Thin film thermocouple cold end temperature error compensation method and related device

Abstract

The invention discloses a thin film thermocouple cold end temperature error compensation method and a related device. The method comprises the following steps: S1, acquiring real-time thermoelectric force of a thin film thermocouple; s2, inputting the real-time thermoelectric force into a pre-trained cold end error compensation model, and calculating to obtain a compensation thermoelectric force; and S3, according to the compensation thermoelectric force obtained through calculation of the cold end error compensation model and the real-time thermoelectric force of the thin film thermocouple, real-time thermoelectric output quantity is obtained through calculation, and the cold end temperature of the thin film thermocouple is compensated in real time through the thermoelectric output quantity. According to the method, the error thermoelectric force under the temperature gradient from the reference temperature to the real-time cold end temperature is taken as an error term, and the error term and the real-time thermoelectric force obtained through the thin film thermocouple are summed, so that the thermal electromotive force of the thin film thermocouple is compensated and corrected, and the method can be suitable for different sensitive material systems, and has a wide application prospect. The compensation of different sensitive systems is in favor of reducing the cold end error of various non-standard film thermocouples in the temperature test on the premise that the process is not changed.

Description

Technical Field

[0001] This application belongs to the field of sensing technology and relates to a method for compensating the cold-end temperature error of a thin-film thermocouple and related devices. Background Art

[0002] High-temperature temperature measurement has important application value in the industrial and military fields. Especially in scenarios such as metallurgy, chemical industry, aerospace, and missile propulsion systems, accurate temperature monitoring directly affects process control, product quality, and equipment safety. The current mainstream thin-film thermocouples have significant advantages compared to traditional thermocouple wires: their miniaturized structure can be deployed through surface deposition without damaging the structure of the object to be measured, and can effectively maintain the strength and dynamic characteristics of the equipment. Therefore, it has become a key research direction for high-temperature measurement. However, the existing technology has the following key problems: First, the thin-film thermocouples lack standardized design and generally lack a compensating wire system to extend the cold-end junction; Second, in extreme working conditions with limited space and complex heat transfer, the cold-end temperature cannot be stably maintained at the reference value, resulting in distortion of the thermoelectromotive force signal and directly affecting the accuracy of temperature measurement. These problems seriously restrict the reliable application of thin-film thermocouples in high-temperature measurement scenarios. Summary of the Invention

[0003] The purpose of this application is to solve the problem of thermoelectromotive force loss caused by the increase in the cold-end temperature of the thin-film thermocouple relative to the reference value, and to provide a method for compensating the cold-end temperature error of the thin-film thermocouple and related devices. This application calculates the cold-end error term through a single-layer perceptron, combines the intermediate temperature law, compensates the thermoelectromotive force value, and improves the temperature measurement accuracy of the thin-film thermocouple.

[0004] To achieve the above purpose, this application adopts the following technical solutions: In the first aspect, this application provides a method for compensating the cold-end temperature error of a thin-film thermocouple, including the following steps: Obtain the real-time thermoelectromotive force of the thin-film thermocouple; Input the real-time thermoelectromotive force into a pre-trained cold-end error compensation model to calculate the compensated thermoelectromotive force; Calculate the real-time thermoelectric output based on the compensated thermoelectromotive force calculated by the cold-end error compensation model and the real-time thermoelectromotive force of the thin-film thermocouple, and use the thermoelectric output to compensate the cold-end temperature of the thin-film thermocouple in real time; The pre-trained cold-end error compensation model is a single-layer perceptron of cold-end error constructed through the Pytorch deep learning framework.

[0005] In the second aspect, this application provides a system for compensating the cold-end temperature error of a thin-film thermocouple, including: A real-time thermoelectromotive force acquisition module for obtaining the real-time thermoelectromotive force of the thin-film thermocouple; A compensated thermoelectric potential calculation module is configured to input the real-time thermoelectric potential into a pre-trained cold-end error compensation model to calculate the compensated thermoelectric potential; A thermoelectric output calculation module is configured to calculate the real-time thermoelectric output based on the compensated thermoelectric potential calculated by the cold-end error compensation model and the real-time thermoelectric potential of the thin-film thermocouple, and use the thermoelectric output to perform real-time compensation on the cold-end temperature of the thin-film thermocouple; The pre-trained cold-end error compensation model is a single-layer perceptron of cold-end error constructed through the Pytorch deep learning framework.

