A pressure sensor calibration method based on environmental disturbance compensation

Abstract

A pressure sensor calibration method based on environmental interference compensation relates to the technical field of pressure sensor calibration, and comprises: collecting the output value and environmental data of a target pressure sensor and marking the collection time, and setting a collection cycle; statistically analyzing the environmental data of the target sensor in each historical collection cycle, and obtaining key values ​​of various types of environmental parameters; constructing a feature database, and performing simulation experiments based on the environmental feature set in the feature database, and obtaining pressure deviation values ​​under different environmental feature set conditions; constructing a deviation output model, and inputting the environmental data of the current collection cycle into the feature database for feature matching, and selecting to perform compensation adjustment on the target pressure sensor based on the deviation output model according to the feature matching result, or selecting to perform flexible compensation adjustment operations on the target pressure sensor, thereby significantly improving the measurement accuracy, stability and adaptability of the pressure sensor.

Description

Technical Field

[0001] The invention relates to the technical field of pressure sensor calibration, and in particular to a pressure sensor calibration method based on environmental interference compensation. Background Art

[0002] CN106404244B "Pressure sensor and its signal calibration method" includes at least two range intervals, and each range interval is equipped with an amplifier with a corresponding amplification factor and an analog-to-digital converter with a corresponding offset; the signal calibration method includes selecting a calibration pressure point within the range interval; performing temperature compensation calibration on the range interval according to the calibration pressure point and the amplification factor and offset in the corresponding range interval, obtaining corresponding calibration equation parameters, and storing them in a corresponding memory; and calibrating and outputting the original pressure signal collected by the pressure MEMS chip according to the corresponding calibration equation parameters in the memory.

[0003] CN105258847B "A method and device for calibrating a pressure sensor" includes establishing and saving an original data table according to the test temperature values ​​and test pressure values ​​corresponding to the set temperature points and set pressure points of the pressure sensor to be calibrated; selecting several groups of test temperature values ​​and test pressure values ​​from the original data table as calibration data according to the temperature zone and pressure range used; obtaining a calibration equation according to the calibration data; and performing pressure calibration on the current ambient temperature and the original pressure output value of the pressure sensor to be calibrated according to the calibration equation to obtain a calibrated pressure output value.

[0004] Pressure sensors are widely used in industrial control, medical equipment, automobile manufacturing and other fields. However, in actual use, due to the influence of factors such as temperature changes, humidity fluctuations and electromagnetic interference, the measurement results of pressure sensors often deviate from the true value, reducing the reliability and accuracy of the system. Most existing calibration methods only compensate for a single factor, which is difficult to fully solve the problem of multiple environmental factors acting simultaneously. Summary of the invention

[0005] In order to solve the above technical problems, the object of the present invention is to provide a pressure sensor calibration method based on environmental interference compensation, comprising the following steps:

[0006] Step s1: Collect the output value of the target pressure sensor and environmental data, mark the collection time, and set the collection cycle;

[0007] Step s2: Statistically analyze the environmental data of the target sensor in each historical collection period to obtain key values ​​of various types of environmental parameters;

[0008] Step s3: construct a feature database, conduct simulation experiments based on the environmental feature set in the feature database, and obtain pressure deviation values ​​under different environmental feature set conditions;

[0009] Step s4: construct a deviation output model, input the environmental data of the current acquisition cycle into the feature database for feature matching, and select compensation adjustment of the target pressure sensor based on the deviation output model according to the feature matching result, or select flexible compensation adjustment operation for the target pressure sensor.

[0010] Furthermore, the process of statistically analyzing the environmental data of the target sensor in each historical collection period and obtaining the key values ​​of various types of environmental parameters includes:

[0011] Acquire the environmental data of the target sensor in each historical collection cycle, extract the numerical time series sequence corresponding to each type of environmental parameter in the environmental data, perform statistical analysis on the numerical time series sequence corresponding to each type of environmental parameter, obtain the occurrence frequency of different numerical values ​​of each type of environmental parameter, compare the occurrence frequency of different numerical values ​​with a preset occurrence frequency threshold, screen out the numerical values ​​of each type of environmental parameter whose occurrence frequency is greater than the occurrence frequency threshold, and mark the numerical values ​​as key numerical values.

