A turbine tester probe temperature measurement error prediction method and system

Abstract

This invention provides a method and system for predicting probe temperature measurement errors in turbine testers, belonging to the field of aero-engine turbine testing and measurement technology. The method includes: determining the sampling range of three design variables—exit Mach number, casing outer surface heat transfer coefficient, and inlet total temperature; generating sample points within the design space using the Latin hypercube method; obtaining the total temperature error of the probe measurement point corresponding to each sample point using gas-thermal coupling simulation; constructing a global Kriging surrogate model based on the design variables and the target value of the temperature measurement error; and performing a global sensitivity analysis of the design space using the global Kriging surrogate model to quantify the influence of each design variable on the total temperature error; inputting the exit Mach number, casing outer surface heat transfer coefficient, and inlet total temperature under actual test conditions into the global Kriging surrogate model to obtain the corresponding predicted probe temperature measurement error value. The method described in this invention can achieve accurate quantification and rapid prediction of temperature measurement errors, improving the accuracy and reliability of turbine test temperature measurement.

Description

Technical Field

[0001] This invention belongs to the field of aero-engine turbine testing and measurement technology, specifically relating to a method and system for predicting the temperature measurement error of a turbine tester probe. Background Technology

[0002] Turbine performance testing is a crucial step in aero-engine development, and the accuracy of its measurements directly impacts the evaluation and optimization of turbine aerodynamic and thermodynamic performance. Temperature is one of the core measurement parameters, especially the temperature distribution in high-temperature regions such as the combustion chamber outlet and turbine inlet, which is of great significance for assessing turbine blade life, cooling efficiency, and overall engine performance. Computational fluid dynamics has played a key role in component and whole-engine simulation, but experimental verification remains an indispensable part of model development.

[0003] In actual turbine tests, a temperature gradient exists between the internal and external environments of the test chamber, resulting in a significant temperature difference between the casing wall and the mainstream combustion gas. This difference, in turn, affects the measurement accuracy of the temperature probe through heat conduction and radiation, introducing a non-negligible steady-state error. Existing research indicates that the steady-state error of the total temperature probe mainly includes three categories: velocity error, thermal conduction error, and radiation error. In non-uniform environments with strong temperature gradients, the internal heat conduction process of the probe is complex, affecting both its dynamic calibration accuracy and transient temperature measurement accuracy. Thermocouples, when measuring air temperature, are subject to strong radiative heat exchange interference, causing readings to deviate from the true temperature by tens of degrees Celsius.

[0004] Regarding the error mechanism of the probe itself, research shows that in low Mach number flow fields, radiation error increases with the distance of the measuring point from the mounting base, becoming the dominant source of error; while under high Mach number conditions, both thermal conductivity error and radiation error decrease with increasing distance. Thermal conductivity and radiation effects significantly reduce the surface temperature of the shield, affecting the actual temperature measurement results at the temperature measuring node. Furthermore, radiation error decreases with increasing airflow velocity and pressure, but increases with increasing airflow temperature and the temperature difference between the airflow temperature and the casing wall.

[0005] Besides the inherent error mechanism of the probe itself, its arrangement and the external environment can also significantly interfere with the measurement results. The blocking effect of the probe support can alter the local flow field structure, and the presence of the probe can cause deviations in wall static pressure, an increase in the outlet airflow angle, and a decrease in efficiency. The probe rake can induce interference from complex shock wave systems, leading to measurement results that deviate significantly from the actual flow conditions.

[0006] More critically, the influence of external boundary conditions is a significant factor. In multi-probe structures, due to the large radial temperature gradient of the support structure, the conduction and radiation errors of probes at different locations differ significantly, with the error of probes near the cryogenic casing being much larger than that of the probe at the center of the channel. The outer wall of the test chamber typically maintains a low temperature, creating a strong internal and external temperature gradient with the high-temperature combustion gas inside. This gradient conducts heat through the solid structure, forcing the temperature of the probe support and shield to be much lower than the total temperature of the combustion gas, significantly increasing conductive heat loss and exacerbating radiative heat transfer, resulting in measurement results that deviate severely from the true values.

