Aero-engine adaptive correction method based on multi-discipline coupling

Abstract

The present application relates to the technical field of aero-engine modeling, and particularly relates to an aero-engine self-adaptive correction method based on multi-disciplinary coupling, comprising: constructing a component-level model of an aero-engine; performing segmented characteristic correction; performing physical characteristic correction; and performing variable-dimension simulation to obtain a final corrected target component-level model. The present application effectively solves the problem that a zero-dimensional performance simulation model cannot represent the radial distribution of flow passage temperature and the distortion of component characteristics, and the poor model accuracy caused by the distortion of structure and flow characteristic parameters in transition state simulation, significantly improves the working condition range of the aero-engine, especially the simulation accuracy and engineering applicability of low working conditions and transition states, and provides a reliable model basis for high-precision fault diagnosis and optimization control.

Description

Technical Field

[0001] This invention relates to the field of aero-engine modeling technology, and specifically to an adaptive correction method for aero-engines based on multidisciplinary coupling. Background Technology

[0002] As the "heart" of an aircraft, the performance and reliability of aero-engines directly affect flight safety and energy efficiency. High-precision mathematical models of aero-engines are the core foundation for achieving advanced technologies such as performance analysis, fault diagnosis, and fault-tolerant control. The simulation accuracy of these models, especially under low-temperature operating conditions, has become a key bottleneck restricting breakthroughs and applications in related technologies.

[0003] Existing engine component-level models suffer from poor simulation accuracy across a wide operating range, facing comprehensive challenges arising from the coupling of multiple disciplines and multiple physics fields. These challenges manifest in three main aspects: First, the characteristic maps of major rotating components suffer from inherent distortion and data gaps. Characteristic maps of components such as compressors and turbines are typically derived from bench tests. Due to the influence of machining tolerances, assembly tolerances, and the installation environment, the on-site characteristics of components inherently differ from the "standard" characteristic maps obtained from the test bench. More importantly, under non-design conditions such as low speed and low flow rates, component characteristics are difficult to accurately obtain experimentally, resulting in low simulation accuracy of models based on "standard" characteristic maps under low operating conditions, severely impacting the model's applicability across a wide operating range. Second, the measured transient state data of the engine are jointly affected by rotational inertia, component volume effects, thermal immersion effects between the metal structure and airflow, and the time delay characteristics of sensor measurements. These structural and flow characteristics are typically characterized by characteristic parameters in component-level models and play a crucial role in dynamic processes with rapidly changing operating conditions. Their inaccuracies can prevent models from accurately simulating the transient response of engines, severely limiting their application value in dynamic data-based fault diagnosis and real-time fault-tolerant control. Furthermore, the uniform flow field assumption in aero-engine thermodynamic models ignores the significant radial temperature gradient that actually exists, making it difficult for simulation results to reflect measurement information from actual sensor locations. This simplification is particularly pronounced under low-performance conditions, causing significant deviations between the core thermodynamic parameters calculated by the model and actual engine data, severely restricting the model's reliability in performance evaluation and fault diagnosis.

[0004] In summary, existing aero-engine models suffer from inherent limitations in three areas: component characteristics, transition state structure, flow characteristic parameter assignment, and flow field description. These limitations result in simulation accuracy under low operating conditions and dynamic processes, as well as radial errors in the model's airflow cross-section parameters, failing to meet the demands of increasingly sophisticated engineering applications. Therefore, an adaptive correction method for aero-engines based on multidisciplinary coupling is needed to address these issues. Summary of the Invention

[0005] To address the inherent limitations of existing aero-engine models in three aspects—component characteristics, transition state structure, flow characteristic parameter assignment, and flow field description—which result in simulation accuracy under low operating conditions and dynamic processes, as well as radial errors in the model's airflow cross-section parameters failing to meet the demands of increasingly sophisticated engineering applications, this invention provides an adaptive correction method for aero-engines based on multidisciplinary coupling. This method constructs a new generation of aero-engine performance simulation models covering a wide range of operating conditions, high precision, and multi-dimensional perspectives, providing reliable support for the intelligent operation and control of high-end equipment and resolving existing problems.

[0006] The first aspect of this invention provides an adaptive correction method for aero-engines based on multidisciplinary coupling. The device employs the following technical solution, including:

[0007] Construct component-level models of aero engines;

[0008] Based on the design parameter range, multiple sets of design parameter combinations corresponding to the design operating point are obtained, and the performance simulation parameters of the engine under each set of design parameter combinations are obtained by combining each set of design parameter combinations with the component-level model.

[0009] The optimization objective is to minimize the maximum relative error between the engine's performance simulation parameters and the performance test parameters corresponding to the design operating point, and an intelligent optimization algorithm is used to obtain the optimal combination of design parameters for the design operating point.

[0010] The optimal design parameter combination is input into the component-level model for simulation to obtain the coupling coefficient corresponding to each characteristic parameter of the compressor and turbine in the characteristic diagram and the target converted speed corresponding to the optimal design parameter combination;

[0011] Based on the actual converted speed and the target converted speed, the correction coefficients corresponding to each characteristic parameter of the compressor and turbine are obtained. Based on the correction coefficients and coupling coefficients, and based on the adaptive segmented optimization strategy of multiple steady-state conditions, the characteristic parameters corresponding to each segment interval of the characteristic map of the compressor and turbine are optimized to obtain the target characteristic parameters. Based on the adaptive splicing strategy and the target characteristic parameters, the corrected characteristic map is obtained. Based on the corrected characteristic map, the first component-level model after segmented characteristic correction is obtained.

[0012] The physical properties of the first component-level model are modified based on inertial effect, thermal soaking effect, volume effect, and time delay effect to obtain the second component-level model;

[0013] Based on the second component-level model, and through the variable-dimensional simulation method, a temperature radial distribution mapping model that maps from zero-dimensional to one-dimensional is established, and the temperature radial distribution mapping model is used as the final corrected target component-level model.

[0014] A further technical solution of the present invention is to construct a component-level model of an aero-engine using a component-level modeling method.

[0015] A further technical solution of the present invention is that the design parameters include: inlet air flow rate, compressor efficiency, compressor pressure ratio, combustion chamber efficiency, total pressure loss of combustion chamber, fuel calorific value, combustion chamber load coefficient, turbine efficiency, shaft mechanical efficiency, total pressure loss of turbine outlet, and total pressure loss of intake duct; the performance simulation parameters include: relative speed, total temperature of compressor outlet, total pressure of compressor outlet, total pressure of turbine outlet, and total temperature of turbine outlet.

[0016] A further technical solution of the present invention is to use the Latin hypercube sampling method to collect multiple sets of design parameter combinations under steady-state conditions from within the design parameter range.

[0017] A further technical solution of the present invention is that the step of obtaining the correction coefficients corresponding to each characteristic parameter of the compressor and turbine based on the actual converted speed and the target converted speed is as follows:

[0018]

[0019] In the formula, A polynomial function with corrected coefficients; For actual converted speed; Calculate the rotational speed for the target; The coefficient of the zeroth term of the polynomial function; The coefficient of the first term of the polynomial function; where is the coefficient of the quadratic term in the polynomial function; The coefficients of the cubic term in the polynomial function; the target converted speed is the corrected converted speed, in hour =1, that is =1.

[0020] A further technical solution of the present invention is as follows: Based on the correction coefficient and coupling coefficient, and using an adaptive piecewise optimization strategy based on multiple steady-state conditions, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are optimized to obtain the target characteristic parameters:

[0021] Based on the position of multiple steady-state conditions on the characteristic diagram, the entire characteristic diagram is divided into several segmented intervals;

[0022] According to the preset range of polynomial coefficients, obtain multiple combinations of polynomial coefficients;

[0023] Based on each set of polynomial coefficient combinations, obtain the correction coefficient corresponding to each characteristic parameter under each set of polynomial coefficient combinations;

[0024] Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and according to the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected according to the adaptive segmented optimization strategy of multiple steady-state conditions to obtain the target characteristic parameters.

