Self-checking method and self-checking device for accelerometer

Abstract

The invention relates to the technical field of accelerometers, in particular to a self-checking method and a self-checking device of an accelerometer, which comprises the steps that a preset vibration signal is generated through a vibration signal module in a sensor module, and the vibration signal module comprises a driving circuit and a vibrator; the method comprises the steps that a vibration signal module is introduced to generate a preset vibration signal, an accelerometer collects vibration response data excited by a vibrator, and whether the working state of the accelerometer is normal or not is judged on the basis of a comparison result of the vibration response data and preset standard parameters. And the dynamic response data of the accelerometer is compared with the preset standard parameters, so that a more real and more comprehensive self-checking scheme is realized, and the reliability and the user credibility of the accelerometer are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of accelerometers, and more specifically, it relates to a self-checking method and a self-checking device for an accelerometer. Background Art

[0002] The single-axis and multi-axis accelerometer chips sold on the market can be divided into two categories: some have a self-checking function, while others do not. The purpose of the self-checking function is to detect the working state of the accelerometer before use to ensure its normal operation. If the detection result is normal, it can be used continuously; if it is abnormal, the accelerometer needs to be repaired or replaced to ensure the measurement accuracy and reliability of the product. For accelerometers without a self-checking function, due to the lack of effective state verification means, their output data is more likely to be incorrect, leading to misjudgment and affecting the system performance.

[0003] Currently, accelerometers with a self-checking function usually adopt an internal self-checking mechanism. The common practice is to apply a DC signal externally to excite the sensitive elements inside the accelerometer, so that it outputs a specific signal, and to judge whether the accelerometer is normal by detecting the presence or absence of this signal. However, this self-checking method has significant drawbacks:

[0004] The self-checking process completely depends on the internal circuit and theoretical signals, and users cannot visually confirm whether the self-check actually occurs, resulting in doubts about the credibility of the results.

[0005] The DC signal excitation only simulates static conditions and fails to test the dynamic response ability of the accelerometer in a real vibration environment, covering up potential faults (such as abnormal damping materials or frequency-selective failures), thus causing misjudgment.

[0006] Some common accelerometers on the market, such as SCL3400, MXC6235xQ, and BMI260 six-axis sensors, etc., do not have a self-checking function. If these products malfunction during application, it is often difficult to detect in the early stage, increasing the system risk. Therefore, there is an urgent need for a self-checking method applicable to single-axis and multi-axis accelerometers, which can overcome the deficiencies of the existing technology and provide a real and reliable detection means. Summary of the Invention

[0007] To this end, the purpose of the present invention is to provide a self-checking method and a self-checking device for an accelerometer. By introducing a vibration signal module to generate a preset vibration signal and combining the comparison of the dynamic response data of the accelerometer with preset standard parameters, a more real and comprehensive self-checking scheme is realized, thereby improving the reliability of the accelerometer and the user's trust.

[0008] To achieve the above purpose, the present invention provides the following technical solutions:

[0009] A self-checking method for an accelerometer, comprising the following steps:

[0010] S1. Generate a preset vibration signal through a vibration signal module inside the sensor module. The vibration signal module includes a driving circuit and an oscillator.

[0011] S2. Collect the vibration response data excited by the oscillator by the accelerometer to obtain a vibration time-domain signal, and its sampling frequency satisfies , where is the highest frequency of the oscillator, is a coefficient greater than 2;

[0012] S3. Process the vibration time-domain signal to obtain characteristic parameters, including:

[0013] S31. Perform low-pass filtering on the time-domain signal, and the cut-off frequency > ;

[0014] S32. Convert the filtered time-domain signal into a frequency-domain signal, and extract at least one of the main frequency energy, harmonic components and signal-to-noise ratio as a characteristic parameter;

[0015] S4. Judge the working state of the accelerometer based on the comparison result between the characteristic parameter and the preset standard parameter.

[0016] The present invention is further set as: The step S4 specifically includes:

[0017] S41. Dynamically adjust the judgment threshold according to environmental parameters;

[0018] S42. When the characteristic parameter first exceeds the judgment threshold, it is marked as initially abnormal, and 3-5 repeated vibration tests are triggered, and each vibration lasts for 10-500 ms;

[0019] 2]S43. Use the least square method to fit the multiple test data, correct the zero offset and sensitivity parameters of the accelerometer, and obtain the corrected calibration parameters;

[0020] S44. Based on the calibration parameters, re-execute steps S1-S3 for verification tests;

[0021] The present invention is further set as: The sampling frequency has a value range of 10 Hz - 20 kHz;

[0022] The cut-off frequency = 1.2× ;

[0023] When the deviation of the characteristic parameters of the verification test exceeds 80% of the determination threshold, an irrecoverable fault warning signal is generated and the self-check function is disabled.

