A pressure sensor temperature compensation system

Abstract

The present invention relates to the technical field of sensors, and specifically to a temperature compensation system for a pressure sensor, which includes a data acquisition unit, a judgment trigger unit, a temperature compensation unit, an optimization analysis unit, and a result output unit. In the present invention, the data received by the data acquisition unit can be received through the data receiving module, and then the threshold judgment module is used to assist in judging whether temperature compensation is required, improving the accuracy of temperature compensation triggering. The model constructed by the model construction module can combine the least squares method and the interpolation method to process the temperature drift of the pressure sensor. The model training module and the combination optimization module can assist in the subsequent optimization of the model, thereby reducing the influence of extreme outliers on the sensor and avoiding the ill-conditioned problem caused by too many fitting times, ensuring the initial compensation effect on the sensor. The compensation analysis module is used to analyze the compensation result, and secondary optimization processing can be carried out after the analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of sensors, and particularly to a temperature compensation system for a pressure sensor. Background Art

[0002] The pressure sensor is one of the most widely used sensors, usually composed of a pressure-sensitive element and a signal processing unit, and is now widely used in various industries. Since the pressure detection and output correction of the pressure sensor rely on resistance, and temperature is a key factor affecting the resistance value, the temperature compensation technology is crucial for the pressure sensor.

[0003] For example, "A Temperature Compensation System and Method for a Pressure Sensor" provided by Chinese Patent Publication No. CN105628266A includes a pressure sensor to be compensated and a transmitter; the pressure sensor to be compensated is a constant-current source-driven silicon piezoresistive pressure sensor; the transmitter includes a constant-current source, an analog compensation module, an analog-to-digital conversion module, and a digital compensation module; the constant-current source is used to provide a driving power supply for the pressure sensor to be compensated; the analog compensation module is used to convert the output voltage of the pressure sensor to be compensated so that the relationship between the converted output voltage and the input voltage of the pressure sensor to be compensated can be fitted by a quadratic polynomial curve under a certain pressure; the analog-to-digital conversion module is used to collect the input voltage and the converted output voltage corresponding to each calibration temperature point at each calibration pressure according to three preset calibration temperature points and two calibration pressures; it is also used to collect the input voltage and the converted output voltage corresponding to the measured pressure; among them, the three calibration temperature points are within the working temperature range of the pressure sensor to be compensated, and the two calibration pressures are within the measurement range of the pressure sensor to be compensated; the digital compensation module is used to solve the quadratic polynomial curve equation representing the relationship between the converted output voltage and the input voltage at each calibration pressure according to the input voltage and the converted output voltage corresponding to each calibration temperature point at each calibration pressure; it is also used to solve the value of the measured pressure according to the input voltage and the converted output voltage corresponding to the measured pressure, the quadratic polynomial curve equation, and the calibration pressure.

[0004] In the above patent, although the problem of low measurement accuracy of the pressure sensor is solved by three calibration temperature points, when using the quadratic polynomial calculation, the least squares method needs to be used to fit the quadratic polynomial curve multiple times, and the curve is easily affected by extreme abnormal points. Moreover, when the fitting times are too high and the epoch interval in data processing is small, ill-conditioned problems are likely to occur, affecting the overall compensation effect.

[0005] In view of this, the present application intends to propose a temperature compensation system for a pressure sensor. Summary of the Invention

[0006] The object of the present invention is to provide a temperature compensation system for a pressure sensor. The data acquisition module can receive the data acquired by the data acquisition unit, and then the threshold judgment module can assist in judging whether temperature compensation is required, improving the accuracy of temperature compensation triggering. The model constructed by the model construction module can combine the least squares method and the interpolation method to process the temperature drift of the pressure sensor. The model training module and the combined optimization module can assist in the subsequent optimization of the model, thereby reducing the influence of the least squares curve by extreme outliers and avoiding the ill-conditioned problem caused by excessive fitting times, ensuring the initial compensation effect on the sensor. The compensation verification module can preliminarily verify the compensation result, thereby ensuring the integrity of the compensation and avoiding compensation omission of data. The compensation analysis module analyzes the compensation result, and after analysis, secondary optimization processing can be performed. After secondary optimization, the pressure sensor can be further optimized. By using zero position compensation and sensitivity compensation, the influence caused by individual differences of a single sensor can be reduced, thereby ensuring the overall use effect of the sensor and reducing the interference of temperature on the use of the sensor to solve the above problems.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A temperature compensation system for a pressure sensor, comprising a data acquisition unit, a judgment trigger unit, a temperature compensation unit, an optimization analysis unit, and a result output unit;

