A multi-range high-precision air pressure detection system and method based on multi-frequency CMUT

Abstract

The present invention belongs to the field of air pressure detection technology, and specifically discloses a multi-range high-precision air pressure detection system and method based on a multi-frequency CMUT. The present invention sets up multiple CMUT sensors with different resonance unit radii, so that the system can work in multiple different frequency ranges at the same time, greatly expanding the air pressure measurement range. By selecting appropriate sensors, the multi-frequency air pressure sensor can cover a wide range of application scenarios from low pressure to high pressure. In low-pressure measurement scenarios, sensors with larger radius have higher sensitivity and can accurately capture tiny air pressure changes. In high-pressure measurement scenarios, sensors with smaller radius have lower sensitivity and can prevent the sensor from over-responding. The present invention uses an automatic sensor switching mechanism to select the sensor that is most suitable for the current air pressure range to work, and at the same time ensures that sensors with larger radius in the high-pressure range will not be selected to work, so as to avoid exceeding the range and causing damage to the CMUT sensor.

Description

Technical Field

[0001] The present invention belongs to the technical field of air pressure detection, and in particular relates to a multi-range high-precision air pressure detection system and method based on a multi-frequency CMUT. Background Art

[0002] Air pressure detection technology has widespread applications in aerospace, petrochemicals, automotive, and industrial control, and air pressure sensors play a vital role in these industries. Compared to traditional air pressure sensors, MEMS air pressure sensors offer significant advantages in size and power consumption, making them particularly suitable for space-constrained devices that require long-term operation. MEMS air pressure sensors also offer high precision and sensitivity, enabling real-time monitoring of minute air pressure changes, ensuring data accuracy and reliability. CMUT, a new resonant sensor based on MEMS technology, converts minute air pressure changes into frequency offset signals. It offers advantages such as high sensitivity, low power consumption, and ease of arraying and integration. CMUT not only enables high-precision air pressure measurement but also has a wide operating frequency range, meeting the needs of diverse applications. Furthermore, CMUTs offer a high signal-to-noise ratio and strong anti-interference capabilities, enabling stable and reliable operation in complex environments.

[0003] However, CMUT sensors also have some limitations. First, although the measurement range of CMUT sensors is already larger than many traditional sensors, when exceeding a certain pressure range, measurement accuracy may decrease due to physical limitations of the sensor structure or nonlinear response. Second, the stability of the CMUT can be affected by external environmental conditions such as temperature, humidity, or airflow changes. These factors can cause frequency offset instability, thereby affecting measurement accuracy.

[0004] Both traditional and MEMS pressure sensors have limitations in terms of range and accuracy. First, the sensor's range is often limited by its physical design and material properties, making it difficult to adapt to various application requirements.

[0005] For example, low-pressure sensors are commonly used in fields such as meteorological monitoring and gas leak detection, medium-pressure sensors are widely used in scenarios such as automotive tire pressure monitoring and industrial process control, and high-pressure sensors are primarily used in high-pressure environments such as aerospace and petrochemicals. However, the measurement ranges of these sensors generally cannot cover low-pressure or high-pressure application requirements. Secondly, pressure sensors are easily affected by environmental factors or by internal system anomalies or failures, resulting in reduced measurement accuracy. These limitations significantly affect the applicability of pressure sensors for high-precision pressure measurement in multiple scenarios.

[0006] To this end, the present invention proposes a multi-range high-precision air pressure detection system and method based on a multi-frequency CMUT. Summary of the Invention

[0007] The purpose of the present invention is to provide a multi-range, high-precision air pressure detection system based on a multi-frequency CMUT. By providing multiple CMUT sensors with different resonant unit radii, the air pressure measurement range can be expanded to cover both low-pressure and high-pressure measurement environments. At the same time, it can effectively reduce errors caused by environmental factors or system failures, thereby improving the air pressure measurement accuracy.

[0008] In order to achieve the above objectives, the technical solutions adopted by the present invention are as follows:

[0009] A multi-range, high-precision air pressure detection system based on a multi-frequency CMUT, including a power management module, a multi-frequency air pressure sensor, an oscillation circuit, an STM32 frequency measurement system, and a host computer system;

[0010] The output end of the power management module is connected to the multi-frequency air pressure sensor and is used to provide a DC bias voltage for the multi-frequency air pressure sensor; the multi-frequency air pressure sensor includes a plurality of CMUT sensors;

[0011] The radius of the resonant unit of each CMUT sensor is different and shows an increasing or decreasing trend in sequence;

[0012] The air pressure measurement range of the CMUT sensor varies with the radius of the resonant unit. The smaller the radius of the resonant unit, the greater the maximum measurable air pressure. The measurement ranges of all CMUT sensors overlap, and the minimum measurable air pressure is 0 Pa.

[0013] Each CMUT sensor is individually connected to a set of oscillation circuits, and each set of oscillation circuits is connected to the input end of the STM32 frequency measurement system. The STM32 frequency measurement system is connected to the host computer system through a wireless communication module.

[0014] In addition, the present invention also provides a multi-range high-precision air pressure detection method based on a multi-frequency CMUT. The method uses the multi-range high-precision air pressure detection system based on a multi-frequency CMUT, and includes the following steps:

[0015] Step 1. Initialize each CMUT sensor to ensure that it is stable and working properly. Select the CMUT sensor with the largest operating threshold from among all CMUT sensors to detect the air pressure value of the test environment as the initial air pressure value.

[0016] Step 2. Compare the measured air pressure value with the operating threshold of each CMUT sensor;

[0017] If the currently measured air pressure value is only less than the maximum operating threshold value of all CMUT sensors, only the CMUT sensor with the maximum operating threshold value is selected for air pressure detection, and the process goes to step 6;

[0018] Otherwise, select two CMUT sensors whose operating thresholds match the measured air pressure values ​​and go to step 3;

[0019] Step 3. Use the STM32 frequency measurement system to collect the frequency data of the two CMUT sensors selected in step 2 in real time. Use the host computer to convert the frequency data of the two CMUT sensors into corresponding air pressure values ​​in real time.

[0020] Continuously monitor the air pressure values ​​measured by the two CMUT sensors and determine whether they match the operating thresholds of the two currently selected CMUT sensors. If so, proceed to step 4; otherwise, return to step 2 and reselect the CMUT sensors.

