A temperature sensor fault detection system, temperature sensor fault-tolerant control method, device and storage medium

Abstract

The present application provides a temperature sensor fault detection system, a temperature sensor fault-tolerant control method, a device, and a storage medium. These methods collect the output electrical signals corresponding to each temperature sensor; construct a state vector of the blade thermal evolution process; and then determine the fitted electrical signals of each temperature sensor on a gas turbine blade based on the state vector and the dynamic mapping relationship between the output electrical signals and the actual blade temperature. The system then obtains thermal parameters of the gas turbine blade, constructs a temperature prediction model for the gas turbine blade based on the thermal parameters, and generates a temperature simulation distribution curve for the gas turbine blade. All fitted electrical signals are coupled and analyzed based on the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signals corresponding to each temperature sensor; and fault-tolerant control is performed on each temperature sensor on the gas turbine blade based on all signal fidelities. The present application's solution can improve the fault tolerance of temperature sensors when the output electrical signals experience nonlinear fluctuations.

Description

Technical Field

[0001] The present application relates to the technical field of sensor fault detection, and more specifically, to a temperature sensor fault detection system, a temperature sensor fault-tolerant control method, a device, and a storage medium. Background Art

[0002] Sensor fault detection is a technology that identifies faults or abnormal conditions by monitoring and analyzing the output signals of sensors. Sensor fault detection relies on real-time analysis of the sensor's monitoring data and comparing it with the output signal under normal working conditions to identify whether the sensor has deviated. Common sensor fault types include sensor failure, drift, short circuit, open circuit, etc. In addition, sensor fault detection technology is widely used, especially in automation, aerospace, medical health, intelligent transportation and other fields. It can not only improve the reliability and safety of the system, but also extend the service life of the equipment and reduce maintenance costs. With the development of artificial intelligence and machine learning technologies, the intelligence level of sensor fault detection has gradually improved, and the accuracy and efficiency of fault diagnosis are also constantly improving.

[0003] Temperature sensor fault detection involves real-time monitoring of the sensor's output signal and analyzing it for anomalies to ensure temperature measurement accuracy and system stability. Common temperature sensor fault types include sensor failure, signal drift, sensor overload, connection failure, and environmental interference. Using different detection and compensation methods for these fault types can effectively improve the system's fault tolerance and reliability. Temperature sensors on gas turbine blades typically measure temperature and convert it into an electrical signal based on the thermoelectric effect. Existing techniques typically detect faults by collecting the sensor's output signal and evaluating overall temperature changes based on a temperature prediction model. However, this approach often fails to fully capture the dynamic mapping between the blade's thermal evolution and the sensor's output signal. (For example, local thermal damage to a blade under transient high-temperature conditions may manifest as only subtle nonlinear fluctuations and thus be insufficiently captured.) This makes it difficult for a single temperature prediction model to accurately reflect local thermal anomalies under complex operating conditions. Therefore, improving the fault tolerance of temperature sensors when the output signal exhibits nonlinear fluctuations has become a challenge facing the industry. Summary of the Invention

[0004] The present application provides a temperature sensor fault detection system, a temperature sensor fault-tolerant control method, a device, and a storage medium, which can detect nonlinear fluctuations of the temperature sensor under the premise of dynamic changes in the thermal evolution process of the blade.

[0005] In a first aspect, the present application provides a temperature sensor fault-tolerant control method, comprising the following steps:

[0006] Collect the output electrical signals corresponding to each temperature sensor on the gas turbine blade;

[0007] Constructing a state vector of the blade thermal evolution process through all output electrical signals, and then determining the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with a dynamic mapping relationship between the output electrical signals and the actual blade temperature;

[0008] Acquiring thermal parameters of a gas turbine blade, constructing a temperature prediction model for the gas turbine blade based on the thermal parameters and geometric structure information of the gas turbine blade, and then generating a temperature simulation distribution curve of the gas turbine blade using the temperature prediction model;

[0009] Perform coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor;

[0010] Fault-tolerant control of each temperature sensor on a gas turbine blade is achieved with full signal fidelity.

[0011] In some embodiments, constructing a state vector of the blade thermal evolution process using all output electrical signals specifically includes:

[0012] determining state characteristics of the electrical signals at different sampling points based on all output electrical signals;

[0013] The state vector of the blade thermal evolution process is constructed based on all state characteristics and the temperature-voltage response characteristics of the thermocouple.

[0014] In some embodiments, determining the fitted electrical signal of each temperature sensor on the gas turbine blade according to the state vector combined with the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade specifically includes:

[0015] generating real temperature information of the gas turbine blades through the state vector;

[0016] The fitting electrical signals of each temperature sensor on the gas turbine blade are determined based on the real temperature information and the dynamic mapping relationship between the output electrical signal and the real temperature of the blade.

[0017] In some embodiments, constructing a temperature prediction model for a gas turbine blade based on the thermal parameters and geometric structure information of the gas turbine blade specifically includes:

[0018] Obtain geometric structure information of gas turbine blades;

[0019] Constructing a heat conduction equation for a gas turbine blade by combining the geometric structure information with the thermal parameters;

[0020] A temperature prediction model for the gas turbine blade is determined according to the heat conduction equation and the boundary conditions of the gas turbine blade.