[0025] A further technical solution of the present invention is subject to the following constraints:

[0026] The efficiency constraints for compressors and turbines are: efficiency increases first and then decreases in the high-speed range, and efficiency continues to decrease in the medium-speed and low-speed ranges.

[0027] The constraints for pressure ratio and equivalent flow rate of adjacent speed lines are: as the speed decreases, both the pressure ratio and equivalent flow rate at the same equivalent flow rate decrease; and the pressure ratio of the low speed line ≤ the pressure ratio of the high speed line, and the linear flow rate of the low speed line ≤ the linear flow rate of the high speed line.

[0028] A further technical solution of the present invention is as follows: Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and according to a preset constraint condition and an adaptive piecewise optimization strategy for multiple steady-state operating conditions, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected to obtain the target characteristic parameters.

[0029] Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and based on the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval are corrected according to the adaptive segmented optimization strategy of multiple steady-state conditions to obtain the initial characteristic parameters.

[0030] Initial characteristic maps are obtained based on initial characteristic parameters; simulation calculations are performed on the component-level model based on the initial characteristic maps to obtain the engine's performance simulation parameters under each combination of design parameters.

[0031] Obtain the relative errors of the engine's performance simulation parameters and performance test parameters under each combination of design parameters;

[0032] Intelligent optimization is performed with the goal of minimizing the maximum relative error among all relative errors. The polynomial coefficient combination corresponding to the minimum maximum relative error is taken as the target polynomial coefficient combination, and the correction coefficient corresponding to the target polynomial coefficient combination is taken as the target correction coefficient.

[0033] Based on the target correction coefficient and coupling coefficient, and according to the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected according to the adaptive segmented optimization of multiple steady-state conditions to obtain the target characteristic parameters.

[0034] A further technical solution of the present invention is that the expression for the target characteristic parameter is:

[0035]

[0036]

[0037]

[0038] In the formula, the characteristic parameters include pressure ratio, efficiency, and converted flow rate. Target pressure ratio; This is the original pressure ratio; The coupling coefficient is the ratio of the pressure. This is the target correction factor corresponding to the pressure ratio; For target efficiency; For original efficiency; The coupling coefficient corresponds to the efficiency. This is the target correction coefficient corresponding to efficiency; Calculate the flow rate for the target; To calculate the flow rate; To calculate the coupling coefficient corresponding to the converted flow rate; The target correction factor is used to calculate the flow rate.

[0039] A further technical solution of the present invention is as follows: the steps of correcting the physical properties of the first component-level model based on inertial effect, thermal immersion effect, volume effect, and time delay effect to obtain the second component-level model are as follows:

[0040] The rotor speed of the first component-level model is corrected based on the inertial effect, the airflow temperature of the first component-level model is corrected based on the hot soaking effect, the outlet flow rate of the sub-component in the first component-level model is corrected based on the volume effect, and the dynamic response of the state parameters of the first component-level model is corrected based on the time delay effect.

[0041] The beneficial effects of this invention are:

[0042] This invention employs a collaborative correction strategy involving multidisciplinary coupling of segmented characteristic correction, physical characteristic correction, and variable-dimensional simulation. Segmented characteristic correction modifies the characteristic map in segments and combines adaptive stitching technology to achieve wide-domain correction. Physical characteristic correction corrects the physical characteristic parameters of inertial effects, thermal immersion effects, volumetric effects, and time delay effects based on the gas path parameters characterized by individual engine differences, reflecting the true physical characteristics. Variable-dimensional simulation establishes a 0-dimensional to 1-dimensional radial temperature distribution mapping model, reconstructing the radial temperature distribution of the flow field section to correct the error of the uniform flow field assumption. This effectively solves the problems of zero-dimensional performance simulation models failing to characterize the radial temperature distribution of the flow channel, component characteristic distortion, and poor model accuracy caused by structural and flow characteristic parameter distortion in transition state simulation. It significantly improves the simulation accuracy and engineering applicability of aero-engines across a wide operating range, especially under low operating conditions and transition states, providing a reliable model foundation for high-precision fault diagnosis and optimal control. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a detailed flowchart of an adaptive correction method for aero-engines based on multidisciplinary coupling, according to the present invention.

[0045] Figure 2 This is a flowchart of the adaptive correction method for aero-engines based on multidisciplinary coupling in an embodiment of the present invention;

[0046] Figure 3 This is a flowchart of the design working point parameter correction method in an embodiment of the present invention;

[0047] Figure 4 This is a flowchart illustrating the correction process of the component-level model using the adaptive segmentation optimization strategy and the adaptive splicing strategy in an embodiment of the present invention.

[0048] Figure 5 This is a flowchart illustrating the physical property correction process in an embodiment of the present invention.

[0049] Figure 6 This is a schematic diagram of the turbine outlet temperature distribution along the blade height in an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram showing the simulation comparison and relative error comparison of the total pressure at the compressor outlet in the transient model in this embodiment of the invention;

[0051] Figure 8 This is a schematic diagram showing the simulation comparison of total turbine outlet pressure and relative error of the transient model in this embodiment of the invention;

[0052] Figure 9 This is a schematic diagram showing the simulation comparison of total outlet temperature and relative error of the compressor in the transition state model in this embodiment of the invention;

[0053] Figure 10 This is a schematic diagram showing the simulation comparison of total turbine outlet temperature and relative error of the transition state model in an embodiment of the present invention;

[0054] Figure 11 This is a schematic diagram showing the simulation comparison of relative rotational speed and relative error of the transition state model in an embodiment of the present invention;

[0055] Figure 12This is a comparison diagram of the total turbine outlet temperature before and after considering the transition state effect in an embodiment of the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] This invention provides an embodiment of an adaptive correction method for aero-engines based on multidisciplinary coupling. The core of this embodiment lies in systematically improving the accuracy and expanding the operating range of the component-level model through a multidisciplinary, hierarchical correction strategy. This embodiment first constructs a component-level model of the aero-engine; based on the component-level model, segmented characteristic correction is performed on the characteristic diagram to improve steady-state simulation accuracy; further, physical characteristic correction is performed, adjusting structural and flow characteristic parameters to improve transient state simulation accuracy; then, variable-dimensional simulation is performed to establish a 0-dimensional to 1-dimensional temperature radial distribution mapping model; finally, a high-precision adaptive correction model for the aero-engine is formed, such as... Figure 1 As shown, this embodiment specifically includes:

[0058] S1. Construct component-level models of aero engines;

[0059] For example, in one specific embodiment, a component-level modeling method is used to construct a component-level model for simulating the performance of a micro engine. This involves decomposing the engine into several sub-components (such as the intake, compressor, combustion chamber, turbine, and exhaust nozzle), establishing a mathematical model based on the aerodynamic and thermodynamic characteristics of each sub-component, and performing coupled iterative calculations using flow, pressure, temperature, and power parameters between the sub-components. Ultimately, this achieves the simulation of the engine's steady-state and transient performance, and the model's accuracy is verified by comparing it with industry-standard commercial software to ensure its reliability.

[0060] In this embodiment, when constructing the component-level model of the aero-engine, to achieve accurate digital mapping of the engine's aero-thermodynamic processes, the core of the model must rely on a series of rigorous physical laws and mathematical algorithms. These algorithmic functions work together to transform the continuous, nonlinear physical processes inside the engine into a computable and iterative closed-loop equation system. Specifically, the model abandons the constant specific heat capacity assumption in traditional simplified models and instead adopts a variable specific heat calculation function to accurately reflect the changes in the thermophysical properties of the working fluid with temperature and composition. For key rotating components such as the compressor and turbine, their performance is described through characteristic maps: the model uses a high-precision interpolation function to achieve continuous querying of characteristic map data, ensuring accurate prediction of component performance over a wide range of operating conditions. When solving the nonlinear system composed of various components, the model uses a variable step-size iterative function. This algorithm can adaptively adjust the step size, balancing solution efficiency and stability, improving the model's convergence ability under extreme conditions and significantly increasing computational speed. Furthermore, the model kernel strictly adheres to fundamental physical processes: isentropic flow calculation functions are used to describe the acceleration and deceleration of airflow in the duct and nozzle; compression process calculation functions, combined with characteristic diagram efficiency and variable specific heat parameters, accurately calculate the compressor's power consumption and outlet section parameters; and expansion process calculation functions are used to determine the turbine's work capacity and outlet section parameters. These algorithmic functions collectively constitute the computational skeleton of the component-level model, providing a reliable value assessment benchmark and iterative foundation for subsequent component and overall modeling.