[0009] The data acquisition unit is used to acquire relevant data on the pressure and temperature changes of the current pressure sensor and the medium;

[0010] The judgment trigger unit is used to receive the signal sent by the data acquisition unit and judge whether the current pressure and temperature require temperature compensation;

[0011] The temperature compensation unit is used to construct a temperature compensation model matching the current pressure sensor, and complete the temperature compensation of the sensor through the training and optimization of the model;

[0012] The optimization analysis unit is used to analyze the compensation result and perform targeted optimization on the data after analysis;

[0013] The result output unit is used to output the final result processed by the optimization analysis unit.

[0014] Preferably, the data acquisition unit includes a pressure data collection module, a temperature data collection module, and a data processing module;

[0015] The pressure data collection module is used to acquire the pressure value applied by the outside world to the current pressure sensor and store the pressure value as a data set;

[0016] The temperature data collection module is used to detect the temperature change of the medium, input the detected temperature into the chip temperature compensation port, and store and backup the obtained change values.

[0017] The data processing module is used to receive the electrical signals sent by the pressure data collection module and the temperature data collection module, preliminarily screen and denoise the signals, and then input the signals into the judgment trigger unit after processing.

[0018] Preferably, the judgment trigger unit includes a data receiving module, a threshold judgment module, and a trigger warning module.

[0019] The data receiving module is used to receive the signals sent by the data acquisition unit, classify the data according to pressure and temperature, and then store them in the database respectively.

[0020] The threshold judgment module is used to calculate the temperature sensitivity coefficient and temperature additional error of the current sensor respectively, compare the calculated values with the set thresholds. If the calculated values do not exceed the thresholds, there is no need to perform temperature compensation on the sensor. If the calculated values exceed the thresholds, a signal is sent to the trigger warning module.

[0021] The trigger warning module is used to receive the signals from the threshold judgment module, trigger the temperature compensation unit to work by using the trigger warning module, and send an alarm signal in the system at the same time.

[0022] Preferably, the temperature compensation unit includes a model construction module, a model training module, a combination optimization module, and a simulation application module.

[0023] The model construction module is used to construct a temperature compensation model and test data that match the current pressure sensor.

[0024] The model training module is used to synchronize the test data, randomly select the data in the dataset as the test data, calculate and process the test data by using the model, and complete the training of the model through multiple iterations.

[0025] The combination optimization module uses the model training module to locate the best particle, obtain the parameter set corresponding to the particle, store and backup the parameters, and use the assignment of the best particle as the initial weight and threshold of the model.

[0026] The simulation application module calls the simulation software to perform a simulation experiment on the optimized model. After the experiment, it verifies the calculation results. If there are errors in the calculation results, multiple experiments are carried out for correction. If the calculation results are correct, the model is loaded into the system for use.

[0027] Preferably, the optimization analysis unit includes a compensation verification module, a compensation analysis module, and a secondary optimization module.

[0028] The compensation verification module is used to calculate the zero drift and sensitivity error of the sensor, compare the data changes before and after compensation according to the calculation results, and quickly verify whether the temperature compensation unit has completed the compensation;

[0029] The compensation analysis module is used to compare and analyze the calculation results of the zero drift and sensitivity error in the compensation verification module with the preset zero drift and sensitivity error ranges, and determine whether the current pressure sensor needs to be optimized again;

[0030] The secondary optimization module is used to input the current detection results into the temperature compensation unit again to optimize the temperature compensation of the pressure sensor again.

[0031] Preferably, in the threshold judgment module, the relative value calculation formulas for calculating the temperature sensitivity coefficient and the temperature additional error are as follows:

[0032]

[0033]

[0034] In the formula, ΔT is the operating temperature change range of the sensor, L(FS) is the full scale of work, and Δx m is the maximum value of the output value corresponding to a certain input calibration value shifting with temperature when the temperature changes by ΔT.