[0021] In some embodiments, performing coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor specifically includes:

[0022] Obtain the temperature-voltage response characteristics of the temperature sensor;

[0023] Converting the temperature simulation distribution curve into a theoretical electrical signal of each temperature sensor according to the temperature-voltage response characteristics of the temperature sensor;

[0024] The signal fidelity of the output electrical signal corresponding to each temperature sensor is determined based on all theoretical electrical signals and all fitted electrical signals.

[0025] In some embodiments, performing fault-tolerant control on each temperature sensor on a gas turbine blade based on all signal fidelity specifically includes:

[0026] Set the fidelity threshold of the temperature sensor electrical signal;

[0027] performing anomaly detection on each temperature sensor on the gas turbine blade based on the fidelity threshold and the signal fidelity of each temperature sensor;

[0028] Fault-tolerant control is performed on each temperature sensor on the gas turbine blade according to the abnormality detection result.

[0029] In some embodiments, the temperature sensor is a thermocouple sensor.

[0030] In a second aspect, the present application provides a temperature sensor fault detection system, which includes a sensor fault-tolerant control unit, wherein the sensor fault-tolerant control unit includes:

[0031] An acquisition module is used to collect the output electrical signals corresponding to each temperature sensor on the gas turbine blade;

[0032] a processing module configured to construct a state vector of the blade thermal evolution process using all output electrical signals, and then determine the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with a dynamic mapping relationship between the output electrical signals and the actual blade temperature;

[0033] The processing module is further configured to obtain thermal parameters of the gas turbine blade, construct a temperature prediction model of the gas turbine blade based on the thermal parameters combined with geometric structure information of the gas turbine blade, and further generate a temperature simulation distribution curve of the gas turbine blade using the temperature prediction model;

[0034] The processing module is further configured to perform coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor;

[0035] An execution module is provided for fault-tolerant control of each temperature sensor on a gas turbine blade according to the overall signal fidelity.

[0036] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned temperature sensor fault-tolerant control method when executing the computer program.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned temperature sensor fault-tolerant control method are implemented.

[0038] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0039] In the temperature sensor fault detection system, temperature sensor fault-tolerant control method, device and storage medium provided by the present application, the output electrical signals corresponding to each temperature sensor on the gas turbine blade are collected; the state vector of the blade thermal evolution process is constructed through all the output electrical signals, and then the fitting electrical signals of each temperature sensor on the gas turbine blade are determined according to the state vector combined with the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade; the thermal parameters of the gas turbine blade are obtained, and a temperature prediction model of the gas turbine blade is constructed based on the thermal parameters combined with the geometric structure information of the gas turbine blade, and then the temperature simulation distribution curve of the gas turbine blade is generated through the temperature prediction model; all the fitting electrical signals are coupled and analyzed according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor; and each temperature sensor on the gas turbine blade is fault-tolerantly controlled based on all the signal fidelities.

[0040] It can be seen that in this application, by collecting the outputs of various temperature sensors on the blade and constructing the state vector of the entire thermal evolution process, the dynamic mapping relationship between the output electrical signal and the actual temperature is used to determine the fitting electrical signal, that is: the state vector integrates multi-point data, so that the multi-dimensional information of the system in the thermal evolution process is effectively reflected, so that the nonlinear fluctuations of the output electrical signal caused by the joint action of geometric structure, thermal parameters and environmental changes can be captured. Secondly, the fitting electrical signal generated by the dynamic mapping relationship can be used as a theoretical reference value, which reflects the real change trend of the blade temperature under ideal conditions, can filter out interference and local noise, highlight abnormal deviations in the sensor output, provide accurate comparison basis for subsequent fault detection, and timely identify nonlinear anomalies in the temperature acquisition process; then, based on the acquired blade thermal parameters A temperature prediction model is constructed based on the geometric structure information. After generating the temperature simulation distribution curve, all the fitted electrical signals are coupled and analyzed to determine the signal fidelity of each sensor output. The temperature simulation distribution curve represents the temperature distribution under the ideal thermal evolution dynamics. By coupling and comparing it with the actual fitted electrical signal, the response of each sensor under complex working conditions is intuitively reflected. Finally, the signal fidelity is used as a comprehensive evaluation indicator to enable the system to detect whether there is data loss, noise increase or slow response in the sensor output signal, so as to timely capture the abnormal changes in the sensor output electrical signal caused by nonlinear fluctuations, and then perform fault-tolerant control on each temperature sensor. In summary, this scheme can improve the fault tolerance of the temperature sensor when the output electrical signal produces nonlinear fluctuations. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a flow chart of a temperature sensor fault-tolerant control method according to some embodiments of the present application;

[0042] Figure 2 is a schematic diagram of a process for determining a temperature prediction model according to some embodiments of the present application;

[0043] Figure 3 is a schematic diagram of anomaly detection process according to some embodiments of the present application;

[0044] Figure 4 is a structural diagram of a sensor fault-tolerant control unit according to some embodiments of the present application;

[0045] Figure 5 This is an internal structure diagram of a computer device for implementing a temperature sensor fault-tolerant control method according to some embodiments of the present application. DETAILED DESCRIPTION

[0046] In order to better understand the technical solution in this embodiment, the technical solution in this embodiment will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0047] refer to Figure 1 , which is a flow chart of a temperature sensor fault-tolerant control method according to some embodiments of the present application. The temperature sensor fault-tolerant control method 100 mainly includes the following steps:

[0048] In step 101, the output electrical signals corresponding to the temperature sensors on the gas turbine blades are collected.