[0061] In this embodiment, component-level modeling is the physical foundation of this method. Its core advantage lies in its modular decomposition to realistically reflect the internal physical processes of the engine. Compared to a single overall performance model, this model deconstructs the engine according to its aerodynamic and thermodynamic functions into core components such as the intake, compressor, combustion chamber, turbine, and nozzle, and includes a shaft system characterizing rotor dynamics. This modeling architecture not only accurately simulates the overall performance of the engine but also clearly reveals the internal working states and interactions of each component under non-design conditions. It provides a scalable modular interface and accurate physical reference for subsequent characteristic map correction, physical characteristic correction, and variable-dimensional simulation, fundamentally ensuring the physical consistency and adaptability of the correction process. The compression components include the fan and compressor, whose function is to increase the pressure of the intake air. In the model, they are described through compression process calculation functions and characteristic maps, with the core being the calculation of the pressure ratio, efficiency, and required power at a given speed and flow rate. The combustion chamber is used for mixing and burning fuel with high-pressure air, converting chemical energy into thermal energy. The model typically characterizes its performance through parameters such as combustion efficiency and total pressure recovery coefficient. The turbine components include a high-pressure turbine and a low-pressure turbine. As the expansion components of the engine, they extract energy from the combustion gases to drive the compressor and fan. In the model, the expansion process is described by calculation functions and their characteristic graphs.

[0062] In this embodiment, the constructed component-level model of the aero-engine possesses both steady-state and transient state simulation capabilities. The core of steady-state modeling lies in solving the common working equations characterizing the engine's equilibrium state. For the turbojet engine selected in this paper, the iterative variable is chosen as the rotor percentage speed. Compressor interpolation auxiliary tool pressure ratio And turbo interpolation auxiliary tool pressure ratio The error equations are constructed as follows: power balance error between the compressor and turbine, continuous compressor flow error, and continuous turbine flow error. A variable step-size iterative algorithm is used to drive the errors to converge to a reasonable range (typically 0.00001), thereby solving for the engine's steady-state operating point under given conditions. For transient state modeling, the key is to introduce a time dimension to simulate the engine's dynamic response process. Specifically, the model uses dynamic step-size control technology to solve the shaft dynamics equations describing the rotor's dynamic characteristics. The calculation process is as follows: In the current calculation step, the torque difference generated by the instantaneous imbalance between turbine power and compressor power consumption is calculated. Based on the torque difference, the rotor acceleration is obtained. Combined with the time step, the rotor speed for the next calculation can be determined. Furthermore, the compressor interpolation tool, pressure ratio, is used as the basis for further calculations. And turbo interpolation auxiliary tool pressure ratio By constructing the compressor flow continuity error and the turbine flow continuity error, the transient operating point of the engine under given conditions can be solved.

[0063] S2. Perform segmentation characteristic correction;

[0064] Specifically, the characteristics of key components such as compressors and turbines exhibit significant "range-specific differences" across a wide operating range. This range-specific characteristic difference is the core reason for the need for segmented correction: traditional overall correction methods cannot adapt to the characteristic patterns of different ranges, easily leading to excessively large local operating condition errors. Therefore, this embodiment employs a segmented optimization strategy based on multiple steady-state operating conditions for adaptive segmentation, optimizing each segment individually to minimize the objective function of each segment. The optimized characteristics of each segment need to be spliced ​​together. Through a mechanism of "after optimizing one segment, locking the speed range of the steady-state point of that segment and using the optimized graph as the initial graph for the next segment optimization; when optimizing the next segment, only correcting the unlocked speed lines," the characteristic graphs are naturally connected, satisfying physical consistency constraints (such as the monotonicity and continuity of the characteristic curves), specifically, as follows: Figure 3As shown, in this embodiment, multiple sets of design parameter combinations corresponding to the design operating point are obtained based on the design parameter range. The performance simulation parameters of the engine at the design operating point under each set of design parameter combinations are obtained by combining each set of design parameter combinations with the component-level model. The optimization objective is to minimize the maximum relative error between the performance simulation parameters of the engine at the design operating point and the performance test parameters corresponding to the design operating point. An intelligent optimization algorithm is used to obtain the optimal design parameter combination for the design operating point. The optimal design parameter combination is input into the component-level model for simulation to obtain the coupling coefficient corresponding to each characteristic parameter of the compressor and turbine in the characteristic diagram and the target converted speed corresponding to the optimal design parameter combination. Correction coefficients corresponding to each characteristic parameter of the compressor and turbine are obtained based on the actual converted speed and the target converted speed. Based on the correction coefficients and coupling coefficients, and based on an adaptive segmented optimization strategy for multiple steady-state operating conditions, the characteristic parameters corresponding to each segment interval of the characteristic diagram for the compressor and turbine are optimized to obtain target characteristic parameters. A corrected characteristic diagram is obtained based on the adaptive splicing strategy and the target characteristic parameters. Finally, a first component-level model with corrected segmented characteristics is obtained based on the corrected characteristic diagram.

[0065] For example, in a specific implementation, the design parameters for the design operating point include: inlet airflow (Ambient.WA), compressor efficiency (HPC.Eff), compressor pressure ratio (HPC.Pr), combustion chamber efficiency (MBurner.Eff), total pressure loss in the combustion chamber (MBurner.PR_Loss), fuel calorific value (MBurner.LHV), combustion chamber load factor (MBurner.PartLoadCoeff), turbine efficiency (HPT.Eff), shaft mechanical efficiency (HPT.ET_Mech), total pressure loss at the turbine outlet (HPT.Pr), and total pressure loss in the intake manifold (Inlet.PI_D); the performance simulation parameters include: relative speed (N), total temperature at the compressor outlet (Tt3), total pressure at the compressor outlet (Pt3), total pressure at the turbine outlet (Pt7), and total temperature at the turbine outlet (Tt7). Design parameters are fundamental to the accuracy of engine wide-range operating condition simulations. Due to manufacturing and assembly tolerances, the design parameters provided by the engine manufacturer will deviate from the actual engine. Furthermore, actual testing cannot guarantee that the engine will operate under the design conditions. Therefore, in this example, 14 steady-state operating points were selected. The performance test parameter vectors for these 14 steady-state operating points are shown in Table 1. The speed at the 14th steady-state operating point is 100%, meaning it is close to the design operating point. The performance test parameters at the 14th steady-state operating point are used as the performance design parameters for the design operating conditions.

[0066] Table 1

[0067]

[0068] In the optimization process, defining the range of design parameters is one of the core prerequisites. The upper and lower limits of the design parameters can be relaxed during the initial setting to encompass the optimal value. After the current optimization is completed, the optimization results are compared with the upper and lower limits. If any parameter is close to the upper or lower limit, the limit should be adjusted accordingly, and then optimized again, dynamically adjusting the parameters. Table 2 shows the range of design parameters in this implementation case.

