A parameter processing method for whole machine performance analysis of ground bench endurance test

Abstract

The application belongs to the field of aero-engine design, and is a parameter processing method for analyzing the overall performance of a ground bench endurance test. First, steady-state test data of the engine endurance test is obtained, and the steady-state performance test results corresponding to the specified speed state are processed. According to the performance calibration test results, the baseline steady-state performance model corresponding to the engine of the bench is formed through identification calculation, and the relative deviation between the performance parameter test value and the baseline performance calculation value is calculated. Through statistical regression, the regression function curve of the performance parameter calculation relative deviation with the test cycle is obtained, and the performance parameter test value of each test is reconstructed in combination with the baseline performance calculation value. Through similarity calculation based on the performance model, the performance parameter conversion value under the standard atmospheric conditions of the static sea level is obtained, and the unified comparison of the performance parameters is realized. The method avoids the similarity conversion deviation caused by the fixed and unchanged dimensionless parameters θ and δ index values, and improves the processing accuracy of the endurance test parameter results.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of aero-engine design, and particularly relates to a parameter processing method for whole-machine performance analysis of ground bench endurance test. BACKGROUND

[0002] In the endurance test process, due to the long time span and large difference in temperature and pressure of the inlet flow, the engine usually repeatedly runs at a certain relatively fixed rotating speed state, and the engine has different degrees of performance attenuation. Since the performance parameter measurement results in the test process are affected by factors such as test sensors, test environment and measurement accuracy, the performance parameter measurement results will have different degrees of deviation. Therefore, it is necessary to reasonably process the performance parameter measurement results, and then effectively evaluate the performance change of the engine.

[0003] The existing technical solution is based on the dimensionless parameter conversion method of the similarity principle, and the test results reflecting the performance parameter changes of the engine in the endurance test process are calculated. The related technical solutions are as follows:

[0004] Step 1: process the engine steady-state performance parameter test results of each endurance test;

[0005] Step 2: according to the calibration test results, the steady-state performance data of the specified engine state is calculated by interpolation;

[0006] Step 3: according to the similarity conversion relationship of the dimensionless parameter, the performance parameter results under the standard sea level condition are calculated;

[0007] Step 4: analyze the performance change of the performance parameter results converted to the static sea level standard day with the test times.

[0008] In the endurance test process, the engine usually repeatedly runs at a certain relatively fixed high rotating speed state. Since the measured parameters of the engine in the test process are affected by factors such as test environment and measurement accuracy, the directly measured performance parameter results will have random errors. In the endurance test process of the engine, the time span is long, the temperature and pressure of the inlet flow change greatly, and there are different degrees of performance attenuation. In the calculation process, the dimensionless parameter conversion is carried out according to the unchanged θ and δ index values in formulas (1-3), which will cause a certain deviation in the conversion results.

[0009]

[0010] Therefore, how to more accurately and effectively compare the performance parameters of the engine in different test times in the endurance test process is a problem to be solved. SUMMARY

[0011] The application aims to provide a parameter processing method for ground bench endurance test whole machine performance analysis, so as to solve the problem of large deviation of the processing result of the steady state performance parameter of the engine under different working conditions in a long time operation process in the prior art, and improve the effectiveness of the data analysis result.

[0012] The technical solution of the application is: a parameter processing method for ground bench endurance test whole machine performance analysis, comprising: obtaining test data of an engine in a stable rotating speed time period, and processing to obtain an engine steady state performance parameter test result of each endurance test; according to a steady state performance test result of a performance calibration test, interpolating to calculate a steady state performance parameter value of a specified engine state; according to a performance calibration test result, calculating a performance correction factor of each component, and combining a original steady state performance model to form a baseline steady state performance model corresponding to the engine; according to a working condition parameter of the current test data, combining the baseline steady state performance model corresponding to the engine, and calculating an engine baseline performance parameter value under a current working condition; according to a current test result and a baseline performance parameter value corresponding to the current test, calculating a relative deviation amount of the performance parameter, and performing statistical regression on the performance parameter deviation amount to obtain a function relationship curve of the test number and the performance parameter calculation relative deviation amount; according to the deviation function relationship curve and the baseline steady state performance model calculation result, the test result of each test parameter in the endurance test process can be reconstructed, the performance calculation value of the reconstructed test data under the standard sea level condition is obtained through the baseline performance model identification calculation, and the unified comparison of the performance parameters is realized.