[0049] In a specific implementation, an electrical signal may be drawn from a temperature sensor inside a gas turbine blade through a high-temperature wire and then collected to obtain the output electrical signal corresponding to each temperature sensor on the gas turbine blade.

[0050] It should be noted that the temperature sensor described in this application is a thermocouple sensor. In addition, the output electrical signal refers to a timing signal composed of thermoelectromotive force generated by collecting the temperature difference between the hot end (installed on the gas turbine blade) and the cold end of the temperature sensor at a set sampling frequency, that is, the output electrical signal includes the thermoelectromotive force of the corresponding temperature sensor at each sampling point; preferably, the output electrical signal of the temperature sensor has a small magnitude, and the output electrical signal can be precisely amplified by a low-noise amplifier or a dedicated thermocouple amplifier chip (such as AD8495).

[0051] In step 102, a state vector of the blade thermal evolution process is constructed through all output electrical signals, and then the fitted electrical signals of each temperature sensor on the gas turbine blade are determined based on the state vector combined with the dynamic mapping relationship between the output electrical signals and the actual blade temperature.

[0052] In some embodiments, constructing the state vector of the blade thermal evolution process using all output electrical signals can be achieved by using the following steps:

[0053] determining state characteristics of the electrical signals at different sampling points based on all output electrical signals;

[0054] The state vector of the blade thermal evolution process is constructed based on all state characteristics and the temperature-voltage response characteristics of the thermocouple.

[0055] In a specific implementation, the state characteristics of the electrical signals at different sampling points can be determined based on all the output electrical signals in the following manner: since the temperature evolution process of the gas turbine blade has the characteristics of time continuity and local change stability, and the output electrical signal of the thermocouple can reflect its temperature change trend to a certain extent, a local time window centered on the sampling point can be constructed, and the statistical features used to describe the signal characteristics can be extracted within the window as the state characteristics of the electrical signal at the corresponding sampling point; for example, first, for each sampling point, a local time window centered on the sampling point is set, and then multiple statistical features in the local time window corresponding to each sampling point are determined. The vector composed of the statistical features as vector elements is used as the state feature of the electrical signal at the corresponding sampling point; preferably, the first-order difference, second-order difference, mean and variance of all thermoelectromotive force in the local time window can be calculated as multiple statistical features in the local time window. In other embodiments, other methods can also be used for implementation, which are not limited here; it should be noted that the first-order difference is used to approximately characterize the temperature change rate and capture the speed of heating / cooling; the second-order difference is used to describe the acceleration of temperature change and identify whether there is an inflection point; the mean is used to reflect the local level of thermoelectromotive force; the variance is used to reflect the volatility of the local thermoelectromotive force.

[0056] It should be noted that the state characteristics described in this application refer to a set of quantitative indicators used to describe the local change trends and characteristics of the output electrical signals of the temperature sensor during the thermal evolution of the gas turbine blades. The state characteristics can be used to describe the statistical changes of the output electrical signals within a short time scale, thereby revealing the potential behavioral patterns of temperature changes.

[0057] In specific implementation, the state vector of the thermal evolution process of the blade is constructed based on all state features combined with the temperature-voltage response characteristics of the thermocouple, that is: the state features of the electrical signal (such as local mean, change rate, etc.) are mapped to corresponding temperature features through the temperature-voltage response characteristics of the thermocouple, and then the temperature features extracted from multiple thermocouple channels are integrated into a state vector of temperature evolution; for example: first, the state features corresponding to each sampling point are obtained, and then the state features of the electrical signal at each sampling point are converted into the state features of the temperature of the blade at the corresponding sampling point in combination with the temperature-voltage response characteristics of the thermocouple, and finally, the set of the state features of the temperature of the blade at all sampling points is used as the state vector of the thermal evolution process of the blade; as a preferred embodiment, the mean value of the electrical signal in the local time window of a sampling point can be used as the state vector of the thermoelectric state. The representative value of the thermocouple output voltage is input into the inverse function of the voltage-temperature response function of the thermocouple to obtain the estimated value of the temperature level corresponding to the sampling point, that is, the temperature level is obtained after the mean value of the electrical signal of the sampling point is converted by the inverse function of the thermocouple; the rate of change of temperature over time is approximately estimated by calculating the difference between the temperature values ​​corresponding to the output voltage values ​​of the thermocouple at two consecutive moments and dividing it by the time interval (sampling frequency); that is, the temperature change corresponding to the thermoelectromotive force at adjacent moments is divided by the time step to obtain the temperature change rate; the second-order difference of the thermoelectromotive force in the local time window is used as an approximate indicator of temperature acceleration, and after substituting it into the inverse function of the thermocouple, the temperature change trend or the acceleration of temperature change can be estimated; that is, the second-order difference of the electrical signal is converted by a nonlinear response function to estimate the temperature change trend.