[0024] The present invention is further configured such that: in the step S1, it further includes detecting the actual vibration amplitude and frequency of the oscillator through a feedback loop, and adjusting the PWM duty cycle or voltage amplitude output by the drive circuit in real time, so that the frequency error of the actual vibration signal is less than 3% and the amplitude error is less than 5%.

[0025] The present invention is further configured such that: in the step S1, it further includes: performing a frequency scan on the oscillator in the range of 50 Hz - 5 kHz, detecting the peak value of the response amplitude of the accelerometer to determine the resonant frequency, and using this resonant frequency as the main frequency of the subsequent vibration signal.

[0026] The present invention is further configured such that: the feedback loop includes:

[0027] Detecting the phase of the drive current through a current sensor connected in series in the oscillator drive circuit, combining the vibration time-domain signal output by the accelerometer, calculating the actual vibration frequency and comparing it with the preset frequency, and dynamically adjusting the duty cycle of the drive signal.

[0028] The present invention is further configured such that: further including after the frequency scan:

[0029] a. With the resonant frequency as the center, using logarithmically spaced frequency points to perform a sine sweep excitation in the range of 0.8 -1.2 range, and the amplitude of the excitation signal is adjusted dynamically according to where is the amplitude of the excitation signal; is the resonant frequency corresponding to the peak value of the drive voltage; is the frequency of the excitation signal;

[0030] b. Synchronously collecting the differential output signal of the accelerometer and the phase difference of the drive current, and generating a response amplitude curve A(f) after removing the carrier interference;

[0031] c. Performing a Hartley transform on the A(f) curve and calculating the standard deviation of the frequency-domain energy distribution as the smoothness index;

[0032] d. Using the dynamic time warping algorithm to calculate the morphological similarity between the measured A(f) curve and the preset standard template curve ;

[0033] e. When satisfying > or < When it is determined that there is an abnormality in the damping material of the accelerometer, where is 1.2 - 1.8; is 0.9 - 0.95; is the preset reference standard deviation.

[0034] The present invention is further configured to: after generating an irrecoverable fault warning signal, encode the fault type as hardware damage or environmental interference overload, and write the fault information into the non-erasable storage area of the accelerometer through an encrypted check code.

[0035] A self-checking device for an accelerometer, comprising:

[0036] A rigid encapsulation housing, which is internally divided into a first cavity and a second cavity;

[0037] A vibrator and a driving circuit board are fixed in the first cavity;

[0038] The second cavity is isolated and installed with an accelerometer and a signal processing module through a metal shielding cover;

[0039] The driving circuit board includes an H-bridge circuit and a closed-loop control chip, and the signal processing module integrates an FFT analysis unit and a calibration database.

[0040] The present invention is further configured to: further include:

[0041] A wireless communication module, which supports receiving external trigger instructions and uploading the self-checking results to a cloud server;

[0042] A multi-axis cooperation interface, which is configured to synchronously trigger the pairwise combination self-checking of a three-axis accelerometer, and generate a global fault alarm signal when the self-checking results of at least two axes are abnormal.

[0043] Comparing with the deficiencies of the prior art, the beneficial effects of the present invention are:

[0044] A real preset vibration signal is generated by a single vibrator inside the sensor module to replace the traditional DC signal excitation. The user can perceive the vibration process and intuitively verify the authenticity of the self-checking. The detection result no longer depends on theoretical assumptions, but is based on the actual dynamic response, significantly improving the reliability and user trust.

[0045] The output of the vibrator is calibrated in real time through closed-loop feedback (frequency error < 3%, amplitude error < 5%), combined with wide-frequency scanning and resonant frequency detection, to ensure the accurate matching of the vibration signal with the characteristics of the accelerometer. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a schematic flow chart of the present invention;

[0047] Figure 2 is a vibration signal start time diagram of the present invention;

[0048] Figure 3 This is the output value diagram of the accelerometer when the first vibrationless signal of the present invention is collected;

[0049] Figure 4 This is the output value diagram of the accelerometer when the first vibration signal of the present invention is collected;

[0050] Figure 5 This is the output value diagram of the accelerometer when the second vibrationless signal of the present invention is collected;

[0051] Figure 6 This is the output value diagram of the accelerometer when the second vibrationless signal of the present invention is collected. Detailed implementation manners

[0052] Refer to Figures 1 to 6 The embodiments of a self-checking method for an accelerometer of the present invention are further described, specifically including:

[0053] The first step: Presetting vibration signal generation and resonance frequency scanning

[0054] The core of this step is to use the vibration signal module integrated inside the sensor module to generate a precisely controllable preset vibration signal, and through the feedback loop and frequency scanning, determine the resonance frequency of the oscillator, laying a foundation for subsequent precise excitation and performance evaluation.

