A method for evaluating the overall stability margin of an aero gas turbine engine

Abstract

The present invention relates to the technical field of aeroengines, and discloses a method for evaluating the overall stability margin of an aero-gas turbine engine. An engine overall model is constructed by analyzing the physical characteristics and working principles of each component of the gas turbine engine. An engine test bench is built for experiments to determine whether the compressor enters the surge state and record the experimental data. Numerical simulation is carried out based on the constructed engine overall model and the simulation data is recorded. The simulation data is compared with the experimental data to verify the accuracy of the model. If accurate, a surge boundary is constructed; if not, the model is optimized. It is judged in real time whether the engine has a stall hysteresis phenomenon, and a normal operation signal or a stall hysteresis signal is generated. If the generated stall hysteresis signal exists, the surge boundary is updated. The stability margin value of the engine is calculated based on the surge boundary, and the overall stability margin of the engine is evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of aeroengines, and particularly relates to a method for evaluating the overall stability margin of an aero-gas turbine engine. Background Art

[0002] In the aviation field, as the core power device of an aircraft, the performance of an aero-gas turbine engine directly affects flight safety and efficiency. The stable operation of the engine is one of the key factors to ensure the reliable flight of the aircraft, and the stability margin is an important indicator to measure the engine's ability to maintain stable operation under different working conditions. Accurately evaluating the overall stability margin of an aero-gas turbine engine is of great significance for improving engine performance and ensuring flight safety.

[0003] Constructing an overall engine model is the basis for evaluating the stability margin. In the prior art, the modeling method is limited by the limitations in understanding the engine interior, as well as the inaccuracy and incompleteness of the modeling data, and cannot accurately reflect the true surge boundary of the engine, thereby affecting the accuracy of the stability margin evaluation.

[0004] On the other hand, during the operation of the engine, the stall hysteresis phenomenon has strong concealment and complexity. Existing monitoring technical means are difficult to capture the occurrence of the stall hysteresis phenomenon in real time and accurately, and it is difficult to quickly and accurately identify the stall hysteresis characteristics from a large amount of monitoring data, and an effective stall hysteresis signal cannot be generated in time, resulting in the engine not being processed in time and effectively when the stall hysteresis occurs, increasing the safety risk of the engine operation.

[0005] The operating conditions of an aero-gas turbine engine will change significantly under different environmental conditions. Most of the existing stability margin evaluation methods are designed based on static working conditions or a limited range of working conditions, lacking the ability to track and adapt to the dynamic changes of the engine operating conditions in real time, resulting in the disconnection between the evaluation results and the actual stable operation ability of the engine, and unable to provide strong guarantee for the reliable operation of the engine.

[0006] In view of the above problems, the present invention proposes a method for evaluating the overall stability margin of an aero-gas turbine engine. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for evaluating the overall stability margin of an aero-gas turbine engine to solve at least one of the above-mentioned prior art problems.

[0008] The present invention provides a method for evaluating the overall stability margin of an aero-gas turbine engine, including the following steps:

[0009] Construct an overall engine model;

[0010] Build an engine test bench for experiments, and at the same time conduct numerical simulations based on the engine's overall model. Record the experimental and simulation data when the compressor enters the surge state and compare them to verify the accuracy of the model. If it is accurate, construct the surge boundary based on the model; if not, optimize the model.

[0011] Specifically, during the experiment, reduce the intake air flow at a fixed step size.

[0012] Obtain the amplitude of pressure fluctuations during the process of reducing the intake air flow. If the amplitude of pressure fluctuations is greater than or equal to the pressure fluctuation threshold, it is determined that the compressor has entered the surge state. Record the experimental intake air flow of the compressor, and collect the outlet pressure value and inlet pressure value of the compressor and perform data processing to obtain the experimental surge pressure ratio.

[0013] Conduct numerical simulations based on the engine's overall model, gradually reduce the intake air flow at the same step size as the experiment, and calculate the Lyapunov exponent inside the compressor during the process of reducing the intake air flow. If the Lyapunov exponent is greater than 0, it is determined that the compressor has entered the surge state, and record the simulated intake air flow of the compressor and the calculated simulated surge pressure ratio.

[0014] Compare and analyze the experimental intake air flow and experimental surge pressure ratio obtained from the experiment with the simulated intake air flow and simulated surge pressure ratio obtained from the simulation, and calculate the relative error of the intake air flow and the relative error of the pressure ratio.

[0015] If the relative error of the intake air flow is less than or equal to the intake air flow error threshold, and the relative error of the pressure ratio is less than or equal to the pressure ratio error threshold, generate a simulation accurate signal.

[0016] Conduct experiments and simulations on the engine under several different working conditions, obtain the number of times the simulation accurate signal is generated and perform data processing to obtain the matching ratio of the engine's overall model.

[0017] If the matching ratio is greater than or equal to the matching ratio threshold, generate a model matching signal; otherwise, generate a model deviation signal.

[0018] If a model matching signal is generated, establish a mathematical model of the engine's surge boundary through the analysis of the simulation results under different working conditions. According to the multiple regression analysis method, fit a large number of simulation results to obtain the analytical expression of the surge boundary.

[0019] Draw the surge boundary on the characteristic curve according to the analytical expression.

[0020] Analyze in real time whether the engine has a stall hysteresis phenomenon. If it appears, generate a stall hysteresis signal; otherwise, generate a normal operation signal, and calculate the stable margin value of the engine's normal operation.