[0021] Step 4. Set the sliding window size and use the sliding window anomaly detection method to retain the valid data measured by the two currently selected CMUT sensors and remove the abnormal data measured by the two CMUT sensors;

[0022] Step 5. Use the weighted summation method to perform data weighted fusion on the outputs of the two CMUT sensors in the working state;

[0023] Step 6. Output the air pressure measurement result.

[0024] The beneficial technical effects of the present invention are:

[0025] As described above, the present invention describes a multi-range, high-precision air pressure detection system based on a multi-frequency CMUT. By setting up multiple CMUT sensors with different resonant unit radii, the system can flexibly and selectively operate in different frequency ranges, greatly expanding the air pressure measurement range. By selecting the appropriate sensor, the multi-frequency air pressure sensor can cover a wide range of application scenarios from low pressure to high pressure. In low-pressure measurement scenarios, CMUT sensors with larger radius have higher sensitivity and can accurately capture tiny air pressure changes. In high-pressure measurement scenarios, CMUT sensors with smaller radius have lower sensitivity, which can prevent the sensor from over-responding. Through the sensor automatic switching mechanism, the sensor that is most suitable for the current air pressure range is selected to work. At the same time, it can also ensure that the CMUT sensor with a larger radius in the high-pressure range will not be selected to work, so as to avoid exceeding the range and causing damage to the CMUT sensor. In addition, the present invention also proposes a multi-range, high-precision air pressure detection method based on a multi-frequency CMUT based on the above system, which effectively improves the accuracy of air pressure measurement. During actual operation, the system will preferentially select the two sensors that are most suitable for the current application scenario to work. By introducing redundant design, the system effectively mitigates the impact of external environmental factors (such as temperature, humidity, and vibration) on sensor measurement accuracy, enhancing the system's fault tolerance and reducing measurement errors due to environmental fluctuations. The system automatically switches sensors according to different pressure ranges and adjusts weight distribution based on actual conditions, thereby optimizing the data fusion process. This adaptive mechanism improves pressure measurement accuracy, ensuring the use of the most appropriate sensor within different pressure ranges, effectively reducing errors and measurement fluctuations, and enhancing the system's reliability and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is an architecture diagram of a multi-range high-precision air pressure detection system in an embodiment of the present invention;

[0027] Figure 2 This is a schematic diagram of the structure of a multi-frequency air pressure sensor in an embodiment of the present invention (taking three frequencies as an example);

[0028] Figure 3 Schematic diagram of the air pressure sensing mechanism of the CMUT sensor in an embodiment of the present invention;

[0029] Figure 4 Schematic diagram of the mechanical mass-spring model of the CMUT sensor in an embodiment of the present invention;

[0030] Figure 5 Schematic diagram of the circuit connection relationship of the power management module in an embodiment of the present invention;

[0031] Figure 6 10.2V bias voltage under different radii of CMUT pressure-frequency shift response model diagram of the embodiment of the present invention;

[0032] Figure 7 Flowchart of the air pressure detection method in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions, and beneficial effects of the present invention more clearly understood, the present invention is further described below in conjunction with specific embodiments and with reference to the accompanying drawings. Certain embodiments of the present invention will be more fully described below with reference to the accompanying drawings, some, but not all, of which are illustrated. The various embodiments of the present invention may be implemented in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention satisfies applicable legal requirements.

[0034] In the description of the present invention, it should be noted that the terms "inner," "outer," "upper," "lower," "front," and "rear," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0035] Example 1

[0036] In an embodiment of the present invention, a multi-range high-precision air pressure detection system and method based on a multi-frequency CMUT is provided. Please refer to Figures 1 to 6 The multi-range high-precision air pressure detection system includes a power management module, a multi-frequency air pressure sensor, an oscillation circuit, an STM32 frequency measurement system, and a host computer system.

[0037] Combine Figure 1 As shown, the output end of the power management module is connected to the multi-frequency air pressure sensor, and the power management module can provide a DC bias voltage for the multi-frequency air pressure sensor.

[0038] Combine Figure 5 As shown in the figure, the power management module uses the LM2733 boost converter and the MAX5433 digital potentiometer to achieve high-precision bias voltage regulation. The power management module consists of multiple identical boost circuits, each connected to a CMUT sensor, providing DC bias voltage for multiple CMUT sensors.

[0039] Figure 5 The structure of a set of boost circuits in the power management module is shown. The specific configuration of the circuit network is as follows:

[0040] Input filter capacitor C1 has a capacitance of 2.2μF and is used to filter high-frequency noise. Boost inductor L1 has an inductance of 10μH and works in conjunction with freewheeling diode D1 (model MBR0540) to achieve the boost function. Output filter capacitor C2 has a capacitance of 4.7μF and stabilizes the output voltage. Compensation capacitor CF has a capacitance of 82pF and is used for phase compensation, thereby improving system stability. Furthermore, voltage divider resistor R2 has a resistance of 4.275kΩ and, together with R1, forms a voltage divider network that transmits the feedback signal to the FB pin of the LM2733 boost converter, forming a closed-loop control loop to ensure stable boost output.

[0041] Among them, the resistor voltage division formula is: .

[0042] In this circuit, R1 represents the output resistance of the MAX5433 digital potentiometer. This digital potentiometer is controlled via SPI communication with an STM32 microcontroller. Specifically, the SCL pin of the MAX5433 digital potentiometer is connected to the PB6 pin of the STM32 microcontroller, and the SDA pin of the MAX5433 digital potentiometer is connected to the PB7 pin of the STM32 microcontroller. This allows for 32 levels of output resistance adjustment, with a maximum value of 100kΩ. Calculations show that the output voltage range of the bias network is 5.7V to 30V, with a voltage adjustment step interval of approximately 0.9V, fully meeting the requirements for high precision and flexibility.

[0043] After comprehensive consideration, it was finally decided to use a fixed bias voltage of 10.2V.

[0044] Combine Figure 1 As shown, the multi-frequency air pressure sensor includes multiple CMUT sensors (three CMUT sensors are used as an example), each with a different resonant unit radius. During the manufacturing process, the CMUT sensor radius can be set over a wide range, initially ranging from 1μm to 100μm, or even larger. In the embodiments of the present invention, three CMUT sensors with decreasing resonant unit radii, specifically 15μm, 12μm, and 10μm, were tested. Theoretical derivation results show that the pressure measurement thresholds of the three sensors are 340kPa, 830kPa, and 1.72MPa, respectively.