[0013] Preferably, the processing method of the engine steady state performance parameter is: performing sliding average on the obtained test data, and performing equal annulus or equal flow average on the cross section temperature and pressure parameter to obtain a steady state performance parameter processing result of the selected engine state.

[0014] Preferably, the interpolation calculation formula of the steady state performance data is:

[0015]

[0016] In formula (4), Y represents a performance parameter, N1r represents a low pressure rotor conversion rotating speed, the subscript start represents an interpolation calculation starting rotating speed, the subscript end represents an interpolation calculation ending rotating speed, and the subscript current represents a current interpolation rotating speed.

[0017] Preferably, the calculation method of the performance correction factor of each component is:

[0018] The engine gas path performance calculation and analysis is performed, and the specific formula is:

[0019] Z=h(X) (5)

[0020] In formula (5), h represents a functional relationship between the measurement parameters and the component performance correction factors, the relationship is expressed by the engine steady-state performance model, Z represents the measurement parameter vector calculated by the steady-state performance model, and X represents the component performance correction factor vector;

[0021] If the component performance changes, the degree of change of the component performance parameter is represented by δ, and a first-order Taylor series expansion of h(x) is performed at the given operating point to obtain:

[0022] h(X+δX)=h(X)+H·δX +HOT (6)

[0023] The influence coefficient matrix H is obtained, and the mathematical expression is:

[0024]

[0025] Neglecting the influence of the high-order terms of the influence coefficient matrix H, we obtain:

[0026] h(X+δX)=h(X)+H·δX (8)

[0027] Through matrix conversion calculation, we obtain:

[0028] δX=(H T H) -1 H T δZ (9)

[0029] The nonlinear equation set is solved by the Newton-Raphson algorithm to obtain the component performance correction factor calculation value that meets the measurement parameter calculation accuracy requirement of the objective function.

[0030] Preferably, the engine baseline performance calculation method under the current operating condition is:

[0031] ParaBase= f(Pamb,Ma,ΔTamb,N1r,AnsynReference) (10)

[0032] In formula (10), Pamb represents the ambient pressure, Ma represents the Mach number, ΔTamb represents the ambient temperature deviation, N1r represents the low-pressure rotor converted speed, AnsynReference is the result calculated according to the performance calibration test data, and f represents the baseline steady-state performance model. The baseline performance parameter value corresponding to the current test operating condition is obtained by calculation of the baseline steady-state performance model, and ParaBase represents the baseline performance calculation value under the current test operating condition.

[0033] Preferably, the calculation formula of the performance parameter relative deviation amount is:

[0034]

[0035] ParaTest represents the test result of the current test, ParaBase represents the baseline performance calculation value under the current test condition, and ParaDev represents the calculation relative deviation amount of the current test result and the baseline performance result.

[0036] Preferably, the calculation formula of the parameter reconstruction result is:

[0037] ParaSmooth = ParaBase + ΔParaSmooth (12)

[0038] In formula (12), ParaSmooth represents the parameter reconstruction result of the current test, ParaBase represents the baseline model calculation value under the current condition, and ΔParaSmooth represents the parameter deviation amount calculated by the deviation fitting regression curve.

[0039] A parameter processing method for overall performance analysis of a ground bench endurance test of an engine is provided. After obtaining test data in a stable speed state time period of the engine, bad points are removed, cross-section averages and sliding averages are used to reduce parameter measurement processing deviation, and interpolation is used to calculate steady-state performance data corresponding to a specified speed state. Then, according to performance calibration test results, performance correction factors of each component are identified and calculated, and a baseline steady-state performance model corresponding to the engine is formed. According to the working condition parameters of the current test data, the baseline performance parameter result under the current working condition is calculated in combination with the baseline steady-state performance model corresponding to the engine, and the relative deviation of the performance parameter test value and the baseline performance calculation value is calculated. The regression function curve of the performance parameter calculation relative deviation amount with the test number is obtained by statistical regression, and then the performance parameter value of each test is reconstructed in combination with the baseline performance model parameter calculation result. According to the reconstructed parameter test result and the baseline steady-state performance model, the corresponding performance correction factors of each component are identified and calculated, and the performance parameter conversion value under the standard sea level condition is calculated by the baseline steady-state performance model. This result can replace the dimensionless parameter conversion result to realize the unified comparison of the performance parameters. This method can significantly reduce the influence of random errors of the measured parameters, effectively reflect the performance influence caused by changes in the incoming flow temperature and pressure and the geometric configuration, avoid the similar conversion deviation caused by the fixed dimensionless parameters θ and δ index values, and improve the processing precision of the endurance test parameter results. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions provided by the present application, the following will briefly introduce the drawings. Obviously, the drawings described below are only some embodiments of the present application.