[0021] Specifically, during the actual operation cycle of the engine, the pressure difference between the inlet and outlet of the compressor is obtained in real time;

[0022] Perform a fast Fourier transform on the obtained pressure difference between the inlet and outlet of the compressor to convert the time-domain signal into a frequency-domain signal, and obtain the frequency-domain amplitude of the pressure difference;

[0023] If there is any discrete frequency point where the frequency-domain amplitude is greater than or equal to the frequency-domain amplitude threshold, generate a suspected stall hysteresis signal;

[0024] If a suspected stall hysteresis signal is received, obtain the response time of the intake air flow to the throttle command. If the response time is less than the response time threshold, generate a normal operation signal; otherwise, generate a stall hysteresis signal;

[0025] If a normal operation signal is received, obtain the real-time pressure ratio and intake air flow of the engine, and combine them to obtain the real-time measurement point at the current measurement time point on the intake air flow-pressure ratio plane;

[0026] Obtain the pressure ratio corresponding to the same intake air flow on the surge boundary and perform data processing to calculate the stability margin value of the engine under normal operation;

[0027] If a stall hysteresis signal is generated, update the surge boundary, and calculate the stability margin value of the engine under the stall hysteresis phenomenon based on the updated surge boundary;

[0028] Specifically, if a stall hysteresis signal is received, extract the pressure difference between the inlet and outlet of the compressor within the Δt time window before and after the generation of the stall hysteresis signal during the actual operation cycle of the engine, perform data analysis and processing, and expand the analytical expression of the surge boundary into a time-varying form to obtain the corrected surge boundary;

[0029] Perform data processing based on the corrected surge boundary and the real-time measurement point to calculate the stability margin value of the engine under the stall hysteresis phenomenon;

[0030] Evaluate the overall stability margin of the engine based on the stability margin value.

[0031] Advantages of the present invention:

[0032] 1. By combining CFD numerical simulation and component-level mathematical models, the present invention constructs a high-precision overall engine model, improving the accuracy of stability margin evaluation. The comparison and verification between bench tests and numerical simulations ensure the reliability of the model, laying a solid foundation for subsequent evaluation work. Based on real-time measurement and analysis technologies, such as fast Fourier transform analysis of pressure difference, intake air flow, and frequency-domain characteristics, the stall hysteresis phenomenon of the engine can be detected in a timely manner, improving the real-time performance and accuracy of the evaluation.

[0033] 2. The present invention dynamically updates the surge boundary through CFD transient simulation and the adaptive Kalman filtering algorithm, making the evaluation method more adaptable to the actual operating conditions of the engine, improving the adaptability and robustness of the stable margin evaluation. Finally, based on the evaluation results of the stable margin values, safety signals, warning signals or danger signals can be generated, providing a strong guarantee for the safe operation of the engine, reducing the operating risk, and improving the safety and reliability of aviation flight. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0035] Figure 1 is a flowchart of a method for evaluating the overall stable margin of an aero-gas turbine engine provided in Embodiment 1 of the present invention;

[0036] Figure 2 is a flowchart of the steps for obtaining the model matching signal and the model deviation signal in the method for evaluating the overall stable margin of an aero-gas turbine engine provided in Embodiment 1 of the present invention;

[0037] Figure 3 is a flowchart of the steps for obtaining the stable margin value of the engine in the normal state in the method for evaluating the overall stable margin of an aero-gas turbine engine provided in Embodiment 1 of the present invention;

[0038] Figure 4 is a program block diagram of an overall stable margin evaluation system for an aero-gas turbine engine provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] To enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0040] Embodiment 1: As Figure 1 shown, a method for evaluating the overall stable margin of an aero-gas turbine engine provided in an embodiment of the present invention specifically includes the following steps:

[0041] S1. By analyzing the physical characteristics and working principles of the components of a gas turbine engine, combined with CFD numerical simulation, component-level mathematical models of each component are constructed. These component-level mathematical models are integrated into an engine overall model, and simulation calculations are realized through the MATLAB platform;

[0042] In some embodiments, according to the physical characteristics and working principles of the components of a gas turbine engine, corresponding component-level mathematical models are established. The components of a gas turbine engine include a compressor, a combustion chamber, a turbine, etc.;

[0043] Specifically, based on the one-dimensional unsteady flow theory, the characteristic line method is used to solve the partial differential equations describing the internal flow of the compressor. Combining with the geometric parameters of the compressor, the geometric parameters include blade shape, number of stages, tip clearance, etc., a component-level mathematical model of the compressor is constructed, and the component-level mathematical model of the compressor is corrected using the CFD numerical simulation results;

[0044] Based on the chemical reaction kinetics mechanism, combined with CFD simulation to analyze the three-dimensional flow field in the combustion chamber, considering the fuel injection, atomization, mixing and combustion processes, by solving the energy conservation, mass conservation and momentum conservation equations, a component-level mathematical model of the combustion chamber is constructed;

[0045] Based on the working principle of the turbine, an aerodynamic thermodynamics model of the turbine is established. Using CFD simulation to analyze the complex flow field in the turbine blade passage, and combining with experimental data to correct the model parameters, a component-level mathematical model of the turbine is obtained;

[0046] It should be noted that the role of the component-level mathematical model of the compressor is to accurately describe the relationship between the pressure ratio, efficiency of the compressor and the intake air flow rate, rotational speed. The component-level mathematical model of the combustion chamber can determine the temperature and pressure distribution at the outlet of the combustion chamber. The role of the component-level mathematical model of the turbine is to accurately describe the relationship between the expansion ratio, efficiency of the turbine and the intake air flow rate, rotational speed;

[0047] Connect and integrate the component-level mathematical models according to the actual structure and working process of the gas turbine engine. Through the mass conservation, energy conservation and momentum conservation equations, the parameters of each component are correlated with each other to construct an engine overall model. An engine overall model platform is built through the system simulation software MATLAB, and the component-level mathematical models are integrated into the platform in the form of modules to realize the simulation calculation of the overall performance of the engine;

[0048] It should be noted that the role of this step is to construct an accurate and reliable engine overall model for simulating the operation of the engine under different working conditions, so as to analyze and evaluate the performance and stability of the engine;

[0049] S2. Build an engine test bench for experiments. By adjusting the intake air flow rate and monitoring the pressure change, determine whether the compressor enters the surge state and record the experimental data. Conduct numerical simulation based on the built engine whole - machine model and determine the surge state during the simulation process, record the simulation data, compare the simulation data with the experimental data, verify the accuracy of the model and generate a model matching signal or a model deviation signal. If a model matching signal is generated, construct a surge boundary. If a model deviation signal is generated, optimize the model;

[0050] As Figure 2 shown, the specific steps for obtaining the model matching signal and the model deviation signal are as follows;