[0058] It should be noted that the state vector described in this application is a multidimensional vector used to describe the thermal evolution of the gas turbine blade at a certain moment, which includes the temperature change rate, the acceleration of the temperature change and the trend of the temperature change.

[0059] In some embodiments, determining the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade can be achieved by the following steps:

[0060] generating real temperature information of the gas turbine blades through the state vector;

[0061] The fitting electrical signals of each temperature sensor on the gas turbine blade are determined based on the real temperature information and the dynamic mapping relationship between the output electrical signal and the real temperature of the blade.

[0062] In a specific implementation, the real temperature information of the gas turbine blade is generated through the state vector, namely:

[0063] The state vector is converted into an estimated result of the actual temperature of the blade at the corresponding moment, thereby obtaining the actual temperature information of the gas turbine blade; for example, the state vector at the current moment can be output according to the adaptive Kalman filter, and then the key component representing the thermal state of the blade in the state vector, namely, the actual temperature value, can be extracted, wherein the extraction process can adopt a nonlinear mapping method, namely, multiple characteristic variables in the state vector (namely, the rate of change of temperature, the acceleration of temperature change and the trend of temperature change) are input into the temperature conversion model to reconstruct the actual temperature value of the current blade position.

[0064] It should be noted that the temperature conversion model refers to a functional model that reconstructs the actual temperature value at the current blade position through a specific mapping function based on multiple thermal characteristic variables in the state vector. Preferably, all characteristic variables in the state vector can be used as input variables, the Python software platform can be run, and the NumPy tool can be called to perform feature construction (such as polynomial combination or cross-term generation). The sklearn tool is then used to perform polynomial fitting to obtain the temperature conversion model. In other embodiments, the temperature conversion model can also be established using neural networks, support vector machines, or other regression modeling techniques, which are not limited in this application.

[0065] It should be noted that the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade refers to the mathematical relationship followed between the output electrical signal of the temperature sensor and the actual temperature of the blade in the gas turbine blade as the blade temperature changes. In specific implementation, experimental data, historical temperature information (such as the temperature obtained by a high-precision temperature sensor) and the electrical signal of the thermocouple can be used to establish a basic mapping function (such as linear regression, polynomial regression, etc.) to obtain the dynamic mapping relationship between the output and the actual temperature of the blade.

[0066] In specific implementation, the fitting electrical signals of each temperature sensor on the gas turbine blade are determined based on the real temperature information combined with the dynamic mapping relationship between the output electrical signal and the real temperature of the blade, that is, a mathematical relationship model between the real temperature of the blade and the output electrical signal of the temperature sensor is established, and the real temperature information of the blade is converted into a corresponding electrical signal according to the model, that is, a fitting electrical signal; as a preferred embodiment, after running Python software, the historical temperature information is used as input and the corresponding actual measured electrical signal is used as output to construct training data, and then the PolynomialFeatures tool is used to expand the real temperature information into a polynomial feature matrix, and then the LinearRegression tool is applied for fitting to obtain the polynomial coefficients in the dynamic mapping relationship between the electrical signal and the real temperature of the blade, and then the real temperature information is input and the software is run to output the fitting electrical signal of the temperature sensor.

[0067] Among them, it should be noted that since the thermal environment at each location may be different, a dynamic mapping relationship between the output electrical signal and the actual temperature of the blade can be established for different temperature sensors separately to generate a fitting electrical signal for each temperature sensor. In other embodiments, other methods can also be used to achieve this, which is not limited here. In addition, the fitting electrical signal refers to an electrical signal that is inferred from the actual temperature estimate of the blade through a mathematical mapping model (such as polynomial fitting, neural network, etc.) based on the thermal evolution state vector of the blade and the response characteristics of the thermocouple.

[0068] In step 103, thermal parameters of the gas turbine blade are obtained, and a temperature prediction model of the gas turbine blade is constructed based on the thermal parameters and geometric structure information of the gas turbine blade. Then, a temperature simulation distribution curve of the gas turbine blade is generated through the temperature prediction model.

[0069] It should be noted that the thermal parameters of the gas turbine blades refer to physical parameters that describe the thermal behavior and performance of the blades in a high-temperature environment, specifically including: thermal conductivity, specific heat capacity, thermal convection coefficient and heat flux density, etc. Preferably, the thermal performance of the gas turbine blades can be subjected to finite element analysis by a computer simulation tool (such as Abaqus), and the thermal conductivity, thermal stress, thermal expansion and temperature distribution of the simulated blades under different operating conditions can be used as the thermal parameters of the gas turbine blades in this application. In other embodiments, other methods can also be used to determine the parameters, which are not limited here.