[0055] A preset vibration signal is generated through the vibration signal module inside the sensor module. The vibration signal module includes a drive circuit and an oscillator.

[0056] The drive circuit adopts a full-bridge (H-bridge) circuit design and uses a closed-loop control chip to achieve precise driving of the oscillator. The H-bridge circuit can provide bidirectional driving ability, improving the driving efficiency and response speed of the oscillator. The closed-loop control chip integrates a PWM (pulse width modulation) controller, which can precisely adjust the duty cycle of the output PWM signal, thereby controlling the effective value of the driving voltage. In addition, the drive circuit also integrates overcurrent protection, overvoltage protection, and short-circuit protection circuits to ensure the safety of the module under abnormal working conditions.

[0057] The oscillator is the actuator of the vibration signal, and a piezoelectric ceramic oscillator or an electromagnetic oscillator is adopted. The piezoelectric ceramic oscillator uses the piezoelectric effect to generate mechanical vibration by applying voltage, and has the advantages of fast response speed, small size, and low power consumption, and is suitable for the generation of high-frequency vibration signals. The electromagnetic oscillator uses electromagnetic force to drive a magnet or coil to generate vibration, and can generate a larger vibration amplitude, and is suitable for the simulation of low-frequency large-amplitude vibration signals. The material of the oscillator selects materials with high elastic modulus and low damping coefficient to improve the vibration efficiency and quality factor. The oscillator is fixed in the first cavity through a precision bonding process or a threaded connection to ensure that the vibration signal can be effectively transmitted to the accelerometer.

[0058] To ensure the accuracy and stability of the vibration signal, the vibration signal module integrates a feedback loop.

[0059] In the feedback loop, a current sensor connected in series in the oscillator drive circuit is used to detect the phase and amplitude of the drive current. The vibration time-domain signal collected by the accelerometer itself is used as part of the feedback signal. The closed-loop control chip in the signal processing module is responsible for receiving the feedback signals from the current sensor and the accelerometer and performing calculations.

[0060] Frequency / Amplitude Adjustment Algorithm:

[0061] By using algorithms such as Fourier transform or zero-crossing detection on the phase change of the drive current detected by the current sensor, the frequency information of the drive current is extracted and used as a reference for the actual vibration frequency of the oscillator. At the same time, combined with the vibration time-domain signal output by the accelerometer, the main frequency component in the vibration time-domain signal is extracted using the Fast Fourier Transform (FFT) or spectrum analysis algorithm to further verify and correct the calculation result of the actual vibration frequency.

[0062] The vibration time-domain signal output by the accelerometer directly reflects the vibration amplitude of the oscillator. By performing peak detection, root mean square (RMS) calculation, or integration operation on the time-domain signal output by the accelerometer, the actual vibration amplitude of the oscillator can be obtained.

[0063] The closed-loop control chip compares the calculated actual vibration frequency and amplitude with the preset frequency and amplitude, and calculates the frequency error and amplitude error. If the frequency error exceeds 3% or the amplitude error exceeds 5%, the closed-loop control chip will adjust the PWM duty cycle or voltage amplitude output by the drive circuit in real time. The specific adjustment strategy can adopt the PID control algorithm. According to the magnitude and change trend of the error, the PWM duty cycle or voltage amplitude is precisely controlled, ultimately making the frequency error of the actual vibration signal less than 3% and the amplitude error less than 5% to ensure that the accuracy of the vibration signal meets the self-test requirements.

[0064] Before the formal self-test, the oscillator needs to be frequency scanned to determine its resonant frequency. The drive circuit performs logarithmic sweep excitation on the oscillator in the frequency range of 50 Hz to 5 kHz, detects the peak value of the response amplitude of the accelerometer to determine the resonant frequency, and uses this resonant frequency as the main frequency of the subsequent vibration signal.

[0065] During the frequency scanning process, the signal processing module collects the output signal of the accelerometer in real time and calculates the response amplitude. The response amplitude can be the peak-to-peak value, root mean square value (RMS), or spectrum amplitude of the accelerometer output signal.