[0069] Table 2

[0070]

[0071] For example, in one specific embodiment, step S21, which aims to minimize the maximum relative error between the engine's performance simulation parameters and performance test parameters, and uses an intelligent optimization algorithm to obtain the optimal design parameter combination for the design operating point, is as follows: In this embodiment, the Latin hypercube sampling method is used to collect multiple sets of design parameter combinations for the design operating point from the design parameter range shown in Table 2. Based on the component-level model of step S1, simulation is performed using multiple design parameter combinations to obtain the engine's performance simulation parameters under each set of design parameter combinations. During the optimization process, the minimum error between the engine's performance simulation parameters and the performance test parameters corresponding to the design operating point is taken as the optimization objective, and an intelligent optimization algorithm is used to obtain the optimal design parameter combination for the design operating point. Specifically, the maximum relative error between the simulation values ​​and test values ​​of relative speed, compressor outlet total temperature, compressor outlet total pressure, turbine outlet total pressure, and turbine outlet total temperature is obtained, and the design parameter combination corresponding to the minimum maximum relative error is taken as the optimal design parameter combination. That is, using the first fitness... The maximum is used to determine the minimum of the maximum relative error, i.e., the first fitness. The expression is:

[0072]

[0073] In the formula, For the first One performance simulation parameter; For the first Each performance test parameter. The design parameters in the optimal design parameter combination replace the original input parameters as the new design parameter input, so that the simulation values ​​of the current engine component-level model match the engine test value results.

[0074] For example, in one specific embodiment, the design parameter optimization in step S21 improves the simulation accuracy of the performance parameters at the design operating point. The remaining operating points, deviating from the design operating point, require interpolation calculations based on the characteristic map during engine component-level model calculations. However, the published characteristic map differs from the actual engine characteristics. Therefore, the characteristic map needs to be corrected to minimize the relative error between the performance simulation parameters and performance test parameters at other steady-state points deviating from the design operating point. The characteristic map of the compressor and turbine includes several converted speed lines, each storing pressure ratio, efficiency, and converted flow rate as discrete points. Specifically, in this embodiment, step S22, obtaining the corrected characteristic map, is as follows:

[0075] Step S221, obtaining the coupling coefficients for each characteristic parameter in the characteristic diagrams of the compressor and turbine, and the target converted speed corresponding to the optimal design parameter combination, involves the following steps: Due to the difference between the published characteristic diagrams and the actual component characteristics, the pressure ratio, efficiency, and converted flow rate are multiplied by the coupling coefficients respectively. The optimal design parameter combination is then input into the component-level model for simulation to obtain the coupling coefficients corresponding to each characteristic parameter of the compressor and turbine in the characteristic diagrams. The purpose of the coupling coefficients is to ensure that the published characteristic diagrams are consistent with the magnitude of the actual engine characteristics. Based on the coupling coefficients, correction coefficients are assigned to each characteristic line. (Correction factor) include: ),in, This is the correction factor corresponding to the compressor's pressure ratio; This is the correction factor corresponding to the compressor efficiency; This is the correction factor corresponding to the converted flow rate of the compressor; This is the correction factor corresponding to the pressure ratio of the turbine; This is the correction factor corresponding to the turbine's efficiency; Let be the correction factor corresponding to the converted flow rate of the turbine, then the correction factor... The expression is:

[0076]

[0077] In the formula, A polynomial function with corrected coefficients; For actual converted speed; Calculate the rotational speed for the target; The coefficient of the zeroth term of the polynomial function; The coefficient of the first term of the polynomial function; where is the coefficient of the quadratic term in the polynomial function; The coefficients of the cubic term in the polynomial function; the target converted speed is the corrected converted speed, in hour =1, that is =1; therefore, steady-state optimization uses 6 sets of polynomial coefficients. , , To optimize variables, the polynomial coefficients are uniformly set to a range of -1 to 1; the target converted speed is the converted speed obtained by inputting the optimal design parameter combination into the component-level model for simulation.

[0078] Step S222, optimizing the characteristic parameters of the compressor and turbine in each segment of the characteristic diagram based on the correction coefficient and coupling coefficient, and using an adaptive segmented optimization strategy for multi-steady-state conditions to obtain the target characteristic parameters, involves: dividing the entire characteristic diagram into several segmented intervals based on the position of the multi-steady-state conditions on the characteristic diagram and using an adaptive segmented optimization strategy for multi-steady-state conditions; obtaining multiple sets of polynomial coefficient combinations of the correction coefficients according to the preset range of polynomial coefficients; obtaining the correction coefficient corresponding to each characteristic parameter under each set of polynomial coefficient combinations; and correcting the characteristic parameters of the compressor and turbine in each segmented interval of the characteristic diagram based on the coupling coefficient and the correction coefficients corresponding to each set of polynomial coefficient combinations, and using a preset constraint condition and an adaptive segmented optimization strategy for multi-steady-state conditions to obtain the target characteristic parameters.

[0079] In this embodiment, the steps for obtaining target characteristic parameters by correcting the characteristic parameters of the compressor and turbine in each segment interval of the characteristic map according to the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and based on preset constraints and an adaptive piecewise optimization strategy for multiple steady-state conditions, are as follows: Initial characteristic parameters are obtained by correcting the characteristic parameters of the compressor and turbine in each segment interval according to the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and based on preset constraints and an adaptive piecewise optimization strategy for multiple steady-state conditions; an initial characteristic map is obtained based on the initial characteristic parameters; and simulation calculations are performed on the component-level model based on the initial characteristic map. The engine's performance simulation parameters under each set of design parameter combinations are obtained; the relative errors of the engine's performance simulation parameters and performance test parameters under each set of design parameter combinations are acquired; intelligent optimization is performed with the minimum maximum relative error among all relative errors as the optimization objective, and the polynomial coefficient combination corresponding to the minimum maximum relative error is taken as the target polynomial coefficient combination, and the correction coefficients corresponding to the target polynomial coefficient combination are taken as the target correction coefficients; based on the target correction coefficients and coupling coefficients, and according to the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected according to the adaptive piecewise optimization strategy of multi-steady-state conditions to obtain the target characteristic parameters. That is, the expression of the corrected target characteristic parameters is:

[0080]

[0081]

[0082]

[0083] In the formula, the characteristic parameters include pressure ratio, efficiency, and converted flow rate. Target pressure ratio; This is the original pressure ratio; The coupling coefficient is the ratio of the pressure. This is the target correction factor corresponding to the pressure ratio; For target efficiency; For original efficiency; The coupling coefficient corresponds to the efficiency. This is the target correction coefficient corresponding to efficiency; Calculate the flow rate for the target; To calculate the flow rate; To calculate the coupling coefficient corresponding to the converted flow rate; The target correction factor is used to calculate the flow rate.

[0084] In this embodiment, during the steady-state optimization process, since each change of 6 correction coefficients corresponds to 6 sets of polynomial coefficients... , , This process corrects a set of compressor and turbine characteristic parameters, i.e., corrects the characteristic plot. Then, using this characteristic plot, all 14 sets of steady-state operating points are run to obtain 14 sets of simulation performance parameters. Each set of simulation performance parameters contains 5 simulation performance parameters and 5 relative error values ​​between the simulation performance parameters and the test performance parameters, totaling 70 relative error values. The maximum relative error value is found among these 70 values, and minimizing this maximum relative error value is used as the optimization objective, using the coefficients of a polynomial function... , , To optimize the variables, the maximum relative error is determined by maximizing the second fitness, i.e., the second fitness. for:

[0085]

[0086] In the formula, For the first The first of the simulation performance parameter combinations for the steady-state operating point One simulation performance parameter; For the first The first of the test performance parameter combinations at the steady-state operating point One test performance parameter.

