Pressure sensor error calibration method and system based on machine learning

Abstract

The invention relates to the technical field of instrument calibration, and particularly discloses a pressure sensor error calibration method and system based on machine learning, and the method comprises the steps: obtaining a training set; wherein the training set comprises multiple groups of sample data; normalizing the training set to obtain a normalized training set; establishing an error calibration model for pressure sensor error calibration; training an error calibration model according to the normalized training set to obtain a trained error calibration model; and inputting the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration so as to output a current calibrated pressure value corresponding to the current measured pressure value. According to the pressure sensor error calibration method based on machine learning provided by the invention, the measurement error of the pressure sensor can be quickly calibrated with high precision, and the overall performance of the pressure sensor is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of instrument calibration, and in particular to a pressure sensor error calibration method and system based on machine learning. Background Art

[0002] Pressure sensor is a device used to measure and convert fluid pressure into standard electrical signals. It is widely used in aerospace, oil drilling, boiler pressure measurement and other industries. Due to the manufacturing process error of the pressure sensor and the temperature drift of the sensor's direct output signal, the output result of the instrument is not accurate enough, and it is difficult to provide reliable data for the production process. The data output by the pressure sensor needs to be calibrated to meet the needs of high-precision industry. High-precision error calibration methods are very important. Summary of the invention

[0003] In response to the existing error calibration needs, the present invention provides a pressure sensor error calibration method and system based on machine learning, which can calibrate the pressure sensor error quickly and with high precision.

[0004] As a first aspect of the present invention, a pressure sensor error calibration method based on machine learning is provided, wherein the pressure sensor is placed in a high and low temperature test chamber, the temperature in the high and low temperature test chamber is controlled within a preset temperature range, so that the operating temperature of the pressure sensor is within the preset temperature range, a plurality of operating temperature values ​​within the preset temperature range are set, and different pressures are sequentially applied to the pressure sensor by a high-precision pressure source at each operating temperature value, so as to obtain a plurality of measured pressure values ​​output by the pressure sensor at each operating temperature value and their corresponding input pressure values; the pressure sensor error calibration method based on machine learning comprises:

[0005] Step S1: obtaining a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value;

[0006] Step S2: normalizing the training set to obtain a normalized training set;

[0007] Step S3: establishing an error calibration model for pressure sensor error calibration;

[0008] Step S4: training the error calibration model according to the normalized training set to obtain a trained error calibration model;

[0009] Step S5: inputting the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output a current calibrated pressure value corresponding to the current measured pressure value.

[0010] Furthermore, the step of normalizing the training set to obtain a normalized training set further includes:

[0011] (1) Normalizing the operating temperature values ​​in the multiple groups of sample data, the formula is as follows:

[0012]

[0013] in, Represents the normalized value of the operating temperature T, T min Represents the minimum operating temperature of the pressure sensor, T max Represents the maximum operating temperature value of the pressure sensor;

[0014] (2) Normalizing the input pressure values ​​in the multiple groups of sample data, the formula is as follows:

[0015]

[0016] in, Represents input pressure value The normalized value, represents the minimum input pressure value among the multiple sets of sample data, represents the maximum input pressure value among the plurality of sets of sample data;

[0017] (3) Normalizing the measured pressure values ​​in the multiple groups of sample data, the formula is as follows:

[0018]

[0019] in, Represents the normalized value of the measured pressure value x, x min represents the minimum measured pressure value in the multiple sets of sample data, x max Represents the maximum measured pressure value in the multiple sets of sample data.

[0020] Furthermore, the establishment of an error calibration model for pressure sensor error calibration also includes:

[0021] (1) Use Gaussian process regression with machine learning algorithm to calibrate the manufacturing error of pressure sensor:

[0022] μ * =K(x * ,X)·K(X,X) -1 ·y

[0023]

[0024] Among them, μ* represents the output value of Gaussian process regression, K(x * ,X) represents the covariance matrix, k rbf represents the covariance matrix K(x * ,X), σ f represents the predicted output range, l represents the function smoothness, X represents the operating temperature value and the measured pressure value in the training set, y represents the input pressure value in the training set, and x * Represents the normalized value corresponding to the new input pressure value;