[0012] Furthermore, the process of building a feature database includes:

[0013] Construct a feature database, randomly select a key value from the key values ​​of each type of environmental parameter to combine, generate an environmental feature set, repeat the above steps until the number of environmental feature sets is equal to the preset simulation value upper limit, and store the environmental feature set in the feature database.

[0014] Furthermore, a simulation experiment is conducted based on the environmental feature set in the feature database to obtain the pressure deviation value under different environmental feature set conditions, including:

[0015] Conduct simulation experiments on the target pressure sensor according to each environmental feature set to obtain the original output value of the target pressure sensor under the conditions of different environmental feature sets, and use a standard pressure source to obtain the standard pressure value corresponding to the original output value under the conditions of different environmental feature sets;

[0016] The original output values ​​under different environmental parameter conditions are preprocessed in data format to obtain the original pressure value, and the original pressure value is compared with the standard pressure value corresponding to the original output value to obtain the pressure deviation value under different environmental feature set conditions.

[0017] Furthermore, the process of building a deviation output model includes:

[0018] Obtaining a standard environmental feature set of a target pressure sensor, marking the standard environmental feature set as judgment data, comparing different environmental feature sets with the judgment data, and obtaining numerical deviation values ​​of various types of environmental parameters in different environmental feature sets;

[0019] A deviation output model is constructed based on a multivariate regression algorithm. The numerical deviation values ​​of each type of environmental parameters in different environmental feature sets and the pressure deviation values ​​corresponding to different environmental feature sets are used as training sets and test sets. The training set is input into the deviation output model for training until the loss function training is stable, and the model parameters are saved. The deviation output model is tested with the test set until it meets the preset requirements, and the deviation output model is obtained.

[0020] Furthermore, the environmental data of the current acquisition period is input into the feature database for feature matching, and the process of selecting to perform compensation adjustment on the target pressure sensor based on the deviation output model according to the feature matching result, or selecting to perform flexible compensation adjustment operation on the target pressure sensor includes:

[0021] Input the environmental data of the target pressure sensor in the current acquisition period into the feature database for feature matching, obtain the similarity between the environmental data of the current acquisition period and each environmental feature set, and compare the similarity corresponding to each environmental feature set with a preset similarity threshold;

[0022] If there is an environmental feature set whose similarity is greater than the similarity threshold, compare the environmental data of the current acquisition period with the judgment data, obtain the numerical deviation value of each type of environmental parameter, input the numerical deviation value of each type of environmental parameter into the deviation output model, obtain the pressure deviation value of the target pressure sensor in the current acquisition period according to the deviation output model, perform data format preprocessing on the output value of the target pressure sensor in the current acquisition period, obtain the pressure value, and compensate and adjust the pressure value according to the pressure deviation value;

[0023] If there is no environmental feature set whose similarity is greater than the similarity threshold, a flexible compensation adjustment operation is performed on the target pressure sensor.

[0024] Furthermore, the process of performing a flexible compensation adjustment operation on the target pressure sensor includes:

[0025] Construct a fuzzy control model, the input of the fuzzy control model is the numerical deviation value of each type of environmental parameter, and the output is the pressure deviation value. According to the numerical deviation values ​​of each type of environmental parameter in different environmental feature sets in the feature database and the pressure deviation values ​​corresponding to the different environmental feature sets, construct a fuzzy adaptive rule control table of the fuzzy control model, set the membership threshold, membership function and continuous interval of pressure deviation value of the fuzzy control model, and divide the continuous interval of pressure deviation value into several fuzzy subsets;

[0026] The numerical deviation values ​​of various types of environmental parameters in the current collection cycle are input into the fuzzy control model. The fuzzy control model uses the numerical deviation values ​​of various types of environmental parameters as evaluation indicators, and obtains the weight matrix of the evaluation indicators. The membership matrix of the evaluation indicators for each fuzzy subset is obtained through fuzzy comprehensive evaluation.