[0087] In this embodiment, the process of optimizing each segment of the characteristic map is as follows: During the optimization of 14 steady-state operating points, since each optimization will run 14 steady-state operating points, each steady-state operating point will generate a set of simulation performance parameters. In each set of simulation performance parameters, 5 simulation performance parameters and 5 test performance parameters have 5 relative error values, for a total of 70 relative error values. The largest of the 70 relative error values ​​is selected as the optimization target. This may ignore other relative error values. Moreover, this optimization method has a large amount of computation and a slow optimization speed. The final optimization may result in a large average relative error value. Therefore, before intelligent optimization, segmented optimization is necessary. This involves dividing the 14 steady-state operating points into three groups based on engine speed. In this example, the 14 steady-state operating points are divided into three groups (the number of segments can be adjusted as needed): steady-state operating points in the high-speed range (points 10-14 in Table 1), steady-state operating points in the medium-speed range (points 5-9 in Table 1), and steady-state operating points in the low-speed range (points 1-4 in Table 1). By optimizing the characteristic parameters for each speed range individually, and then stitching together the characteristic graphs of all speed ranges, the computational load is reduced, the optimization speed is accelerated, and the average relative error is reduced.

[0088] In this embodiment, the constraints include: efficiency constraints for the compressor and turbine, and constraints on the pressure ratio and converted flow rate of adjacent speed lines. The efficiency constraints for the compressor and turbine are: efficiency increases first and then decreases in the high-speed range, while efficiency continuously decreases in the medium-speed and low-speed ranges. The purpose of these efficiency constraints is to prevent physical deviations in the optimized characteristic curve (such as abnormal efficiency changes or pressure ratio / flow rate line intersections), to perform shape verification, and to penalize unreasonable characteristics. Specifically, this embodiment performs efficiency characteristic pattern verification by extracting the maximum efficiency value for each converted speed line, verifying the rationality of the maximum efficiency value changing with speed, and eliminating any values ​​if the maximum efficiency value of a certain speed line does not meet the efficiency constraints by using a penalty function to maximize the objective function. The constraints for the pressure ratio and equivalent flow rate of adjacent speed lines are as follows: as the speed decreases, both the pressure ratio and equivalent flow rate at the same equivalent flow rate decrease; and the pressure ratio of the low speed line ≤ the pressure ratio of the high speed line and the flow rate of the low speed line ≤ the flow rate of the high speed line. If an adjacent line pair violates the rule that "the pressure ratio of the low speed line ≤ the pressure ratio of the high speed line and the flow rate of the low speed line ≤ the flow rate of the high speed line", then a penalty function is executed to maximize the objective function and ensure that the optimized characteristic diagram conforms to physical laws.

[0089] Step S223: The steps for stitching together the target characteristic parameters based on the adaptive stitching strategy to obtain the corrected characteristic map, and obtaining the first component-level model after segmented characteristic correction based on the corrected characteristic map are as follows:

[0090] In this embodiment, the compressor characteristic plot has 14 conversion speed lines ranging from 0.35 to 1, with intervals of 0.05. The turbine characteristic plot has 13 conversion speed lines ranging from 0.4 to 1, with intervals of 0.05. Specifically, after each segmented interval optimization, it is necessary to determine the range of conversion speed lines for the next segmented interval optimization to avoid the next segmented interval optimization overwriting the results of the previous segmented interval optimization. Figure 4 As shown, since the segmented interval optimization in this example proceeds from the high-speed segmented interval to the low-speed segmented interval, it is only necessary to determine which two converted speed lines on the characteristic diagram the speed of the minimum steady-state operating point in the high-speed interval, i.e., the 10th steady-state operating point, falls between. In this example, after the optimization of the first segmented interval of the characteristic diagram, the speed of the 10th steady-state operating point falls between the converted speed lines corresponding to the 9th and 10th steady-state operating points of the compressor, and between the converted speed lines corresponding to the 8th and 9th sets of design parameters on the turbine's characteristic diagram. In the next segmented interval optimization, the range of converted speed lines to be optimized needs to be controlled. The range of converted speed lines corresponding to the 1st to 14th steady-state operating points of the compressor and the 1st to 13th steady-state operating points of the turbine needs to be adjusted to the range corresponding to the 1st to 9th steady-state operating points of the compressor and the 1st to 8th steady-state operating points of the turbine. This ensures that subsequent optimizations maintain the same converted speed lines for all steady-state operating points from the previous segmented interval optimization, optimizing only the converted speed lines and keeping the subsequent segmented interval optimization steps unchanged. This speed line-locked stitching logic is used to adaptively stitch the characteristic maps, resulting in a final corrected set of compressor and turbine characteristic maps. Finally, replacing the original characteristic map with the corrected one completes the multi-steady-state operating point segmented optimization, significantly improving the steady-state calculation accuracy of the engine component-level model. This also serves as a crucial bridge connecting the "design operating point baseline" and "transient state effect correction." The preceding design operating point optimization only calibrates a single operating condition, while the segmented characteristic map correction, based on the design operating point baseline, extends the accuracy to a wide operating condition steady-state range. Furthermore, the corrected characteristic map is a necessary input condition for transient dynamic effect correction.

[0091] Thus, the first component-level model after segmentation characteristic correction is obtained based on the corrected characteristic map.

[0092] S3. Perform physical property correction;

[0093] Specifically, optimizing the steady-state component characteristic diagram of the engine can significantly improve the simulation accuracy of the model at the steady-state operating point. However, the transient process does not run along the steady-state operating line, meaning that the simulation accuracy of the transient process cannot be guaranteed. To improve the simulation accuracy of the transient process, it is necessary to model four key dynamic effects accordingly and use an intelligent optimization algorithm to find the core physical parameters affecting these four dynamic effects. This allows for accurate correction of the physical characteristic parameters of the engine structure and flow characteristics, enabling the transient model to truly reflect the combined effect of each dynamic effect and significantly improve the simulation accuracy of the transient process. In this embodiment, the physical characteristics of the first component-level model are corrected based on inertial effects, thermal soaking effects, volume effects, and time delay effects to obtain the second component-level model.

[0094] For example, in one specific embodiment, the rotor speed of the first component-level model is corrected based on the inertial effect, the airflow temperature of the first component-level model is corrected based on the hot immersion effect, the outlet flow rate of the sub-component in the first component-level model is corrected based on the volume effect, and the dynamic response of the state parameters of the first component-level model is corrected based on the time delay effect.

[0095] S31. Rotor speed of the first component-level model corrected based on inertial effect:

[0096] In this embodiment, inertia is the most fundamental factor affecting the engine's transient performance. When the fuel flow rate changes, the rotor generates additional residual power due to inertia, causing an imbalance between the compressor's power consumption and the coaxial turbine's output power, thus inducing a dynamic process of speed change. The expression for power balance is:

[0097]

[0098] in, This is the remaining power; To generate power for the turbine, The compressor consumes power. The power balance expression holds true under both steady-state and transient operating conditions of the engine. Under steady-state conditions... The value is 0. However, during the engine's transient state, when the fuel supply changes, due to the rotor's inertia, the power of the coaxially connected turbine and compressor is no longer balanced. If the value is not 0, then the expression for the rotor acceleration is:

[0099]

[0100] In the formula, This refers to the rotor acceleration; This is the remaining power; For rotational inertia, Rotational speed;

[0101] The expression for updating the rotational speed is:

[0102]

[0103] in, for Rotational speed at any given moment; for Rotational speed at any given moment; This refers to the rotor acceleration; For time step; when calculating Rotation speed at any given moment As known quantities, the rotational speed at each moment is calculated sequentially according to the equation until the transition process ends.

[0104] S32. Correcting the airflow temperature of the first component-level model based on the thermal soaking effect:

[0105] In this embodiment, the heat immersion effect occurs during the engine's transient operation, where the airflow temperature is dynamically changing. Due to the thermal inertia of metallic materials, the temperature response of engine structural components significantly lags behind the airflow. This temperature difference drives unsteady heat exchange between the airflow and the engine body, thus delaying the transient temperature change of the airflow. The heat immersion effect significantly alters the ideal temperature distribution, disrupts the thermal matching relationship between components, and consequently causes systematic deviations in various measured parameters. To quantify this impact, the heat immersion effect modeling typically employs a lumped parameter method, involving corresponding physical characteristic parameters. Specifically, each key component is simplified into a control body with heat storage capacity. Based on the principle of energy conservation, the transient heat exchange between the control body and the flowing airflow is calculated, thereby obtaining correction values ​​for the control body temperature and airflow temperature, thus achieving an effective characterization of its dynamic thermal response. The expression for the heat exchange between the airflow and the engine body is:

[0106]

[0107] In the formula, To exchange heat, The heat transfer coefficient, For equivalent heat transfer area, It is a time constant. For the first time when the transition state effect is not considered The gas temperature at that moment; Before considering the transition state effect, the first The material temperature at a given moment; Let be the time step. The heat transfer coefficient between the airflow and the engine body is determined by considering the effects of heat conduction and convection; therefore, the expression for the heat transfer coefficient is:

[0108]

[0109] In the formula, The heat transfer coefficient, The thermal conductivity of the engine material. The convective heat transfer coefficient between the airflow and the material. This is the equivalent heat transfer length.

