A piezoelectric thin film sensor algorithm calibration system and application thereof

Abstract

The application discloses a piezoelectric film sensor algorithm calibration system and application thereof, and particularly relates to the technical field of sensors, which comprises a pressure input unit, an output end of the pressure input unit is electrically connected with an input end of a sensor unit, an output end of the sensor unit is electrically connected with an input end of a pressure acquisition unit, an input end of the pressure acquisition unit is electrically connected with an output end of the pressure input unit, and an output end of the pressure input unit is electrically connected with an input end of a waveform processing unit. The application adopts the waveform processing unit to realize the oscillation caused by the dispersion effect in the shock wave waveform in the pressure input process, so that the waveform has obvious wave peaks, the impact force in the input process is more accurate, the influence of the chaotic waveform on the input pressure value measurement is reduced, the calibration algorithm is adopted to realize the rapid calibration process of the sensor itself, and the accuracy calibration confirmation when leaving the factory and the re-calibration after the sensor maintenance are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sensors, and more particularly, to a piezoelectric film sensor algorithm calibration system and application thereof. BACKGROUND

[0002] The piezoelectric film has unique characteristics, and as a dynamic strain sensor, is very suitable for application in monitoring life signals on the surface of human skin or implanted in the human body. Some film elements are sensitive enough to detect human pulse through a coat.

[0003] When a piezoelectric polyvinylidene fluoride (PVDF) polymer film (piezoelectric film) is stretched or bent, an electric signal (charge or voltage) is generated between the upper and lower electrode surfaces of the film, and is proportional to the deformation of stretching or bending. Generally, piezoelectric materials are sensitive to pressure, but for piezoelectric films, a small force applied in the longitudinal direction will generate a large stress in the transverse direction, and if the same force is applied to a large area of the film, the stress generated will be much smaller.

[0004] The sensor may have changes in parts or test use environment when it is shipped and after maintenance, resulting in a certain deviation in the data obtained by testing, and thus there is a significant difference between the test results and the actual results, so the sensor needs to be calibrated to ensure that the measurement results output by the sensor are the same as the data input by the test, and the temperature of the test environment may be different when the sensor is used, and the resistance of the internal components of the sensor changes due to the different temperatures, so the temperature has a great influence on the results obtained by the sensor in the test process, and thus a piezoelectric film sensor algorithm calibration system and application thereof are needed to solve the above problems. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, the present application provides a piezoelectric film sensor algorithm calibration system and application thereof, and the technical problems to be solved by the present application are that the sensor may have changes in parts or test environment when it is shipped and after maintenance, resulting in a certain deviation in the data obtained by testing, and thus there is a significant difference between the test results and the actual results, so the sensor needs to be calibrated to ensure that the measurement results output by the sensor are the same as the data input by the test, and the temperature of the test environment may be different when the sensor is used, and the resistance of the internal components of the sensor changes due to the different temperatures, so the temperature has a great influence on the results obtained by the sensor in the test process.

[0006] In order to achieve the above object, the present application provides the following technical scheme: a piezoelectric film sensor algorithm calibration system, comprising a pressure input unit, an output end of the pressure input unit being electrically connected with an input end of a sensor unit, an output end of the sensor unit being electrically connected with an input end of a pressure acquisition unit, the input end of the pressure acquisition unit being electrically connected with the output end of the pressure input unit, the output end of the pressure input unit being electrically connected with an input end of a waveform processing unit.

[0007] An output end of the waveform processing unit is electrically connected with an input end of the sensor unit, an output end of the sensor unit is electrically connected with an input end of a calibration unit, the input end of the calibration unit is electrically connected with an output end of the pressure acquisition unit, an input end of the sensor unit is electrically connected with an output end of a temperature control unit, an output end of the temperature control unit is electrically connected with an input end of a temperature acquisition unit.

[0008] An output end of the temperature acquisition unit is electrically connected with an input end of the calibration unit, an output end of the calibration unit is electrically connected with an input end of a compensation unit, an output end of the compensation unit is electrically connected with an input end of the calibration unit, an input end of the compensation unit is electrically connected with an output end of the temperature acquisition unit.