[0027] The pressure deviation value of the evaluation index is obtained according to the membership matrix and the weight matrix, and the pressure value is compensated and adjusted according to the pressure deviation value.

[0028] Furthermore, the process of obtaining the weight matrix of the evaluation index includes:

[0029] The pressure deviation values ​​corresponding to several environmental feature sets are obtained from the feature database as sample data, the numerical deviation values ​​of each type of environmental parameters in each environmental feature set in the sample data are obtained, the numerical deviation values ​​of each type of environmental parameters in each environmental feature set in the sample data are used as input parameters, the pressure deviation values ​​corresponding to each environmental feature set in the sample data are used as output values, the Sobol index is used to perform main effect sensitivity analysis, the main effect Sobol index of each type of environmental parameter is obtained, the weight coefficient of each type of environmental parameter is set according to the main effect Sobol index of each type of environmental parameter, and the weight matrix of the evaluation index is obtained according to the weight coefficient of each type of environmental parameter.

[0030] Compared with the prior art, the pressure sensor calibration method based on environmental interference compensation provided by the present invention has the following detailed beneficial effects:

[0031] 1. Improve measurement accuracy

[0032] Comprehensive consideration of multiple environmental factors: The present invention can comprehensively consider the impact of multiple environmental factors on the output of the pressure sensor by collecting and analyzing multiple environmental parameters (such as temperature, humidity, electromagnetic field strength, etc.) and building a feature database. Compared with the traditional single-factor calibration method, the present invention can more accurately compensate for environmental interference and improve measurement accuracy.

[0033] Dynamic compensation: Dynamic compensation is achieved by monitoring the current environmental conditions in real time and selecting the appropriate compensation strategy based on the feature matching results. This allows the sensor to maintain high accuracy under different environmental conditions, especially in scenarios with frequent environmental changes.

[0034] 2. Enhance system stability

[0035] Characteristic database: Build a characteristic database and conduct simulation experiments to obtain pressure deviation values ​​under different environmental characteristic set conditions. This not only provides rich calibration data, but also ensures the stability and reliability of the system under various environmental conditions.

[0036] Fuzzy control model: When no similar environmental feature set is found in the feature database, the fuzzy control model is used for flexible compensation. The fuzzy control model can handle uncertainty and complexity, further enhancing the stability and robustness of the system.

[0037] 3. Reduce calibration costs

[0038] Automated calibration: This invention greatly reduces the need for human intervention and manual calibration by automatically collecting environmental data and sensor output values, automatically building a feature database, and automatically selecting compensation strategies. This not only saves time and labor costs, but also improves the efficiency of calibration.

[0039] Reduced maintenance frequency: Through dynamic compensation and flexible compensation, the sensor can maintain high measurement accuracy under various environmental conditions, reducing the need for frequent calibration and maintenance caused by environmental changes, and reducing long-term operating costs.

[0040] 4. Improve adaptability

[0041] Multivariate regression algorithm: The use of multivariate regression algorithm to build a deviation output model can handle the complex relationship between multiple environmental parameters and improve the adaptability and generalization ability of the model. This enables the sensor to perform well in different application scenarios.

[0042] Flexible compensation: Flexible compensation through fuzzy control model can handle unforeseen environmental conditions and special situations, improving the adaptability and flexibility of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a schematic diagram of a pressure sensor calibration method based on environmental interference compensation according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0045] like Figure 1 As shown, a pressure sensor calibration method based on environmental interference compensation includes the following steps:

[0046] Step s1: Collect the output value of the target pressure sensor and environmental data, mark the collection time, and set the collection cycle;

[0047] Step s2: Statistically analyze the environmental data of the target sensor in each historical collection period to obtain key values ​​of various types of environmental parameters;

[0048] Step s3: construct a feature database, conduct simulation experiments based on the environmental feature set in the feature database, and obtain pressure deviation values ​​under different environmental feature set conditions;

[0049] Step s4: construct a deviation output model, input the environmental data of the current acquisition cycle into the feature database for feature matching, and select compensation adjustment of the target pressure sensor based on the deviation output model according to the feature matching result, or select flexible compensation adjustment operation for the target pressure sensor.