[0110] When performing simulation calculations for the engine's non-design operating points (all steady-state operating points except the 14th steady-state operating point), the following formula is used to... Make corrections:

[0111]

[0112] In the formula, The convective heat transfer coefficient between airflow and material at non-design operating conditions; The convective heat transfer coefficient between the airflow and the material at the design operating point; The specific heat capacity of airflow at constant pressure under non-design operating conditions; The specific heat capacity of the airflow at constant pressure under the design operating condition; This refers to the gas flow rate at non-design operating conditions. The gas flow rate is given at the design operating point; after calculating the heat exchange, the change in material temperature is calculated using the following formula:

[0113]

[0114] In the formula, This refers to the change in material temperature. The specific heat capacity of metallic materials. The equivalent mass of the component;

[0115] Therefore, the material temperature at the current moment, considering the hot-dip effect, is calculated as follows:

[0116]

[0117] In the formula, To account for the transition state effect and the hot soaking effect, after the first The material temperature at a given moment; For the first time when the transition state effect is not considered The material temperature at a given moment;

[0118] The current enthalpy of the gas flow is:

[0119]

[0120] In the formula, When considering the transition state effect, the first The enthalpy of the gas flow at each moment; For the first time when the transition state effect is not considered The enthalpy value of the gas flow at each moment.

[0121] Based on the enthalpy of the gas flow and the gas-oil ratio, the actual gas flow temperature at the current moment, considering the hot immersion effect, can be calculated using the variable specific heat algorithm:

[0122]

[0123] In the formula, For the first The actual gas flow temperature at a given moment; The temperature is calculated using the enthalpy and the oil-gas ratio.

[0124] S33. Correcting the outlet flow rate of sub-components in the first-component-level model based on the volumetric effect:

[0125] In this embodiment, during the engine's transient operation, parameters such as temperature, pressure, and flow rate within the flow channel are all dynamically changing. Due to the volumetric inertia of the connecting pipes and cavities between the compressor, combustion chamber, and turbine, a difference occurs between the instantaneous flow rates flowing into and out of the cavities, resulting in an unbalanced flow rate. This phenomenon causes the continuity condition of flow between the inlet and outlet of each component to no longer hold during transient processes, requiring specific corrections to the outlet flow rates of the compressor, combustion chamber, and turbine using a volumetric effect model. When using component-level models for transient performance simulation, the influence of the volumetric effect needs to be considered. According to the mass conservation equation, the unbalanced flow rate... The calculation formula is as follows:

[0126]

[0127] In the formula, For unbalanced traffic, For component volume, For the first The inlet pressure of the component at any given moment; For the first The inlet pressure of the component at any given moment; This is the universal gas constant; The absolute temperature of the gas at the outlet of the cavity;

[0128] The actual outflow rate of the component at the current moment is:

[0129]

[0130] In the formula, To account for the transition state effect, the first The actual outflow rate of the component at any given moment; For the first time when the transition state effect is not considered The actual outflow of the component at any given time.

[0131] S34. Dynamic response of state parameters of the first component-level model based on time delay effect:

[0132] In this embodiment, during the engine transient process, it is typically assumed that the fuel supply and aerodynamic parameters will respond instantaneously. However, in reality, due to the inherent measurement lag of the sensor and signal transmission delay, the measured feedback value cannot change synchronously with the actual physical quantity, resulting in a time difference between the two. This dynamic response delay phenomenon caused by the measurement and transmission links is known as the time lag effect. This embodiment establishes a time lag effect model based on the hysteresis phenomenon in high and low pressure rotor speed measurement using a first-order inertial characterization. The transfer function between the speed sensor input and output is:

[0133]

[0134] In the formula, These are the input parameters for the sensor; To account for the time delay effect in the sensor's output parameters, The characteristic time constant is related to the sensor structure; For the Laplace operator.

[0135] The correction formula for the current rotor physical speed is as follows:

[0136]

[0137] In the formula, For the first The physical rotational speed of the rotor at any given moment; For the first time when the transition state effect is not considered The physical rotational speed of the rotor at any given moment; For the first time when the transition state effect is not considered The physical rotational speed of the rotor at any given moment; The exponential decay coefficient; This is the proportional correction factor.

[0138] in, and The expression is:

[0139]

[0140]

[0141] In the formula, For time step.

[0142] It should be noted that the four dynamic effect models of engine transition states involve several physical characteristic parameters with clear physical meanings. Due to the complexity of the engine structure and the large number of components, the theoretical design values ​​of these parameters are usually difficult to determine precisely, and even engines of the same model have individual differences. In the rotor dynamics model, the moment of inertia ( The calculation of the moment of inertia depends on the mass distribution of each component and its relative position to the axis of rotation. Actual components are often irregular geometries, and the accurate calculation of their mass and center of mass involves complex integration processes, leading to uncertainty in the calculated moment of inertia. For the hot-dip effect, the equivalent mass of the components involved... Equivalent heat transfer area Equivalent heat transfer length Due to factors such as ambiguous component boundary delineation and irregular geometric shapes, accurate measurement or calculation is difficult. Component volume in the volumetric effect. Characteristic time constant in time delay effect Similarly, due to unclear structural boundaries, complex flow paths, and sensor layout and structure, it is difficult to obtain precise values. Although physical characteristic parameters are difficult to calculate directly and accurately, reasonable approximate ranges can be obtained through engineering estimation methods based on the overall size and weight of the engine, as well as manufacturer design parameters. Based on this, by setting physically feasible intervals for each parameter and employing intelligent optimization methods to comprehensively identify and correct combinations of key physical characteristic parameters in the model, the relative error between simulation results and experimental data over the entire time series can be effectively reduced, thereby significantly improving the overall prediction accuracy of the transition state model. For example... Figure 5 As shown, the steps for correcting the transition state effect are as follows:

[0143] The uncertain combination of physical property parameters in the transition state effect model is set as the variable to be optimized, for example:

[0144]

[0145] The optimization variables and variables in this example are as shown in Table 3 (the variables and ranges can be adjusted according to the actual situation).

[0146] Table 3

[0147]

[0148] The expression for the optimization objective is defined as follows:

[0149]

[0150] In the formula, The target value; Variables to be optimized The i-th performance simulation parameter is obtained by performing transient performance simulation after inputting the first component-level model. is the i-th performance test parameter in the performance test parameters after noise filtering, and t represents the t-th time of the engine transient state simulation. This represents the entire transition state simulation time interval.

[0151] Based on the intelligent optimization algorithm, the optimal combination of physical characteristic parameters is found, and then the corresponding physical characteristic parameters of the dynamic effect model in the original engine component-level model are replaced with it. The relative error between the transition state simulation data and the test data is checked to see if it meets the requirements until the requirements are met or the number of attempts is limited, and the optimal combination of physical characteristic parameters is output.

[0152] Thus, by using the model corrected at the design operating point as a benchmark, performing steady-state characteristic correction, and then, based on the corrected characteristic diagram, considering the transient physical effects correction in a series correction logic, accurate matching between the model's full-condition simulation data and the test full-condition data can be achieved.