[0070] In some embodiments, reference Figure 2 As shown in FIG. 1 , this figure is a schematic diagram of a process for determining a temperature prediction model according to some embodiments of the present application. Constructing a temperature prediction model for a gas turbine blade based on the thermal parameters combined with geometric structure information of the gas turbine blade can be achieved by the following steps:

[0071] First, obtain the geometric structure information of the gas turbine blade;

[0072] Then, a heat conduction equation of the gas turbine blade is constructed by combining the geometric structure information with the thermal parameters;

[0073] Finally, a temperature prediction model of the gas turbine blade is determined according to the heat conduction equation and the boundary conditions of the gas turbine blade.

[0074] It should be noted that the geometric structure information of the gas turbine blade in this application is data used to describe the shape, size, and other characteristics of the blade. The geometric characteristics play a vital role in the working performance of the blade in the gas turbine (including thermal distribution, aerodynamic performance, etc.), specifically including the overall shape and size of the blade (length, width, thickness, degree of curvature) and the cross-sectional shape of the blade (blade cross-sectional profile) and the aerodynamic characteristics of the blade (bending angle and surface smoothness); preferably, the geometric structure information of the blade can be designed and extracted through computer-aided design modeling software (such as AutoCAD, SolidWorks, etc.), for example: first, use computer-aided design modeling software to design a three-dimensional model of the gas turbine blade, then extract the geometric dimensions (length, width, thickness, etc.), cross-sectional shape and blade twist angle data in the model, and finally, derive the geometric structure information of the gas turbine blade through the computer-aided design modeling software. In other embodiments, other methods can also be used to obtain it, which is not limited here.

[0075] In a specific implementation, the geometric structure information is combined with the thermal parameters to construct a heat conduction equation for the gas turbine blade. Specifically, the physical heat conduction process is combined with the heat conduction equation and material thermal properties of the gas turbine blade to form a mathematical model that describes the heat distribution, conduction, and change of the blade during operation, serving as the heat conduction equation for the gas turbine blade. As a preferred embodiment, the heat conduction equation for the gas turbine blade can be constructed by combining the Fourier heat conduction law in the prior art with the geometric structure information and the thermal parameters. The Fourier heat conduction law states that the rate of heat conduction in a material is proportional to the temperature gradient within the region, and that heat flow always flows in the direction of decreasing temperature. Therefore, the geometric structure information can be used as the spatial domain of the Fourier heat conduction law, while the thermal parameters reflect the inherent thermal properties of the blade material. The heat conduction equation for the gas turbine blade can be constructed using the Fourier heat conduction law. In other embodiments, other methods can also be used, which are not limited here.

[0076] It should be noted that the heat conduction equation is a mathematical expression that describes the heat transfer and temperature change behavior of gas turbine blades during operation. Based on Fourier's law of heat conduction, the thermal properties of the material (such as thermal conductivity) are coupled with the geometric structural characteristics of the blade to form a partial differential equation that describes the temperature propagation law of the gas turbine blade in the spatial and temporal dimensions.

[0077] It should be noted that the boundary conditions of the gas turbine blades described in this application refer to the constraints that describe the heat exchange between the blades and the external environment (such as airflow, combustion gas, etc.) and the temperature distribution of the blades themselves when solving the heat conduction equation. The boundary conditions specifically include: temperature boundary conditions (such as the surface temperature of the blade in contact with the gas), heat flow boundary conditions (describing the heat exchange process between the blade surface and heat sources such as airflow and combustion gas) and initial conditions (temperature distribution of the blades at the initial moment). As a preferred embodiment, a temperature prediction model of the gas turbine blades is determined based on the heat conduction equation and the boundary conditions of the gas turbine blades, that is, a mathematical model composed of the boundary conditions of the gas turbine blades as the constraints of the heat conduction equation is used as the temperature prediction model of the gas turbine blades.

[0078] In addition, it should be noted that the temperature prediction model refers to a simulation model constructed based on the heat conduction equation, combining the thermal parameters (thermal conductivity, specific heat capacity, heat flux density, etc.) of the gas turbine blades, geometric structure information (three-dimensional shape, size, etc.) and thermal boundary conditions (temperature, heat flow). The temperature prediction model can be used to predict the spatial and temporal temperature distribution of the blades under different operating conditions.

[0079] In a specific implementation, the temperature simulation distribution curve of the gas turbine blade is generated by the temperature prediction model in the following manner: first, a three-dimensional model of the gas turbine blade is constructed using AutoCAD software; then, the three-dimensional model is automatically divided into multiple grids using ANSYS Fluent; then, the temperature value in each grid is initialized based on the initial conditions in the boundary conditions; second, the heat conduction equation is discretized using the finite element method in the prior art to obtain an algebraic equation in each grid; third, the temperature boundary conditions and the heat flow boundary conditions are encoded (for example, for the heat flow boundary conditions on the blade surface, the heat flow magnitude can be set or the temperature can be directly set), and the encoding is applied to the algebraic equation; the discretized algebraic equation system is solved, and then the temperature value of each grid is obtained from ANSYS Fluent; finally, the temperature values ​​at different time nodes and locations are extracted from the obtained temperature data; and the temperature simulation distribution curve of the gas turbine blade is generated using an interpolation method in the prior art (such as linear interpolation, spline interpolation, etc.).