[0050] It should be further explained that, in the specific implementation process, the process of statistically analyzing the environmental data of the target sensor in each historical collection period and obtaining the key values ​​of various types of environmental parameters includes:

[0051] Acquire the environmental data of the target sensor in each historical collection cycle, extract the numerical time series sequence corresponding to each type of environmental parameter in the environmental data, the environmental parameters include temperature, humidity, electromagnetic field strength, etc., perform statistical analysis on the numerical time series sequence corresponding to each type of environmental parameter, obtain the occurrence frequency of different numerical values ​​of each type of environmental parameter, compare the occurrence frequency of different numerical values ​​with a preset occurrence frequency threshold, screen out the numerical values ​​of each type of environmental parameter whose occurrence frequency is greater than the occurrence frequency threshold, and mark the numerical values ​​as key numerical values.

[0052] It should be further explained that, in the specific implementation process, the process of building a feature database includes:

[0053] Construct a feature database, randomly select a key value from the key values ​​of each type of environmental parameter to combine, generate an environmental feature set, repeat the above steps until the number of environmental feature sets is equal to the preset simulation value upper limit, and store the environmental feature set in the feature database.

[0054] It should be further explained that, in the specific implementation process, the simulation experiment is carried out based on the environmental feature set in the feature database, and the process of obtaining the pressure deviation value under different environmental feature set conditions includes:

[0055] Conduct simulation experiments on the target pressure sensor according to each environmental feature set to obtain the original output value (usually voltage or digital signal) of the target pressure sensor under different environmental feature set conditions, and use a standard pressure source (such as a precision pressure gauge) to obtain the standard pressure value corresponding to the original output value under different environmental feature set conditions (the environmental parameters include temperature, humidity, electromagnetic field strength, etc.), and the known pressure value is used to calibrate and verify the accuracy of the sensor;

[0056] The original output values ​​under different environmental parameter conditions are preprocessed in data format to obtain the original pressure value, and the original pressure value is compared with the standard pressure value corresponding to the original output value to obtain the pressure deviation value under different environmental feature set conditions.

[0057] It should be further explained that, in the specific implementation process, the process of building the deviation output model includes:

[0058] Obtaining a standard environmental feature set of a target pressure sensor, marking the standard environmental feature set as judgment data, comparing different environmental feature sets with the judgment data, and obtaining numerical deviation values ​​of various types of environmental parameters in different environmental feature sets;

[0059] A deviation output model is constructed based on a multivariate regression algorithm. The numerical deviation values ​​of each type of environmental parameters in different environmental feature sets and the pressure deviation values ​​corresponding to different environmental feature sets are used as training sets and test sets. The training set is input into the deviation output model for training until the loss function training is stable, and the model parameters are saved. The deviation output model is tested with the test set until it meets the preset requirements, and the deviation output model is obtained.

[0060] It should be further explained that in the specific implementation process, the specific structure of the multivariate regression algorithm is , where Y is the dependent variable, is the intercept term, , , ..., Respectively represent each independent variable , , ..., The associated coefficient, Represents the error term. The "loss function" specifically used in the present invention is Huber Loss. Huber Loss combines the advantages of mean square error MSE and mean absolute error MAE. It is similar to mean square error MSE (smooth) when the error is small, and more like mean absolute error MAE (linear) when the error is large. Therefore, it can converge quickly when the error is small, and reduce excessive punishment when encountering outliers. Since Huber Loss performs well in the small error area, it helps to improve the accuracy of the model. At the same time, it takes a linear penalty for large errors, reducing the impact of outliers on the model. Therefore, Huber Loss is used as the loss function in the present invention.