[0153] To illustrate whether optimizing transient effect parameters improves model simulation accuracy, three methods were compared. Method 1 uses the moment of inertia provided by the designer, Method 2 optimizes only the moment of inertia, and Method 3 optimizes the parameters of four transient effects. The model simulation process is a rapid acceleration and deceleration process between idle and maximum states, with a time step of 0.03 s. The maximum relative errors of the three methods for the five measurement parameters are shown in Table 4. The parameter comparison and relative error of Method 3 at each time series point are shown in Table 4. Figure 7-12 As shown, Figure 7 middle P t3 This is the total outlet pressure of the compressor. Figure 8 middle P t7 The total pressure at the turbine outlet. Figure 9 middle T t3 This refers to the total temperature at the compressor outlet. Figure 10 middle T t7 The total temperature at the turbine outlet. Figure 11 In this context, N represents the relative rotational speed. Figure 12 middle, T t7 This is the total temperature at the turbine outlet.

[0154] Compared with Method 1, the moment of inertia of Method 2 is calculated from the engine design reference value of 0.00616. Optimized to 0.00556 The changes were relatively small, which is related to the fact that the moment of inertia is an indirect measurement, and its measurement value itself has a certain degree of reliability. The maximum relative error decreased from 14.5% to 14.2%, a reduction of 0.3%, indicating a slight improvement in accuracy. The maximum error values ​​all occurred at the temperature parameter T. t7 The maximum error reduction is directly related to the objective of the optimization algorithm, which is the maximum relative error of all time series points across the five parameters. For the other four parameters... N , P t3 , P t7 and T t3 The maximum relative errors changed by -0.4%, -0.5%, -0.1%, and 0.2%, respectively. The average maximum relative error of the five parameters decreased by 0.22%, indicating that optimizing the moment of inertia can play a certain role in reducing transient errors.

[0155] The biggest difference between Method 3 and Method 2 lies in the temperature. T t3 and T t7 The relative errors decreased by 1.3% and 4.5% respectively, showing relatively large changes. The maximum relative error for all parameters was 12.9%, a reduction of 1.3% compared to Method 2, and a greater improvement in accuracy compared to optimizing only the moment of inertia. N , P t3 and P t7 The maximum relative errors changed by 1.4%, -0.4%, and 1.1% respectively, and the average maximum relative error of the five parameters decreased by 0.74%, indicating that optimizing the parameters of the four transient effects yields better results than optimizing only the moment of inertia. Furthermore, according to... Figures 7-11 The test data shows that the engine speed and pressure can stabilize in a relatively short time, but the two temperature parameters are difficult to stabilize within the current control plan. This indicates that the transient effect of the two temperature parameters is most pronounced in the micro turbojet engine, especially... T t7 Due to factors such as heat exchange, temperature changes exhibit a strong lag. Compared to Method 2, Method 3 shows a significant reduction in relative temperature error, while errors in rotational speed and pressure increase. This is related to the selection of the maximum error for the five parameters during optimization. Within a certain number of optimization iterations, optimization is prioritized for the stronger heat immersion effect, while rotational speed and pressure are considered less. If the model does not consider the heat immersion effect, the temperature will exhibit rapid changes, such as... Figure 12 As shown, Figure 12 middle, T t7 The total temperature at the turbine outlet is determined according to methods 2 and 3, based on the parameters.T t7 In comparison, especially at the end of rapid deceleration, Method 3, after considering the heat immersion effect, produces a slower temperature change in the model simulation, preventing it from dropping too quickly and resulting in higher simulation accuracy.

[0156] Table 4

[0157]

[0158] S4. Perform variable-dimensional simulation to obtain the final corrected target component-level model;

[0159] Specifically, the second component-level model, after performance characteristic map segmentation and physical characteristic corrections, expands the model's application range (especially in low-temperature conditions) and exhibits good accuracy in both steady and transient states. However, it still assumes a uniform flow field (0-dimensional) for the outlet temperatures of each component. In the overall gas path, the turbine section contributes most significantly to the "non-uniformity" of the temperature field: after mixing with the guide vane / rotor and secondary flow, the hot spot in the upstream combustion chamber still forms a significant radial temperature gradient and peak migration at the turbine outlet. In contrast, the radial temperature difference between the compressor and most pipe sections is generally weaker and more easily smoothed out by the mixing process. Therefore, this embodiment, based on the second component-level model, establishes a zero-dimensional to one-dimensional temperature radial distribution mapping model using a variable-dimensional simulation method, and uses this model as the final corrected target component-level model.

[0160] For example, such as Figure 6 As shown, in one specific embodiment, priority is given to modeling the turbine outlet temperature distribution to achieve the maximum correction effect on key error sources at the lowest cost. In engineering, turbine exhaust temperature measurement points are often arranged at multiple points along the circumference of the casing, and radially located in the upper-middle blade height (approximately 50%–80% of the blade height) near the blade tip to balance thermal load sensitivity and installation feasibility. Sensor readings are essentially "sampled values ​​at local blade heights," which are not comparable to 0-dimensional uniform temperatures, easily amplifying the deviation between the model and the actual measurement. To ensure that the simulated temperature corresponds one-to-one with the actual sensor height, thereby improving the accuracy of error assessment and correction, this embodiment introduces a variable-dimensional simulation method for temperature distribution. The assumption of a uniform temperature field (0-dimensional) taken from the cross-section is mapped to a one-dimensional temperature distribution along the blade height based on the radial temperature distribution coefficient (RTDF). Using the variable-dimensional simulation method of this embodiment, the temperature at the corresponding height can be compared with the actual sensor installation location to obtain a higher accuracy error between the temperature sampled value and the simulated value. The variable-dimensional simulation method for turbine exhaust temperature distribution is as follows: Figure 6As shown, a typical high-pressure turbine blade is divided into multiple segments along its height. Here, four segments are used as an example, including five nodes (blade root, 25% blade height, 50% blade height, 75% blade height, and blade tip). Measured data shows that the gas temperature is lowest at the blade tip and root, and highest at 75% blade height. To simplify the model design, it is assumed that the gas temperature changes linearly along the blade height from the point of maximum temperature to the blade tip and root. Through this simplification, a one-dimensional radial temperature field can be constructed based on a zero-dimensional component model, achieving a 0-dimensional to 1-dimensional variable-dimensional mapping.

[0161] First, the combustion chamber inlet and outlet temperatures are obtained using the engine's second-component-level model (0-dimensional). A reference temperature is defined based on the combustion chamber's temperature rise:

[0162]

[0163] In the formula, This refers to the combustion chamber inlet temperature. This refers to the combustion chamber outlet temperature. This is a reference temperature.

[0164] Based on the radial temperature distribution coefficient obtained from the experiment, the highest combustion gas temperature at the blade height is calculated as follows:

[0165]

[0166] In the formula, This represents the highest combustion temperature on the blade. For reference temperature; is the radial temperature distribution coefficient.

[0167] At the location with the highest exhaust gas temperature at the blade height, the lowest exhaust gas temperature at the blade root and blade tip can be obtained as follows:

[0168]

[0169] In the formula, The lowest combustion temperature at the leaf root and leaf tip; The combustion chamber outlet temperature is used; linear interpolation is applied between the blade root (0) and 75% blade height, and between 75% blade height and blade tip (100% blade height) to obtain the combustion gas temperature at five nodes. Thus, originally a single The uniform cross section described was replaced with a discrete one-dimensional temperature field varying along the blade height. This allows for a simplified characterization of the non-uniform flow field at the turbine exit. Based on a high-precision component-level model that performs segmented characteristic correction and physical characteristic correction, and using the radial temperature gradient distribution obtained from variable-dimensional simulation, the assumption of a uniform flow field can be corrected. This upgrades the dimensionality of the high-precision aero-engine component-level model, addressing the shortcomings of zero-dimensional simulation. It provides more realistic thermodynamic conditions for early-stage engine performance prediction, timing-optimized control strategy analysis, and later-stage engine health monitoring and fault diagnosis.