[0009] As a further scheme of the present application: the pressure acquisition unit comprises an input pressure detection unit and an output pressure detection unit;

[0010] The input pressure detection unit is used for detecting the pressure applied to the sensor unit during the pressure input process;

[0011] The output pressure detection unit is used for detecting the pressure at the sensor unit.

[0012] As a further scheme of the present application: the waveform processing unit adopts a waveform shaper, and the function of the waveform processing is to eliminate the oscillation caused by the dispersion effect in the shock wave waveform during the pressure output process, so that the waveform has a clear wave crest.

[0013] As a further scheme of the present application: the sensor unit comprises a data detection unit, a charge amplification unit and an indication unit, an output end of the data detection unit being electrically connected with an input end of the charge amplification unit, an output end of the charge amplification unit being electrically connected with an input end of the indication unit.

[0014] As a further scheme of the present application: the data detection unit is used for converting the received pressure into corresponding electrical signals when the sensor unit is impacted;

[0015] The charge amplification unit is used for amplifying the obtained electrical signals;

[0016] The indication unit is used for recording the collected input pressure data, output pressure data and electrical signal data.

[0017] As a further scheme of the present application: the calibration unit is used for primary sensor calibration according to obtained pressure data and electrical signal data, and secondary sensor calibration according to compensated electrical signal data and pressure data when temperature changes.

[0018] As a further scheme of the present application: the primary sensor calibration comprises the following steps:

[0019] receiving pressure detection data, sensor data and temperature detection data;

[0020] sensor calibration processing;

[0021] judging whether in room temperature state;

[0022] if in room temperature state, completing primary sensor calibration;

[0023] if not in room temperature state, transmitting data;

[0024] the sensor calibration process is:

[0025] piezoelectric coefficient d of sensor 33 is: d 33 = Q / F,

[0026] wherein Q is piezoelectric charge, F is output force, and U(t) is voltage signal after conversion and amplification by charge amplification unit, and the expression of charge Q obtained by backstepping voltage signal is:

[0027] Q = U(t)d1 / K

[0028] wherein d1 is sensitivity of charge amplification unit, K is output gain of charge amplification unit, and output force F is given by output signal ε:

[0029] F = EAε(t) = EAηU

[0030] wherein E is elastic modulus of test transmission rod, A is transmission rod area, and η is conversion coefficient between voltage and output signal;

[0031] the final piezoelectric coefficient is:

[0032] As a further scheme of the present application: the compensation unit is used for compensation processing on electrical signal data detected when not in room temperature state.

[0033] As a further scheme of the present application: the compensation processing comprises the following steps:

[0034] receiving data;

[0035] judging whether it is in room temperature state;

[0036] if not in room temperature state, establishing neural network compensation model;

[0037] carrying out data compensation;

[0038] sending data;

[0039] the received data includes data after initial sensor calibration and electric signal data measured in non-room temperature environment;

[0040] the data compensation process is:

[0041] the compensation characteristic function is y=f -1 (M P ,M t )=x, y is output after compensation,

[0042] fuzzy neural network is adopted for calculation, in which membership function in each variable is μ Ai (x i ), and the membership function adopts Gaussian function:

[0043] in which c and σ are respectively center and width of the membership function,

[0044] I represents input information, O represents output information, and input-output relationship of each layer is:

[0045]

[0046]

[0047] ω is connection weight between two formulas,

[0048] let expected output be y d (t)

[0049] output of the network is y(t)

[0050] network learning makes minimum, and network weight is adjusted according to the following formula:

[0051]

[0052] adjustment formula of the Gaussian function parameter is:

[0053]

[0054] in the formula, η is learning rate, and β is momentum parameter.

[0055] learning rule of the neural network is obtained as:

[0056]

[0057] where the training is until |y i -y di |<Eend, the approximation of f -1 (v) is complete.