[0061] It should be further explained that, in the specific implementation process, the environmental data of the current acquisition period is input into the feature database for feature matching, and the process of selecting to compensate and adjust the target pressure sensor based on the deviation output model according to the feature matching result, or selecting to perform a flexible compensation adjustment operation on the target pressure sensor includes:

[0062] Input the environmental data of the target pressure sensor in the current acquisition period into the feature database for feature matching, obtain the similarity between the environmental data of the current acquisition period and each environmental feature set, and compare the similarity corresponding to each environmental feature set with a preset similarity threshold;

[0063] The calculation formula for obtaining the similarity between the environmental data of the current collection period and each environmental feature set is:

[0064] ;

[0065] in, Indicates similarity, Indicates the value of environmental parameter type i at time t in the environmental data of the current collection cycle, represents the value of environmental parameter type i at the tth moment in the environmental feature set, N1 represents the total number of moments in the collection cycle, and N2 represents the total number of environmental parameter types;

[0066] If there is an environmental feature set whose similarity is greater than the similarity threshold, compare the environmental data of the current acquisition period with the judgment data, obtain the numerical deviation value of each type of environmental parameter, input the numerical deviation value of each type of environmental parameter into the deviation output model, obtain the pressure deviation value of the target pressure sensor in the current acquisition period according to the deviation output model, perform data format preprocessing on the output value of the target pressure sensor in the current acquisition period, obtain the pressure value, and compensate and adjust the pressure value according to the pressure deviation value;

[0067] If there is no environmental feature set whose similarity is greater than the similarity threshold, a flexible compensation adjustment operation is performed on the target pressure sensor.

[0068] It should be further explained that, in a specific implementation process, the process of performing flexible compensation adjustment operation on the target pressure sensor includes:

[0069] Construct a fuzzy control model, the input of the fuzzy control model is the numerical deviation value of each type of environmental parameter, and the output is the pressure deviation value. According to the numerical deviation values ​​of each type of environmental parameter in different environmental feature sets in the feature database and the pressure deviation values ​​corresponding to the different environmental feature sets, construct a fuzzy adaptive rule control table of the fuzzy control model, set the membership threshold, membership function and continuous interval of pressure deviation value of the fuzzy control model, and divide the continuous interval of pressure deviation value into several fuzzy subsets;

[0070] The numerical deviation values ​​of various types of environmental parameters in the current collection cycle are input into the fuzzy control model. The fuzzy control model uses the numerical deviation values ​​of various types of environmental parameters as evaluation indicators, and obtains the weight matrix of the evaluation indicators. The membership matrix of the evaluation indicators for each fuzzy subset is obtained through fuzzy comprehensive evaluation.

[0071] The pressure deviation value of the evaluation index is obtained according to the membership matrix and the weight matrix, and the pressure value is compensated and adjusted according to the pressure deviation value.

[0072] It should be further explained that, in the specific implementation process, the process of obtaining the pressure deviation value of the evaluation index according to the membership matrix and the weight matrix includes:

[0073] The weight matrix and the membership matrix of the evaluation index are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation index, the membership of the evaluation index for different pressure deviation values ​​is obtained according to the fuzzy comprehensive evaluation matrix, the pressure deviation value with the highest membership corresponding to the evaluation index is screened out, and the pressure deviation value with the highest membership corresponding to the evaluation index is used as the pressure deviation value of the evaluation index;

[0074] Wherein, the formula is:

[0075] ;

[0076] in, is the fuzzy comprehensive evaluation matrix of the evaluation index, is the weight matrix of the evaluation index, is the membership matrix, " represents the multiplication of the weight matrix of the evaluation index and the elements at the corresponding positions of the membership matrix, It is a weighted parameter for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0077] It should be further explained that, in the specific implementation process, the process of obtaining the weight matrix of the evaluation index includes:

[0078] The pressure deviation values ​​corresponding to several environmental feature sets are obtained from the feature database as sample data, the numerical deviation values ​​of each type of environmental parameters in each environmental feature set in the sample data are obtained, the numerical deviation values ​​of each type of environmental parameters in each environmental feature set in the sample data are used as input parameters, the pressure deviation values ​​corresponding to each environmental feature set in the sample data are used as output values, the Sobol index is used to perform main effect sensitivity analysis, the main effect Sobol index of each type of environmental parameter is obtained, the weight coefficient of each type of environmental parameter is set according to the main effect Sobol index of each type of environmental parameter, and the weight matrix of the evaluation index is obtained according to the weight coefficient of each type of environmental parameter.