[0035] Preferably, when constructing the model, the following steps are further included:

[0036] S1. Use calibration tests to detect relevant values, set different temperature measurement conditions at a step of 10°C, set pressure measurement points, measure the sampling points three times and take the average value, and calculate by combining the current ambient temperature with the function coefficient to obtain the output voltage corresponding to the calibrated pressure under this temperature condition, as the output voltage data set;

[0037] S2. Divide the temperature range by the cubic spline interpolation method, select 3 kinds of calibrated voltage outputs with the smallest difference degree from the output voltage data set corresponding to the output voltage of the pressure to be measured, which is the voltage data set, and calculate the corresponding calibrated pressure by combining the voltage data set coefficient;

[0038] S3. Use Newton interpolation method to establish the functional relationship between the output pressure F and the calibrated temperature vector t, and substitute the output voltage into the function to obtain the final pressure to be measured.

[0039] Preferably, the expression of the temperature compensation model is as follows:

[0040] F = f(p, t);

[0041] p' i = e' i F 3 + g'i F 2 +h' i F + q' i ;

[0042] Wherein, e' i , g' i , h' i , q' i are the function coefficients of each order, i is the row number, p is the pressure, and t is the temperature.

[0043] Preferably, in the model training module, when training the model, the measured data of the pressure sensor is initially solved to obtain initial parameters, the error is calculated, and when the error reaches the minimum, the optimal particle corresponding to the optimal function is found through iteration.

[0044] Preferably, in the compensation analysis module, the change range of zero drift is (-1.75, 1.89), and the sensitivity error range is (0, 0.3).

[0045] Advantages of the present invention:

[0046] 1. In the present invention, the data acquisition unit can receive the data obtained by the data acquisition unit through the data receiving module, and then the threshold judgment module can assist in judging whether temperature compensation is required, improving the accuracy of temperature compensation triggering. The model constructed by the model construction module can combine the least squares method and the interpolation method to process the temperature drift of the pressure sensor. The model training module and the combined optimization module can assist in the subsequent optimization of the model, thereby reducing the influence of the least squares curve by extreme outliers and avoiding the ill-conditioned problem caused by too many fitting times, ensuring the preliminary compensation effect on the sensor.

[0047] 2. In the present invention, the compensation verification module can preliminarily verify the compensation result, and judge whether the compensation is completed through the data change before and after compensation, thereby ensuring the integrity of the compensation and avoiding the interference of the subsequent sensor caused by the omission of data compensation. The compensation analysis module analyzes the compensation result, and after analysis, secondary optimization processing can be carried out. When performing secondary optimization, zero position compensation and sensitivity compensation are used to reduce the influence caused by the individual differences of single sensors, thereby ensuring the overall use effect of the sensors and reducing the interference of temperature on the use of the sensors. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is the system block diagram of a temperature compensation system for a pressure sensor according to the present invention;

[0049] Figure 2 is the system diagram of the data acquisition unit of a temperature compensation system for a pressure sensor according to the present invention;

[0050] Figure 3This is a system diagram of the judgment trigger unit of a temperature compensation system for a pressure sensor according to the present invention.

[0051] In the figure: 1. Data acquisition unit; 11. Pressure data collection module; 12. Temperature data collection module; 13. Data processing module; 2. Judgment trigger unit; 21. Data reception module; 22. Threshold judgment module; 23. Trigger warning module; 3. Temperature compensation unit; 31. Model construction module; 32. Model training module; 33. Combined optimization module; 34. Simulation application module; 4. Optimization analysis unit; 41. Compensation verification module; 42. Compensation analysis module; 43. Secondary optimization module; 5. Result output unit. Detailed implementation manners

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] Please refer to Figures 1-3 , this design proposes an implementation manner, a temperature compensation system for a pressure sensor, including a data acquisition unit 1, a judgment trigger unit 2, a temperature compensation unit 3, an optimization analysis unit 4, and a result output unit 5;

[0055] The data acquisition unit 1 is used to acquire relevant data on the pressure and temperature changes of the current pressure sensor and the medium;

[0056] The judgment trigger unit 2 is used to receive the signal sent by the data acquisition unit 1 and judge whether temperature compensation is required for the current pressure and temperature;

[0057] The temperature compensation unit 3 is used to construct a temperature compensation model matching the current pressure sensor and complete the temperature compensation of the sensor through training and optimization of the model;

[0058] The optimization analysis unit 4 is used to analyze the compensation result and perform targeted optimization on the data through the analysis;

[0059] The result output unit 5 is used to output the final result processed by the optimization analysis unit 4.