[0051] In some embodiments, build a dedicated engine test bench, equipped with an intake air flow rate regulating device and a pressure sensor. The intake air flow rate regulating device can accurately control the intake air flow rate of the engine. The pressure sensor is installed in the compressor flow passage and at the inlet and outlet positions of the compressor, and is used to monitor the pressure value in the compressor flow passage, the inlet pressure value and the outlet pressure value of the compressor in real - time;

[0052] According to the design operating condition range of the engine, based on any operating condition, set the initial intake air flow rate as , close to the maximum intake air flow rate during the normal operation of the engine. During the experiment, gradually reduce the intake air flow rate in a fixed step size ΔQ. After each adjustment of the intake air flow rate, wait for the engine to run stably to ensure that all parameters reach a steady state, and then adjust the intake air flow rate again;

[0053] It should be noted that the operating condition represents the operating state and working conditions of the engine, involving various parameters and conditions during the engine operation, including rotational speed, temperature, and pressure;

[0054] Take several acquisition time points during the boundary analysis period. The time interval between adjacent acquisition time points is the same. At the acquisition time points, collect the pressure value in the flow passage through the pressure sensor installed in the compressor flow passage;

[0055] It should be noted that the boundary analysis period is a time period for collecting and analyzing the boundary of the engine experimental data;

[0056] Take the difference and absolute - value processing of the pressure values collected at adjacent acquisition time points to obtain the pressure fluctuation amplitude, and compare the obtained pressure fluctuation amplitude with the pressure fluctuation threshold;

[0057] If the pressure fluctuation amplitude is greater than or equal to the pressure fluctuation threshold, it indicates that obvious air - flow oscillation occurs inside the compressor, determine that the compressor enters the surge state, and record the experimental intake air flow rate of the compressor , and collect the compressor outlet pressure value and the inlet pressure value through the pressure sensors installed at the inlet and outlet positions of the compressor, perform a ratio process on the obtained outlet pressure value and inlet pressure value to obtain the experimental surge pressure ratio ;

[0058] It should be noted that the role of conducting actual experiments through the test bench is to obtain the operating data of the engine under different intake airflows, providing a practical basis for judging the surge state and verifying the model;

[0059] Based on the obtained engine whole - machine model platform, simulate the corresponding working conditions of the engine during the experiment, start from the set value of the experiment, and gradually decrease it with the same step size ΔQ as the experiment for numerical simulation calculation. At each intake airflow working condition, by solving the three - dimensional Reynolds - averaged N - S equations, obtain the internal flow field distribution of the compressor, including parameters such as velocity, pressure, and temperature. Based on the non - linear dynamics theory, calculate the Lyapunov exponent inside the compressor. When the Lyapunov exponent is greater than 0, it indicates that the internal airflow of the compressor is in an unstable state, determine that the compressor enters the surge state, and record the simulated intake airflow of the compressor and the calculated simulated surge pressure ratio ; It should be noted that the role of using the pressure fluctuation amplitude and the Lyapunov exponent to judge whether the compressor enters the surge state from the experimental and simulation perspectives respectively is to ensure the accuracy of surge determination;

[0060] Compare the experimental intake airflow obtained from the experiment, the experimental surge pressure ratio with the simulated intake airflow and the simulated surge pressure ratio obtained from the simulation respectively. Through the formula

[0061]

[0062] calculate the relative error of intake airflow and the relative error of pressure ratio , and compare the obtained relative error of intake airflow and relative error of pressure ratio with the intake airflow error threshold and the pressure ratio error threshold respectively;

[0063] If the relative error of intake airflow is greater than the intake airflow error threshold, or the relative error of pressure ratio is greater than the pressure ratio error threshold, it indicates that the numerical simulation results deviate significantly from the experimental results;

[0064] If the relative error of intake airflow is less than or equal to the intake airflow error threshold, and the relative error of pressure ratio is less than or equal to the pressure ratio error threshold, it indicates that the numerical simulation results are in good agreement with the experimental results, and generate a simulation accurate signal;

[0065] The engine is experimented and simulated under several different working conditions, the number of times of generating accurate simulation signals is obtained and processed by taking the ratio with the total number of working conditions of the experiment and simulation, and the matching ratio of the engine's overall model is obtained;

[0066] If the matching ratio is greater than or equal to the matching ratio threshold, it is judged that the simulation result of the engine's overall model is accurate, and a model matching signal is generated;

[0067] If the matching ratio is less than the matching ratio threshold, it is judged that the simulation result of the engine's overall model is inaccurate, and a model deviation signal is generated; It should be noted that the role of comparing the intake air flow rate, surge pressure ratio, and calculating the relative error obtained from the experiment and simulation is to evaluate the accuracy of the model;

[0068] Based on the generated model matching signal, through the analysis of the simulation results under different working conditions and combined with the nonlinear dynamics theory, a mathematical model of the engine's surge boundary is established, and the surge boundary is expressed as a function of the intake air flow rate Q and pressure ratio π , according to the multiple regression analysis method, a large number of simulation results are fitted to obtain the analytical expression of the surge boundary;

[0069] According to the simulation results, the intake air flow rate - pressure ratio characteristic curve of the engine is drawn by the data processing software Origin, and the continuous surge boundary is drawn on the characteristic curve according to the analytical expression of the surge boundary;

[0070] It should be noted that the role of establishing the mathematical model of the engine's surge boundary and drawing the surge boundary is to provide a reference for the subsequent calculation of the stability margin value;

[0071] Based on the generated model deviation signal, a hybrid optimization strategy of genetic algorithm and sequential quadratic programming is adopted. Taking the experimental surge pressure ratio as the optimization target, the fitness function is defined, the parameter search range is set on the engine's overall model platform, global search is carried out through the genetic algorithm, and local refinement correction is carried out through sequential quadratic programming to update the engine's overall model;

[0072] Based on the updated model, the simulation calculation of the surge boundary is carried out again and compared with the experimental results until the matching ratio of the engine's overall model is greater than or equal to the matching ratio threshold;