[0079] It should be further explained that, in the specific implementation process, the process of using the Sobol index for main effect sensitivity analysis includes:

[0080] Calculate the total variance V(Y) of the output values;

[0081] ;

[0082] in, is the mean of the output values, is the mean of the squares of the output values;

[0083] Calculate each input parameter The main effect variance ;

[0084] ;

[0085] in, is given The conditional expectation of the output value under the condition of yes The probability density function of

[0086] Calculate each input parameter Sobol index of main effects ;

[0087] ;

[0088] Indicates input parameters The contribution to the output variance.

[0089] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A pressure sensor calibration method based on environmental interference compensation, characterized in that: The following steps are involved: Step s1: Collect the output value of the target pressure sensor and environmental data, mark the collection time, and set the collection cycle; Step s2: Statistically analyze the environmental data of the target sensor in each historical collection period to obtain key values ​​of various types of environmental parameters; Step s3: construct a feature database, conduct simulation experiments based on the environmental feature set in the feature database, and obtain pressure deviation values ​​under different environmental feature set conditions; Step s4: construct a deviation output model, input the environmental data of the current acquisition period into the feature database for feature matching, and select compensation adjustment of the target pressure sensor based on the deviation output model according to the feature matching result, or select flexible compensation adjustment operation for the target pressure sensor; The process of statistically analyzing the environmental data of the target sensor in each historical collection period and obtaining the key values ​​of various types of environmental parameters includes: Acquire the environmental data of the target sensor in each historical collection cycle, extract the numerical time series sequence corresponding to each type of environmental parameter in the environmental data, perform statistical analysis on the numerical time series sequence corresponding to each type of environmental parameter, obtain the occurrence frequency of different numerical values ​​of each type of environmental parameter, compare the occurrence frequency of different numerical values ​​with a preset occurrence frequency threshold, screen out the numerical values ​​of each type of environmental parameter whose occurrence frequency is greater than the occurrence frequency threshold, and mark the numerical values ​​as key numerical values; The process of building a feature database includes: Construct a feature database, randomly select a key value from the key values ​​of each type of environmental parameter to combine, generate an environmental feature set, repeat the above steps until the number of environmental feature sets is equal to the preset upper limit of the simulation value, and store the environmental feature set in the feature database; Based on the environmental feature set in the feature database, a simulation experiment is conducted to obtain the pressure deviation value under different environmental feature set conditions. The process includes: Conduct simulation experiments on the target pressure sensor according to each environmental feature set to obtain the original output value of the target pressure sensor under the conditions of different environmental feature sets, and use a standard pressure source to obtain the standard pressure value corresponding to the original output value under the conditions of different environmental feature sets; The original output values ​​under different environmental parameter conditions are preprocessed in data format to obtain the original pressure value, and the original pressure value is compared with the standard pressure value corresponding to the original output value to obtain the pressure deviation value under different environmental feature set conditions; The process of building a bias output model includes: Obtaining a standard environmental feature set of a target pressure sensor, marking the standard environmental feature set as judgment data, comparing different environmental feature sets with the judgment data, and obtaining numerical deviation values ​​of various types of environmental parameters in different environmental feature sets; Construct a deviation output model based on a multivariate regression algorithm, use the numerical deviation values ​​of each type of environmental parameters in different environmental feature sets and the pressure deviation values ​​corresponding to different environmental feature sets as training sets and test sets, input the training set into the deviation output model for training until the loss function training is stable, save the model parameters, test the deviation output model through the test set until it meets the preset requirements, and obtain the deviation output model; The process of inputting the environmental data of the current acquisition period into the feature database for feature matching, and selecting to perform compensation adjustment on the target pressure sensor based on the deviation output model according to the feature matching result, or selecting to perform flexible compensation adjustment operation on the target pressure sensor includes: Input the environmental data of the target pressure sensor in the current acquisition period into the feature database for feature matching, obtain the similarity between the environmental data of the current acquisition period and each environmental feature set, and compare the similarity corresponding to each environmental feature set with a preset similarity threshold; The calculation formula for obtaining the similarity between the environmental data of the current collection period and each environmental feature set is: ; in, Indicates similarity, Indicates the value of environmental parameter type i at time t in the environmental data of the current collection cycle, represents the value of environmental parameter type i at the tth moment in the environmental feature set, N1 represents the total number of moments in the collection cycle, and N2 represents the total number of environmental parameter types; If there is an environmental feature set whose similarity is greater than the similarity threshold, compare the environmental data of the current acquisition period with the judgment data, obtain the numerical deviation value of each type of environmental parameter, input the numerical deviation value of each type of environmental parameter into the deviation output model, obtain the pressure deviation value of the target pressure sensor in the current acquisition period according to the deviation output model, perform data format preprocessing on the output value of the target pressure sensor in the current acquisition period, obtain the pressure value, and compensate and adjust the pressure value according to the pressure deviation value; If there is no environmental feature set whose similarity is greater than the similarity threshold, a flexible compensation adjustment operation is performed on the target pressure sensor.