[0045] This air pressure detection system is highly flexible and adaptable. The number of CMUT sensors in the system can be freely set according to actual needs and can be flexibly adjusted according to specific air pressure detection scenarios to meet diverse application requirements. At the same time, the radius change trend of the CMUT sensor's resonant unit can be flexibly selected according to actual conditions. It can be set to either monotonically increasing or monotonically decreasing, and the radius range can also be flexibly adjusted. This flexible radius setting method enables the air pressure detection system to better adapt to air pressure detection tasks in different situations and achieve accurate and reliable air pressure detection.

[0046] Combine Figure 2 As shown in Figure 1, a CMUT sensor consists of hundreds or thousands of tiny resonant units connected in parallel. Each resonant unit consists of a top electrode, a resonant membrane, a vacuum cavity, an insulating layer, and a bottom electrode.

[0047] The operating principle of a CMUT sensor is based on the resonance effect of a resonant membrane at a specific frequency. When external air pressure is applied to the surface of the resonant membrane, the physical shape of the resonant membrane changes, causing a slight shift in the resonant frequency.

[0048] like Figure 3 As shown, the bias voltage (V dc ) is added through electrostatic force This further enhances the resonant performance and pressure sensitivity of the resonant membrane. Under the combined effects of bias voltage and external air pressure, the membrane deforms and changes its resonant frequency. By precisely measuring this frequency change, the corresponding air pressure value can be inferred, enabling highly accurate pressure detection.

[0049] Combine Figure 3 As shown, is the radius of the resonant membrane, represents the radial position, represents the thickness of the resonant film, represents the cavity height, Indicates the deflection of the membrane after bending and deformation. When the membrane is subjected to force, its resonant frequency It can be expressed as:

[0050] .

[0051] in, is the density of the membrane, is Poisson's ratio, is Young's modulus.

[0052] In order to study the electrostatic characteristics of CMUT sensors, they are usually approximated as parallel plate capacitors, such as Figure 4 shown. represents the equivalent spring constant of the resonant membrane, Represents the average displacement of the CMUT resonant membrane. Electrostatic force and spring force (restoring force of CMUT resonant membrane) are respectively expressed as:

[0053] , .

[0054] in represents the initial equivalent cavity height of the CMUT sensor, A represents the electrode area of ​​the CMUT sensor, V represents the selected bias operating voltage, represents the dielectric constant of vacuum, represents the relative dielectric constant of the vacuum cavity, where The value of is 1.

[0055] according to Figure 3 As shown, It can be expressed as:

[0056] .

[0057] in, represents the relative dielectric constant of the resonant film, represents the relative dielectric constant of the insulating layer, represents the vacuum chamber height, Indicates the insulation layer height.

[0058] when and When the forces are equal, the upper electrode reaches a state of force equilibrium and remains stationary. The maximum displacement and breakdown voltage of the membrane during operation are obtained after solving:

[0059] , .

[0060] The bending displacement equation of the circular membrane of the CMUT sensor under the action of electrostatic force and external air pressure is given by the following equation:

[0061] .

[0062] in, The radial position of the thin plate The deformation of is the electrostatic force per unit area applied on the membrane, is the external air pressure value, is the bending stiffness of the plate.

[0063] Ignoring higher-order terms in the Taylor expansion, it can be simplified to:

[0064] .

[0065] in,

[0066] .

[0067] The displacement of the CMUT circular membrane under the action of air pressure is obtained by solving:

[0068] .

[0069] make , solve The value is the pressure measurement limit threshold corresponding to the CMUT sensor.

[0070] Finally, according to the law of conservation of energy, the resonant frequency of the CMUT sensor is obtained as:

[0071] .

[0072] in,

[0073] ;

[0074] ;

[0075] in,

[0076] .

[0077] The pressure sensitivity is defined as:

[0078] .

[0079] The above theoretical derivation shows that the CMUT sensor's pressure measurement threshold is closely related to its sensitivity and the radius of the resonant unit. This shows that the multi-frequency CMUT design can effectively meet the needs of high and low pressure detection.

[0080] by Figure 2 Take the three-frequency CMUT pressure sensor shown in the figure as an example. The sensor is composed of three CMUT sensors with different radii. According to the CMUT theory derivation formula, when the bias voltage is 10.2V, the pressure measurement limit thresholds of these three sensors are 340kPa, 830kPa and 1.72MPa respectively (as shown in Figure 2). Figure 6 shown).

[0081] Multi-frequency pressure sensors can operate simultaneously in multiple frequency ranges. By selecting the appropriate sensor, they can cover a wide range of application scenarios, from low to high pressure. Furthermore, CMUT sensors of varying radii have varying sensitivities, perfectly suited to both high and low pressure monitoring needs. In low-pressure measurement scenarios, CMUT sensors with larger radii have higher sensitivity, accurately capturing minute pressure changes. In high-pressure measurement scenarios, CMUT sensors with smaller radii have lower sensitivity, preventing over-response. By properly selecting the radius of the CMUT sensor's resonant unit, sensor performance can be flexibly adjusted to meet various pressure measurement needs, further expanding its application scenarios.

[0082] Combine Figure 1 As shown, each CMUT sensor is individually connected to a set of oscillation circuits.

[0083] The oscillation circuit uses a dual-op-amp Colpitts oscillator circuit, constructed with two operational amplifier stages: an AD8045 and an OPA699. The AD8045 performs initial signal amplification, with a gain set to 4.5. The OPA699, with a gain set to 4.2, also performs signal limiting, strictly limiting the output voltage range to -0.3V to 3.3V to prevent damage to the input port of the STM32 frequency measurement system. After two stages of amplification, the signal passes through a MAX4201 buffer to eliminate the effects of parasitic loads.

[0084] In addition, a high-pass filter is provided in the circuit, which is used for positive phase shift compensation to solve the negative phase shift problem in the circuit, ensuring that the total phase of the oscillation circuit is zero, thereby achieving stable and reliable frequency tracking function.

[0085] At the same time, the oscillation circuit can also adopt a Colpitts oscillation circuit based on BJT (transistor) and MOSFET (field effect transistor).

[0086] Combine Figure 1 As shown, each oscillation circuit is connected to the input end of the STM32 frequency measurement system.