[0025] (2) According to the change DT of the output characteristic curve of the pressure sensor at different temperatures, the temperature drift of the pressure sensor is calibrated:

[0026]

[0027] Where T represents the current operating temperature of the pressure sensor, T i represents the previous working temperature value of the current working temperature value T in the training set, T i+1 represents the next working temperature value after the current working temperature value T in the training set, Represents the operating temperature value T i The pressure sensor output characteristic curve is as follows: +1 Represents the operating temperature value T i+1 The pressure sensor output characteristic curve below, DT represents T i+1 and T i The corresponding change in the pressure sensor output characteristic curve, T c Represents the reference temperature value;

[0028] (3) Calibrate the pressure sensor by combining the manufacturing error and temperature drift of the pressure sensor through linear interpolation:

[0029]

[0030] Where P' represents the final pressure sensor calibration result, Represents the operating temperature value T i The corresponding output value of Gaussian process regression is, Represents the operating temperature value T i+1 The corresponding output value of Gaussian process regression, FullScale represents the measurement range of the pressure sensor.

[0031] Furthermore, in step S4 and step S5, the symmetric positive definite covariance matrix K is decomposed into two lower triangular matrices L and L using Cholesky decomposition. T The product of:

[0032] K=L·L T

[0033] Among them, L and L T represents two lower triangular matrices;

[0034] Let α = L -T ·(L -1 ·y) into the formula μ * =K(x * ,X)·K(X,X) -1 ·y, to obtain

[0035] As another aspect of the present invention, a pressure sensor error calibration device based on machine learning is provided, which is used to implement the pressure sensor error calibration method based on machine learning as described above, and the pressure sensor error calibration device based on machine learning includes:

[0036] An acquisition module, used to acquire a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value;

[0037] A normalization module, used for normalizing the training set to obtain a normalized training set;

[0038] Establishing a module for establishing an error calibration model for pressure sensor error calibration;

[0039] A training module, used for training the error calibration model according to the normalized training set to obtain a trained error calibration model;

[0040] The calibration module is used to input the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output the current calibrated pressure value corresponding to the current measured pressure value.

[0041] As another aspect of the present invention, a pressure sensor error calibration system based on machine learning is provided, comprising: a computer, a pressure sensor signal acquisition device, a high and low temperature test chamber and a high-precision pressure source, the computer comprising the pressure sensor error calibration device based on machine learning as described above, the computer being connected to the pressure sensor signal acquisition device and the high and low temperature test chamber respectively, the pressure sensor signal acquisition device being connected to the high and low temperature test chamber, the computer controlling the temperature in the high and low temperature test chamber within a preset temperature range, the high-precision pressure source being used to apply different pressures to the pressure sensor in the high and low temperature test chamber in turn at each operating temperature value, so that the pressure sensor signal acquisition device can collect multiple measured pressure values ​​output by the pressure sensor, the pressure sensor signal acquisition device uploads the collected multiple measured pressure values ​​to the computer, and the computer can complete the pressure sensor error calibration process based on machine learning.

[0042] The present invention provides a pressure sensor error calibration method based on machine learning, which has the following beneficial effects: the error calibration method is essentially a software compensation method. Starting from the physical source of the measurement error generated by the pressure sensor, combined with the current advanced machine learning algorithm, it can well solve the data calibration problems of various types of pressure sensors and improve the overall performance of the pressure sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0044] Figure 1 A flow chart of the pressure sensor error calibration method based on machine learning provided by the present invention.

[0045] Figure 2 A flowchart of a specific implementation of the pressure sensor error calibration method based on machine learning provided by the present invention.

[0046] Figure 3 This is a structural diagram of the error calibration model provided by the present invention.