[0080] It should be noted that the solution to the discretized algebraic equation group can be achieved by using numerical methods for solving linear or nonlinear equation groups (such as Gaussian elimination method, conjugate gradient method, etc.); in addition, the temperature simulation distribution curve of the gas turbine blade can be displayed in a graphical manner, such as an isotherm diagram, a temperature distribution diagram, etc., by using data visualization tools (such as MATLAB, the Matplotlib library in Python, or the post-processing module of ANSYS). In other embodiments, other methods can also be used for implementation, which are not limited here; in addition, the temperature simulation distribution curve refers to a curve of the temperature change over time of the gas turbine blade obtained based on the numerical solution of the heat conduction equation under ideal thermal boundary conditions.

[0081] In step 104, coupling analysis is performed on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor.

[0082] In some embodiments, coupling analysis is performed on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor by using the following steps:

[0083] Obtain the temperature-voltage response characteristics of the temperature sensor;

[0084] Converting the temperature simulation distribution curve into a theoretical electrical signal of each temperature sensor according to the temperature-voltage response characteristics of the temperature sensor;

[0085] The signal fidelity of the output electrical signal corresponding to each temperature sensor is determined based on all theoretical electrical signals and all fitted electrical signals.

[0086] It should be noted that the temperature-voltage response characteristic of the temperature sensor refers to the conversion relationship between the thermoelectromotive force output by the thermocouple at different temperatures and its corresponding temperature value. It is a static characteristic of the thermocouple material itself and can usually be obtained by consulting a standard graduation table (such as a standard graduation table for K-type and J-type thermocouples) or through experimental calibration. For example, when the ambient temperature is known (for example, controlled by a standard temperature control device), the thermocouple can be placed under different known temperature conditions, and then the corresponding thermocouple output voltage is recorded. Then, based on multiple temperature-voltage sample points, a response function of the thermocouple is fitted or established (such as cubic interpolation, polynomial function, table lookup method, etc.), and finally, the response function is used as the temperature-voltage response characteristic of the temperature sensor.

[0087] In specific implementation, the temperature simulation distribution curve is converted into the theoretical electrical signal of each temperature sensor through the temperature-voltage response characteristics of the temperature sensor. This can be achieved in the following manner: first, based on the layout position of each temperature sensor in the three-dimensional model of the gas turbine blade, the temperature data of the corresponding position in the temperature simulation distribution curve that changes with time is extracted; second, based on the type of thermocouple (such as K type, J type, etc.) used by each temperature sensor, its standard temperature-voltage response characteristics are obtained (for example, the national standard thermocouple graduation table can be consulted or a calibrated response function can be used); then, based on the temperature value at each time point, the theoretical temperature value of each temperature sensor at each sampling point is converted into its corresponding thermoelectromotive force point by point by substituting the response function; and then, the set of all thermoelectromotive forces sorted in the time order of the sampling points is used as the theoretical electrical signal of the corresponding temperature sensor.

[0088] In specific implementation, the signal fidelity of the output electrical signal corresponding to each temperature sensor is determined based on all theoretical electrical signals and all fitted electrical signals, that is, the deviation between the theoretical electrical signal and the fitted electrical signal of each temperature sensor is used as the signal fidelity of the corresponding temperature sensor; for example, first, for each temperature sensor, the theoretical electrical signal and the fitted electrical signal corresponding to each temperature sensor are obtained, and then the absolute value of the thermoelectromotive force residual of the theoretical electrical signal and the fitted electrical signal at the corresponding sampling point in the same time series is determined, and the obtained result is used as the signal fidelity of the output electrical signal of each temperature sensor at the corresponding sampling point.

[0089] It should be noted that the theoretical electrical signal described in this application is a predicted signal derived from a theoretical model, which represents the electrical signal that the temperature sensor should output under a specific environment; the fitted electrical signal is an electrical signal obtained through some fitting processes (such as regression analysis) to match the sensor behavior in the actual environment. This application quantifies the error of the sensor output by calculating the deviation (residual) between the theoretical electrical signal and the fitted electrical signal; the smaller the error, the more accurate the signal and the higher the signal fidelity; conversely, the larger the error, the lower the signal fidelity, indicating inaccurate measurement. Therefore, the signal fidelity described in this application refers to the degree of matching between the actual output electrical signal of the temperature sensor and the theoretically predicted electrical signal, and the accuracy and reliability of the electrical signal output by the temperature sensor can be quantified by the signal fidelity.

[0090] In step 105 , each temperature sensor on the gas turbine blade is subjected to fault-tolerant control based on all signal fidelity.

[0091] In some embodiments, fault-tolerant control of each temperature sensor on a gas turbine blade based on overall signal fidelity may be achieved by:

[0092] Set the fidelity threshold of the temperature sensor electrical signal;

[0093] performing anomaly detection on each temperature sensor on the gas turbine blade based on the fidelity threshold and the signal fidelity of each temperature sensor;

[0094] Fault-tolerant control is performed on each temperature sensor on the gas turbine blade according to the abnormality detection result.