[0087] The STM32 frequency measurement system uses an STM32F407 microcontroller. Configured in external counter mode, it collects data from multiple sensors separately and calculates the real-time frequency data of the CMUT sensors by accurately counting the rising edges of the input signals. This design ensures that the system can quickly respond to and process changes in the CMUT sensor frequency.

[0088] Combine Figure 1 As shown in the figure, the STM32 frequency measurement system is connected to the host computer system through a wireless communication module.

[0089] The wireless communication module uses the HC-05 Bluetooth module, which uses serial communication to wirelessly transmit frequency data from the STM32 frequency measurement system to the host computer system. The RX pin of the HC-05 Bluetooth module is connected to the PA9 pin of the STM32 frequency measurement system, and the TX pin is connected to the PA10 pin. The communication baud rate is set to 115200 to ensure efficient and stable data transmission.

[0090] The host computer system displays frequency data in real time and converts it into air pressure values. It can display data from multiple CMUT sensor channels individually, allowing users to flexibly select channels and adjust related settings. Furthermore, the host computer system transmits commands to the STM32 frequency measurement system via a wireless transmission module. Based on these commands, the STM32 frequency measurement system adjusts the bias voltage on and off, thereby controlling the operating status of the CMUT sensors.

[0091] Specifically, when the bias voltage is set to 10.2V, the CMUT sensor is in working state; when the bias voltage is 0V or in the off state, the CMUT sensor is also in the off state. By adjusting the bias voltage, it is possible to ensure that certain CMUT sensors with larger radii within the high-voltage range are not selected for operation, thereby avoiding exceeding the range and causing sensor damage.

[0092] The signal processing flow of the multi-range high-precision air pressure detection system is as follows: First, the power management module can provide a DC bias voltage for the multi-frequency air pressure sensor. The DC bias voltage can cause the resonant membrane to produce static pre-deformation. By adjusting the DC bias voltage, the sensitivity of the multi-frequency air pressure sensor can be precisely controlled, so that it can adapt to various different air pressure measurement needs; secondly, the multi-frequency air pressure sensor includes multiple CMUT sensors with different radii. An oscillation circuit is built based on the parallel resonance point of the CMUT. By meeting the Barkhausen oscillation condition, the oscillation circuit can track the resonant frequency in real time, thereby achieving high-precision frequency acquisition; then, the STM32 frequency measurement system obtains the collected frequency data and then uploads it to the host computer system via the wireless communication module; finally, the host computer system converts the received frequency data into air pressure values ​​in real time and displays them based on the predetermined pressure-frequency response relationship, thereby achieving high-precision air pressure measurement and display functions.

[0093] Example 2

[0094] Combine Figure 7 As shown, the present invention further proposes a multi-range high-precision air pressure detection method based on a multi-frequency CMUT, which applies the multi-range high-precision air pressure detection system based on a multi-frequency CMUT in the above embodiment 1.

[0095] The multi-range high-precision air pressure detection method based on the multi-frequency CMUT in this embodiment includes the following steps:

[0096] Step 1. Initialize each CMUT sensor to ensure that it is stable and working properly. Select the CMUT sensor with the largest working threshold from all CMUT sensors to detect the air pressure value of the test environment as the initial air pressure value.

[0097] Initialize each CMUT sensor and test all CMUT sensors in the same environment.

[0098] The CMUT sensor with the highest operating threshold is selected to measure the ambient air pressure, and the initial pressure value obtained serves as the basic reference for subsequent measurements and data processing. The CMUT with the highest operating threshold has a wider pressure measurement range and can maintain stable and accurate measurement performance across a wide pressure range. By measuring the ambient air pressure using this CMUT, an accurate initial pressure value can be obtained. This initial pressure value serves as the basic reference for subsequent measurements and data processing, providing an important starting point for the normal operation and accurate measurement of the entire pressure measurement system.

[0099] Step 2. Compare the measured air pressure value with the operating threshold of each CMUT sensor. Based on the current measured air pressure value, select the appropriate sensor from multiple CMUT sensors for subsequent data collection.

[0100] Specifically, the number of CMUT sensors is set to n, where n is a natural number, and it is assumed that the radius of the resonance units of all CMUT sensors shows a decreasing trend from the 1st to the nth CMUT sensors.

[0101] In this embodiment, n is greater than or equal to 3. Of course, n can also be 2, which will not be repeated here.

[0102] For the kth CMUT sensor, its working threshold Set to 85% of the kth CMUT sensor pressure limit threshold, 1≤k≤n, then each working threshold satisfies <… …< .

[0103] If the air pressure value is only less than the maximum working threshold of all CMUT sensors, that is, when the air pressure value is between and When the pressure is between 0 and 1, only the CMUT sensor n with the maximum working threshold is selected for air pressure detection. In this case, steps 3, 4 and 5 are not involved, and the process goes directly to step 6. After step 6, the pressure measurement value is output. .

[0104] When the air pressure is between and When the pressure is between 0 and 1, the first n-1 sensors all exceed the operating threshold, and only sensor n is within the operating threshold range. To protect the sensors from damage in high-pressure environments and ensure the reliability of the detection system, only sensor n is selected to perform the air pressure detection operation.

[0105] If the air pressure value is not less than the maximum working threshold value of all CMUT sensors, two CMUT sensors whose working threshold values ​​are adapted to the initial ambient air pressure value are selected for air pressure detection. and When the value is between , the i-th CMUT sensor and the i+1-th CMUT sensor are enabled for air pressure detection, and go to step 3, where 1≤i≤n-1, and let , It is not a physical quantity that actually exists. To facilitate the subsequent explanation of the relevant content, its value is set to 0Pa.

[0106] When the air pressure is between and When the pressure is between 1 and 2, the first i-1 sensors cannot be used due to exceeding the operating threshold. Considering that the more sensitive CMUT sensors have greater measurement accuracy in response to air pressure changes, and that subsequent collaborative testing, measurement data cross-validation, and data fusion can enhance detection accuracy, the i-th CMUT sensor and the i+1-th CMUT sensor are selected for air pressure testing. Using two CMUT sensors for air pressure testing provides a certain degree of redundancy for the system. By real-time monitoring and comparison of the data from the two sensors, sensor failures or anomalies can be promptly detected and addressed, improving the reliability and stability of the system.