[0047] Figure 4 This is a structural diagram of the pressure sensor error calibration system based on machine learning provided by the present invention. DETAILED DESCRIPTION

[0048] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0050] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0051] In this embodiment, a pressure sensor error calibration method based on machine learning is provided. Figure 4 As shown, the pressure sensor is placed in a high and low temperature test box, and the temperature in the high and low temperature test box is controlled within a preset temperature range, so that the operating temperature of the pressure sensor is within the preset temperature range, and multiple operating temperature values ​​within the preset temperature range are set. At each operating temperature value, different pressures are applied to the pressure sensor in turn through a high-precision pressure source to obtain multiple measured pressure values ​​output by the pressure sensor at each operating temperature value and their corresponding input pressure values; as shown Figure 1 As shown, the pressure sensor error calibration method based on machine learning includes:

[0052] Step S1: obtaining a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value;

[0053] Specifically, in step S1, when acquiring multiple groups of sample data, starting from the lowest temperature point or the highest temperature point in the working range of the pressure sensor, the pressure sensor is boosted from zero to full scale or reduced from full scale to zero at a certain temperature interval within the working range. During the boost and reduce process, each sampling point is separated by a certain pressure, and the temperature conditions, input pressure values ​​and sensor output signal values ​​are recorded.

[0054] Step S2: performing normalization processing on the training set to obtain a normalized training set; the normalization processing can enhance the versatility and stability of the data calibration method;

[0055] Preferably, the step of normalizing the training set to obtain a normalized training set further includes:

[0056] (1) Normalizing the operating temperature values ​​in the multiple groups of sample data, the formula is as follows:

[0057]

[0058] in, Represents the normalized value of the operating temperature T, T min Represents the minimum operating temperature of the pressure sensor, T max Represents the maximum operating temperature value of the pressure sensor;

[0059] (2) Normalizing the input pressure values ​​in the multiple groups of sample data, the formula is as follows:

[0060]

[0061] in, Represents input pressure value The normalized value, represents the minimum input pressure value among the multiple sets of sample data, represents the maximum input pressure value among the plurality of sets of sample data;

[0062] (3) Normalizing the measured pressure values ​​in the multiple groups of sample data, the formula is as follows:

[0063]

[0064] in, Represents the normalized value of the measured pressure value x, x min represents the minimum measured pressure value in the multiple sets of sample data, x max Represents the maximum measured pressure value in the multiple sets of sample data.

[0065] Step S3: establishing an error calibration model for pressure sensor error calibration using a machine learning algorithm according to the physical characteristics of the pressure sensor;

[0066] It should be noted that the errors of pressure sensors mainly come from two sources: manufacturing process and ambient temperature. After the pressure sensor output characteristic curve is manufactured, the actual structural parameters of the sensor are not completely consistent with the theoretical design due to the error of the manufacturing process, and there is an inherent nonlinear error. This nonlinear error is even more unpredictable at different operating temperatures. These two types of errors are determined after the sensor is manufactured and are unique physical characteristics of each sensor.

[0067] Preferably, if Figure 2-3 As shown, the error calibration model established for pressure sensor error calibration also includes:

[0068] (1) Use Gaussian Process Regression (GPR) of machine learning algorithm to calibrate the manufacturing error of pressure sensor:

[0069] μ * =K(x * ,X)·K(X,X) -1 ·y

[0070]

[0071] Among them, μ * represents the output value of Gaussian process regression, K(x * ,X) represents the covariance matrix, k rbf represents the covariance matrix K(x * ,X), σ f represents the predicted output range, l represents the function smoothness, X represents the operating temperature value and the measured pressure value in the training set, y represents the input pressure value in the training set, and x * Represents the normalized value corresponding to the new input pressure value;

[0072] (2) According to the change DT of the output characteristic curve of the pressure sensor at different temperatures, the temperature drift of the pressure sensor is calibrated:

[0073]

[0074] Where T represents the current operating temperature of the pressure sensor, T i represents the previous working temperature value of the current working temperature value T in the training set, T i+1 represents the next working temperature value after the current working temperature value T in the training set, Represents the operating temperature value T i The pressure sensor output characteristic curve is as follows: Represents the operating temperature value T i+1The pressure sensor output characteristic curve below, DT represents T i+1 and T i The corresponding change in the pressure sensor output characteristic curve, T c Represents the reference temperature value, usually 20℃;

[0075] (3) Calibrate the pressure sensor by combining the manufacturing error and temperature drift of the pressure sensor through linear interpolation:

[0076]

[0077] Where P' represents the final pressure sensor calibration result, Represents the operating temperature value T i The corresponding output value of Gaussian process regression is, Represents the operating temperature value T i+1 The corresponding output value of Gaussian process regression, FullScale represents the measurement range of the pressure sensor.