[0007] In summary, the accuracy of temperature measurement in turbine testing is influenced by both the internal error mechanism of the probe and the external flow field environment. The internal and external temperature gradients directly affect the magnitude of heat conduction and radiation errors by altering the heat transfer state, leading to deviations between the measured temperature and the actual total airflow temperature. However, systematic research on the mechanisms of influence of internal and external temperature gradients is still lacking, especially under conditions of combined inlet total temperature, external wall heat transfer coefficient, and outlet Mach number. The evolution of temperature measurement errors and the quantitative relationships of key factors require further investigation. Therefore, it is necessary to conduct research on the influence of internal and external temperature gradients on temperature measurement errors, revealing the mechanisms and contributions of each parameter, and providing theoretical basis and data support for temperature measurement error correction and test condition optimization in turbine testing. Summary of the Invention

[0008] To address the unclear influence of internal and external temperature gradients on probe temperature measurement errors, this invention provides a method for predicting probe temperature measurement errors in turbine testers that considers radiation effects. This method can rapidly predict corresponding probe temperature measurement errors based on different test conditions, thereby achieving effective correction of measured temperatures under various conditions. Furthermore, the turbine tester probe temperature measurement error analysis method provides a reliable theoretical basis and technical support for temperature measurement error correction and test condition optimization in turbine testing. This method overcomes the limitation of traditional analyses that ignore the actual influence of internal and external temperature gradients. By establishing a tester simulation model that considers the coupling effect of internal and external temperature gradients and clearly defining the calculation method for probe temperature measurement errors, it achieves accurate analysis of temperature measurement errors.

[0009] To achieve the above objectives, in a first aspect, the present invention provides a method for predicting the temperature measurement error of a turbine tester probe, comprising the following steps: S1. Determine the sampling range of the three design variables: outlet Mach number, heat transfer coefficient of outer surface of casing, and inlet total temperature. Use the Latin hypercube method to generate sample points in the design space, perform gas-thermal coupling simulation analysis, and obtain the total temperature error of the probe measurement points corresponding to each sample point. S2. Based on the dataset consisting of design variables and target values ​​of temperature measurement error, construct a global Kriging surrogate model, and use the global Kriging surrogate model to perform a global sensitivity analysis on the design space to quantify the influence of each design variable on the total temperature error. S3. Input the exit Mach number, heat transfer coefficient of the outer surface of the casing, and total inlet temperature of the actual test conditions into the global Kriging proxy model to obtain the corresponding probe temperature measurement error prediction value.

[0010] Furthermore, after constructing the global Kriging proxy model in step S2, leave-one-out cross-validation is used to evaluate the model's accuracy. If the model's accuracy is... Then the model accuracy is determined to meet the requirements. Then, sampling continues within the design space, and gas-thermal coupling simulation analysis is performed to obtain the total temperature error of the probe measurement points corresponding to each sample point; until the model accuracy meets the requirements.

[0011] Furthermore, the global sensitivity analysis described in step S2 includes variable variance contribution rate analysis and main effect curve analysis. The variance contribution rate clarifies the influence weight of each design variable on the total temperature error, and the main effect curve analyzes the correlation trend between each design variable and the total temperature error.

[0012] Furthermore, gas-thermal coupling simulation analysis is performed to obtain the total temperature error of the probe measurement points corresponding to each sample point, including the following steps: S11. Perform full-circulation flow field simulation of the turbine tester to obtain the flow field parameters of the characteristic sections; S12. Extract a local simulation model of 1 / 4 of the full ring scale of the area near the probe in the turbine tester; S13. The total temperature and total pressure of the characteristic section obtained from the whole-machine full-circulation flow field simulation are used as the inlet boundary conditions of the calculation domain of the local simulation model, and the outlet boundary of the local simulation model is set as static pressure condition; a third type of boundary condition combining temperature and convective heat transfer coefficient is set on the outer surface of the casing, the outer surface of the hub and the outer surface of the probe of the turbine tester, while considering the influence of radiative heat transfer. S14. Define the calculation method for the total temperature probe temperature measurement error, and calculate the temperature measurement error of the probe measurement point under each test condition based on the calculation method.