[0041] Figure 1 It is a schematic diagram of the overall process of the present application.

[0042] Figure 2 A schematic diagram of a fitting regression result for calculating a relative deviation amount of a parameter of the present application;

[0043] Figure 3 A schematic diagram of a result converted to a standard sky standard sky for a measurement parameter result of the present application. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme in the present application will be described in more detail below with reference to the drawings in the present application.

[0045] A parameter processing method for analyzing the performance of an entire machine in a ground bench endurance test, comprising the following steps:

[0046] Step S100, obtaining test data in a stable speed state time period of an engine, and processing to obtain an engine steady-state performance parameter test result corresponding to each test in the endurance test;

[0047] The whole machine test parameters include environmental pressure, inlet total pressure, inlet total temperature, low-pressure rotor speed, high-pressure rotor speed, thrust, air flow, fuel flow, fan outlet total temperature and total pressure, high-pressure compressor outlet total temperature and total pressure, high-pressure turbine outlet total temperature, and low-pressure turbine outlet total temperature and total pressure.

[0048] The processing method of the engine steady-state performance parameters is as follows: the obtained test data is subjected to bad point elimination, the cross-section temperature and pressure parameters are subjected to equal ring surface or equal flow average, the data at different collection times are subjected to sliding average processing to remove random errors in the performance parameter measurement results, so as to ensure the accuracy of subsequent calculation and analysis results.

[0049] Step S200, interpolating and calculating steady-state performance data of a specified engine state according to the steady-state performance test result of the current test;

[0050] The interpolation calculation formula of the steady-state performance parameters is as follows:

[0051]

[0052] In formula (1), Y represents the performance parameter, N1r represents the low-pressure rotor converted speed, the subscript start represents the interpolation calculation starting speed, the subscript end represents the interpolation calculation ending speed, and the subscript current represents the current interpolation speed.

[0053] The steady-state performance data corresponding to the specified engine state is obtained by interpolating and processing the engine steady-state performance parameter test results.

[0054] Step S300, according to the performance calibration test results, identify and calculate the performance correction factors of each component, and combine the original steady-state performance model to form the baseline steady-state performance model corresponding to the engine;

[0055] The calculation method of the performance correction factor of each component is as follows:

[0056] The engine gas path performance calculation and analysis is performed, and the specific formula is as follows:

[0057] Z = h (X) (2)

[0058] In formula (2), h represents the functional relationship between the measurement parameters and the performance correction factors of the components (the steady-state performance model of the engine), Z represents the measurement parameter vector, and X represents the performance correction factor vector of the components;

[0059] The functional relationship between the measurement parameters and the performance correction factors of the components is described using the steady-state performance model of the engine. According to the required measurement parameter test results, the corresponding performance correction factors of the components can be calculated through the identification algorithm.

[0060] The steady-state performance data obtained through step S200 form the measurement parameter vector Z.

[0061] If the performance of the components changes, the parameter change degree is represented by δ, and the first-order Taylor series expansion of h (x) is performed at the given working point, and the following is obtained:

[0062] h (X + δX) = h (X) + H · δX + HOT (3)

[0063] The influence parameter matrix H is obtained, and its mathematical expression is as follows:

[0064]

[0065] The high-order term influence of the influence parameter matrix H is ignored, and the following is obtained:

[0066] h (X + δX) = h (X) + H · δX (5)

[0067] Through matrix conversion calculation, the following is obtained:

[0068] δX = (H T H) -1 H T δZ (6)

[0069] The Newton-Raphson algorithm is used to solve the nonlinear equation set, and the calculated values of the performance correction factors of each component that meet the measurement parameter calculation accuracy requirements of the objective function are obtained.

[0070] The component performance correction factor vector X is solved by formulas (2) to (6) to obtain the performance correction factors of each engine component corresponding to the test results. The engine components include components such as fan, compressor and turbine.