[0073] It should be noted that the role of this step is to verify the accuracy of the engine's overall model, optimize the inaccurate model, and then establish an accurate engine surge boundary based on the simulation data of the engine's overall model, providing a reliable basis for the evaluation of the engine's stability margin;

[0074] S3. By measuring the compressor inlet and outlet pressures and the airflow velocity in the engine in real time, calculate the pressure difference and the intake air flow rate, and use the fast Fourier transform to analyze the frequency domain characteristics of the pressure difference to determine whether the engine has a stall hysteresis phenomenon, generating a normal operation signal or a stall hysteresis signal. If a normal operation signal is generated, calculate the stability margin value of the engine under normal conditions;

[0075] As Figure 3 shown, the specific steps for obtaining the stability margin value of the engine under normal conditions are as follows;

[0076] In some embodiments, during the actual operation cycle of the engine, the compressor inlet pressure is measured in real time at the measurement time points by a pressure sensor and the outlet pressure , where the actual operation cycle of the engine represents a working period after the engine is actually put into operation. There are several measurement time points within the actual operation cycle of the engine, and the time intervals between adjacent measurement time points are all the same. t represents the sequence number of the measurement time point within the actual operation cycle of the engine;

[0077] Perform a difference operation on the measured compressor inlet pressure and the outlet pressure to obtain the pressure difference ;

[0078] Similarly, during the actual operation cycle of the engine, the airflow velocity in the engine is measured in real time at the measurement time points by a hot-wire anemometer . Combining with the cross-sectional area A of the engine measurement section, through the flow formula

[0079]

[0080] calculate the intake air flow rate of the engine at the measurement time point ;

[0081] It should be noted that measuring the compressor inlet and outlet pressures and the airflow velocity in the engine in real time and calculating the pressure difference and the intake air flow rate serve to provide data support for judging the operating state of the engine;

[0082] Perform a fast Fourier transform on the obtained pressure difference , and through the formula

[0083]

[0084] convert the time domain signal to the frequency domain signal to obtain the frequency domain amplitude of the pressure difference , where j represents the imaginary unit, , N represents the number of discrete frequency points, is the discrete frequency point, = 0, 1, 2 ……, N - 1;

[0085] Compare the obtained frequency-domain amplitude with the frequency-domain amplitude threshold;

[0086] If the frequency-domain amplitudes of all discrete frequency points are all less than the frequency-domain amplitude threshold, it indicates that the engine is operating normally at this time, and a normal operation signal is generated;

[0087] If there exists any discrete frequency point such that the frequency-domain amplitude is greater than or equal to the frequency-domain amplitude threshold, it indicates that a stall hysteresis phenomenon may occur at this time, and a suspected stall hysteresis signal is generated;

[0088] It should be noted that by performing a fast Fourier transform on the pressure difference to convert the time-domain signal into a frequency-domain signal, and through the comparison of the frequency-domain amplitude with the threshold, it is possible to preliminarily determine whether a stall hysteresis phenomenon may occur in the engine;

[0089] Based on the generated suspected stall hysteresis signal, obtain the response time of the intake air flow to the throttle command. The response time represents the time it takes for the intake air flow of the engine to adjust from the current state to the state corresponding to the new throttle command;

[0090] If the response time is less than the response time threshold, it is determined that the engine does not have a stall hysteresis phenomenon, and a normal operation signal is generated;

[0091] If the response time is greater than or equal to the response time threshold, it is determined that the engine has a stall hysteresis phenomenon, and a stall hysteresis signal is generated;

[0092] It should be noted that for the suspected stall hysteresis phenomenon, analyzing the response time of the intake air flow to the throttle command serves to determine whether the engine actually has a stall hysteresis;

[0093] Based on the generated normal operation signal, obtain the inlet pressure and outlet pressure of the compressor at the current measurement time point and perform a ratio processing to obtain the real-time pressure ratio , obtain the intake air flow of the engine at the current measurement time point , and combine the obtained intake air flow with the real-time pressure ratio to obtain the real-time measurement point at the current measurement time point on the intake air flow - pressure ratio plane;

[0094] Calculate the distance between the real-time measurement point and the surge boundary on the intake air flow - pressure ratio plane. Specifically, obtain the pressure ratio corresponding to the same intake air flow as the real-time measurement point on the surge boundary as , through the formula

[0095]

[0096] the stable margin value SM of the engine under normal operating conditions is calculated;

[0097] It should be noted that the purpose of this step is to monitor the operating state of the engine in real time, detect the stall hysteresis phenomenon in a timely manner, and calculate its stable margin value during the normal operation of the engine to ensure the safe operation of the engine;

[0098] S4. If the generated stall hysteresis signal is received, perform time-frequency characteristic analysis on the pressure difference signals before and after the stall hysteresis signal, estimate and update the key parameters in the surge boundary mathematical model by combining CFD transient simulation and the adaptive Kalman filtering algorithm, and calculate the stable margin value of the engine under the stall hysteresis phenomenon according to the updated surge boundary;

[0099] In some embodiments, if the generated stall hysteresis signal is received, extract the pressure difference within the Δt time window before and after the generation of the stall hysteresis signal during the actual operation cycle of the engine , analyze the time-frequency characteristics of the pressure difference signal through wavelet transform, and the formula for wavelet transform is

[0100]

[0101] where represents the input pressure difference signal, is the wavelet basis function, a represents the scale parameter, b represents the translation parameter, represents the wavelet coefficient obtained by wavelet transform;

[0102] Calculate the wavelet coefficient energy in different frequency bands based on the wavelet coefficients obtained by wavelet transform, the overall energy distribution characteristics, combine the CFD transient simulation results to obtain simulation data, and fuse the pressure difference signal with the simulation data to obtain the stall hysteresis characteristic data;

[0103] It should be noted that the purpose of performing wavelet transform on the pressure difference before and after the stall hysteresis signal and analyzing its energy distribution characteristics is to obtain more comprehensive stall hysteresis characteristic data;

[0104] Based on the fused stall hysteresis characteristic data, use the adaptive Kalman filtering algorithm to estimate the key parameters in the surge boundary mathematical model. Specifically, expand the intake flow - pressure ratio coefficient term in the surge boundary analytical expression into a time-varying form:

[0105]

[0106] Among them, the coefficients a(t), b(t), and c(t) are all updated by fitting the stall hysteresis characteristic data. The objective function is constructed by using the historical data within the sliding time window, and the coefficients are optimized by the gradient descent method;

[0107] It should be noted that the purpose of estimating and updating the key parameters of the surge boundary mathematical model is to make the corrected surge boundary reflect the influence of engine performance degradation and component wear;

[0108] Integrate the real-time data interface in the Origin platform to dynamically update the intake air flow - pressure ratio characteristic curve. For the stall hysteresis region, the fractal interpolation algorithm is used to enhance the resolution of the curve in the nonlinear region to obtain the corrected surge boundary, calculate the Hausdorff distance between the corrected surge boundary and the real-time measurement points, and quantify the stability margin value of the engine

[0109]

[0110] Calculate the stability margin value SM of the engine under the stall hysteresis phenomenon, where k is the number of measurement time points within the Δt time window before and after generating the stall hysteresis signal, represents the real-time pressure ratio of the compressor at the i-th measurement time point within the time window, represents the intake air flow on the surge boundary at the i-th measurement time point within the time window the corresponding pressure ratio, i = 1, 2, 3, ……, k;

[0111] It should be noted that the function of this step is to deeply analyze the characteristics of the stall hysteresis phenomenon, correct the surge boundary, improve the accuracy of the stability margin value calculation, and better evaluate the stability of the engine under the stall hysteresis condition;

[0112] S5. Based on the obtained engine stability margin value, evaluate the overall engine stability margin, and generate a safety signal, a warning signal, or a danger signal for the engine according to the evaluation result;

[0113] In some embodiments, the overall engine stability margin is evaluated according to the calculated engine stability margin value;

[0114] Specifically, if the engine stability margin value is greater than or equal to the warning threshold, evaluate the overall engine stability margin as good and generate a safety signal;

[0115] If the engine stability margin value is greater than or equal to the danger threshold and less than the warning threshold, evaluate the overall engine stability margin as ordinary and generate a warning signal;

[0116] If the engine stability margin value is less than the danger threshold, evaluate the overall engine stability margin as dangerous and generate a danger signal;

[0117] It should be noted that the function of this step is to classify the stable state of the engine into three levels: good, normal, and dangerous according to the stable margin value of the engine, generate corresponding signals, timely detect the unstable state of the engine, and provide a decision-making basis for the operation management and maintenance of the engine;

[0118] The technical solution of the embodiment of the present invention is as follows: By analyzing the physical characteristics and working principles of each component of the gas turbine engine, combined with CFD numerical simulation, a component-level mathematical model of each component is constructed. These component-level mathematical models are integrated into an engine whole-machine model. Through the MATLAB platform, simulation calculation is realized, an engine test bench is constructed for experiments. By adjusting the intake air flow and monitoring the pressure change, it is judged whether the compressor enters the surge state and the experimental data is recorded. Based on the constructed engine whole-machine model, numerical simulation is carried out and the surge state during the simulation process is judged, and the simulation data is recorded. The simulation data is compared with the experimental data to verify the accuracy of the model. If it is accurate, the surge boundary is constructed. If it is inaccurate, the model is optimized. By measuring the inlet and outlet pressures of the compressor and the air flow velocity in the engine in real time, the pressure difference and the intake air flow are calculated, and the frequency domain characteristics of the pressure difference are analyzed by using the fast Fourier transform to judge whether the engine has a stall hysteresis phenomenon, and a normal operation signal or a stall hysteresis signal is generated. If a normal operation signal is generated, the stable margin value of the engine is calculated. If the generated stall hysteresis signal is received, the time-frequency characteristics of the pressure difference signal before and after the stall hysteresis signal are analyzed, and the key parameters in the surge boundary mathematical model are estimated and updated by combining CFD transient simulation and the adaptive Kalman filter algorithm. According to the updated surge boundary, the stable margin value of the engine is calculated. Based on the obtained stable margin value of the engine, the overall engine stable margin is evaluated, and according to the evaluation result, a safety signal, a warning signal or a danger signal of the engine is generated.

[0119] Embodiment 2: As Figure 4 shown, a method for evaluating the overall stable margin of an aero gas turbine engine provided by an embodiment of the present invention specifically includes the following steps:

[0120] Overall machine modeling module: Construct an engine overall machine model;

[0121] According to the physical characteristics and working principles of each component of the gas turbine engine, a corresponding component-level mathematical model is established. Each component of the gas turbine engine includes a compressor, a combustion chamber, a turbine, etc.;

[0122] Specifically, based on the one-dimensional unsteady flow theory, the characteristic line method is used to solve the partial differential equations describing the internal flow of the compressor. Combining with the geometric parameters of the compressor, including blade shape, number of stages, tip clearance, etc., a component-level mathematical model of the compressor is constructed, and the component-level mathematical model of the compressor is corrected using the CFD numerical simulation results;

[0123] Based on the chemical reaction kinetics mechanism, combined with CFD simulation to analyze the three-dimensional flow field in the combustion chamber, considering the fuel injection, atomization, mixing and combustion processes, a component-level mathematical model of the combustion chamber is constructed by solving the energy conservation, mass conservation and momentum conservation equations;

[0124] Based on the working principle of the turbine, an aerodynamic and thermodynamic model of the turbine is established. The CFD simulation is used to analyze the complex flow field in the turbine blade passage, and the model parameters are corrected in combination with the experimental data to obtain the component-level mathematical model of the turbine;

[0125] Connect and integrate the component-level mathematical models according to the actual structure and working process of the gas turbine engine. Through the mass conservation, energy conservation and momentum conservation equations, the parameters of each component are correlated with each other to construct an engine whole-machine model. An engine whole-machine model platform is built through the system simulation software MATLAB, and the component-level mathematical models are integrated into the platform in the form of modules to realize the simulation calculation of the overall performance of the engine;