[0006] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0007] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0008] In a fifth aspect, the present application provides a computer program product including computer instructions. The processor of the computer device reads the computer instructions, and the processor of the computer device executes the computer instructions to implement the steps of the above method.

[0009] Compared with the prior art, the present application has the following beneficial effects: Under the condition that the cold-end temperature of the thin-film thermocouple in the non-standard material system can be obtained, the present invention calculates the error thermoelectric potential under the temperature gradient from the reference temperature to the real-time cold-end temperature, takes this thermoelectric potential value as an error term, and sums it with the real-time thermoelectric potential obtained by the thin-film thermocouple, so as to compensate and correct the thermoelectromotive force of the thin-film thermocouple. This method can be applied to different sensitive material systems. Under the premise that the compensation process remains unchanged for different sensitive systems, only the corresponding model needs to be re-trained, which is beneficial to reducing the cold-end error of various non-standard thin-film thermocouples in temperature measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 It is a flowchart of the method of the present application.

[0012] Figure 2 This is the schematic diagram of the system of this application.

[0013] Figure 3 This is the flowchart of cold junction error compensation.

[0014] Figure 4 This is the comparison before and after compensation of material system A.

[0015] Figure 5 This is the comparison before and after compensation of material system B.

[0016] Figure 6 This is the comparison before and after compensation of material system C.

[0017] Figure 7 This is the comparison before and after compensation of material system D. Detailed implementation manners

[0018] To make the objectives, technical solutions and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some but not all of the embodiments of this application. Usually, the components of the embodiments of this application described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of this application claimed, but merely represents selected embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of this application without creative efforts shall fall within the scope of protection of this application.

[0020] It should be noted that: like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0021] In the description of the embodiments of this application, it should be noted that if terms such as "upper", "lower", "horizontal", "inner", etc. are used to indicate the orientation or positional relationship, it is based on the orientation or positional relationship shown in the accompanying drawings or the orientation or positional relationship in which the inventive product is usually placed during use. It is only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation of this application. In addition, terms such as "first", "second", etc. are only used for descriptive distinction and cannot be construed as indicating or implying relative importance.

[0022] In addition, when the term "horizontal" appears, it does not mean that the component is required to be absolutely horizontal, but it can be slightly inclined. For example, "horizontal" only means that its direction is more horizontal relative to "vertical", and it does not mean that the structure must be completely horizontal, but it can be slightly inclined.

[0023] In the description of the embodiments of the present application, it should also be noted that unless otherwise clearly specified and limited, when the terms "arranged", "installed", "connected", and "coupled" appear, they should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0024] The following further describes the present application in detail with reference to the drawings: See Figure 1 , the embodiments of the present application disclose a method for compensating the cold-end temperature error of a thin-film thermocouple, including the following steps: S1 Obtain the real-time thermoelectromotive force of the thin-film thermocouple; S2 Input the real-time thermoelectromotive force into a pre-trained cold-end error compensation model to calculate the compensated thermoelectromotive force; In practical applications, input the real-time thermoelectromotive force into a pre-trained cold-end error compensation model to calculate the compensated thermoelectromotive force, specifically as follows: S201 Input the actually measured room temperature as the cold-end temperature into the cold-end error compensation model, and replace the original cold-end temperature data with the hot-end temperature input; S202 The cold-end error compensation model outputs the theoretical thermoelectromotive force difference corresponding to the room temperature condition according to the cold-end temperature data and the hot-end temperature data, which is the compensated thermoelectromotive force caused by the cold-end temperature fluctuation.