[0060] The data acquisition unit 1 includes a pressure data collection module 11, a temperature data collection module 12, and a data processing module 13. The pressure data collection module 11 is used to obtain the pressure value applied to the current pressure sensor from the outside and store the pressure value as a data set. The temperature data collection module 12 is used to detect the temperature change of the medium, input the detected temperature into the chip temperature compensation port, and store and back up the obtained change value. The data processing module 13 is used to receive the electrical signals sent by the pressure data collection module 11 and the temperature data collection module 12, perform preliminary screening and noise reduction processing on the signals, and then input the processed signals into the judgment trigger unit 2.

[0061] The judgment trigger unit 2 includes a data receiving module 21, a threshold judgment module 22, and a trigger warning module 23. The data receiving module 21 is used to receive the signals sent by the data acquisition unit 1, classify the data according to pressure and temperature, and then store them in the database respectively. The threshold judgment module 22 is used to calculate the temperature sensitivity coefficient and temperature additional error of the current sensor respectively, compare the calculated values with the set thresholds. If the calculated values do not exceed the thresholds, there is no need to perform temperature compensation on the sensor. If the calculated values exceed the thresholds, a signal is sent to the trigger warning module 23.

[0062] In the threshold judgment module 22, the relative value calculation formulas for calculating the temperature sensitivity coefficient and temperature additional error are as follows:

[0063]

[0064]

[0065] In the formula, ΔT is the working temperature change range of the sensor, L(FS) is the full workload range, and Δx m is the maximum value of the output value corresponding to a certain input calibration value shifting with temperature when the temperature changes by ΔT.

[0066] The trigger warning module 23 is used to receive the signals from the threshold judgment module 22, trigger the temperature compensation unit 3 to work by using the trigger warning module 23, and send an alarm signal in the system at the same time.

[0067] In this embodiment, through the pressure data collection module 11 and the temperature data collection module 12, the pressure and temperature data of the current pressure sensor can be collected respectively, avoiding data chaos and loss caused by mutual mixing of data. Input the detected temperature into the chip temperature compensation port. When calculating chip compensation, this temperature value can be input for comprehensive calculation twice, which can eliminate the temperature difference between the backend conditioning chip and the front-end chip, and then reduce the data difference between individual pressure sensors. The data processing module 13 can remove the white noise inside the signals, thereby reducing the interference of noise on the effective data during data transmission and reception, and ensuring the accurate transmission of effective information.

[0068] After receiving the data, the data receiving module 21 traverses and checks it to promptly find out if there is any data leakage. In case of data leakage, it sends a signal to the control terminal and uses the data processing module 13 to resend the data, thus reducing the interference caused by data omission during subsequent model establishment and testing. The threshold judgment module 22 can quickly calculate the temperature sensitivity coefficient and temperature additional error of the sensor, compare the calculated value with the threshold. When the calculated value exceeds the threshold, it is determined that the current sensor needs temperature compensation. Different thresholds can be set to adjust the moment when the sensor needs temperature compensation, thereby improving the applicability of the entire system. Triggering the warning module 23 can control the use of the temperature compensation unit 3. Through the warning signal sent by triggering the warning module 23, the system can capture the working state of the current sensor, track and mark the sensor, facilitating subsequent query of relevant data after the system performs temperature compensation;

[0069] The result output unit 5 first stores the results processed by the optimization analysis unit 4 in the local database, then backs up the data in the cloud database, and then generates a report on the relevant data for users or managers to log in to the system.

[0070] Embodiment 2

[0071] Please refer to Figures 1-3 , this design proposes an implementation method, a pressure sensor temperature compensation system. The temperature compensation unit 3 includes a model construction module 31, a model training module 32, a combined optimization module 33, and a simulation application module 34;

[0072] The model construction module 31 is used to construct a temperature compensation model and test data that match the current pressure sensor; the model training module 32 is used to synchronize the test data, randomly select data from the data set as test data, use the model to perform calculation processing on the test data, and complete the training of the model through multiple iterations. When training the model, perform an initial solution on the data measured by the pressure sensor to obtain initial parameters, calculate the error. When the error reaches the minimum, find the best particle corresponding to the optimal function through iteration;

[0073] The combined optimization module 33 uses the model training module 32 to locate the best particle, obtains the parameter set corresponding to the particle, stores and backs up the parameters, and uses the assignment of the best particle as the initial weight and threshold of the model; the simulation application module 34 calls the simulation software to perform a simulation experiment on the optimized model. After the experiment is completed, verify the calculation result. If there is an error in the calculation result, perform multiple experiments for correction. If the calculation result is correct, load the model into the system for use.