[0058] The present application has the advantage that:

[0059] 1、The present application adopts the waveform processing unit, realizes the oscillation caused by the dispersion effect in the shock wave waveform in the pressure input process, makes the waveform have obvious wave crest, makes the impact force in the input process more accurate, reduces the influence of the chaotic waveform on the input pressure value measurement, adopts the calibration algorithm, realizes the rapid calibration process of the sensor itself, realizes the accuracy calibration confirmation at the factory and the re-calibration after the sensor maintenance;

[0060] 2、The present application adopts the temperature control unit and the temperature acquisition unit, the temperature control unit is used for controlling the temperature of the calibration environment, realizes the calibration process at different temperatures, the temperature acquisition unit can acquire the temperature at the sensor, the data detected by the sensor at different temperatures and the data detected by the sensor at normal temperature are different, through the change of the calibration process data of the same pressure at different temperatures, the neural network model is used to compensate the zero point data in each measurement value, realizes that the results measured at various temperature conditions are more accurate and close to the actual value, makes the overall calibration process more ideal, and ensures that the sensor remains accurate under the influence of different temperatures, and through the neural network model, a high-precision simulation model can be obtained, the whole calculation process is stable, and the convergence speed is fast, so that the use range of the calibrated sensor is wider, and the present application is suitable for promotion. BRIEF DESCRIPTION OF DRAWINGS

[0061] Figure 1 It is a system connection schematic diagram of the present application;

[0062] Figure 2 It is a sensor unit internal structure schematic diagram of the present application;

[0063] Figure 3 It is a pressure acquisition unit internal schematic diagram of the present application;

[0064] Figure 4 It is a data compensation process schematic diagram of the present application;

[0065] Figure 5 It is a sensor calibration process schematic diagram of the present application;

[0066] In the diagram: 1. Pressure input unit; 2. Waveform processing unit; 3. Sensor unit; 31. Data detection unit; 32. Charge amplification unit; 33. Oscilloscope unit; 4. Pressure acquisition unit; 41. Input pressure detection unit; 42. Output pressure detection unit; 5. Temperature control unit; 6. Temperature acquisition unit; 7. Calibration unit; 8. Compensation unit. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Example 1:

[0069] like Figures 1-5 As shown, the present invention provides a piezoelectric thin film sensor algorithm calibration system, including a pressure input unit 1, the output end of the pressure input unit 1 being electrically connected to the input end of the sensor unit 3, the output end of the sensor unit 3 being electrically connected to the input end of the pressure acquisition unit 4, the input end of the pressure acquisition unit 4 being electrically connected to the output end of the pressure input unit 1, and the output end of the pressure input unit 1 being electrically connected to the input end of the waveform processing unit 2.

[0070] The output terminal of waveform processing unit 2 is electrically connected to the input terminal of sensor unit 3. The output terminal of sensor unit 3 is electrically connected to the input terminal of calibration unit 7. The input terminal of calibration unit 7 is electrically connected to the output terminal of pressure acquisition unit 4. The input terminal of sensor unit 3 is electrically connected to the output terminal of temperature control unit 5. The output terminal of temperature control unit 5 is electrically connected to the input terminal of temperature acquisition unit 6.

[0071] The output terminal of the temperature acquisition unit 6 is electrically connected to the input terminal of the calibration unit 7. The output terminal of the calibration unit 7 is electrically connected to the input terminal of the compensation unit 8. The output terminal of the compensation unit 8 is electrically connected to the input terminal of the calibration unit 7. The input terminal of the compensation unit 8 is electrically connected to the output terminal of the temperature acquisition unit 6.

[0072] As shown in the figure, the pressure acquisition unit 4 includes an input pressure detection unit 41 and an output pressure detection unit 42.

[0073] The input pressure detection unit 41 is used to detect the pressure applied to the sensor unit 3 during the pressure input process;

[0074] The output pressure detection unit 42 is used to detect the pressure at the sensor unit 3.

[0075] As shown in the figure, the waveform processing unit 2 adopts a waveform shaper, and the function of the waveform processing is to eliminate the oscillation caused by the dispersion effect in the shock wave waveform in the pressure output process, so that the waveform has a clear wave crest.

[0076] As shown in the figure, the sensor unit 3 includes a data detection unit 31, a charge amplification unit 32 and a display unit 33, the output end of the data detection unit 31 is electrically connected with the input end of the charge amplification unit 32, and the output end of the charge amplification unit 32 is electrically connected with the input end of the display unit 33.

[0077] As shown in the figure, the data detection unit 31 is used to convert the received pressure into corresponding electrical signals when impacted;

[0078] The charge amplification unit 32 is used to amplify the obtained electrical signals;

[0079] The display unit 33 is used to record the collected input pressure data, output pressure data and electrical signal data.