[0025] It should be noted that the pre-trained cold-end error compensation model is a single-layer perceptron for cold-end error constructed through the Pytorch deep learning framework. The specific methods for constructing and training the cold-end error compensation model are as follows: Step 1 Use the Pytorch deep learning framework to construct a single-layer perceptron; the input layer has 2 nodes (hot-end temperature, cold-end temperature), the hidden layer has 64 nodes, the output layer has 1 node (thermoelectromotive force), the activation function is selected as ReLU, the optimizer is selected as Adam, and the initial learning rate is 0.03; In practical applications, the single-layer perceptron can be replaced with the following variants: Variant 1: Introduce a residual connection to construct a perceptron with a depth of 3 layers, adjust the number of hidden layer nodes to 128 - 64 - 32, and change the activation function to Leaky ReLU to alleviate the vanishing gradient.

[0026] Variant 2: Integrate the LSTM module, and the input features are extended to a time series (including the temperature history data of the previous 5 seconds), which is applicable to strong transient heat transfer scenarios.

[0027] In step 2, the thin-film thermocouple of the target material system is preliminarily calibrated to obtain calibration data; the calibration data includes the hot-end temperature, the cold-end temperature, and the thermoelectric potential; the hot-end temperature and the cold-end temperature are input features, and the thermoelectric potential is the output feature; the calibration data is divided into a training set and a test set according to a ratio of 9:1, the batch size is set to 64, and the training cycle is 500 times. After verifying that the mean square error (MSE) of the model is lower than 0.01 through the test set, the model weights are saved. In step 3, the calibration data is normalized according to two physical quantities, temperature and potential. The temperature data is linearly mapped from 0 to 1000 °C to [-1, 1], and the thermoelectric potential is standardized according to the sensor range (-10 mV to 50 mV). The normalized data is input into a single-layer perceptron for training; and it is verified with the test set, and the hyperparameters are adjusted until convergence.

[0028] In practical applications, the specific method for preliminarily calibrating the thin-film thermocouple of the target material system is as follows: Place the hot end of the thin-film thermocouple into the muffle furnace. The heating rate of the muffle furnace is 1 °C / min, and the temperature range for calibrating the hot end is 30 to 1000 °C. The cold end is placed in the laboratory air environment, and the thermoelectric signal of the thin-film thermocouple is input into the data collector. Deploy a standard K-type thermocouple near the hot and cold nodes of the thin-film thermocouple, and input the temperature signal of the standard K-type thermocouple into the data collector. Use the laboratory room temperature as the cold-end reference temperature of the thin-film thermocouple.

[0029] It should be noted that for different substrate materials (such as ceramics, nickel-based alloys), adjust the temperature gradient in the calibration step: Ceramic materials: The upper limit of the calibration temperature is increased to 1200 °C, and the heating rate is reduced to 0.5 °C / min to prevent thermal stress cracking.

[0030] Nickel-based alloys: Add transient temperature shock tests (such as a step change from 500 °C to 800 °C), and collect dynamic response data to enhance the robustness of the model.

[0031] S3 Calculate the real-time thermoelectric output based on the compensated thermoelectric potential calculated by the cold-end error compensation model and the real-time thermoelectric potential of the thin-film thermocouple, and use the thermoelectric output to perform real-time compensation on the cold-end temperature of the thin-film thermocouple. It should be noted that the specific method for calculating the real-time thermoelectric output based on the cold-end error term and the real-time thermoelectric potential data of the thin-film thermocouple is as follows: Add the compensated thermoelectromotive force calculated by the cold-end error compensation model to the actually measured thermoelectromotive force to obtain the corrected thermoelectric output. The calculation method is as follows:

[0032] The corrected thermoelectromotive force is equivalent to the ideal output with the cold-end temperature constant at 20°C.

[0033] The system is deployed on an embedded device (such as Raspberry Pi 4B), connected to the data acquisition module through the GPIO interface, and the model inference latency is less than 10 ms, meeting the real-time compensation requirements.