[0078] It should be noted that during the calibration process, the temperature drift trend is introduced by linear interpolation to compensate for the error caused by the change in working temperature. The temperature drift trend between the two reference curves is introduced by linear interpolation according to the temperature distribution, and the influence of the ambient temperature on the sensor is restored to obtain the final calibration result.

[0079] Step S4: training the error calibration model according to the normalized training set to obtain a trained error calibration model; the trained error calibration model can be used to calibrate the error of the pressure sensor so that it can achieve high-precision output under different working environments.

[0080] Preferably, in step S3 and step S4, after the error calibration model is established, several pressure sensor output characteristic curves D at different operating temperatures are trained according to the operating temperature range of the pressure sensor, and their changes DT at different temperatures are calculated. The input of the error calibration model is the operating temperature value and the measured pressure value, and the output of the error calibration model is the pressure sensor calibration result.

[0081] Step S5: inputting the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output a current calibrated pressure value corresponding to the current measured pressure value.

[0082] Preferably, in step S4 and step S5, the symmetric positive definite covariance matrix K is decomposed into two lower triangular matrices L and L using Cholesky decomposition. T The product of avoids the situation where the matrix is ​​close to singularity and ensures the stability of calculation:

[0083] K=L·LT

[0084] Among them, L and L T represents two lower triangular matrices;

[0085] Let α = L -T ·(L -1 ·y) into the formula μ * =K(x * ,X)·K(X,X) -1 ·y, to obtain Simplify the calculation of forward calibration.

[0086] Specifically, in step S5, two pre-trained sensor output characteristic curves at adjacent working temperatures are first selected according to the current working temperature T. and

[0087] Then use α to perform forward calibration to avoid inverting the covariance matrix and perform Gaussian process regression to obtain the result after manufacturing error calibration. and

[0088]

[0089] Finally, linear interpolation is used to introduce temperature drift,

[0090] Finally, the output of the forward calibration of the error calibration model is: P'=FullScale*μ', where FullScale is the measurement range of the pressure sensor.

[0091] As another embodiment of the present invention, a pressure sensor error calibration device based on machine learning is provided, which is used to implement the pressure sensor error calibration method based on machine learning as described above. The pressure sensor error calibration device based on machine learning includes:

[0092] An acquisition module, used to acquire a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value;

[0093] A normalization module, used for normalizing the training set to obtain a normalized training set;

[0094] Establishing a module for establishing an error calibration model for pressure sensor error calibration;

[0095] A training module, used for training the error calibration model according to the normalized training set to obtain a trained error calibration model;

[0096] The calibration module is used to input the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output the current calibrated pressure value corresponding to the current measured pressure value.

[0097] As another embodiment of the present invention, a pressure sensor error calibration system based on machine learning is provided. Figure 4 As shown, it includes: a computer, a pressure sensor signal acquisition device, a high and low temperature test chamber and a high-precision pressure source, the computer includes the pressure sensor error calibration device based on machine learning mentioned above, the computer is connected to the pressure sensor signal acquisition device and the high and low temperature test chamber respectively, the pressure sensor signal acquisition device is connected to the high and low temperature test chamber, the computer controls the temperature in the high and low temperature test chamber within a preset temperature range, the high-precision pressure source is used to apply different pressures to the pressure sensor in the high and low temperature test chamber in turn at each working temperature value, so that the pressure sensor signal acquisition device can collect multiple measured pressure values ​​output by the pressure sensor, the pressure sensor signal acquisition device uploads the collected multiple measured pressure values ​​to the computer, and the computer can complete the pressure sensor error calibration process based on machine learning.

[0098] Specifically, a standard pressure source with higher precision than the pressure sensor is used to provide a reference for calibration, and a pressure sensor signal acquisition device is used to obtain the output signal of the pressure sensor in the high and low temperature test chamber. The high-precision pressure source, pressure sensor signal acquisition device and high and low temperature test chamber are all automatically controlled by a computer.

[0099] A specific implementation case is provided below to explain in detail the specific implementation process of the pressure sensor error calibration method based on machine learning provided by the present invention.