[0080] As shown in the figure, the calibration unit 7 is used to perform primary sensor calibration according to the obtained pressure data and electrical signal data, and when the temperature changes, perform secondary sensor calibration according to the compensated electrical signal data and pressure data.

[0081] As shown in the figure, the primary sensor calibration includes the following steps:

[0082] Receiving pressure detection data, sensor data and temperature detection data;

[0083] Sensor calibration processing;

[0084] Judging whether it is in a room temperature state;

[0085] If it is in a room temperature state, the primary sensor calibration is completed;

[0086] If it is not in a room temperature state, the data is transmitted;

[0087] As shown in the figure: the compensation unit 8 is used to compensate the detected electrical signal data when it is not in a room temperature state.

[0088] As shown in the figure, the compensation processing includes the following steps:

[0089] Receiving data;

[0090] Judging whether it is in a room temperature state;

[0091] If it is not in a room temperature state, a neural network compensation model is established;

[0092] Compensating the data;

[0093] Sending data;

[0094] The received data includes data after initial sensor calibration and electrical signal data measured in a non-chamber temperature environment.

[0095] Embodiment two:

[0096] A piezoelectric film sensor algorithm calibration system, comprising a pressure input unit 1, an output end of the pressure input unit 1 being electrically connected with an input end of a sensor unit 3, an output end of the sensor unit 3 being electrically connected with an input end of a pressure acquisition unit 4, an input end of the pressure acquisition unit 4 being electrically connected with an output end of the pressure input unit 1, and an output end of the pressure input unit 1 being electrically connected with an input end of a waveform processing unit 2.

[0097] An output end of the waveform processing unit 2 is electrically connected with an input end of the sensor unit 3, an output end of the sensor unit 3 is electrically connected with an input end of a calibration unit 7, an input end of the calibration unit 7 is electrically connected with an output end of the pressure acquisition unit 4, an input end of the sensor unit 3 is electrically connected with an output end of a temperature control unit 5, and an output end of the temperature control unit 5 is electrically connected with an input end of a temperature acquisition unit 6.

[0098] An output end of the temperature acquisition unit 6 is electrically connected with an input end of the calibration unit 7, an output end of the calibration unit 7 is electrically connected with an input end of a compensation unit 8, an output end of the compensation unit 8 is electrically connected with an input end of the calibration unit 7, and an input end of the compensation unit 8 is electrically connected with an output end of the temperature acquisition unit 6.

[0099] As shown in the figure, the pressure acquisition unit 4 comprises an input pressure detection unit 41 and an output pressure detection unit 42;

[0100] The input pressure detection unit 41 is used for detecting the pressure applied to the sensor unit 3 during the pressure input process.

[0101] The output pressure detection unit 42 is used for detecting the pressure at the sensor unit 3.

[0102] As shown in the figure, the waveform processing unit 2 adopts a waveform shaper, and the function of the waveform processing is to eliminate the oscillation caused by the dispersion effect in the shock wave waveform during the pressure output process, so that the waveform has a clear wave crest.

[0103] As shown in the figure, the sensor unit 3 comprises a data detection unit 31, a charge amplification unit 32 and a display unit 33, an output end of the data detection unit 31 being electrically connected with an input end of the charge amplification unit 32, and an output end of the charge amplification unit 32 being electrically connected with an input end of the display unit 33.

[0104] As shown in the figure, the data detection unit 31 is used for converting the received pressure into a corresponding electrical signal when an impact is received.

[0105] The charge amplification unit 32 is used to amplify the obtained electrical signal;

[0106] The demonstration unit 33 is used to record the collected input pressure data, output pressure data and electrical signal data.

[0107] As shown in the figure, the calibration unit 7 is used to perform primary sensor calibration according to the obtained pressure data and electrical signal data, and secondary sensor calibration according to the compensated electrical signal data and pressure data when the temperature changes.