[0034] Such as Figure 2 As shown, the embodiment of the present application discloses a cold-end temperature error compensation system for thin-film thermocouples, including: A real-time thermoelectromotive force acquisition module for acquiring the real-time thermoelectromotive force of the thin-film thermocouple; A compensated thermoelectromotive force calculation module for inputting the real-time thermoelectromotive force into a pre-trained cold-end error compensation model to calculate the compensated thermoelectromotive force; A thermoelectric output calculation module for calculating the real-time thermoelectric output according to the compensated thermoelectromotive force calculated by the cold-end error compensation model and the real-time thermoelectromotive force of the thin-film thermocouple, and using the thermoelectric output to perform real-time compensation on the cold-end temperature of the thin-film thermocouple; The pre-trained cold-end error compensation model is a single-layer perceptron of cold-end error constructed through the Pytorch deep learning framework.

[0035] In practical applications, the system of the present application is integrated into an industrial PLC (Siemens S7-1500) and interacts with the DCS system through the OPC UA protocol.

[0036] Anti-interference design: Add an RC low-pass filter (cutoff frequency 1 kHz) at the front end of the data acquisition module to suppress the interference of electromagnetic noise on the thermoelectromotive force signal.

[0037] Adaptive compensation: Add a temperature sensor array to monitor the actual cold-end temperature. When the cold-end temperature deviates from 20°C, dynamically update the model input parameters to achieve closed-loop compensation.

[0038] Such as Figure 3 As shown, the present application discloses a method for compensating the cold-end temperature error of a thin-film thermocouple based on a single-layer perceptron, including the following steps: S1. Preliminarily calibrate the thin-film thermocouple of the target material system S101. Place the hot end of the thin-film thermocouple into a muffle furnace (Nabertherm, P300 LHT 2-17, Germany), set the heating rate of the muffle furnace to 1°C / min, and calibrate the temperature range of the hot end to be 30~1000°C, with the cold end placed in the laboratory air environment.

[0039] S102. The thin-film thermocouple inputs the thermoelectric signal into a data collector (HIOKI, LR8431-30, Japan) through copper leads for storage. At the same time, a standard K-type thermocouple (OMEGA, U.S.A) is deployed near the hot and cold nodes of the thin-film thermocouple. Its leads are also connected to the data collector and converted into real-time temperature signals.

[0040] S103. Measure the indoor temperature of the laboratory (20 °C) as the cold-end reference temperature of the thin-film thermocouple.

[0041] S2. Construct and train a single-layer perceptron for compensating the cold-end error S201. Using the Pytorch deep learning framework, set the number of input features to 2, the number of hidden layers to 64, the number of output features to 1, the initial learning rate to 0.03, select Adam as the optimizer, and select RELU as the activation function.

[0042] S202. The calibration data includes the hot-end temperature, cold-end temperature, and thermoelectromotive force. Among them, the two temperatures are input features, and the thermoelectromotive force is the output feature. The collected calibration data is divided into two parts: a training set and a test set, with a ratio of approximately 9:1.

[0043] S203. Normalize the calibration data according to two physical quantities: temperature and potential, and input the normalized data into the single-layer perceptron for training. Then use the test set for verification, and further adjust the model hyperparameters until the model error shown by the test set is small enough.

[0044] S204. After the model training is completed, save the model.

[0045] S3. Calculate the cold-end error term using the single-layer perceptron S301. Load the saved model, use the measured room temperature (20 °C) to replace the cold-end temperature input during the model training process, and use the cold-end temperature during the training process to replace the hot-end temperature input.

[0046] S302. The model calculates and outputs the thermoelectromotive force, and the thermoelectromotive force at this time is the value of the thermoelectromotive force that needs to be compensated.

[0047] S4. Compensate the cold-end error S401. Add the thermoelectromotive force calculated and output by the model to the thermoelectromotive force data obtained by real-time calibration to obtain the compensated thermoelectromotive force. This compensated thermoelectromotive force is the thermoelectric output without cold-end error when the cold-end temperature is fixed at room temperature.

[0048] The principle of this application is as follows: The process of generating thermoelectromotive force described by the Seebeck effect is as follows:

[0049] Seebeck coefficient is related to the characteristics of the material system, and the hot-end temperature is the measured target temperature, and the cold-end temperature instability will lead to thermoelectric potential and calibration of appears error. The single-layer perceptron of the present invention learns the mapping relationship from the cold-end temperature and the hot-end temperature input to the thermoelectric potential output, avoiding the complex Seebeck coefficient function of measurement and calculation relationship, resulting in a problem of poor compensation effect.