2. A pressure sensor calibration method based on environmental disturbance compensation according to claim 1, characterized in that: The process of performing a flexible compensation adjustment operation on the target pressure sensor includes: Construct a fuzzy control model, the input of the fuzzy control model is the numerical deviation value of each type of environmental parameter, and the output is the pressure deviation value. According to the numerical deviation values ​​of each type of environmental parameter in different environmental feature sets in the feature database and the pressure deviation values ​​corresponding to the different environmental feature sets, construct a fuzzy adaptive rule control table of the fuzzy control model, set the membership threshold, membership function and continuous interval of pressure deviation value of the fuzzy control model, and divide the continuous interval of pressure deviation value into several fuzzy subsets; The numerical deviation values ​​of various types of environmental parameters in the current collection cycle are input into the fuzzy control model. The fuzzy control model uses the numerical deviation values ​​of various types of environmental parameters as evaluation indicators, and obtains the weight matrix of the evaluation indicators. The membership matrix of the evaluation indicators for each fuzzy subset is obtained through fuzzy comprehensive evaluation. The pressure deviation value of the evaluation index is obtained according to the membership matrix and the weight matrix, and the pressure value is compensated and adjusted according to the pressure deviation value.

3. A pressure sensor calibration method based on environmental disturbance compensation according to claim 2, characterized in that: The process of obtaining the weight matrix of the evaluation index includes: Obtain pressure deviation values ​​corresponding to several environmental feature sets from the feature database as sample data, obtain numerical deviation values ​​of various types of environmental parameters in various environmental feature sets in the sample data, use numerical deviation values ​​of various types of environmental parameters in various environmental feature sets in the sample data as input parameters, use pressure deviation values ​​corresponding to various environmental feature sets in the sample data as output values, use Sobol index to perform main effect sensitivity analysis, obtain main effect Sobol index of various types of environmental parameters, set weight coefficients of various types of environmental parameters according to main effect Sobol index of various types of environmental parameters, and obtain weight matrix of evaluation index according to weight coefficients of various types of environmental parameters; The process of using the Sobol index for main effect sensitivity analysis includes: Calculate the total variance V(Y) of the output values; ； in, is the mean of the output values, is the mean of the squares of the output values; Calculate each input parameter The main effect variance ; ； in, is given The conditional expectation of the output value under the condition of yes The probability density function of Calculate each input parameter Sobol index of main effects ; ； Indicates input parameters The contribution to the output variance.

Images