[0013] Furthermore, the method for calculating the total temperature probe measurement error in step S14 includes: taking the average total temperature of adjacent fluid segments as the reference real fluid temperature at a position where the probe is offset by a set angle in the circumferential direction and is not affected by its flow. ; Calculate the average temperature of multiple solid measurement points on the probe. The difference between the reference real fluid temperature and the average temperature of the probe solid measurement point is defined as the total temperature error of the probe measurement point. ,Right now .

[0014] Furthermore, in S13, the DTRM radiation model is used to calculate radiative heat transfer, and the local simulation model uses the SST k-ω turbulence model to solve the steady-state Reynolds time-averaged NS equations.

[0015] Furthermore, the computational domain of the local simulation model includes the solid casing of the test instrument, the internal fluid domain, and the solid domain of the probe. The probe base fits into the outer surface of the test instrument casing through a groove to simulate the actual installation state of the probe.

[0016] Secondly, the present invention can also provide a turbine tester probe temperature measurement error prediction system, including a total temperature error acquisition module, a quantization module and a prediction module; The total temperature error acquisition module is used to determine the sampling range of three design variables: outlet Mach number, heat transfer coefficient of outer surface of casing and inlet total temperature. It uses the Latin hypercube method to generate sample points in the design space, performs gas-thermal coupling simulation analysis, and obtains the total temperature error of the probe measurement points corresponding to each sample point. The quantification module constructs a global Kriging surrogate model based on a dataset consisting of design variables and target values ​​of temperature measurement error. It then uses this global Kriging surrogate model to perform a global sensitivity analysis of the design space, quantifying the degree of influence of each design variable on the total temperature error. The prediction module is used to input the exit Mach number, heat transfer coefficient of the outer surface of the casing, and total inlet temperature of the actual test conditions into the global Kriging proxy model to obtain the corresponding probe temperature measurement error prediction value.

[0017] The total temperature error acquisition module includes a flow field simulation unit and an error analysis unit; The flow field simulation unit is used to carry out full-circulation flow field simulation of the turbine tester to obtain flow field parameters of characteristic sections. It can also extract the area near the probe to establish a local simulation model at a 1 / 4 full-circulation scale, set the model boundary conditions, and carry out CFD numerical calculations. The error analysis unit is used to define the calculation method for the total temperature probe temperature measurement error and obtain the temperature measurement error of each sample point based on the calculation results of the flow field simulation module.

[0018] Thirdly, the present invention also provides a computer device, including a processor and a memory, wherein the memory is used to store a computer executable program, the processor reads part or all of the computer executable program from the memory and executes it, and the processor can realize the above-mentioned method for predicting the temperature measurement error of a turbine tester probe when executing part or all of the computer executable program.

[0019] Simultaneously, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it can realize the above-mentioned method for predicting the temperature measurement error of a turbine tester probe.

[0020] Compared with the prior art, the present invention has at least the following beneficial effects: Through the above method, this invention realizes an integrated approach for analyzing and predicting the temperature measurement error of turbine test probes considering the influence of radiation. Sample points are generated within the design space using the Latin hypercube sampling method, and error data is obtained through CFD calculations. Then, an error correlation correction model based on the Kriging surrogate model is established, enabling rapid prediction of temperature measurement errors. This method effectively solves the key technical problem of efficiently obtaining the total temperature probe measurement error in complex environments with internal and external temperature gradients and the influence of radiative heat transfer, avoiding the limitations of repeated high-cost numerical simulations in traditional methods.

[0021] Based on the conjugate heat transfer and radiation coupling model, a simulation model of the test instrument that can realistically reflect the effect of internal and external temperature gradients was constructed, and the calculation method of probe temperature measurement error was clearly defined, providing a reliable physical basis for error analysis.