[0074] When constructing the model, the following steps are further included:

[0075] Step 1: Detect relevant values using calibration tests. Set different temperature measurement conditions with a step of 10°C, set pressure measurement points, measure the sampling points three times and take the average value. Combine the current ambient temperature with the function coefficients to calculate the output voltage corresponding to the calibrated pressure under this temperature condition, and use it as the output voltage dataset.

[0076] Step 2: Divide the temperature range by the cubic spline interpolation method. Select the 3 calibrated voltage outputs with the smallest difference from the output voltage corresponding to the pressure to be measured in the output voltage dataset, which is the voltage dataset. Combine the voltage dataset coefficients to calculate the corresponding calibrated pressure.

[0077] Step 3: Use Newton interpolation method to establish the functional relationship between the output pressure F and the calibrated temperature vector t, and substitute the output voltage into the function to obtain the final pressure to be measured.

[0078] The expression of the temperature compensation model is as follows:

[0079] F = f(p, t);

[0080] p' i = e' i F 3 + g' i F 2 + h' i F + q' i ;

[0081] In the formula, e' i , g' i , h' i , q' i are the function coefficients of each order, i is the number of rows, p is the pressure, and t is the temperature.

[0082] In this embodiment, the model construction module 31 can quickly construct the temperature compensation model and test data. Through multiple iterations and calculations, the training of the compensation model can be completed, making the model highly matched with the current pressure sensor. Using the initial parameters obtained from the initial solution of the measured data as the base values, and using them as the reference basis to assist in the calculation of errors, quickly find the moment with the smallest error, locate the current moment, and thus find the best particle through iterative processing. By obtaining the parameter set of the best particle and assigning values, the model is constrained, thereby realizing the optimization processing of the model and reducing the interference of abnormal points on the use of the model.

[0083] When constructing the model, the temperature interval is divided into 10℃, and the measurement conditions at different temperatures are set. The three average values ​​of the sampling points can reduce the impact of randomness on the measurement of the sampling points and ensure the accuracy of the sampling points. The output voltage under the current temperature conditions can be obtained through the sampling points. The voltage data set generated by the three calibration voltage outputs can reduce the difference between the pressure to be measured and the output voltage, thereby reducing the mutual interference between the two sets of data. The calibrated pressure value calculated by the above steps is relatively accurate. The Newton difference method can be used to establish the relationship between the output pressure and the calibration temperature vector, which can assist in the calculation of the pressure to be measured, provide a data basis for the subsequent model construction, and ensure the model's processing effect on the output pressure and temperature.

[0084] Embodiment 3

[0085] See also Figures 1-3 ,This design proposes an implementation method, a temperature compensation system for a pressure sensor, the optimization and analysis unit 4 includes a compensation verification module 41, a compensation analysis module 42 and a secondary optimization module 43;

[0086] The compensation verification module 41 is used to calculate the zero drift and sensitivity error of the sensor, and compare the data changes before and after compensation based on the calculation results to quickly verify whether the temperature compensation unit 3 has completed compensation;

[0087] The compensation analysis module 42 is used to compare and analyze the zero drift and sensitivity error calculation results in the compensation verification module 41 with the preset zero drift and sensitivity error ranges to determine whether the current pressure sensor needs to be optimized again;

[0088] The secondary optimization module 43 is used to input the current detection result into the temperature compensation unit 3 again to optimize the temperature compensation of the pressure sensor again.

[0089] In this embodiment, the compensation verification module 41 can calculate the compensation result of the pressure sensor, the zero drift and the sensitivity error of the sensor in sequence, and assist in verification based on the data difference before and after compensation to avoid compensation omissions in some data, thereby ensuring the overall compensation effect of the pressure sensor. The trigger adjustment of the secondary optimization can be defined by presetting the zero drift and sensitivity error range. When the zero drift and sensitivity compensation do not exceed the predetermined range, only one optimization is required. When the zero drift and sensitivity compensation exceed the predetermined range, the secondary optimization of the temperature compensation of the pressure sensor is triggered in time. It is judged that the zero drift at 25°C is 4.32 and the sensitivity error is 0.09; the zero drift at 35°C is 1.25 and the sensitivity error is 0; the zero drift at 45°C is -0.65 and the sensitivity error is 0. The secondary optimization module 43 can realize the call of the temperature compensation unit 3, so as to facilitate the use of the temperature compensation unit 3 to perform compensation optimization on the current pressure sensor again.