[0107] The above sensor selection process is driven by the sensor automatic switching mechanism. Figure 7 As shown, there is a one-to-one correspondence between sensor combinations and air pressure ranges within the system. The varying number of sensors in the system determines the variety of sensor combinations that can be formed. The automatic sensor switching mechanism flexibly calls upon appropriate sensor resources based on actual detection needs. When the air pressure in the detection environment exceeds or falls below the current pressure range, the automatic sensor switching mechanism responds quickly, adjusting the sensor combination involved in the detection and adjusting the weights according to pre-set rules. This effectively improves the accuracy of air pressure detection and ensures the accuracy and reliability of the test results.

[0108] When the air pressure detection system is running, each CMUT sensor monitors the air pressure value at all times. When the air pressure value detected by a CMUT sensor exceeds or falls below the current air pressure range, the air pressure detection system will immediately initiate an automatic response program.

[0109] After the response program is initiated, the system automatically switches to a sensor operating mode appropriate for the current air pressure conditions. During this mode switching process, the number of CMUT sensors typically remains unchanged, with only two CMUT sensors always active, except for the final, specific scenario in the highest air pressure range. However, within each operating mode, the weight of each CMUT sensor in the data analysis process is adjusted based on factors such as the current air pressure conditions and the sensor's own performance. This automated switching mechanism empowers the system with robust environmental adaptability, ensuring stable and efficient air pressure detection, maintaining the system's precise detection performance, and ultimately outputting more accurate and reliable air pressure values.

[0110] Step 3. Use the STM32 frequency measurement system to collect the frequency data of the two CMUT sensors selected in step 2 in real time. Use the host computer to convert the frequency data of the two CMUT sensors into corresponding air pressure values ​​in real time.

[0111] The STM32 frequency measurement system synchronously collects data from both sensors, for example, every 20ms. The STM32 timer is used to synchronize the data, ensuring consistent sampling rates for both CMUT sensors. Hardware synchronization signals are used to time-align the data, effectively avoiding data deviations caused by timing differences.

[0112] The collected frequency data is transmitted to the host computer in real time through the wireless communication module. The host computer has a preset air pressure frequency response model. The model uses the internal algorithm to calculate and convert the received frequency data in real time to obtain the corresponding air pressure value. Figure 6 As shown, the pressure-frequency offset response model diagrams of three CMUTs with different radii are presented.

[0113] The system monitors the air pressure value measured by either of the two CMUT sensors in real time and determines whether it matches the operating threshold of the two currently selected CMUT sensors. If so, the system proceeds to step 4; otherwise, it returns to step 2 and reselects a CMUT sensor. In other words, when a sensor exceeds or falls below the current air pressure range, the system adjusts the sensor operating combination.

[0114] The system is capable of adaptively switching sensors based on different air pressure ranges. This mechanism ensures measurement accuracy while preventing CMUT sensor failure due to overload and breakdown, effectively protecting the device. During pressure measurement, each sensor updates its pressure value in real time. Based on the current real-time pressure value, the system automatically and accurately switches to the sensor combination that best suits the pressure conditions. This ensures the accuracy and reliability of the system's output data, improving the performance and stability of the entire air pressure detection system.

[0115] Step 4. Set the sliding window size and use the sliding window anomaly detection method to retain the valid data measured by the two currently selected CMUT sensors and remove the abnormal data measured by the two CMUT sensors;

[0116] Each sliding window is set to contain N pressure data points. This window is dedicated to processing pressure data measured by the CMUT sensors. Two data buffers are prepared, one for the i-th CMUT sensor and the other for the i+1-th CMUT sensor, to store the data within the sliding window. At the beginning of data acquisition, continuously acquire pressure data from both sensors and store them in the corresponding buffers in chronological order.

[0117] When new data is collected, for each CMUT sensor, the newly collected data is added to the end of its corresponding buffer. If the buffer contains less than N data, the existing data at the beginning of the buffer is copied to make up N data. If the buffer contains more than N data, the data stored earliest in the buffer (i.e., at the beginning of the buffer) is removed to ensure that the buffer always contains the latest N data.

[0118] The sliding window operates by moving forward one data at a time. Every time a new data enters, an anomaly detection is performed on the data in the sliding window.

[0119] For the i-th CMUT sensor, let the N data in the sliding window (i.e., in the buffer) be , and the mean of N data is .

[0120] in, .

[0121] Calculate the data mean in the sliding window of the i-th sensor and the i+1-th sensor respectively and , and calculate the absolute value of their difference .

[0122] in, .

[0123] like , then it is determined that the data in the current sliding window is normal.

[0124] like , then the data in the current sliding window is judged to be abnormal.

[0125] In the air pressure detection system, Indicates the threshold of the one with the larger working threshold among the two currently enabled CMUT sensors. for 0.005 times of the maximum pressure. for 0.005 times of the original value can effectively filter out noise interference, ensure the measurement signal is true and reliable, and meet the system accuracy requirements.

[0126] For windows that are judged to be normal, the system will extract the current sliding window average data from the buffer areas of the two sensors. 、 After the extraction is completed, this set of data is transferred to step 5 for data fusion processing. In step 5, the system uses the weighted summation method to perform weighted fusion on the relevant data including this set of data, and finally outputs the current air pressure measurement result to ensure the accuracy and reliability of the air pressure measurement data.

[0127] During the barometric pressure measurement process, a sliding window technique is used to dynamically average the data of the most recent N sampling points. This result is then applied to the final barometric pressure fusion output process. This effectively filters out high-frequency random noise introduced by factors such as circuit thermal noise, mechanical vibration, and electromagnetic interference, significantly reducing signal fluctuations and the standard deviation of the barometric pressure readings. This significantly enhances the stability of the output data and reliably ensures long-term accuracy.

[0128] In this embodiment, N is set to 5, and a smaller window size is used to ensure the system's ability to quickly track instantaneous changes in air pressure. Regardless of whether the air pressure rises or drops suddenly, the system can respond promptly. This setting effectively reduces signal latency and successfully avoids the signal lag caused by overly large windows in traditional large-window filtering methods, enabling air pressure measurements to more accurately reflect real-time conditions.

[0129] For windows identified as having data anomalies, data at the ends of the two sensor data buffers within that window is discarded. Since the previous sliding windows all performed normally, and the current sliding window exhibits an anomaly, it can be reasonably inferred that the anomaly lies with the newly entered data (i.e., the data at the ends of the two sensor data buffers). Therefore, to ensure data accuracy and reliability, data at the ends of the two sensor data buffers within that window must be discarded.