[0022] The analysis and prediction method of this invention can quickly obtain the probe temperature measurement error according to different test conditions, and accurately correct the measured temperature accordingly, which significantly improves the accuracy and reliability of temperature measurement in turbine tests. It can be used to quantify the impact of typical test conditions such as inlet total temperature, casing heat dissipation intensity and outlet Mach number on temperature measurement accuracy, provide a basis for temperature measurement error correction and test condition optimization in turbine tests, and provide more accurate test data support for the performance evaluation and cooling design of turbine hot-end components. Attached Figure Description

[0023] Figure 1 This is a simulation model of the testing equipment; Figure 2 Definition of error for multi-point shielded total temperature probe; Figure 3 Establish a framework for the agency model and sensitivity analysis; Figure 4 Cross-validation for the surrogate model; Figure 5 Contribution rate of variance for each variable; Figure 6 The main effect curve. Detailed Implementation

[0024] To more intuitively demonstrate the objectives, technical solutions, and advantages of the present invention compared to the turbine tester probe temperature measurement error analysis and prediction method, the present invention will be further described in detail below with reference to a specific embodiment and the accompanying drawings.

[0025] However, it is worth noting that the embodiments described below are only introductions to some implementations of this application, not all implementations. Similarly, the accompanying drawings are merely illustrative in nature and are intended to explain this application, not to limit it. All other implementations obtained by those skilled in the art within the scope of this application without inventive effort are within the protection scope of this application.

[0026] All terms used herein (including technical and scientific terms) have the meanings commonly understood by one of ordinary skill in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid manner.

[0027] According to an embodiment of the first aspect of the present invention, firstly, a full-circulation flow field simulation of the entire test instrument is conducted to obtain the flow field parameters of the characteristic cross-section. For example... Figure 1 As shown, to accurately analyze the measurement error introduced by the probe measuring points, a local model was constructed near the probe for refined study. This local model contains three computational domains: the solid casing of the test chamber, the internal fluid domain, and the solid domain of the probe. In the assembly relationship, the probe base achieves a tight fit with the outer surface of the test chamber casing through a groove to realistically reflect the installation state in the actual test. The total temperature and total pressure of the characteristic section obtained from the simulation of the whole-circuit model are used as the inlet boundary conditions of the local model's computational domain, and the outlet boundary is set as a static pressure condition. To simulate the internal and external temperature gradients of the test chamber casing under different operating conditions, boundary conditions combining temperature and convective heat transfer coefficients are applied to the outer surface of the casing, the outer surface of the hub, and the probe extension surface.

[0028] A characteristic section refers to a representative cross-section selected in the flow channel of a turbine tester, typically located upstream of the probe installation position. Flow field parameters (such as total temperature and total pressure) on the characteristic section reflect the inlet conditions of the probe's environment. The computational domain is the physical space defined during numerical simulation, including both fluid and solid regions. Within this domain, fluid and energy conservation equations are solved to obtain the flow field and temperature distribution.

[0029] In this application, the total temperature and total pressure of the characteristic section obtained from the whole machine simulation are used as the inlet boundary conditions in the local simulation model, which ensures the physical connection between the local model and the overall flow field. By setting the third type of boundary conditions that combine temperature and convective heat transfer coefficient on the outer surface of the casing, the outer surface of the hub, and the outer surface of the probe, and by setting the surface emissivity on the probe and the test chamber casing to consider the influence of radiative heat transfer, the heat conduction, convection and radiative heat transfer processes experienced by the probe under the internal and external temperature gradients can be comprehensively and accurately simulated.

[0030] This application refines the computational domain of the local simulation model into a solid casing of the test chamber, an internal fluid domain, and a solid probe domain. It also accurately simulates the fit between the probe base and the outer surface of the test chamber casing via grooves, enabling the model to comprehensively capture the thermodynamic behavior of the probe under complex thermal environments. Specifically, the introduction of the solid casing allows the model to consider the solid thermal conduction effect caused by the temperature gradient between the casing and the mainstream combustion gas, avoiding temperature gradient distortion caused by neglecting the solid structure. The complete simulation of the internal fluid domain ensures accurate reproduction of the convective heat transfer of the fluid to the probe and the influence of flow field parameters on temperature measurement. The introduction of the solid probe domain directly simulates the probe's own thermal behavior, including its internal heat conduction and heat transfer with the fluid. The simulation of the probe base's fit with the outer surface of the test chamber casing via grooves accurately reproduces the thermal contact and conduction path between the probe and the casing during actual installation, ensuring the realism of heat transfer from the casing to the probe base and the internal heat conduction of the probe.