[0071] Step S400: Based on the operating parameters of the current test data and the corresponding baseline steady-state performance model of the engine, calculate the engine baseline performance parameter values ​​under the current test conditions. The calculation method for the engine baseline performance parameter values ​​under the current operating conditions is as follows:

[0072] ParaBase= f(Pamb,Ma,ΔTamb,N1r,AnsynReference) (7)

[0073] In formula (7), Pamb represents the environmental pressure of the current test, Ma represents the Mach number of the current test (Ma = 0), ΔTamb represents the environmental temperature deviation of the current test, N1r represents the converted speed of the low-pressure rotor, AnsynReference is the performance correction factor of each component identified and calculated based on the performance calibration test data, f is the baseline steady-state performance model, and ParaBase represents the baseline performance calculation value under the current test conditions. The baseline performance parameter values ​​corresponding to the current test data are obtained by calculating through the baseline steady-state performance model.

[0074] Step S500: Based on the current test results and the baseline performance parameter values ​​corresponding to the current test, calculate the relative deviation of the performance parameters, and perform statistical regression on the deviation results to obtain the functional relationship curve between the number of tests and the calculated relative deviation of the performance parameters.

[0075] The formula for calculating the relative deviation of performance parameters is:

[0076]

[0077] In formula (8), ParaTest represents the result of the current test run, ParaBase represents the baseline performance parameter value under the current operating conditions, and ParaDev represents the calculated relative deviation between the current test result and the baseline performance parameter value.

[0078] Step S600: Based on the deviation function relationship curve and the calculation results of the baseline steady-state performance model, obtain the performance parameter reconstruction results for each test during the sustained test.

[0079] The formula for calculating the parameter reconstruction result is:

[0080] ΔParaSmooth= f(test run number)*ParaBase (9)

[0081] ParaSmooth = ParaBase + ΔParaSmooth (10)

[0082] ParaSmooth in formula (9)~(10) represents the parameter reconstruction result of the current test, ParaBase represents the calculated value of the baseline steady-state performance model under the current working condition, and ΔParaSmooth represents the performance parameter deviation amount calculated through the deviation fitting curve, wherein f represents the regression function of the performance parameter calculation relative deviation amount obtained through statistical analysis.

[0083] In step S700, according to the reconstructed test data, the corresponding performance correction factor of each component is calculated through the baseline steady-state performance model identification, and the performance parameter value of the current test data under the standard sea level condition is calculated by combining the baseline steady-state performance model. The result can be used for engine performance parameter attenuation change analysis.

[0084] Reference point performance parameter = f (Pamb, Ma, ΔTamb, N1r, AnsynCurrent) (11)

[0085] In formula (11), Pamb represents the environmental pressure of the standard sea level, Ma represents the Mach number (Ma=0), ΔTamb represents the environmental temperature deviation of the standard sea level, N1r represents the low-pressure rotor conversion speed, AnsynCurrent is the performance correction factor of each component identified and calculated according to the current test data, and f is the baseline steady-state performance model.

[0086] After obtaining test data during the engine's stable speed period, this application reduces deviations in parameter measurement results through bad point elimination, cross-sectional averaging, and moving average, and interpolates to calculate steady-state performance data corresponding to a specified speed state. Then, based on the performance calibration test results, it identifies and calculates performance correction factors for each component and forms a baseline steady-state performance model for the engine. Based on the operating parameters of the current test data and the baseline steady-state performance model, it calculates the baseline performance parameters under the current operating conditions, and then calculates the relative deviation between the test and calculated performance parameters. Through statistical regression, it obtains the regression function curve of the relative deviation of the performance parameter calculation with the number of test runs, and then reconstructs the performance parameter values ​​for each test run based on the baseline performance model parameter calculation results. Based on the reconstructed parameter test results and the baseline steady-state performance model, it identifies and calculates the corresponding performance correction factors for each component, and calculates the converted performance parameters under the standard day conditions at a stationary sea level using the baseline steady-state performance model. This result can replace the dimensionless parameter conversion result, achieving a unified comparison of performance parameters. This method can significantly reduce the impact of random errors in parameter measurement, effectively reflect the performance impact caused by changes in incoming flow temperature and pressure and geometric configuration, avoid similarity conversion deviations caused by the fixed values ​​of dimensionless parameters θ and δ, and improve the processing accuracy of test parameter results during long-term commissioning.