[0108] As shown in the figure, the primary sensor calibration includes the following steps:

[0109] Receiving pressure detection data, sensor data and temperature detection data;

[0110] Sensor calibration processing;

[0111] Determine whether it is in room temperature state;

[0112] If it is in room temperature state, the primary sensor calibration is completed;

[0113] If it is not in room temperature state, transmit data;

[0114] The sensor calibration process is:

[0115] The piezoelectric coefficient d of the sensor 33 is: d 33 = Q / F,

[0116] Where Q is the piezoelectric charge, F is the output force, and U(t) is the voltage signal after amplification and conversion by the charge amplification unit. The expression for the charge Q obtained by back-propagating the voltage signal is:

[0117] Q = U(t) d1 / K

[0118] Where d1 is the sensitivity of the charge amplification unit, K is the output gain of the charge amplification unit, and the output force F is given by the output signal ε:

[0119] F = EAε(t) = EAηU

[0120] Where E is the elastic modulus of the test transmission rod, A is the area of the transmission rod, and η is the conversion coefficient between voltage and output signal.

[0121] The final piezoelectric coefficient is:

[0122] As shown in the figure: the compensation unit 8 is used to compensate the detected electrical signal data when it is not in room temperature state.

[0123] As shown in the figure, the compensation processing includes the following steps:

[0124] receiving data;

[0125] judging whether in room temperature state;

[0126] if not in room temperature state, establishing neural network compensation model;

[0127] carrying out data compensation;

[0128] sending data;

[0129] The received data includes data after initial sensor calibration and electrical signal data measured in a non-room temperature environment.

[0130] Embodiment three:

[0131] A piezoelectric thin film sensor algorithm calibration system, comprising a pressure input unit 1, an output end of the pressure input unit 1 being electrically connected with an input end of a sensor unit 3, an output end of the sensor unit 3 being electrically connected with an input end of a pressure acquisition unit 4, an input end of the pressure acquisition unit 4 being electrically connected with an output end of the pressure input unit 1, and an output end of the pressure input unit 1 being electrically connected with an input end of a waveform processing unit 2.

[0132] An output end of the waveform processing unit 2 is electrically connected with an input end of the sensor unit 3, an output end of the sensor unit 3 is electrically connected with an input end of a calibration unit 7, an input end of the calibration unit 7 is electrically connected with an output end of the pressure acquisition unit 4, an input end of the sensor unit 3 is electrically connected with an output end of a temperature control unit 5, and an output end of the temperature control unit 5 is electrically connected with an input end of a temperature acquisition unit 6.

[0133] An output end of the temperature acquisition unit 6 is electrically connected with an input end of the calibration unit 7, an output end of the calibration unit 7 is electrically connected with an input end of a compensation unit 8, an output end of the compensation unit 8 is electrically connected with an input end of the calibration unit 7, and an input end of the compensation unit 8 is electrically connected with an output end of the temperature acquisition unit 6.

[0134] As shown in the figure, the pressure acquisition unit 4 comprises an input pressure detection unit 41 and an output pressure detection unit 42.

[0135] The input pressure detection unit 41 is used for detecting the pressure applied to the sensor unit 3 during the pressure input process.

[0136] The output pressure detection unit 42 is used for detecting the pressure at the sensor unit 3.

[0137] As shown in the figure, the waveform processing unit 2 adopts a waveform shaper, and the function of the waveform processing is to eliminate the oscillation caused by the dispersion effect in the shock wave waveform during the pressure output process, so that the waveform has a clear wave crest.

[0138] As shown in the figure, the sensor unit 3 includes a data detection unit 31, a charge amplification unit 32 and a display unit 33, the output end of the data detection unit 31 is electrically connected with the input end of the charge amplification unit 32, and the output end of the charge amplification unit 32 is electrically connected with the input end of the display unit 33.

[0139] As shown in the figure, the data detection unit 31 is used to convert the received pressure into corresponding electrical signal when impacted;

[0140] The charge amplification unit 32 is used to amplify the obtained electrical signal;

[0141] The display unit 33 is used to record the collected input pressure data, output pressure data and electrical signal data.

[0142] As shown in the figure, the calibration unit 7 is used to perform primary sensor calibration according to the obtained pressure data and electrical signal data, and perform secondary sensor calibration according to the compensated electrical signal data and pressure data when the temperature changes.