[0031] This invention analyzes a shielded total temperature probe configured with six measuring points. For example... Figure 2 As shown in Figure (a), the shielded total temperature probe consists of temperature measuring points, a probe support, a shield, and a mounting base. To improve the accuracy of temperature measurement, each temperature measuring point is equipped with a shield with a consistent geometric configuration. The inlet diameter of the shield is 4.0 mm, the diameter of the thermocouple junction is 1.0 mm, and the diameter of the thermocouple wire is 1 mm. The probe is filled with a cement-based material to block the fluid mixing effect between different shields. To account for the influence of radiation, a surface emissivity is set on the probe surface.

[0032] In this invention, the local model of the experimental device is simulated in three dimensions using the commercial software CFX to solve the steady-state Reynolds time-averaged Navier-Stokes equations. Due to SST... k-ω The model is highly practical in predicting heat transfer; the SST turbulence model is selected. k-ω Model. Based on fluid-structure interaction, a radiation model is also considered. This is due to the optical thickness of the gas flowing through the shielding chamber. =0.0006 ,in The absorption coefficient is set to 0.2m according to the ANSYS help documentation. -1 , The length of the shielded chamber is 0.003 m, much less than 1, indicating that the fluid is "optically thin," allowing radiation energy to easily penetrate it with almost no absorption. It can also be considered transparent, and its radiation effect can be ignored. In this case, only radiative heat transfer between solid surfaces is considered within the shield, while the absorption, emission, and scattering of radiation by the medium (fluid) are treated as "attenuation" and "enhancement" along the radiation propagation path. Therefore, the DTRM (Discrete Transfer Radiation Model) radiation model is selected for calculation. This application uses the DTRM radiation model to calculate radiative heat transfer. The DTRM model can meticulously track the transmission path of radiative energy in complex geometries (such as probes, casings, and hubs) and high-temperature combustion media, considering absorption, scattering, and wall reflection. It can accurately capture the transmission of radiative energy in complex geometries and non-uniform media, avoiding deviations in radiation error calculations. The SST k-ω model combines the advantages of the k-ε model in regions far from the wall with the accuracy of the k-ω model in regions near the wall. It is particularly suitable for flow field environments with complex boundary layers, adverse pressure gradients, and flow separation inside turbines. Using the SST k-ω turbulence model to solve the steady-state Reynolds time-averaged Navier-Stokes equations can accurately handle complex flow phenomena such as boundary layer separation and turbulent mixing in turbine flow fields, ensuring the accuracy of flow field parameter solutions. In the process of analyzing the temperature measurement error of the turbine test probe, it can more realistically reflect the heat exchange and fluid dynamic interaction between the probe and the surrounding environment.

[0033] To investigate the combined effects of flow, thermal conduction, and radiation on the temperature error at the measuring point under internal and external temperature gradients, Figure 2 As shown in Figure (b), the following error analysis method is defined: at a position where the probe is offset 5º circumferentially and is not affected by its flow, take the adjacent fluid line segment. L O1O6 The average total temperature is used as a reference for the actual fluid temperature. T L Simultaneously, the average temperature of the six solid measurement points on the probe was calculated. T M .Will T L and T M The difference is defined as the total temperature error at the probe measurement point. E T See equation (1).

[0034] (1) By offsetting the probe 5° circumferentially and ensuring it is unaffected by flow disturbances, the measured fluid temperature is made as close as possible to the actual fluid temperature, avoiding the influence of probe body disturbances on the local flow field. This circumferential offset avoids areas directly affected by the probe support or measuring point. At this reference position, the average total temperature of adjacent fluid line segments is taken as the reference actual fluid temperature. T L The impact of local temperature fluctuations is reduced through averaging.

[0035] According to an embodiment of a second aspect of the present invention, a method for predicting the temperature measurement error of a turbine tester probe is provided.

[0036] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. It should be noted that implementations not illustrated or described in the drawings or the main text of the specification are forms known to those skilled in the art and are not described in detail. Furthermore, the definitions of the components described above are not limited to the various specific structures, shapes, or methods mentioned in the embodiments, and those skilled in the art can easily modify or substitute them.