[0087] As a specific implementation method, baseline performance test data is obtained through performance calibration test. Based on the engine's 1st to 500th bench endurance test data, test data of engine state N1r = 90% is obtained through formula (1). By selecting engine operating parameters, high and low pressure speeds, fuel flow, total temperature and pressure at high pressure turbine outlet, total temperature and pressure at low pressure turbine outlet, total temperature and pressure at fan outlet, total temperature and pressure at compressor inlet and outlet, and other performance measurement parameters, the performance correction factors of each component corresponding to the calibration test data are calculated through formulas (2) to (6). The baseline performance parameter values ​​under each actual test operating condition are calculated through formula (7). The relative deviation of the performance parameters for each test is calculated using formula (8). Figure 2 The figure shows the trend of the calculated relative deviation of the low-pressure turbine outlet total temperature T5 with the number of test runs. Figure 2 The curve formed is a deviation fitting regression curve in the form of a quadratic polynomial function, and the discrete points around the curve represent the relative deviations of the performance parameters; the actual measured parameters and reconstructed measured parameters of T5 at different test runs are as follows: Figure 2 As shown, the calculated results of the reconstructed measurement parameters under standard-day conditions at static sea level are as follows: Figure 3 As shown.

[0088] Table 1. Actual and reconstructed measurements of T5

[0089]

[0090] The T5 performance parameter calculation deviation curve fitting formula is as follows:

[0091] y = 0.000004652029x + 0.005085790751x - 0.004445847293 2

[0092] R 2 = 0.964986718768

[0093] In the above formula, x represents the test number, and y represents the performance parameter calculation relative deviation.

[0094] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.​

Claims

1. A parameter processing method for performance analysis of a ground-based long-term test bench, characterized in that, include: The test data during the stable speed operation period of the engine is obtained and processed to obtain the steady-state performance parameter test results of each test of the engine during the long-term test, including: performing a sliding average on the obtained test data, removing bad parameter values ​​from the cross-sectional temperature and pressure parameters, and performing equal toroidal or equal flow average to obtain the steady-state performance parameter processing results of the selected engine state. Based on the performance calibration test results, the steady-state performance test results at a specified engine speed are calculated by interpolation for each sustained test. Based on the performance calibration test results, the performance correction factors of each component are identified and calculated, and the baseline steady-state performance model of the engine is constructed by combining the original steady-state performance model. Based on the operating parameters of the current test data and combined with the corresponding baseline steady-state performance model of the engine, the baseline performance parameter values ​​of the engine under the current operating conditions are calculated. Based on the results of the current test run and the baseline performance parameter values ​​corresponding to the current test run, the relative deviation of the performance parameters is calculated, and statistical regression is performed on the deviation to obtain the functional relationship curve between the number of tests and the relative deviation of the performance parameters. The parameter reconstruction results for each test during the sustained test were calculated based on the functional relationship curves and the baseline steady-state performance model. Based on the reconstructed test results, the corresponding performance correction factors for each component are obtained through identification and calculation. Combined with the baseline steady-state performance model, the performance parameter values ​​of the current test data under standard-day conditions at static sea level are calculated. The formula for calculating the relative deviation of the performance parameters is as follows: ParaTest represents the results of the current test run, and ParaBase represents the baseline performance calculation value under the current test run conditions. The calculation formula for the parameter reconstruction is as follows: ΔParaSmooth = f(number of test runs) * ParaBase Where f represents the regression function of the relative deviation of the performance parameter ParaDev as a function of the number of test runs; ParaSmooth=ParaBase+ΔParaSmooth Where ParaSmooth represents the parameter reconstruction result of the current test run, and ΔParaSmooth represents the calculated value of the regression function deviation corresponding to the selected test run.

2. The parameter processing method for overall performance analysis during long-term ground test runs as described in claim 1, characterized in that, The formula for interpolation calculation is: Where Y represents the performance parameter, N1r represents the converted speed of the low-pressure rotor, the subscript start represents the starting speed of the interpolation calculation, the subscript end represents the ending speed of the interpolation calculation, and the subscript current represents the current interpolation speed.

3. The parameter processing method for overall performance analysis during long-term ground test runs as described in claim 1, characterized in that, in: ParaBase=F(Pamb,Ma,ΔTamb,Nlr,AnsynReference) In the formula, Pamb represents the ambient pressure of the current test run, Ma represents the Mach number, ΔTamb represents the ambient temperature deviation of the current test run, N1r represents the converted speed of the low-pressure rotor, F represents the baseline steady-state performance model of the engine, AnsynReference is the performance correction factor of each component identified and calculated based on the performance calibration test data, and ParaBase represents the baseline performance calculation value under the current test run conditions.

Images