[0170] In summary, through step S2 (characteristic map segmentation correction), step S3 (physical characteristic correction), and step S4 (variable-dimensional simulation), an adaptive correction method for aero-engines based on multidisciplinary coupling was constructed. This method, while preserving the computational efficiency and engineering deployability of the component-level model, collaboratively suppresses distortions in key rotating components, transition state deviations caused by individual differences in structural and flow parameters, and turbine back-end temperature radial distribution errors caused by the uniform flow field assumption. This achieves adaptive high-precision correction over a wide operating range (especially low operating conditions). Therefore, this invention significantly improves the model's ability to represent the real physical characteristics of engines, enabling simulation outputs to be consistent with sensor placement and sampling height, thus facilitating engine performance evaluation and fault diagnosis.

[0171] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive correction method for aero-engines based on multidisciplinary coupling, characterized in that, include: Construct component-level models of aero engines; Based on the design parameter range, multiple sets of design parameter combinations corresponding to the design operating point are obtained. Each set of design parameter combinations is input into the component-level model to obtain the engine performance simulation parameters under each set of design parameter combinations. The optimization objective is to minimize the maximum relative error between the engine's performance simulation parameters and the performance test parameters corresponding to the design operating point, and an intelligent optimization algorithm is used to obtain the optimal combination of design parameters for the design operating point. The optimal design parameter combination is input into the component-level model for simulation to obtain the coupling coefficient corresponding to each characteristic parameter of the compressor and turbine in the characteristic diagram and the target converted speed corresponding to the optimal design parameter combination; Based on the actual converted speed and the target converted speed, the correction coefficients corresponding to each characteristic parameter of the compressor and turbine are obtained. Based on the correction coefficients and coupling coefficients, and based on the adaptive segmented optimization strategy of multiple steady-state conditions, the characteristic parameters corresponding to each segment interval of the characteristic map of the compressor and turbine are optimized to obtain the target characteristic parameters. Based on the adaptive splicing strategy and the target characteristic parameters, the corrected characteristic map is obtained. Based on the corrected characteristic map, the first component-level model after segmented characteristic correction is obtained. The adaptive splicing strategy is as follows: After the characteristic parameters corresponding to each segment interval are optimized, the range of the converted speed line when optimizing the next segment interval is determined. That is, the two converted speed lines where the speed of the minimum speed operating point in the steady-state operating point of the high speed interval is located are determined. This ensures that when performing subsequent optimization, the converted speed lines where all steady-state operating points of the previous segment interval are located will remain unchanged. Only the subsequent converted speed lines will be optimized, and the subsequent segment interval optimization steps will remain unchanged until the last segment interval is optimized. This achieves the adaptive splicing of characteristic maps to obtain characteristic maps according to the splicing logic of speed line locking. The physical properties of the first component-level model are modified based on inertial effect, thermal soaking effect, volume effect, and time delay effect to obtain the second component-level model; Based on the second component-level model, and through the variable-dimensional simulation method, a temperature radial distribution mapping model that maps from zero-dimensional to one-dimensional is established, and the temperature radial distribution mapping model is used as the final corrected target component-level model.

2. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, A component-level modeling method is used to construct a component-level model of an aero-engine.

3. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, Design parameters include: inlet airflow, compressor efficiency, compressor pressure ratio, combustion chamber efficiency, total pressure loss of combustion chamber, fuel calorific value, combustion chamber load coefficient, turbine efficiency, shaft mechanical efficiency, total pressure loss of turbine outlet, and total pressure loss of intake duct; performance simulation parameters include: relative speed, total temperature of compressor outlet, total pressure of compressor outlet, total pressure of turbine outlet, and total temperature of turbine outlet.

4. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, The Latin hypercube sampling method is used to collect multiple sets of design parameter combinations under steady-state conditions from within the design parameter range.

5. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, The steps to obtain the correction coefficients for each characteristic parameter of the compressor and turbine based on the actual and target converted speeds are as follows: In the formula, A polynomial function with corrected coefficients; For actual converted speed; Calculate the rotational speed for the target; The coefficient of the zeroth term of the polynomial function; The coefficient of the first term of the polynomial function; where is the coefficient of the quadratic term in the polynomial function; The coefficients of the cubic terms of the polynomial function; The target converted speed is the corrected converted speed, in hour =1, that is =1.

6. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, The steps to obtain the target characteristic parameters by optimizing the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram based on the correction coefficient and coupling coefficient and the adaptive piecewise optimization strategy based on multiple steady-state conditions are as follows: Based on the position of multiple steady-state conditions on the characteristic diagram, the entire characteristic diagram is divided into several segmented intervals; According to the preset range of polynomial coefficients, obtain multiple combinations of polynomial coefficients; Based on each set of polynomial coefficient combinations, obtain the correction coefficient corresponding to each characteristic parameter under each set of polynomial coefficient combinations; Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and according to the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected according to the adaptive segmented optimization strategy of multiple steady-state conditions to obtain the target characteristic parameters.

7. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 6, characterized in that, The constraints are: The efficiency constraints for compressors and turbines are: efficiency increases first and then decreases in the high-speed range, and efficiency continues to decrease in the medium-speed and low-speed ranges. The constraints for pressure ratio and equivalent flow rate of adjacent speed lines are: as the speed decreases, both the pressure ratio and equivalent flow rate at the same equivalent flow rate decrease; and the pressure ratio of the low speed line ≤ the pressure ratio of the high speed line, and the linear flow rate of the low speed line ≤ the linear flow rate of the high speed line.

8. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 6, characterized in that, The steps to obtain the target characteristic parameters are as follows: Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and according to the preset constraints and the adaptive piecewise optimization strategy for multiple steady-state conditions, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected respectively. Based on the coupling coefficient and the correction coefficient corresponding to each polynomial coefficient combination, and based on the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval are corrected according to the adaptive segmented optimization strategy of multiple steady-state conditions to obtain the initial characteristic parameters. Initial characteristic maps are obtained based on initial characteristic parameters; simulation calculations are performed on the component-level model based on the initial characteristic maps to obtain the engine's performance simulation parameters under each combination of design parameters. Obtain the relative errors of the engine's performance simulation parameters and performance test parameters under each combination of design parameters; Intelligent optimization is performed with the goal of minimizing the maximum relative error among all relative errors. The polynomial coefficient combination corresponding to the minimum maximum relative error is taken as the target polynomial coefficient combination, and the correction coefficient corresponding to the target polynomial coefficient combination is taken as the target correction coefficient. Based on the target correction coefficient and coupling coefficient, and according to the preset constraints, the characteristic parameters of the compressor and turbine in each segment interval of the characteristic diagram are corrected according to the adaptive segmented optimization of multiple steady-state conditions to obtain the target characteristic parameters.

9. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 8, characterized in that, The expression for the target characteristic parameter is: In the formula, the characteristic parameters include pressure ratio, efficiency, and converted flow rate. Target pressure ratio; This is the original pressure ratio; The coupling coefficient is the ratio of the pressure. This is the target correction factor corresponding to the pressure ratio; For target efficiency; For original efficiency; The coupling coefficient corresponds to the efficiency. This is the target correction coefficient corresponding to efficiency; Calculate the flow rate for the target; To calculate the flow rate; To calculate the coupling coefficient corresponding to the converted flow rate; The target correction factor is used to calculate the flow rate.

10. The adaptive correction method for aero-engines based on multidisciplinary coupling according to claim 1, characterized in that, The steps for correcting the physical properties of the first component-level model based on inertial effects, thermal soaking effects, volume effects, and time delay effects to obtain the second component-level model are as follows: The rotor speed of the first component-level model is corrected based on the inertial effect, the airflow temperature of the first component-level model is corrected based on the hot soaking effect, the outlet flow rate of the sub-component in the first component-level model is corrected based on the volume effect, and the dynamic response of the state parameters of the first component-level model is corrected based on the time delay effect.

Images