[0037] To efficiently obtain the total temperature error of measurement nodes under different internal and external temperature gradients, this invention establishes a correlation correction model for the total temperature error. First, the design variables are defined. x 1~ x Based on the sampling range of 3, a modeling and sensitivity analysis framework based on the Kriging surrogate model is constructed, such as... Figure 3 As shown. The implementation steps of this framework are as follows: Step S1: Generate uniformly distributed sample points within the design space using the Latin hypercube sampling method; Step S2: Perform gas-thermal coupling simulation analysis, and obtain the total temperature error at the probe measurement point using the error analysis method described above. E T Obtain the corresponding target value for temperature measurement error; Step S3: Based on the dataset of design variables and target values ​​of temperature measurement error, establish a global Kriging surrogate model; Step S4: Verify the model accuracy using leave-one-out cross-validation. If R... 2 If the accuracy is greater than 0.9, it proves that the model has high accuracy, and proceed to the next step. If it does not meet the requirements, continue sampling in the sample space and execute steps S1 to S3, and verify the model accuracy to ensure that the accuracy of the surrogate model meets the requirements.

[0038] Step S5: Finally, use a surrogate model that meets the accuracy requirements to perform sensitivity analysis of the design space, quantify the influence of each design variable on the total temperature error, analyze the contribution of variables, and perform main effect curve analysis.

[0039] A surrogate model was built using 30 samples. The accuracy of the cross-validation model is shown in [link to cross-validation]. Figure 4 The predicted values ​​and the actual values ​​are closely distributed along the 45° line. y = x Nearby, and R The value of 2² = 0.998 indicates that the surrogate model has high predictive accuracy and can be reliably used for subsequent sensitivity analysis. Based on this, a global sensitivity analysis method was used to quantify the contribution of each design variable to changes in performance indicators. Figure 5 The variance contribution rates of three design variables are shown, thus clarifying the importance of each design variable to the system performance. (Inlet total temperature) x 3 has the greatest impact on the total temperature error, accounting for 79.2% of the variance contribution; the heat transfer coefficient of the outer surface of the casing. x The variance contribution was second, accounting for 14.7%. Meanwhile, there was a strong interaction between the total inlet temperature and the heat transfer coefficient of the outer surface of the casing, while the interaction between the outlet Mach number and the heat transfer coefficient of the outer surface of the casing was the weakest. Figure 6 The main effect curves shown indicate that both the inlet total temperature and the heat transfer coefficient of the casing's outer surface are positively correlated with the total temperature error. Further comparison reveals that the growth rate of the inlet total temperature curve is significantly higher than that of the casing's outer surface heat transfer coefficient curve, indicating that the increase in inlet total temperature has a more significant impact on the increase in temperature error at the measuring point. The outlet Mach number is negatively correlated with the total temperature error.

[0040] After establishing the above proxy model, given any inlet total temperature, outer casing surface heat transfer coefficient, and outlet Mach number, the total temperature error can be directly predicted by inputting it into the proxy model, and then the total temperature can be corrected.

[0041] The present invention can also provide a turbine tester probe temperature measurement error prediction system, including a total temperature error acquisition module, a quantization module and a prediction module; The total temperature error acquisition module is used to determine the sampling range of three design variables: outlet Mach number, heat transfer coefficient of outer surface of casing and inlet total temperature. It uses the Latin hypercube method to generate sample points in the design space, performs gas-thermal coupling simulation analysis, and obtains the total temperature error of the probe measurement points corresponding to each sample point. The quantification module constructs a global Kriging surrogate model based on a dataset consisting of design variables and target values ​​of temperature measurement error. It then uses this global Kriging surrogate model to perform a global sensitivity analysis of the design space, quantifying the degree of influence of each design variable on the total temperature error. The prediction module is used to input the exit Mach number, heat transfer coefficient of the outer surface of the casing, and total inlet temperature of the actual test conditions into the global Kriging proxy model to obtain the corresponding probe temperature measurement error prediction value.

[0042] The total temperature error acquisition module includes a flow field simulation unit and an error analysis unit; The flow field simulation unit is used to carry out full-circulation flow field simulation of the turbine tester to obtain flow field parameters of characteristic sections. It can also extract the area near the probe to establish a local simulation model at a 1 / 4 full-circulation scale, set the model boundary conditions, and carry out CFD numerical calculations. The error analysis unit is used to define the calculation method for the total temperature probe temperature measurement error and obtain the temperature measurement error of each sample point based on the calculation results of the flow field simulation module.