[0126] Boundary analysis module: Build an engine test bench for experiments, and at the same time perform numerical simulations according to the engine whole-machine model. Record the experimental and simulation data when the compressor enters the surge state and compare them to verify the accuracy of the model. If it is accurate, construct the surge boundary according to the model. If it is inaccurate, optimize the model;

[0127] Build a dedicated engine test bench, equipped with an intake air flow regulating device and a pressure sensor. The intake air flow regulating device can accurately control the intake air flow of the engine. The pressure sensor is installed in the compressor flow passage and at the inlet and outlet positions of the compressor to monitor the pressure value in the compressor flow passage, the inlet pressure value and the outlet pressure value of the compressor in real time;

[0128] According to the design operating condition range of the engine, based on any operating condition, set the initial intake air flow as , close to the maximum intake air flow during the normal operation of the engine. During the experiment, gradually reduce the intake air flow at a fixed step size ΔQ. After each adjustment of the intake air flow, wait for the engine to run stably to ensure that all parameters reach a steady state, and then adjust the intake air flow again;

[0129] During the boundary analysis period, several acquisition time points are taken, and the interval duration between adjacent acquisition time points is the same. At the acquisition time points, the pressure values in the compressor flow passage are collected through pressure sensors installed in the compressor flow passage;

[0130] Take the difference between the pressure values collected at adjacent acquisition time points and take the absolute value to obtain the pressure fluctuation amplitude, and compare the obtained pressure fluctuation amplitude with the pressure fluctuation threshold;

[0131] If the pressure fluctuation amplitude is greater than or equal to the pressure fluctuation threshold, it indicates that obvious airflow oscillation occurs inside the compressor, and it is determined that the compressor enters the surge state, and the experimental intake air flow rate of the compressor is recorded , and collect the compressor outlet pressure value and the inlet pressure value through pressure sensors installed at the inlet and outlet positions of the compressor, and perform a ratio process on the obtained outlet pressure value and inlet pressure value to obtain the experimental surge pressure ratio ;

[0132] Based on the obtained engine whole - machine model platform, simulate the corresponding working conditions of the engine during the experiment. Starting from the set in the experiment, gradually decrease the intake air flow rate with the same step size ΔQ as in the experiment, and perform numerical simulation calculations. At each intake air flow rate working condition, by solving the three - dimensional Reynolds - averaged N - S equations, obtain the flow field distribution inside the compressor, including parameters such as velocity, pressure, and temperature. Based on the nonlinear dynamics theory, calculate the Lyapunov exponent inside the compressor. When the Lyapunov exponent is greater than 0, it indicates that the airflow inside the compressor is in an unstable state, and it is determined that the compressor enters the surge state, and record the simulated intake air flow rate of the compressor and the calculated simulated surge pressure ratio ;

[0133] Compare the experimental intake air flow rate and the experimental surge pressure ratio obtained from the experiment with the simulated intake air flow rate and the simulated surge pressure ratio obtained from the simulation respectively, and perform a comparative analysis. Through the formula

[0134]

[0135] calculate the relative error of the intake air flow rate and the relative error of the pressure ratio , and compare the obtained relative error of the intake air flow rate and the relative error of the pressure ratio with the intake air flow rate error threshold and the pressure ratio error threshold respectively;

[0136] If the relative error of the intake air flow rate is greater than the intake air flow rate error threshold, or the relative error of the pressure ratio is greater than the pressure ratio error threshold, it indicates that the numerical simulation results deviate significantly from the experimental results;

[0137] If the relative error of the intake air flow is less than or equal to the intake air flow error threshold, and the relative error of the pressure ratio is less than or equal to the pressure ratio error threshold, it indicates that the numerical simulation results are in good agreement with the experimental results, and a simulation accurate signal is generated;

[0138] The engine is experimented and simulated under several different working conditions, the number of times of generating the simulation accurate signal is obtained and processed by taking the ratio with the total number of working conditions of the experiment and simulation, and the matching ratio of the engine's overall model is obtained;

[0139] If the matching ratio is greater than or equal to the matching ratio threshold, it is judged that the simulation results of the engine's overall model are accurate, and a model matching signal is generated;

[0140] If the matching ratio is less than the matching ratio threshold, it is judged that the simulation results of the engine's overall model are inaccurate, and a model deviation signal is generated;

[0141] Based on the generated model matching signal, through the analysis of the simulation results under different working conditions, combined with the nonlinear dynamics theory, a mathematical model of the engine surge boundary is established, and the surge boundary is expressed as a function of the intake air flow Q and the pressure ratio π , according to the multiple regression analysis method, a large number of simulation results are fitted to obtain the analytical expression of the surge boundary;

[0142] According to the simulation results, the intake air flow - pressure ratio characteristic curve of the engine is drawn by the data processing software Origin, and the continuous surge boundary is drawn on the characteristic curve according to the analytical expression of the surge boundary;

[0143] Based on the generated model deviation signal, a hybrid optimization strategy of genetic algorithm and sequential quadratic programming is adopted. The fitness function is defined with the experimental surge pressure ratio as the optimization target. The parameter search range is set on the engine's overall model platform. Global search is carried out by the genetic algorithm, and local refinement correction is carried out by sequential quadratic programming to update the engine's overall model;

[0144] Based on the updated model, the simulation calculation of the surge boundary is carried out again and compared with the experimental results until the matching ratio of the engine's overall model is greater than or equal to the matching ratio threshold;

[0145] Hysteresis judgment module: Analyze in real time whether the engine has a stall hysteresis phenomenon. If it appears, a stall hysteresis signal is generated. Otherwise, a normal operation signal is generated, and the stable margin value of the engine's normal operation is calculated;

[0146] During the actual operation cycle of the engine, the inlet pressure of the compressor is measured in real time by a pressure sensor at the measurement time point and the outlet pressure , where the actual operating cycle of the engine represents a period of time in the working state after the engine is actually put into operation. There are several measurement time points within the actual operating cycle of the engine, and the time intervals between adjacent measurement time points are all the same. t represents the sequence number of the measurement time point within the actual operating cycle of the engine;

[0147] Take the difference between the measured inlet pressure of the compressor and the outlet pressure to obtain the pressure difference ;

