressure turbine flow health parameter W LT,cor and the low-pressure turbine efficiency health parameter η LT perturbation; The Markov chain at the i-th moment is expressed as: Δ (i) :={Δ1 (i) ,Δ2 (i) ,…,Δ8 (i) } T ; Define the random variable Δ j The probability density function of is: in, is the random variable Δ j The probability density function of Data sampling is performed to obtain random perturbation combination samples of the key performance indicators, which are expressed as:

Images