[0043] On the other hand, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the implementation of the turbine tester probe temperature measurement error prediction method described in the present invention.

[0044] The present invention can also provide a computer device, including a processor and a memory, wherein the memory is used to store a computer executable program, the processor reads the computer executable program from the memory and executes it, and the processor can implement the turbine tester probe temperature measurement error prediction method described in the present invention when executing the computer executable program.

[0045] The computer device may be a laptop, tablet, desktop computer, or workstation.

[0046] The processor can be a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or an off-the-shelf programmable gate array (FPGA).

[0047] The memory described in this invention can be an internal storage unit of a laptop, tablet, desktop computer, or workstation, such as memory or hard disk; or it can be an external storage unit, such as a portable hard disk or flash memory card.

[0048] Computer-readable storage media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer-readable storage media can include: read-only memory (ROM), random access memory (RAM), solid-state drives (SSDs), or optical discs, etc. Random access memory can include resistive random access memory (ReRAM) and dynamic random access memory (DRAM).

[0049] In summary, this invention provides a method for predicting probe temperature measurement error in a turbine tester. It establishes a full-circulation flow field simulation model of the entire tester, obtains flow field parameters at characteristic sections, and extracts a 1 / 4 full-circulation scale local fine model from the region near the probe. The total temperature and total pressure obtained from the full-circulation simulation are used as inlet boundary conditions for the local model. Boundary conditions combining temperature and convective heat transfer coefficients are set on the outer surface of the casing, the outer surface of the hub, and the protruding surface of the probe, while also considering the influence of radiative heat transfer. A method for calculating the probe temperature measurement error is defined, and the measurement error under different test conditions is evaluated and analyzed. Based on this, the outlet Mach number, the heat transfer coefficient of the casing outer surface, and the inlet total temperature are used as design variables. A Latin hypercube sampling method is used to generate sample points, and a global Kriging surrogate model is established using CFD calculations. Global sensitivity analysis is used to quantify the influence of each variable on the measurement error, enabling rapid prediction of the measurement error under different operating conditions.

[0050] This invention achieves an integrated design for probe temperature measurement error analysis and prediction in turbine test equipment. It constructs a simulation model that accurately reflects the coupling effect of internal and external temperature gradients and radiative heat transfer, and establishes an error correlation correction model through a surrogate model, overcoming the limitation of traditional analysis that ignores the actual influence of internal and external temperature gradients. This method can quickly obtain probe temperature measurement errors based on test conditions and achieve accurate correction of the measured temperature, effectively solving the key technical problem of efficient acquisition of temperature measurement errors under complex environments. It significantly improves the accuracy and reliability of temperature measurement in turbine testing, providing more accurate test data support for the performance evaluation and cooling design of turbine hot-end components, and has good engineering application value.

[0051] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. It should be noted that implementations not illustrated or described in the drawings or the main text of the specification are forms known to those skilled in the art and are not described in detail. Furthermore, the definitions of the components described above are not limited to the various specific structures, shapes, or methods mentioned in the embodiments, and those skilled in the art can easily modify or substitute them.

Claims

1. A turbine tester probe temperature measurement error prediction method, characterized by, Includes the following steps: S1. Determine the sampling range of the three design variables: outlet Mach number, heat transfer coefficient of outer surface of casing, and inlet total temperature. Use the Latin hypercube method to generate sample points in the design space, perform gas-thermal coupling simulation analysis, and obtain the total temperature error of the probe measurement points corresponding to each sample point. S2. Based on the dataset consisting of design variables and target values ​​of temperature measurement error, construct a global Kriging surrogate model, and use the global Kriging surrogate model to perform a global sensitivity analysis on the design space to quantify the influence of each design variable on the total temperature error. S3. Input the exit Mach number, heat transfer coefficient of the outer surface of the casing, and total inlet temperature of the actual test conditions into the global Kriging proxy model to obtain the corresponding probe temperature measurement error prediction value.