[0050] Combined with the intermediate temperature law:

[0051] wherein, represents the error-term thermoelectric potential for compensation, and its theoretical calculation is as follows:

[0052] Actually, this process is calculated by the single-layer perceptron input with room temperature and the cold-end temperature input.

[0053] Model verification: The embodiments of the present application use four sensitive material systems of indium oxide and its dopants (represented by ABCD respectively) as the verification of the present application.

[0054] Four groups of data are respectively calibrated, and four different single-layer perceptron models are trained. These four models are verified by the test set, and the average relative errors of the learned corresponding thermoelectric potentials are 0.21880%, 0.53927%, 1.20635%, and 0.87627% respectively. At the same time, the fitting indexes of the four models are as follows: Table 1 shows the accuracy of the four trained models in learning thermoelectric output

[0055] Among them, the four indexes are mean squared error: MSE (Mean Squared Error), root mean squared error: RMSE (Root Mean Squard Error), mean absolute error: MAE (Mean Absolute Error), and coefficient of determination: R2 (R-Square).

[0056] It can be considered that the model has a relatively high accuracy in learning the mapping relationship between temperature and thermoelectric potential and can be used for calculating the compensated thermoelectric potential.

[0057] The comparison before and after compensation can be seen Figure 4 、 Figure 5, Figure 6 and Figure 7 , the thermoelectric curve before compensation is a dash line, and the thermoelectric curve after compensation is a solid line. During the calibration process of the four material systems, the cold-end temperature is between 20°C and 180°C. Under this condition, the measurement accuracy of the maximum elevated temperature is improved by 6.67%. Actually, as the cold-end temperature further rises, the cold-end error will be greater, and the effect of accuracy improvement will be more obvious.

[0058] A computer device provided by an embodiment of the present application. The computer device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0059] The computer program can be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present application.

[0060] The computer device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device may include, but is not limited to, a processor and a memory.

[0061] The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0062] The memory can be used to store the computer program and / or module. The processor realizes various functions of the computer device by running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory.

[0063] If the modules / units integrated in the computer device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of this application, it can also be completed by a computer program instructing the relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0064] The embodiment of this application also provides a computer program product or computer program. The computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes Figure 1 the methods provided in the various alternative manners in

[0065] Therefore, details will not be elaborated here.

[0066] In the embodiments of the present application, the term "module" or "unit" refers to a computer program with a predetermined function or a part of a computer program, which works together with other related parts to achieve a predetermined goal, and can be fully or partially implemented by using software, hardware (such as a processing circuit or a memory), or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of the overall module or unit that includes the functions of the module or unit.

[0067] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in this description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0068] The methods and related devices provided in the embodiments of the present application are described with reference to the method flowcharts and / or structural schematic diagrams provided in the embodiments of the present application. Specifically, each process and / or block of the method flowchart and / or structural schematic diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable devices generate a device for implementing the functions specified in one process Figure 1 one process or multiple processes and / or structural schematic Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, or are transmitted through a computer-readable storage medium. Computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The instruction device implements the functions in one process Figure 1 one process or multiple processes and / or structural schematic Figure 1The functions specified in one or more boxes. These computer program instructions can also be loaded onto a computer or other programmable device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing in the process Figure 1 One process or more processes and / or structural diagrams show the steps of the functions specified in one or more boxes.

[0069] The steps in the method embodiments of this application can be adjusted, combined, and deleted according to actual needs.

[0070] The modules in the device embodiments of this application can be combined, divided, and deleted according to actual needs.

[0071] The foregoing disclosure is only for the preferred embodiments of this application. Of course, it cannot be used to limit the scope of rights of this application. Therefore, equivalent changes made according to the claims of this application still fall within the scope covered by this application.