[0148] Similarly, within the actual operating cycle of the engine, the air flow velocity in the engine is measured in real time at the measurement time points by a hot-wire anemometer , combined with the cross-sectional area A of the engine measurement section, and through the flow formula

[0149]

[0150] calculate the intake air flow rate of the engine at the measurement time points ;

[0151] Perform a fast Fourier transform on the obtained pressure difference through the formula

[0152]

[0153] to convert the time-domain signal into a frequency-domain signal and obtain the frequency-domain amplitude of the pressure difference , where j represents the imaginary unit, , N represents the number of discrete frequency points, is the discrete frequency point, = 0, 1, 2..., N - 1;

[0154] Compare the obtained frequency-domain amplitude with the frequency-domain amplitude threshold;

[0155] If the frequency-domain amplitudes of all discrete frequency points are all less than the frequency-domain amplitude threshold, it indicates that the engine is operating normally at this time, and a normal operation signal is generated;

[0156] If there is any discrete frequency point such that the frequency-domain amplitude is greater than or equal to the frequency-domain amplitude threshold, it indicates that a stall hysteresis phenomenon may occur at this time, and a suspected stall hysteresis signal is generated;

[0157] Based on the generated suspected stall hysteresis signal, obtain the response time of the intake air flow to the throttle command. The response time represents the time after the engine receives a change in the throttle command, and its intake air flow The time taken to adjust from the current state to the state corresponding to the new throttle command;

[0158] If the response time is less than the response time threshold, it is determined that the engine does not exhibit stall hysteresis, and a normal operation signal is generated;

[0159] If the response time is greater than or equal to the response time threshold, it is determined that the engine exhibits stall hysteresis, and a stall hysteresis signal is generated;

[0160] Based on the generated normal operation signal, obtain the inlet pressure and outlet pressure of the compressor at the current measurement time point and perform a ratio process to obtain the real-time pressure ratio , obtain the intake air flow rate of the engine at the current measurement time point , combine the obtained intake air flow rate with the real-time pressure ratio , to obtain the real-time measurement point at the current measurement time point on the intake air flow rate - pressure ratio plane;

[0161] Calculate the distance between the real-time measurement point and the surge boundary on the intake air flow rate - pressure ratio plane. Specifically, obtain the pressure ratio corresponding to the same intake air flow rate as the real-time measurement point on the surge boundary which is , through the formula

[0162]

[0163] Calculate the stability margin value SM of the engine in the normal operation state;

[0164] Hysteresis optimization module: If a stall hysteresis signal is generated, update the surge boundary, and calculate the stability margin value of the engine under the stall hysteresis phenomenon based on the updated surge boundary;

[0165] If the generated stall hysteresis signal is received, extract the pressure difference within the Δt time window before and after the generation of the stall hysteresis signal during the actual operation cycle of the engine , analyze the time-frequency characteristics of the pressure difference signal through wavelet transform. The formula for wavelet transform is

[0166]

[0167] where represents the input pressure difference signal, is the wavelet basis function, a represents the scale parameter, b represents the translation parameter, represents the wavelet coefficient obtained by wavelet transform;

[0168] Calculate the wavelet coefficient energy in different frequency bands based on the wavelet coefficients obtained by wavelet transform, the overall energy distribution characteristics, combine the CFD transient simulation results to obtain simulation data, and use data assimilation technology to process the pressure difference signal Fuse with the analog data to obtain the stall hysteresis characteristic data;

[0169] Based on the fused stall hysteresis characteristic data, use the adaptive Kalman filter algorithm to estimate the key parameters in the surge boundary mathematical model. Specifically, expand the intake air flow - pressure ratio coefficient term in the analytical expression of the surge boundary into a time - varying form:

[0170]

[0171] Among them, the coefficients a(t), b(t), and c(t) are all updated by fitting the stall hysteresis characteristic data. Construct an objective function through the historical data within the sliding time window, and optimize the coefficients through the gradient descent method;

[0172] Integrate a real - time data interface in the Origin platform to dynamically update the intake air flow - pressure ratio characteristic curve. For the stall hysteresis region, use the fractal interpolation algorithm to enhance the resolution of the curve in the non - linear region, obtain the corrected surge boundary, calculate the Hausdorff distance between the corrected surge boundary and the real - time measurement points, and quantify the stability margin value of the engine

[0173]

[0174] Calculate the stability margin value SM of the engine under the stall hysteresis phenomenon, where k is the number of measurement time points within the Δt time window before and after generating the stall hysteresis signal, represents the real - time pressure ratio of the compressor at the i - th measurement time point within the time window, represents the intake air flow on the surge boundary at the i - th measurement time point within the time window corresponding pressure ratio, i = 1, 2, 3, ……, k;

[0175] Margin evaluation module: Evaluate the overall stability margin of the engine according to the stability margin value;

[0176] Evaluate the overall stability margin of the engine according to the calculated engine stability margin value;

[0177] Specifically, if the stability margin value of the engine is greater than or equal to the warning threshold, evaluate the overall stability margin of the engine as good and generate a safety signal;

[0178] If the stability margin value of the engine is greater than or equal to the danger threshold and less than the warning threshold, evaluate the overall stability margin of the engine as ordinary and generate a warning signal;

[0179] If the stability margin value of the engine is less than the danger threshold, evaluate the overall stability margin of the engine as dangerous and generate a danger signal.