[0130] If the i-th and i+1-th sensors performing air pressure detection show m consecutive (e.g., 10) sliding window anomalies, the system determines that a sensor is abnormal and performs calibration. If the current sensor number i is less than n-1, calibration can be performed; if i ≥ n-1, calibration cannot be performed due to the lack of a calibration reference sensor, and the air pressure detection system is deemed out of range and unusable.

[0131] The data processing mechanism of this air pressure detection system specifically stipulates that the sensor is only calibrated when 10 consecutive sliding window anomalies occur. This is based on a comprehensive consideration of the sensor's operating characteristics and possible interference factors. The sensor's frequency output may be affected by power supply ripple or electromagnetic interference from the surrounding environment, resulting in spikes or frequency fluctuations. In such cases, the sensor itself may not be experiencing a substantial anomaly, and the anomalies caused by these interferences will usually recover quickly. Therefore, calibrating the sensor based solely on anomalies in a single or a few sliding windows may result in unnecessary calibration operations, affecting system stability and detection efficiency.

[0132] However, if 10 consecutive sliding window anomalies occur, it means the sensor has been exhibiting abnormal behavior for an extended period of time. This indicates that the sensor is unlikely to recover on its own and is likely to be malfunctioning. Calibrating the sensor at this time can promptly identify and resolve sensor issues, ensuring the system can consistently and accurately detect air pressure, effectively improving the reliability of test results and overall system performance.

[0133] Step 4 uses sliding window detection to dynamically analyze historical data and observe the long-term trend differences between the two CMUT sensors. The sliding window can effectively capture local features in the data and accurately identify abnormal data by setting a reasonable threshold, thereby ensuring the reliability and accuracy of the data. In the air pressure detection method, the sliding window anomaly detection adopts a multi-dimensional multi-sensor cross-validation method. On the one hand, the average value of different sensor data is compared (calculated). If the difference exceeds the threshold, an anomaly is identified, enabling horizontal cross-validation of data between sensors. Furthermore, tracking the status of multiple consecutive data sets (10 consecutive anomalies constitute calibration) allows for vertical cross-validation of the data time series. Combining horizontal and vertical cross-validation can promptly identify sensor operating deviations or data anomalies, allowing corrections through calibration and other operations to ensure accurate and reliable pressure measurement data.

[0134] The calibration method is as follows: select the i+2th sensor that is not performing the air pressure detection task to measure the current air pressure. During the measurement process, the measured value needs to be continuously monitored until it reaches a stable state.

[0135] After the air pressure measurement value of the i+2th sensor stabilizes, the measurement result is compared and analyzed with the measurement values ​​of the i-th sensor and the i+1-th sensor that are performing the air pressure detection task.

[0136] When the following conditions are met at the same time, it is determined that the i-th sensor is abnormal or faulty:

[0137] , that is, the measurement value of the i-th sensor and the measured value of the i+2th sensor The difference, compared with the measurement value of the i+1th sensor and The difference is more than 10 times.

[0138] , indicating that the difference between the measured values ​​of the i+1th sensor and the i+2th sensor is within the specified threshold within 0.005 times of the , indicating that the difference between the measured values ​​of the i-th sensor and the i+2-th sensor exceeds 0.005 times of.

[0139] Once the i-th sensor is determined to be abnormal or faulty, the i+1-th sensor is used to calibrate it. Specifically, the current pressure offset is obtained. , a data processing algorithm is introduced to correct the measurement value of the i-th sensor.

[0140] Corrected air pressure after calibration Calculated as follows:

[0141] .

[0142] in is the original measurement value of the i-th sensor before calibration. After this calculation, the calibrated measurement value is obtained , all subsequent measurements of the sensor will be based on this correction value.

[0143] When the following conditions are met at the same time, it is determined that the i-th sensor is abnormal or faulty:

[0144] , that is, the measurement value of the i-th sensor and the measured value of the i+2th sensor The difference, compared with the measurement value of the i+1th sensor and The difference is less than 1 / 10.

[0145] , indicating that the difference between the measured values ​​of the i-th sensor and the i+2-th sensor is within the specified threshold within 0.005 times of the , indicating that the difference between the measured values ​​of the i+1th sensor and the i+2th sensor exceeds 0.005 times of.

[0146] Once the i+1th sensor is determined to be abnormal or faulty, the i-th sensor is used to calibrate it. Specifically, the current pressure offset is obtained. , a data processing algorithm is introduced to correct the measurement value of the i-th sensor.

[0147] Corrected air pressure after calibration Calculated as follows:

[0148] .

[0149] in is the original measurement value of the i-th sensor before calibration. After this calculation, the calibrated measurement value is obtained , all subsequent measurements of the sensor will be based on this correction value.

[0150] If the above two conditions are not met, calibration cannot be performed and the air pressure detection system is judged to be faulty.

[0151] Under normal circumstances, if the measurement value of a sensor deviates significantly from the measurement values ​​of the other two sensors, it can be determined that the sensor may be abnormal. In the above judgment conditions, the ratio of the difference between the two sensors and the other sensor is greater than 10 or less than 1 / 10, and the smaller the difference, the better. The maximum difference is within 0.005 times of the range, and the difference is not within this range as the basis for judgment. 0.005 times of the original value, which can effectively filter out noise interference, ensure the measurement signal is true and reliable, and meet the system accuracy requirements.

[0152] When a sensor detects an anomaly, it is calibrated using the pressure measurements of another normally functioning sensor in the system. The measurement of the normally functioning sensor serves as the calibration benchmark for the abnormal sensor. This is because the normally functioning sensor has higher sensitivity and more accurate measurement results than the (i+2)th sensor. Furthermore, the normally functioning sensor and the abnormal sensor were previously performing measurements simultaneously, so they were exposed to essentially the same operating environmental conditions, such as temperature, humidity, and electromagnetic interference, resulting in greater consistency in their measurement characteristics. This calibration method effectively corrects for any measurement deviations in the abnormal sensor, ensuring the accuracy and consistency of the measurement data from each sensor, thereby improving the measurement accuracy and reliability of the entire air pressure detection system.

[0153] If the calibration is completed, the data in the two sensor buffers will be cleared, and the current sensor will continue to be used for air pressure detection.