[0100] 1. Taking the operating temperature range of the pressure sensor -40-85℃ as an example, the full scale of the pressure sensor is 130kPa. The pressure sensor is collected at different operating temperature values ​​T of -40℃, -10℃, 10℃, 20℃, 50℃, and 85℃. i Under each temperature condition, the measured pressure P is applied to the pressure sensor by a high-precision pressure source. j , from 0% of the pressure sensor range to 100% by 5%, a total of 21 pressure points, collect each (T i ,P j ) The measured pressure value P output by the pressure sensor under the combination n ; Divided into 6 groups of data according to temperature conditions D(Ti ,P j ,P n ), where i∈(-40℃, -10℃, 10℃, 20℃, 50℃, 85℃); j∈1,2,3…,21; n∈1,2,3…,21; the pressure sensor output characteristic curve at [-10,50]℃ is selected as the training set.

[0101] 2. For the working temperature value T i , Input pressure value P j Normalized with the measured pressure value Pn;

[0102] For example, the operating temperature value T i The normalization formula is as follows:

[0103]

[0104] 3. According to the two error sources of the pressure sensor, the error calibration model is established using the machine learning algorithm, and then the error calibration model is trained based on the pressure sensor output characteristic curve sampled at [-40, 10, 20, 85] ℃. Using α = L -T ·(L -1 y) to perform forward calculations and avoid matrix inversion.

[0105] 4. After the error calibration model training is completed, the output value P' of the error calibration model is:

[0106]

[0107] 5. Input the current measured pressure value and the current working temperature value output by the pressure sensor into the trained error calibration model for error calibration; wherein, the current measured pressure value and the current working temperature value are normalized and standardized based on the pressure sensor output characteristic curve sampled at 20°C, simplified to (μ, T); then, according to the working temperature value T in (μ, T), two pre-trained pressure sensor output characteristic curves D and DT under adjacent temperature conditions are selected;

[0108] If the data to be calibrated is sampled at a temperature of T = -10°C, the model selects the pressure sensor output characteristic curve D at -40°C -40 And the pressure sensor output characteristic curve D at 10°C 10 .

[0109] D -40 and D 10 Perform Gaussian process regression prediction with μ respectively, Get the output value μ of Gaussian process regression at -40℃ -40and the output value μ of Gaussian process regression at 10°C 10 ;

[0110] According to the actual temperature distribution, the linear interpolation method is used to calculate the change DT of the pressure sensor output characteristic curve corresponding to -40℃ and 10℃. -40～10 Calculate μ';

[0111]

[0112] Where A is the proportionality coefficient;

[0113] Map μ' back to the measuring range of the pressure sensor, that is, obtain the calibrated pressure value P', P'=130 kPa*μ'; P' is the current calibrated pressure value corresponding to the current measured pressure value.

[0114] The present invention provides a pressure sensor error calibration method based on machine learning, which is essentially a software compensation method. Starting from the physical source of measurement errors generated by pressure sensors and combining with currently advanced machine learning algorithms, it can well solve the data calibration problems of various types of pressure sensors and improve the overall performance of pressure sensors.

[0115] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A pressure sensor error calibration method based on machine learning, characterized in that: The pressure sensor is placed in a high and low temperature test box, and the temperature in the high and low temperature test box is controlled within a preset temperature range, so that the working temperature of the pressure sensor is within the preset temperature range, and multiple working temperature values ​​within the preset temperature range are set. At each working temperature value, different pressures are applied to the pressure sensor in sequence through a high-precision pressure source to obtain multiple measured pressure values ​​output by the pressure sensor at each working temperature value and their corresponding input pressure values; the pressure sensor error calibration method based on machine learning includes: Step S1: obtaining a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value; Step S2: normalizing the training set to obtain a normalized training set; Step S3: establishing an error calibration model for pressure sensor error calibration; Step S4: training the error calibration model according to the normalized training set to obtain a trained error calibration model; Step S5: inputting the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output a current calibrated pressure value corresponding to the current measured pressure value.

2. The pressure sensor error calibration method based on machine learning according to claim 1 is characterized in that: The step of normalizing the training set to obtain a normalized training set further includes: (1) Normalizing the operating temperature values ​​in the multiple groups of sample data, the formula is as follows: in, Represents the normalized value of the operating temperature T, T min Represents the minimum operating temperature of the pressure sensor, T max Represents the maximum operating temperature value of the pressure sensor; (2) Normalizing the input pressure values ​​in the multiple groups of sample data, the formula is as follows: in, Represents input pressure value The normalized value, represents the minimum input pressure value among the multiple sets of sample data, represents the maximum input pressure value among the plurality of sets of sample data; (3) Normalizing the measured pressure values ​​in the multiple groups of sample data, the formula is as follows: in, Represents the normalized value of the measured pressure value x, x min represents the minimum measured pressure value in the multiple sets of sample data, x max Represents the maximum measured pressure value in the multiple sets of sample data.