[0143] As shown in the figure, the primary sensor calibration includes the following steps:

[0144] Receiving pressure detection data, sensor data and temperature detection data;

[0145] Sensor calibration processing;

[0146] Judging whether it is in room temperature state;

[0147] If it is in room temperature state, the primary sensor calibration is completed;

[0148] If it is not in room temperature state, the data is transmitted;

[0149] As shown in the figure: the compensation unit 8 is used to compensate the detected electrical signal data when it is not in room temperature state.

[0150] As shown in the figure, the compensation processing includes the following steps:

[0151] Receiving data;

[0152] Judging whether it is in room temperature state;

[0153] If it is not in room temperature state, the neural network compensation model is established;

[0154] Compensating data;

[0155] Sending data;

[0156] The received data includes the data after the primary sensor calibration and the electrical signal data measured in non-room temperature environment;

[0157] The data compensation process is:

[0158] The compensation characteristic function is y = f -1 (M P ,M t ) = x, y is the output after compensation,

[0159] The fuzzy neural network is used to calculate, wherein the membership function in each variable is μ Ai (x i ), and the membership function is a Gaussian function

[0160] Wherein c and sigma are the center and width of the membership function respectively,

[0161] I represents input information, O represents output information, and the input and output relationship of each layer is:

[0162]

[0163] ω is the connection weight between the two formulas,

[0164] Supposing that the expected output is y d (t)

[0165] The output of the network is y(t)

[0166] The network learning makes minimum, and the network weight is adjusted according to the following formula:

[0167]

[0168] The adjustment formula of the Gaussian function parameter is:

[0169]

[0170] In the formula, η is a learning rate, and β is a momentum parameter.

[0171] The learning rule of the neural network is obtained as:

[0172]

[0173] Wherein training is performed until |y i -y di | <E ends, and the approximation to f -1 (v) is completed.

[0174] In conclusion, in the present application:

[0175] The present application realizes the oscillation caused by the dispersion effect in the shock wave waveform during the pressure input process by adopting a waveform processing unit, makes the waveform have obvious wave crest, makes the impact force of the input process more accurate, reduces the influence of the chaotic waveform on the input pressure value measurement, simultaneously adopts a calibration algorithm, realizes the rapid calibration process of the sensor itself, realizes the accuracy calibration confirmation at the factory and the re-calibration after the sensor maintenance.

[0176] The present application realizes the calibration process at different temperatures by adopting a temperature control unit and a temperature acquisition unit, the temperature control unit is used for controlling the temperature of the calibration environment, simultaneously the temperature acquisition unit can acquire the temperature at the sensor, the difference between the data detected by the sensor at different temperatures and the data detected by the sensor at normal temperature, through the change of the calibration process data of multiple groups at different temperatures and the same pressure, adopts a neural network model to realize the compensation of the zero point data in each measurement value, realizes that the result measured at multiple temperature conditions is more accurate compared with the actual value, makes the overall calibration process more ideal and sufficient, and ensures that the sensor remains accurate under the influence of different temperatures, and through the neural network model, a simulation model with high precision can be obtained, the whole calculation process is stable, the convergence speed is fast, the use range of the calibrated sensor is more extensive, and it is suitable for promotion.

[0177] Secondly: the present application discloses only the structure related to the present disclosure in the embodiment, other structures can refer to the usual design, under the condition of no conflict, the same embodiment and different embodiments of the present application can be combined with each other;

[0178] Finally: the above only describes the preferred embodiments of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A piezoelectric thin film sensor algorithm calibration system comprising a pressure input unit (1), characterized in that: The output end of the pressure input unit (1) is electrically connected with the input end of the sensor unit (3), the output end of the sensor unit (3) is electrically connected with the input end of the pressure acquisition unit (4), the input end of the pressure acquisition unit (4) is electrically connected with the output end of the pressure input unit (1), and the output end of the pressure input unit (1) is electrically connected with the input end of the waveform processing unit (2); The output end of the waveform processing unit (2) is electrically connected with the input end of the sensor unit (3), the output end of the sensor unit (3) is electrically connected with the input end of the calibration unit (7), the input end of the calibration unit (7) is electrically connected with the output end of the pressure acquisition unit (4), the input end of the sensor unit (3) is electrically connected with the output end of the temperature control unit (5), and the output end of the temperature control unit (5) is electrically connected with the input end of the temperature acquisition unit (6); The output end of the temperature acquisition unit (6) is electrically connected with the input end of the calibration unit (7), the output end of the calibration unit (7) is electrically connected with the input end of the compensation unit (8), the output end of the compensation unit (8) is electrically connected with the input end of the calibration unit (7), and the input end of the compensation unit (8) is electrically connected with the output end of the temperature acquisition unit (6); The calibration unit (7) is used for primary sensor calibration according to obtained pressure data and electrical signal data, and secondary sensor calibration according to compensated electrical signal data and pressure data when the calibration temperature changes.