[0095] In specific implementation, the fidelity threshold of the temperature sensor electrical signal can be set according to the accuracy requirements of the system. For example, when the gas turbine has high requirements for temperature monitoring of the blades and is more sensitive to thermal damage detection, a smaller fidelity threshold is set to capture possible faults or damage earlier; when the gas turbine has low requirements for temperature monitoring of the blades and is more tolerant to local damage to the temperature sensor, a larger fidelity threshold is set to avoid frequent false alarms.

[0096] For specific implementation, refer to Figure 3 As shown, this figure is a schematic diagram of the abnormality detection process shown in some embodiments of the present application. The abnormality detection of each temperature sensor on the gas turbine blade based on the fidelity threshold and the signal fidelity of each temperature sensor can be implemented in the following manner, namely: first, the signal fidelity at each sampling point of each temperature sensor is obtained, and then a sampling point is selected. When the signal fidelity of the temperature sensor corresponding to the sampling point is greater than or equal to the fidelity threshold, it indicates that the corresponding temperature sensor is normal. When the signal fidelity of the temperature sensor corresponding to the sampling point is less than the fidelity threshold, it is determined that the corresponding temperature sensor is abnormal, thereby obtaining the abnormality detection results of all temperature sensors; as a preferred embodiment, according to Fault-tolerant control of each temperature sensor on the gas turbine blade according to the abnormality detection result can be achieved in the following manner, namely: when the temperature sensor is detected as abnormal, fault-tolerant control is performed on the corresponding temperature sensor; for example: for a temperature sensor determined to be abnormal, its corresponding output electrical signal is automatically removed from the data fusion module to prevent abnormal data from interfering with the overall temperature monitoring. At the same time, the temperature simulation distribution curve generated by the temperature prediction model is used as a reference to replace the data corresponding to the sensor. The output data of the adjacent normal sensors can also be used to correct the data of the temperature sensor determined to be abnormal through interpolation method. In other embodiments, other methods can also be used for control, which are not limited here.

[0097] In addition, in another aspect of the present application, in some embodiments, the present application provides a temperature sensor fault detection system, including a sensor fault-tolerant control unit, referring to Figure 4, which is a schematic diagram of the structure of a sensor fault-tolerant control unit according to some embodiments of the present application. The sensor fault-tolerant control unit 200 includes: an acquisition module 201, a processing module 202, and an execution module 203, which are described as follows:

[0098] Acquisition module 201, in this application, acquisition module 201 is mainly used to collect output electrical signals corresponding to each temperature sensor on the gas turbine blade;

[0099] Processing module 202, in this application, is mainly used to construct a state vector of the blade thermal evolution process through all output electrical signals, and then determine the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with the dynamic mapping relationship between the output electrical signals and the actual blade temperature;

[0100] In addition, the processing module 202 in the present application is further configured to obtain thermal parameters of the gas turbine blade, construct a temperature prediction model of the gas turbine blade based on the thermal parameters combined with geometric structure information of the gas turbine blade, and then generate a temperature simulation distribution curve of the gas turbine blade using the temperature prediction model;

[0101] In addition, the processing module 202 in the present application is further configured to perform coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor;

[0102] The execution module 203 in this application is mainly used to perform fault-tolerant control on each temperature sensor on the gas turbine blade according to all signal fidelity.

[0103] In addition, the present application also provides a computer device, which includes a memory and a processor, wherein the memory stores code, and the processor is configured to obtain the code and execute the above-mentioned temperature sensor fault-tolerant control method.

[0104] In some embodiments, reference Figure 5 , which is an internal structure diagram of a computer device for implementing a temperature sensor fault-tolerant control method according to some embodiments of the present application. The temperature sensor fault-tolerant control method in the above embodiment can be Figure 5 The computer device 300 shown in FIG. 1 is implemented as shown in FIG. 1 , and the computer device 300 includes at least one processor 301 , a communication bus 302 , a memory 303 , and at least one communication interface 304 .

[0105] The processor 301 may be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC) or one or more processors for controlling the execution of the temperature sensor fault-tolerant control method of the present application.

[0106] The communication bus 302 is used to transmit information between the above components.

[0107] Memory 303 may be, but is not limited to, a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disk storage, an optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer. Memory 303 may exist independently and be connected to processor 301 via communication bus 302. Memory 303 may also be integrated with processor 301.

[0108] Memory 303 is used to store program code for executing the solution of the present application, and is controlled by processor 301 for execution. Processor 301 is used to execute the program code stored in memory 303. The program code may include one or more software modules. The temperature sensor fault-tolerant control method in the above embodiment can be implemented by processor 301 and one or more software modules in the program code in memory 303.

[0109] The communication interface 304 uses any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0110] In a specific implementation, as an example, a computer device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. A processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0111] The aforementioned computer device may be a general-purpose computer device or a dedicated computer device. In a specific implementation, the computer device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of this application do not limit the type of computer device.

[0112] In addition, the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned temperature sensor fault-tolerant control method is implemented.