3. The pressure sensor error calibration method based on machine learning according to claim 1 is characterized in that: The method of establishing an error calibration model for pressure sensor error calibration also includes: (1) Use Gaussian process regression with machine learning algorithm to calibrate the manufacturing error of pressure sensor: μ * =K(x * ,X)·K(X,X) -1 ·y Among them, μ * represents the output value of Gaussian process regression, K(x * ,X) represents the covariance matrix, k rbf represents the covariance matrix K(x * ,X), σ f represents the predicted output range, l represents the function smoothness, X represents the operating temperature value and the measured pressure value in the training set, y represents the input pressure value in the training set, and x * Represents the normalized value corresponding to the new input pressure value; (2) According to the change DT of the output characteristic curve of the pressure sensor at different temperatures, the temperature drift of the pressure sensor is calibrated: Where T represents the current operating temperature of the pressure sensor, T i represents the previous working temperature value of the current working temperature value T in the training set, T i+1 represents the next working temperature value after the current working temperature value T in the training set, Represents the operating temperature value T i The pressure sensor output characteristic curve is as follows: Represents the operating temperature value T i+1 The pressure sensor output characteristic curve below, DT represents T i+1 and T i The corresponding change in the pressure sensor output characteristic curve, T c Represents the reference temperature value; (3) Calibrate the pressure sensor by combining the manufacturing error and temperature drift of the pressure sensor through linear interpolation: Where P' represents the final pressure sensor calibration result, Represents the operating temperature value T i The corresponding output value of Gaussian process regression is, Represents the operating temperature value T i+1 The corresponding output value of Gaussian process regression, FullScale represents the measurement range of the pressure sensor.

4. The pressure sensor error calibration method based on machine learning according to claim 3 is characterized in that: In step S4 and step S5, the symmetric positive definite covariance matrix K is decomposed into two lower triangular matrices L and L using Cholesky decomposition. T The product of: K=L·L T Among them, L and L T represents two lower triangular matrices; Let α = L -T ·(L -1 ·y) into the formula μ * =K(x * ,X)·K(X,X) -1 ·y, to obtain 5. A pressure sensor error calibration device based on machine learning, used to implement the pressure sensor error calibration method based on machine learning according to any one of claims 1 to 4, characterized in that: The pressure sensor error calibration device based on machine learning includes: An acquisition module, used to acquire a training set; wherein the training set includes multiple groups of sample data, each group of sample data includes an operating temperature value, an input pressure value under the operating temperature value, and a measured pressure value corresponding to the input pressure value; A normalization module, used for normalizing the training set to obtain a normalized training set; Establishing a module for establishing an error calibration model for pressure sensor error calibration; A training module, used for training the error calibration model according to the normalized training set to obtain a trained error calibration model; The calibration module is used to input the current working temperature value and the current measured pressure value into the trained error calibration model for error calibration, so as to output the current calibrated pressure value corresponding to the current measured pressure value.

6. A pressure sensor error calibration system based on machine learning, characterized in that: include: A computer, a pressure sensor signal acquisition device, a high and low temperature test chamber and a high-precision pressure source, the computer includes the pressure sensor error calibration device based on machine learning as described in claim 5, the computer is respectively connected to the pressure sensor signal acquisition device and the high and low temperature test chamber, the pressure sensor signal acquisition device is connected to the high and low temperature test chamber, the computer controls the temperature in the high and low temperature test chamber within a preset temperature range, the high-precision pressure source is used to apply different pressures to the pressure sensor in the high and low temperature test chamber in turn at each working temperature value, so that the pressure sensor signal acquisition device can collect multiple measured pressure values ​​output by the pressure sensor, the pressure sensor signal acquisition device uploads the collected multiple measured pressure values ​​to the computer, and the computer can complete the pressure sensor error calibration process based on machine learning.

Images