Claims

1. A method for compensating the cold-end temperature error of a thin-film thermocouple, characterized in that: The following steps are involved: Get the real-time thermoelectric potential of thin film thermocouple; The real-time thermoelectric potential is input into the pre-trained cold-end error compensation model to calculate the compensated thermoelectric potential; The real-time thermoelectric output is calculated based on the compensated thermoelectric potential calculated by the cold-end error compensation model and the real-time thermoelectric potential of the thin-film thermocouple, and the cold-end temperature of the thin-film thermocouple is compensated in real time using the thermoelectric output; The pre-trained cold end error compensation model is a single-layer perceptron for cold end error built through a Pytorch deep learning framework.

2. The thin film thermocouple cold end temperature error compensation method according to claim 1, characterized in that: The construction and training method of the cold end error compensation model is as follows: Use the Pytorch deep learning framework, select Adam as the optimizer, select RELU as the activation function, and build a single-layer perceptron; Preliminarily calibrating the thin film thermocouple of the target material system to obtain calibration data; the calibration data includes hot end temperature, cold end temperature and thermoelectric potential; the hot end temperature and cold end temperature are input characteristics, and the thermoelectric potential is an output characteristic; Divide the data into training set and test set in a ratio of 9:1; The calibration data is normalized according to the two physical quantities of temperature and electric potential, and the normalized data is input into the single-layer perceptron for training; it is verified with the test set, and the hyperparameters are adjusted until convergence.

3. The thin film thermocouple cold end temperature error compensation method according to claim 1, characterized in that: The preliminary calibration of the thin film thermocouple of the target material system includes: The hot end of the thin film thermocouple is placed in a muffle furnace, the cold end is placed in the laboratory air environment, and the thermoelectric signal of the thin film thermocouple is input into a data acquisition device; Deploy standard K-type thermocouples near the hot node and the cold node of the thin film thermocouple, and input the temperature signal of the standard K-type thermocouple into the data acquisition device; The laboratory room temperature is used as the cold end reference temperature of the thin film thermocouple.

4. The thin film thermocouple cold end temperature error compensation method according to claim 3, characterized in that: The heating rate of the muffle furnace is 1°C / min, and the temperature range of the hot end is calibrated to be 30-1000°C.

5. The thin film thermocouple cold end temperature error compensation method according to claim 1, characterized in that: The step of inputting the real-time thermoelectric potential into a pre-trained cold-end error compensation model to calculate the compensated thermoelectric potential includes: The actual measured room temperature is used as the cold end temperature input into the cold end error compensation model, and the original cold end temperature data is replaced with the hot end temperature input; The cold-end error compensation model outputs the theoretical thermoelectric potential difference under room temperature conditions according to the cold-end temperature data and the hot-end temperature data, which is the compensated thermoelectric potential caused by the cold-end temperature fluctuation.

6. The thin film thermocouple cold end temperature error compensation method according to claim 1, characterized in that: The real-time thermoelectric output is calculated based on the cold end error term and the real-time thermoelectric potential data of the thin film thermocouple, including: The compensated thermoelectric potential calculated by the cold-end error compensation model is added to the actual measured thermoelectric potential to obtain the corrected thermoelectric output.

7. A thin film thermocouple cold end temperature error compensation system, characterized in that: include: A real-time thermoelectric potential acquisition module is used to acquire the real-time thermoelectric potential of the thin film thermocouple; A compensation thermoelectric potential calculation module is used to input the real-time thermoelectric potential into a pre-trained cold-end error compensation model to calculate the compensation thermoelectric potential; Thermoelectric output calculation module, used to calculate the real-time thermoelectric output according to the compensated thermoelectric potential calculated by the cold-end error compensation model and the real-time thermoelectric potential of the thin-film thermocouple, and use the thermoelectric output to perform real-time compensation for the cold-end temperature of the thin-film thermocouple; The pre-trained cold end error compensation model is a single-layer perceptron for cold end error built through a Pytorch deep learning framework.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. 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 method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes computer instructions. A processor of a computer device reads the computer instructions, and the processor of the computer device executes the computer instructions to implement the steps of the method according to any one of claims 1 to 6.

Images