[0113] In summary, in the temperature sensor fault detection system, temperature sensor fault-tolerant control method, device and storage medium disclosed in the embodiments of the present application, the output electrical signals corresponding to each temperature sensor on the gas turbine blade are collected; the state vector of the blade thermal evolution process is constructed through all the output electrical signals, and then the fitted electrical signals of each temperature sensor on the gas turbine blade are determined according to the state vector combined with the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade; the thermal parameters of the gas turbine blade are obtained, and a temperature prediction model of the gas turbine blade is constructed based on the thermal parameters combined with the geometric structure information of the gas turbine blade, and then the temperature simulation distribution curve of the gas turbine blade is generated through the temperature prediction model; all the fitted electrical signals are coupled and analyzed according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor; each temperature sensor on the gas turbine blade is fault-tolerantly controlled based on all the signal fidelities; the fault tolerance rate of the temperature sensor can be improved when the output electrical signal produces nonlinear fluctuations.

[0114] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0115] Obviously, those skilled in the art may make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if such changes and modifications fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such changes and modifications.

Claims

1. A temperature sensor fault-tolerant control method, used in a temperature sensor fault detection system to perform fault-tolerant control on a temperature sensor, characterized in that: The method comprises the following steps: Collect the output electrical signals corresponding to each temperature sensor on the gas turbine blade; Constructing a state vector of the blade thermal evolution process through all output electrical signals, and then determining the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with a dynamic mapping relationship between the output electrical signals and the actual blade temperature; Acquiring thermal parameters of a gas turbine blade, constructing a temperature prediction model for the gas turbine blade based on the thermal parameters and geometric structure information of the gas turbine blade, and then generating a temperature simulation distribution curve of the gas turbine blade using the temperature prediction model; Perform coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor; Fault-tolerant control of each temperature sensor on the gas turbine blades based on overall signal fidelity; The coupling analysis of all fitted electrical signals according to the temperature simulation distribution curve is performed to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor, specifically including: Obtain the temperature-voltage response characteristics of the temperature sensor; Converting the temperature simulation distribution curve into a theoretical electrical signal of each temperature sensor according to the temperature-voltage response characteristics of the temperature sensor; Determine the signal fidelity of the output electrical signal corresponding to each temperature sensor based on all theoretical electrical signals and all fitted electrical signals; Among them, fault-tolerant control of each temperature sensor on the gas turbine blade based on the fidelity of all signals specifically includes: Set the fidelity threshold of the temperature sensor electrical signal; performing anomaly detection on each temperature sensor on the gas turbine blade based on the fidelity threshold and the signal fidelity of each temperature sensor; Fault-tolerant control is performed on each temperature sensor on the gas turbine blade according to the abnormality detection result.

2. The method according to claim 1, wherein The state vector of the blade thermal evolution process is constructed through all output electrical signals, including: determining state characteristics of the electrical signals at different sampling points based on all output electrical signals; The state vector of the blade thermal evolution process is constructed based on all state characteristics and the temperature-voltage response characteristics of the thermocouple.

3. The method according to claim 1, wherein Determining the fitted electrical signals of each temperature sensor on the gas turbine blade according to the state vector combined with the dynamic mapping relationship between the output electrical signal and the actual temperature of the blade specifically includes: generating real temperature information of the gas turbine blades through the state vector; The fitting electrical signals of each temperature sensor on the gas turbine blade are determined based on the real temperature information and the dynamic mapping relationship between the output electrical signal and the real temperature of the blade.

4. The method according to claim 1, wherein Constructing a temperature prediction model for a gas turbine blade based on the thermal parameters and geometric structure information of the gas turbine blade specifically includes: Obtain geometric structure information of gas turbine blades; Constructing a heat conduction equation for a gas turbine blade by combining the geometric structure information with the thermal parameters; A temperature prediction model for the gas turbine blade is determined according to the heat conduction equation and the boundary conditions of the gas turbine blade.

5. The method according to claim 1, wherein The temperature sensor is a thermocouple sensor.

6. A temperature sensor fault detection system, comprising a sensor fault-tolerant control unit, which performs fault-tolerant control using the method according to any one of claims 1 to 5, characterized in that: The sensor fault-tolerant control unit comprises: An acquisition module is used to collect the output electrical signals corresponding to each temperature sensor on the gas turbine blade; a processing module configured to construct a state vector of the blade thermal evolution process using all output electrical signals, and then determine the fitted electrical signals of each temperature sensor on the gas turbine blade based on the state vector combined with a dynamic mapping relationship between the output electrical signals and the actual blade temperature; The processing module is further configured to obtain thermal parameters of the gas turbine blade, construct a temperature prediction model of the gas turbine blade based on the thermal parameters combined with geometric structure information of the gas turbine blade, and further generate a temperature simulation distribution curve of the gas turbine blade using the temperature prediction model; The processing module is further configured to perform coupling analysis on all fitted electrical signals according to the temperature simulation distribution curve to obtain the signal fidelity of the output electrical signal corresponding to each temperature sensor; An execution module is provided for fault-tolerant control of each temperature sensor on a gas turbine blade according to the overall signal fidelity.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the temperature sensor fault-tolerant control method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the temperature sensor fault-tolerant control method according to any one of claims 1 to 5 are implemented.

Images