[0154] If the subsequent air pressure detection status is normal and the calibration procedure is not triggered, the sensor calibration is considered successful and the subsequent sensor remains in normal operation. If the calibration procedure is triggered again, the sensor is determined to be faulty. At this time, the system will automatically mark the sensor as faulty and immediately activate the i+2th sensor to replace the faulty sensor to perform the task. At the same time, the subsequent sensor automatic switching mechanism in the system will make corresponding parameter adjustments and logic adaptations based on this change, and will no longer call the faulty sensor to ensure the continuity and accuracy of air pressure detection.

[0155] During the calibration process, calibrating the base pressure deviation is essentially calibrating the base frequency offset. A pressure sensor's base frequency offset can be attributed to factors such as manufacturing process variations, resonant structure degradation, and circuit parameter drift. However, due to the inherent structural characteristics of the pressure sensor, the physical model between air pressure and frequency offset remains stable. When the frequency offset is caused by circuit aging or environmental disturbances, and there are no irreversible changes in the resonant geometry or material constitutive relations, digital compensation is used to calculate and apply an appropriate compensation amount to restore the base frequency to its nominal value. In this case, no modification to the physical model is required. If the sensor still exhibits abnormal behavior after calibration, the frequency offset is likely caused by structural damage such as a fractured resonator or degraded material elastic modulus. Once such structural damage occurs, the linearity and sensitivity of the physical model are likely to change, necessitating prompt sensor replacement.

[0156] Step 5. Use the weighted summation method to perform data fusion processing on the output of the selected CMUT sensor. Assuming that the two sensors currently in working state are sensor a and sensor b, and satisfy 1≤a<b≤n, the weighted summation method is used to fuse the output of the two CMUT sensors in working state to obtain the fused air pressure value. .

[0157] ;

[0158] in, is the weight coefficient of the a-th CMUT sensor, is the air pressure measurement value returned by the a-th CMUT sensor through the sliding window, is the weight coefficient of the b-th CMUT sensor, is the air pressure measurement value returned by the b-th CMUT sensor through the sliding window.

[0159] Since CMUT sensors with higher sensitivity exhibit higher measurement accuracy in response to air pressure changes, they are given a larger weight. CMUT sensors with lower sensitivity perform poorly in high-pressure or low-pressure environments, and their weights are set to smaller values.

[0160] The corresponding weights are reasonably allocated based on the sensitivity of each CMUT sensor. The specific weight allocation mechanism is as follows: First, the sensitivity of each CMUT sensor is calculated according to the sensitivity calculation formula. 、 Then, the weight ratio is determined based on the sensitivity, where the weight calculation formula is: , .

[0161] In step 5, the system adopts a multi-sensor weighted fusion mode. When the CMUT sensors are operating normally, the fusion algorithm fully utilizes the air pressure data measured by each CMUT sensor for weighted fusion. Simultaneously, combined with the system's automatic sensor switching mechanism, the sensor combination and weighting coefficients are dynamically adjusted based on real-time air pressure values, thereby ensuring the accuracy and reliability of air pressure measurements.

[0162] Step 6. Output the air pressure measurement result. If only the nth sensor is currently in working state, the system outputs the air pressure measurement result of the nth sensor. If two sensors are currently in operation, the system will output the air pressure measurement result after weighted fusion of the two sensors. .

[0163] The present invention greatly expands the air pressure measurement range. By selecting appropriate sensors, the multi-frequency air pressure sensor can cover a wide range of application scenarios from low pressure to high pressure. Through the automatic sensor switching mechanism, it is ensured that the CMUT sensor with a larger radius in the high pressure range will not be selected for operation, so as to avoid exceeding the range and causing damage to the CMUT sensor. In addition, the present invention effectively improves the accuracy of air pressure measurement through cross-validation of multiple sensor data. The introduction of system redundancy design can reduce the impact of the external environment on the measurement accuracy of the sensor, enhance the fault tolerance of the system, and reduce measurement errors caused by changes in the external environment. The system can automatically switch sensors according to the set working threshold and adjust the weight distribution according to different application scenarios, thereby optimizing the data fusion process. This adaptive mechanism improves the accuracy of air pressure measurement, ensures the use of the most suitable sensor in different air pressure ranges, effectively reduces errors and measurement fluctuations, and improves the reliability and adaptability of the system.

[0164] Of course, the specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-range high-precision air pressure detection system based on a multi-frequency CMUT, characterized in that: Including power management module, multi-frequency air pressure sensor, oscillation circuit, STM32 frequency measurement system and host computer system; The output end of the power management module is connected to the multi-frequency air pressure sensor and is used to provide a DC bias voltage for the multi-frequency air pressure sensor; wherein the multi-frequency air pressure sensor includes a plurality of CMUT sensors; The radius of the resonant unit of each CMUT sensor is different and shows an increasing or decreasing trend in sequence; The air pressure measurement range of each CMUT sensor varies with the radius of the resonant unit. The smaller the radius of the resonant unit, the greater the maximum measurable air pressure. The measurement ranges of all CMUT sensors overlap, and the minimum measurement pressure is 0 Pa. Each CMUT sensor is individually connected to a set of oscillation circuits, and each set of oscillation circuits is connected to the input end of the STM32 frequency measurement system. The STM32 frequency measurement system is connected to the host computer system via wireless communication.

2. The multi-range, high-precision air pressure detection system based on a multi-frequency CMUT according to claim 1, characterized in that: The radius of the resonance unit of the CMUT sensor ranges from 1 μm to 100 μm.