[0180] The above has described in detail an embodiment of the present invention, but the above content is only a preferred embodiment of the present invention and cannot be considered as defining the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A method for evaluating the overall stability margin of an aero gas turbine engine, characterized in that It includes the following steps: Based on the physical characteristics and working principles of the components of a gas turbine engine, establish component-level mathematical models. Based on one-dimensional unsteady flow theory, use the method of characteristics to solve the partial differential equations describing the internal flow of the compressor, and combine the geometric parameters of the compressor to construct the component-level mathematical model of the compressor. Based on the chemical reaction kinetics mechanism, combine CFD simulation to analyze the three-dimensional flow field in the combustion chamber, and construct the component-level mathematical model of the combustion chamber by solving the equations of energy conservation, mass conservation, and momentum conservation. Based on the working principle of the turbine, establish the aerodynamic and thermodynamic model of the turbine, use CFD simulation to analyze the complex flow field in the turbine blade passage, and obtain the component-level mathematical model of the turbine; Connect and integrate the component-level mathematical models according to the actual structure and working process of the gas turbine engine to construct an engine whole-machine model; Construct an engine test bench for experiments, and at the same time conduct numerical simulations according to the engine whole-machine model, record the experimental and simulation data when the compressor enters the surge state and compare them to verify the accuracy of the engine whole-machine model. If it is accurate, construct a surge boundary according to the engine whole-machine model. If it is not accurate, optimize the engine whole-machine model; Analyze in real time whether the engine has a stall hysteresis phenomenon. If it occurs, generate a stall hysteresis signal. Otherwise, generate a normal operation signal, and calculate the stability margin value of the engine during normal operation; If the generated stall hysteresis signal is received, conduct time-frequency characteristic analysis on the pressure difference signal before and after the stall hysteresis signal, estimate and update the key parameters in the surge boundary mathematical model by combining CFD transient simulation and the adaptive Kalman filter algorithm, and calculate the stability margin value of the engine when the stall hysteresis phenomenon occurs according to the updated surge boundary; Evaluate the overall engine stability margin according to the stability margin value; 2. The method for evaluating the overall stability margin of an aero gas turbine engine according to claim 1, wherein The method of constructing the surge boundary is as follows: If a model matching signal is generated, through the analysis of the simulation results under different working conditions, establish a mathematical model of the engine surge boundary, and fit a large number of simulation results according to the multiple regression analysis method to obtain the analytical expression of the surge boundary; Draw the surge boundary on the characteristic curve according to the analytical expression; 3. The method for evaluating the overall stability margin of an aero gas turbine engine according to claim 2, characterized in that, The method of obtaining the model matching signal is as follows: Conduct experiments and simulations on the engine under several different working conditions, obtain the number of times the simulation accurate signal is generated and conduct data processing to obtain the matching ratio of the engine whole-machine model; If the matching ratio is greater than or equal to the matching ratio threshold, generate a model matching signal. Otherwise, generate a model deviation signal; 4. A method for evaluating the overall stability margin of an aero gas turbine engine according to claim 3, characterized in that, The method of obtaining the simulation accurate signal is as follows: Conduct comparative analysis on the experimental intake air flow rate and experimental surge pressure ratio obtained from the experiment with the simulated intake air flow rate and simulated surge pressure ratio obtained from the simulation, and calculate the relative error of the intake air flow rate and the relative error of the pressure ratio; If the relative error of the intake air flow rate is less than or equal to the intake air flow rate error threshold, and the relative error of the pressure ratio is less than or equal to the pressure ratio error threshold, generate a simulation accurate signal; 5. The method for evaluating the overall stability margin of an aero gas turbine engine according to claim 4, wherein, The method of obtaining the experimental intake air flow rate and experimental surge pressure ratio is as follows: During the experiment, reduce the intake air flow rate at a fixed step size; Obtain the amplitude of the pressure fluctuation during the process of reducing the intake air flow. If the amplitude of the pressure fluctuation is greater than or equal to the pressure fluctuation threshold, it is determined that the compressor enters the surge state. Record the experimental intake air flow of the compressor, collect the outlet pressure value and the inlet pressure value of the compressor, and perform data processing to obtain the experimental surge pressure ratio.

6. The method for evaluating the overall stability margin of an aero gas turbine engine according to claim 4, characterized in that, The method for obtaining the simulated intake air flow and the simulated surge pressure ratio is as follows: Perform numerical simulation according to the engine overall model, gradually reduce the intake air flow at the same step size as the experiment, calculate the Lyapunov exponent inside the compressor during the process of reducing the intake air flow. If the Lyapunov exponent is greater than 0, it is determined that the compressor enters the surge state, and record the simulated intake air flow of the compressor and the calculated simulated surge pressure ratio.

7. A method for evaluating the overall stability margin of an aero gas turbine engine according to claim 1, characterized in that, The method for obtaining the stability margin value during the normal operation of the engine is as follows: If a normal operation signal is received, obtain the real-time pressure ratio and intake air flow of the engine, and combine them to obtain the real-time measurement point on the intake air flow - pressure ratio plane at the current measurement time point; Obtain the pressure ratio corresponding to the same intake air flow as the real-time measurement point on the surge boundary and perform data processing to calculate the stability margin value of the engine in the normal operation state.

8. A method for evaluating the overall stability margin of an aero gas turbine engine according to claim 7, characterized in that The method for obtaining the normal operation signal is as follows: If a suspected stall hysteresis signal is received, obtain the response time of the intake air flow to the throttle command. If the response time is less than the response time threshold, generate a normal operation signal; otherwise, generate a stall hysteresis signal.

9. The method for evaluating the overall stability margin of an aero gas turbine engine according to claim 8, wherein, The method for obtaining the suspected stall hysteresis signal is as follows: During the actual operation cycle of the engine, obtain the pressure difference between the inlet and outlet of the compressor in real time; Perform a fast Fourier transform on the obtained pressure difference between the inlet and outlet of the compressor to convert the time-domain signal into a frequency-domain signal and obtain the frequency-domain amplitude of the pressure difference; If there is any discrete frequency point such that the frequency-domain amplitude is greater than or equal to the frequency-domain amplitude threshold, generate a suspected stall hysteresis signal.

10. A method for evaluating the overall stability margin of an aero gas turbine engine according to claim 1, characterized in that, The method for obtaining the stability margin value under the stall hysteresis phenomenon is as follows: If a stall hysteresis signal is received, extract the pressure difference between the inlet and outlet of the compressor within the Δt time window before and after the generation of the stall hysteresis signal during the actual operation cycle of the engine, perform data analysis and processing, and expand the analytical expression of the surge boundary into a time-varying form to obtain the corrected surge boundary; Perform data processing based on the corrected surge boundary and the real-time measurement point to calculate the stability margin value of the engine under the stall hysteresis phenomenon.

Images