2. The piezoelectric thin film sensor algorithm calibration system according to claim 1, characterized in that: The pressure acquisition unit (4) comprises an input pressure detection unit (41) and an output pressure detection unit (42); The input pressure detection unit (41) is used for detecting the pressure applied to the sensor unit (3) during the pressure input process; The output pressure detection unit (42) is used for detecting the pressure at the sensor unit (3).

3. The piezoelectric thin film sensor algorithm calibration system of claim 1, wherein: The waveform processing unit (2) adopts a waveform shaper, and the waveform processing function is to eliminate the oscillation caused by the dispersion effect in the shock wave waveform during the pressure input process, so that the waveform has a clear wave crest.

4. The piezoelectric thin film sensor algorithm calibration system of claim 1, wherein: The sensor unit (3) comprises a data detection unit (31), a charge amplification unit (32) and an indication unit (33), the output end of the data detection unit (31) is electrically connected with the input end of the charge amplification unit (32), and the output end of the charge amplification unit (32) is electrically connected with the input end of the indication unit (33).

5. The piezoelectric thin film sensor algorithm calibration system of claim 4, wherein: The data detection unit (31) is used for converting the received pressure into corresponding electrical signal when impacted; The charge amplification unit (32) is used for amplifying the obtained electrical signal; The indication unit (33) is used for recording the collected input pressure data, output pressure data and electrical signal data.

6. The piezoelectric thin film sensor algorithm calibration system of claim 1, wherein, The primary sensor calibration comprises the following steps: Receiving pressure detection data, sensor data and temperature detection data; Sensor calibration processing; Judging whether it is in room temperature state; If it is in room temperature state, the primary sensor calibration is completed; If it is not in room temperature state, data is transmitted; The sensor calibration process is: Piezoelectric coefficient d of the sensor 33 is: d 33 = Q / F, Wherein, Q is piezoelectric charge, F is output force, U(t) is voltage signal amplified by charge amplifier unit, and the expression of charge Q obtained by backstepping voltage signal is: Q=U(t)d1 / K Wherein, d1 is sensitivity of charge amplifier unit, K is output gain of charge amplifier unit, and output force F is given by output signal ε: F=EAε(t)=EAη1U Wherein, E is elastic modulus of transmission rod for testing, A is area of transmission rod, and η1 is conversion coefficient between voltage and output signal; The final piezoelectric coefficient is:

7. The piezoelectric thin film sensor algorithm calibration system of claim 1, wherein: The compensation unit (8) is used for compensating the electric signal data obtained in a state not at room temperature.

8. The piezoelectric thin film sensor algorithm calibration system of claim 7, wherein, The compensation process comprises the following steps: receiving data; judging whether it is at room temperature or not; if not, establishing a neural network compensation model; compensating data; sending data; The received data comprises data after initial sensor calibration and electric signal data measured in a non-room temperature environment; The data compensation process is: The compensation characteristic function is y = f -1 (M P ,M t ) = x, y is the output after compensation, A fuzzy neural network calculation is used, in which the membership function in each variable is μ Ai (x i ), and the membership function is a Gaussian function: Wherein, c and σ are center and width of membership function respectively, I represents input information, O represents output information, and input-output relationship of each layer is: ω is connection weight between the two formulas, Let the desired output be y d (t) Output of the network is y(t) Network learning enables The network weights are adjusted according to the following equation: The adjustment formula of Gaussian function parameters is: Wherein, η2 is learning rate, and β is momentum parameter, The learning rule of neural network is: where the training is until |y i - y di | < E end, completion of the approximation of f -1 (v).

9. Application of a piezoelectric film sensor algorithm calibration system according to any one of claims 1-8 in sensor calibration.

Images