2. The turbine tester probe temperature measurement error prediction method of claim 1, wherein, After constructing the global Kriging proxy model in step S2, leave-one-out cross-validation is used to evaluate the model's accuracy. If the model's accuracy is... Then the model accuracy is determined to meet the requirements. Then, sampling continues within the design space, and gas-thermal coupling simulation analysis is performed to obtain the total temperature error of the probe measurement points corresponding to each sample point; until the model accuracy meets the requirements.

3. The turbine tester probe temperature measurement error prediction method of claim 1, wherein, The global sensitivity analysis described in step S2 includes variable variance contribution rate analysis and main effect curve analysis. The variance contribution rate clarifies the influence weight of each design variable on the total temperature error, and the main effect curve analyzes the correlation trend between each design variable and the total temperature error.

4. The turbine tester probe temperature measurement error prediction method of claim 1, wherein Perform gas-thermal coupling simulation analysis to obtain the total temperature error of the probe measurement points corresponding to each sample point. Includes the following steps: Includes the following steps: S11. Perform full-circulation flow field simulation of the turbine tester to obtain the flow field parameters of the characteristic sections; S12. Extract a local simulation model of 1 / 4 of the full ring scale of the area near the probe in the turbine tester; S13. The total temperature and total pressure of the characteristic section obtained from the whole-machine full-circulation flow field simulation are used as the inlet boundary conditions of the calculation domain of the local simulation model, and the outlet boundary of the local simulation model is set as static pressure condition; a third type of boundary condition combining temperature and convective heat transfer coefficient is set on the outer surface of the casing, the outer surface of the hub and the outer surface of the probe of the turbine tester, while considering the influence of radiative heat transfer. S14. Define the calculation method for the total temperature probe temperature measurement error, and calculate the temperature measurement error of the probe measurement point under each test condition based on the calculation method.

5. The turbine tester probe temperature measurement error prediction method of claim 4, wherein, The method for calculating the total temperature probe measurement error in step S14 includes: taking the average total temperature of adjacent fluid segments as the reference real fluid temperature at a position where the probe is offset by a set angle in the circumferential direction and is not affected by its flow. ; Calculate the average temperature of multiple solid measurement points on the probe. The difference between the reference real fluid temperature and the average temperature of the probe solid measurement point is defined as the total temperature error of the probe measurement point. ,Right now .

6. The turbine tester probe temperature measurement error prediction method of claim 4, wherein, In S13, the DTRM radiation model is used to calculate radiative heat transfer, and the local simulation model uses the SST k-ω turbulence model to solve the steady-state Reynolds time-averaged NS equations.

7. The turbine tester probe temperature measurement error prediction method of claim 4, wherein, The computational domain of the local simulation model includes the solid casing of the test instrument, the internal fluid domain, and the solid domain of the probe. The probe base fits into the outer surface of the test instrument casing through a groove to simulate the actual installation state of the probe.

8. A turbine tester probe temperature measurement error prediction system characterized by, It includes a total temperature error acquisition module, a quantization module, and a prediction module; The total temperature error acquisition module is used to determine the sampling range of three design variables: outlet Mach number, heat transfer coefficient of outer surface of casing and inlet total temperature. It uses the Latin hypercube method to generate sample points in the design space, performs gas-thermal coupling simulation analysis, and obtains the total temperature error of the probe measurement points corresponding to each sample point. The quantification module constructs a global Kriging surrogate model based on a dataset consisting of design variables and target values ​​of temperature measurement error. It then uses this global Kriging surrogate model to perform a global sensitivity analysis of the design space, quantifying the degree of influence of each design variable on the total temperature error. The prediction module is used to input the exit Mach number, heat transfer coefficient of the outer surface of the casing, and total inlet temperature of the actual test conditions into the global Kriging proxy model to obtain the corresponding probe temperature measurement error prediction value.

9. A computer device, comprising: It includes a processor and a memory, the memory being used to store a computer-executable program, the processor reading part or all of the computer-executable program from the memory and executing it, and when the processor executes part or all of the computer-executable program, it can implement the turbine tester probe temperature measurement error prediction method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, A computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of a method for predicting the temperature measurement error of a turbine tester probe as described in any one of claims 1-7.

Images