[0090] In the present invention, the pressure and temperature data of the sensor are acquired by the data acquisition unit 1, the data is preliminarily processed by the data processing module 13, and then the processed data is input into the judgment trigger unit 2, the threshold judgment module 22 can calculate the temperature sensitivity coefficient and the temperature additional error of the sensor, and the calculation result is used as the control value. After the data receiving module 21 receives the data, the control value in the threshold judgment module 22 is used to compare and judge with the threshold. When the threshold is not exceeded, the trigger warning module 23 does not work. When the threshold is exceeded, the trigger warning module 23 enables the system warning and sends a trigger signal to the temperature compensation unit 3. Since the model construction module 31 can complete the preliminary construction of the test data and the model, the basic formation of the model can be achieved. Through multiple iterative processing of the model training module 32, the training of the model is gradually completed. After training, the model can be optimized by using the combination optimization module 33. After optimization and verification, the model is loaded into the system for application through the simulation application module 34. The applied model can perform temperature compensation processing on the data of the current sensor;

[0091] The temperature compensation result enters the optimization analysis unit 4, and the zero drift and sensitivity error of the sensor are calculated by the compensation verification module 41. The data changes before and after compensation are compared according to the calculation results. If there is no obvious change in the data before and after compensation, the data changes under the next temperature condition are compared. If there is no obvious change in three consecutive sets of data, it is judged that compensation has not occurred, and the temperature compensation unit 3 is abnormal. The use of the temperature compensation unit 3 is interrupted, and a signal is sent to the safety officer or background maintenance personnel of the system. The staff debugs the temperature compensation unit 3 and relieves the abnormal state of the temperature compensation unit 3. If the data changes before and after compensation are obvious, it is judged that compensation has occurred. At this time, the current compensation result is analyzed by the compensation analysis module 42, and whether the current pressure sensor needs secondary optimization is determined according to whether the zero drift and sensitivity compensation exceed the predetermined range. When the zero drift and sensitivity compensation exceed the predetermined range, secondary optimization is required. At this time, the secondary optimization module 43 is used to input the data into the temperature compensation unit 3 for processing again. If the zero drift and sensitivity compensation do not exceed the predetermined range, no secondary optimization is required. At this time, the compensation result of the pressure sensor is sent to the result output unit 5, and a result report is generated in the result output unit 5. The result report can be viewed at any time.

[0092] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A pressure sensor temperature compensation system, characterized in that: It comprises a data acquisition unit (1), a judgment trigger unit (2), a temperature compensation unit (3), an optimization analysis unit (4) and a result output unit (5); The data acquisition unit (1) is used to acquire data related to the pressure and temperature changes of the current pressure sensor and the medium; The judgment trigger unit (2) is used to receive the signal sent by the data acquisition unit (1) to judge whether the current pressure and temperature need to be compensated; The temperature compensation unit (3) is used to construct a temperature compensation model that matches the current pressure sensor, and completes the temperature compensation of the sensor by training and optimizing the model; The temperature compensation unit (3) comprises a model building module (31), a model training module (32), a combination optimization module (33) and a simulation application module (34); The model building module (31) is used to build a temperature compensation model and test data matching the current pressure sensor; The model training module (32) is used to synchronize test data, randomly select data from the data set as test data, use the model to perform calculations on the test data, and complete model training through multiple iterations; The combination optimization module (33) uses the model training module (32) to locate the best particle, obtains a parameter set corresponding to the particle, stores and backs up the parameters, and uses the value assigned to the best particle as the initial weight and threshold of the model; The simulation application module (34) calls the simulation software to perform a simulation experiment on the optimized model, and verifies the calculation result after the experiment. If the calculation result is wrong, multiple experiments are performed to correct it. If the calculation result is correct, the model is loaded into the system for use; The optimization analysis unit (4) is used to analyze the compensation results and optimize the data in a targeted manner after the analysis; The result output unit (5) is used to output the final result after being processed by the optimization analysis unit (4).