3. A multi-range high-precision air pressure detection method based on a multi-frequency CMUT, using the multi-range high-precision air pressure detection system based on a multi-frequency CMUT according to any one of claims 1 to 2, characterized in that: The multi-range high-precision air pressure detection method based on the multi-frequency CMUT comprises the following steps: Step 1. Initialize each CMUT sensor to ensure that it is stable and working properly. Select the CMUT sensor with the largest operating threshold from among all CMUT sensors to detect the air pressure value of the test environment as the initial air pressure value. Step 2. Compare the measured air pressure value with the operating threshold of each CMUT sensor; If the currently measured air pressure value is only less than the maximum operating threshold value of all CMUT sensors, only the CMUT sensor with the maximum operating threshold value is selected for air pressure detection, and the process goes to step 6; Otherwise, select two CMUT sensors whose operating thresholds match the measured air pressure values ​​and go to step 3; Step 3. Use the STM32 frequency measurement system to collect the frequency data of the two CMUT sensors selected in step 2 in real time. Use the host computer to convert the frequency data of the two CMUT sensors into corresponding air pressure values ​​in real time. Continuously monitor the air pressure values ​​measured by the two CMUT sensors and determine whether they match the operating thresholds of the two currently selected CMUT sensors. If so, proceed to step 4; otherwise, return to step 2 and reselect the CMUT sensors. Step 4. Set the sliding window size and use the sliding window anomaly detection method to retain the valid data measured by the two currently selected CMUT sensors and remove the abnormal data measured by the two CMUT sensors; Step 5. Use the weighted summation method to perform data weighted fusion on the outputs of the two CMUT sensors in the working state; Step 6. Output the air pressure measurement result.

4. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 3, characterized in that: The step 2 is specifically as follows: The number of CMUT sensors is set to n, where n is a natural number, and it is assumed that the radius of the resonant units of all CMUT sensors shows a decreasing trend from the first to the nth CMUT sensor; For the kth CMUT sensor, its working threshold Set to 85% of the kth CMUT sensor pressure limit threshold, 1≤k≤n, then each working threshold satisfies <… …< ; When the air pressure is between and When the value is between , the i-th CMUT sensor and the i+1-th CMUT sensor are enabled for air pressure detection, where 1≤i≤n-1, and ; When the air pressure is between and When , only the nth sensor is enabled for air pressure detection, and in this case it does not participate in the processing of steps 3 to 5, and the air pressure measurement value is directly output after measurement and calculation.

5. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 3, characterized in that: In step 3, the sampling time of the STM32 frequency measurement system is set, and a set of data is collected at each sampling moment, each set of data including the frequency data measured by the two currently selected CMUT sensors; The STM32 frequency measurement system transmits the collected frequency data to the host computer in real time; The host computer converts the received frequency data into the corresponding air pressure value in real time according to the preset air pressure frequency response model; Real-time monitoring of the air pressure value measured by any one of the two CMUT sensors, and determining whether the measured air pressure value matches the operating thresholds of the two CMUT sensors currently used; Assume that the i-th CMUT sensor and the i+1-th CMUT sensor are currently used for air pressure detection; If the air pressure values ​​measured by the two CMUT sensors are both between the operating thresholds of the (i-1)th CMUT sensor and the (i)th CMUT sensor, go to step 4; otherwise, go to step 2 and reselect the CMUT sensor.

6. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 3, characterized in that: The step 4 is specifically as follows: Assume that the i-th CMUT sensor and the i+1-th CMUT sensor are currently used for air pressure detection; Configure independent data buffers for the i-th CMUT sensor and the i+1-th CMUT sensor, respectively. The capacity of each buffer is fixed to N consecutive air pressure values, forming a sliding time window. Synchronously obtain the real-time air pressure measurement data of the i-th and i+1-th CMUT sensors and write the data to the end of the corresponding buffer in chronological order. When the amount of data in the buffer is less than N, the existing data at the beginning of the buffer is copied until N data are added; when the amount of data in the buffer exceeds N, the oldest historical data at the beginning of the buffer is deleted to keep the window length constant at the N latest data; When the amount of data in the buffer exceeds N, the arithmetic mean of the data in the sliding window of the i-th and i+1-th CMUT sensors is calculated after each new data is added. and , and calculate the absolute value of the difference between the two ; like , determine that the data in the current sliding window is normal, and extract the average data of the current sliding window 、 , go to step 5 for data fusion processing; if , determine that the current sliding window data is abnormal, and remove the latest data at the end of the i-th and i+1-th CMUT sensor buffers.

7. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 3, characterized in that: In step 4, if m consecutive sliding windows are determined to be abnormal, the system determines that a sensor is abnormal and performs calibration; At this time, if the calibration conditions are met, calibration is performed; otherwise, the air pressure detection system is determined to be out of range and unusable; if the calibration is successful, the current sensor continues to work; otherwise, an adapted sensor is selected to replace the abnormal sensor.

8. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 7, characterized in that: In step 4, it is assumed that the i-th CMUT sensor and the i+1-th CMUT sensor are currently used for air pressure detection; If i ≥ n-1, calibration cannot be performed due to the lack of a calibration reference sensor, and the air pressure detection system is deemed to be out of range and unusable. If the current sensor number i < n-1, calibration can be performed and the following calibration process is executed. The calibration process is as follows: Select the (i+2)th sensor that is not performing air pressure detection to measure the current air pressure, continuously monitor the measured value until it reaches a stable state, and compare and analyze its measurement result with the measurement values ​​of the (i)th and (i+1)th sensors that are performing air pressure detection; like and , , then the i-th sensor is judged to be abnormal or faulty, and the i+1-th sensor is used to calibrate the i-th sensor; Corrected air pressure after calibration Calculated as follows: ； in, is the original measurement value of the i-th sensor before calibration; The calibrated measurement value is obtained through this calculation , the subsequent measurement value of the i-th sensor is based on this correction value; like and , , then the i+1th sensor is judged to be abnormal or faulty, and the i-th sensor is used to calibrate the i+1th sensor; Corrected air pressure after calibration Calculated as follows: ； in is the original measurement value of the i-th sensor before calibration; The calibrated measurement value is obtained through this calculation , the subsequent measurement values ​​of the i+1th sensor are based on this correction value.

9. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 3, characterized in that: The step 5 is specifically as follows: Assuming that the two sensors currently in working state are sensors a and b, and satisfy 1≤a<b≤n, the weighted summation method is used to fuse the outputs of the two CMUT sensors in working state to obtain the fused air pressure value. ; ； in, 、 are the weight coefficients of the ath and bth CMUT sensors respectively, 、 They are the air pressure measurement values ​​returned by the ath and bth CMUT sensors through the sliding window respectively.

10. The multi-range high-precision air pressure detection method based on multi-frequency CMUT according to claim 9, characterized in that: In step 5, the corresponding weights are reasonably allocated based on the sensitivity of the two CMUT sensors. The weight allocation mechanism is as follows: First, according to the sensitivity calculation formula, calculate the sensitivity of each CMUT sensor separately 、 ; The weight ratio is determined based on sensitivity, where the weight calculation formula is: , .

Images