2. A pressure sensor temperature compensation system according to claim 1, characterized in that: The data acquisition unit (1) comprises a pressure data collection module (11), a temperature data collection module (12) and a data processing module (13); The pressure data collection module (11) is used to obtain the pressure value applied by the outside world to the current pressure sensor, and store the pressure value as a data set; The temperature data collection module (12) is used to detect the temperature change of the medium, input the detected temperature into the chip temperature compensation port, and store and back up the acquired change value; The data processing module (13) is used to receive the electrical signals sent by the pressure data collection module (11) and the temperature data collection module (12), perform preliminary screening and noise reduction on the signals, and input the signals into the judgment trigger unit (2) after processing.

3. A pressure sensor temperature compensation system according to claim 1, characterized in that: The judgment trigger unit (2) comprises a data receiving module (21), a threshold judgment module (22) and a trigger warning module (23); The data receiving module (21) is used to receive the signal sent by the data acquisition unit (1), classify the data according to pressure and temperature, and then store them in a database respectively; The threshold judgment module (22) is used to calculate the temperature sensitivity coefficient and the temperature additional error of the current sensor respectively, and compare the calculated value with the set threshold value. If the calculated value does not exceed the threshold value, there is no need to perform temperature compensation on the sensor. If the calculated value exceeds the threshold value, a signal is sent to the trigger warning module (23); The trigger warning module (23) is used to receive a signal from the threshold judgment module (22), and to trigger the temperature compensation unit (3) to work by using the trigger warning module (23), while sending an alarm signal in the system.

4. A pressure sensor temperature compensation system according to claim 1, characterized in that: The optimization analysis unit (4) comprises a compensation verification module (41), a compensation analysis module (42) and a secondary optimization module (43); The compensation verification module (41) is used to calculate the zero drift and sensitivity error of the sensor, compare the data changes before and after compensation based on the calculation results, and quickly verify whether the temperature compensation unit (3) has completed compensation; The compensation analysis module (42) is used to compare and analyze the zero drift and sensitivity error calculation results in the compensation verification module (41) with the preset zero drift and sensitivity error ranges to determine whether the current pressure sensor needs to be optimized again; The secondary optimization module (43) is used to input the current detection result into the temperature compensation unit (3) again, and optimize the temperature compensation of the pressure sensor again.

5. A pressure sensor temperature compensation system according to claim 3, characterized in that: The relative value calculation formula for calculating the temperature sensitivity coefficient and the temperature additional error in the threshold judgment module (22) is as follows: ； ； In the formula, is the sensor operating temperature range, is the working range, When the temperature changes When a certain input calibration value corresponds to the maximum value of the output value shifted by temperature.

6. A pressure sensor temperature compensation system according to claim 1, characterized in that: When building the model, the following steps are also included: S1. Use calibration test to detect relevant values, set different temperature measurement conditions with a step of 10°C, set pressure measurement points, measure the sampling points three times and take the average value, calculate the current ambient temperature in combination with the function coefficient, and obtain the output voltage corresponding to the calibration pressure under the temperature condition as the output voltage data set; S2. Divide the temperature interval by cubic spline interpolation method, select three calibration voltage outputs with the smallest difference between the output voltage and the output voltage corresponding to the pressure to be measured in the output voltage data set, that is, the voltage data set, and calculate the corresponding calibration pressure in combination with the coefficient of the voltage data set; S3. Use Newton interpolation method to establish the functional relationship between the output pressure F and the calibration temperature vector t, and substitute the output voltage into the function to obtain the final pressure to be measured.

7. A pressure sensor temperature compensation system according to claim 1, characterized in that: The expression of the temperature compensation model is as follows: ； ； In the formula, , , , are the coefficients of each order function, i is the number of rows, p is the pressure, and t is the temperature.

8. A pressure sensor temperature compensation system according to claim 1, characterized in that: In the model training module (32), when training the model, the pressure sensor measurement data is initially solved to obtain initial parameters, and the error is calculated. When the error reaches the minimum, the best particle corresponding to the optimal function is found through iteration.

9. A pressure sensor temperature compensation system according to claim 4, characterized in that: In the compensation analysis module (42), the variation range of the zero point drift is (-1.75, 1.89), and the sensitivity error range is (0, 0.3).

Images