[0066] The signal processing module analyzes the acceleration response amplitudes collected during the frequency sweep process and searches for the peaks of the response amplitudes. The frequency corresponding to the peak is the resonant frequency of the oscillator. The resonant frequency characterizes the frequency point at which the oscillator has the highest vibration efficiency at a specific frequency. Using this frequency as the main frequency of the subsequent vibration signal can improve the utilization rate of vibration energy and the self-test sensitivity.

[0067] After determining the resonant frequency, in order to more finely evaluate the performance of the accelerometer, further frequency sweep detection was carried out:

[0068] a. Centered on the resonant frequency , logarithmic frequency interval points are used to perform a sine sweep excitation in the range of 0.8 - 1.2 . This frequency range covers the frequency response characteristics near the resonant frequency, enabling a more comprehensive evaluation of the performance of the accelerometer near the resonant frequency. Moreover, using logarithmic frequency interval points for the sweep makes the frequency points denser near the resonant frequency, allowing for more precise capture of the details of the frequency response curve.

[0069] The amplitude of the excitation signal is dynamically adjusted according to , where is the amplitude of the excitation signal, which refers to the peak driving voltage output by the driving circuit to the oscillator; is the resonant frequency corresponding to the peak driving voltage; is the frequency of the excitation signal, which refers to the frequency of the current sweep point; 0.2 is the scaling factor used to control the amplitude dynamic adjustment range.

[0070] This formula indicates that near the resonant frequency , the amplitude of the excitation signal approaches the value. As the excitation frequency deviates from the resonant frequency , the amplitude V of the excitation signal will gradually increase. This dynamic adjustment strategy aims to compensate for the reduced vibration efficiency of the oscillator at non-resonant frequencies, ensuring that the accelerometer can obtain a relatively constant excitation intensity throughout the sweep range, thereby improving the signal-to-noise ratio and credibility of the frequency response curve A(f).

[0071] b. Generation of the response amplitude curve A(f):

[0072] The signal processing module synchronously collects the differential output signal of the accelerometer and the phase difference signal of the drive current. The accelerometer usually adopts differential output, which can effectively suppress common-mode noise and improve the anti-interference ability of the signal; collecting the differential output signal of the accelerometer can obtain a purer vibration response signal; the drive current signal is detected by a current sensor, and the phase information of the drive current is extracted through a phase demodulation circuit or a digital signal processing algorithm. The drive current phase difference signal reflects the phase relationship between the oscillator drive signal and the vibration response, which helps to analyze the dynamic characteristics of the oscillator.

[0073] The response amplitude curve A(f) after removing carrier interference: Since the excitation signal is a sine wave, the output signal of the accelerometer and the drive current signal both contain carrier components. In order to obtain the true vibration response amplitude, it is necessary to remove the carrier interference. The methods for removing carrier interference include: synchronous demodulation, envelope detection, digital filtering, etc. The quadrature demodulation algorithm can be used to demodulate the accelerometer output signal and the drive current signal to the baseband by using two reference signals with the same frequency and phase and orthogonal to the excitation signal, so as to remove the carrier interference.

[0074] After removing the carrier interference, the response amplitudes corresponding to each frequency point are obtained. Arranging these response amplitudes in the order of frequency can generate the response amplitude curve A(f), and the A(f) curve reflects the sensitivity characteristics of the accelerometer at different frequencies.

[0075] c. Perform the Hartley transform on the A(f) curve: The signal processing module performs the Hartley transform on the generated response amplitude curve A(f). The Hartley transform is an integral transform in the real number domain, similar to the Fourier transform, but its result is a real number and the calculation efficiency is higher, especially suitable for the spectral analysis of real number signals.

[0076] Calculate the standard deviation of the frequency-domain energy distribution As the smoothness index; the standard deviation Characterizes the degree of dispersion of the frequency-domain energy distribution. For an ideal accelerometer, its frequency response curve A(f) should be smooth and the frequency-domain energy distribution should be concentrated.

[0077] d. The signal processing module uses the dynamic time warping (DTW) algorithm to calculate the morphological similarity between the measured A(f) curve and the preset standard template curve ; The DTW algorithm can effectively compare the shape similarity between the measured A(f) curve and the ideal frequency response curve. Even if there are slight offsets or stretches of the measured curve on the frequency axis, it can accurately evaluate its morphological similarity.

[0078] The preset standard template curve represents the shape of the frequency response curve of an ideal accelerometer, which can be obtained through methods such as theoretical simulation, experimental calibration, or historical data statistics. The standard template curve should have characteristics such as smoothness, single peak value, symmetry, etc., reflecting the frequency response characteristics of the accelerometer under normal working conditions.

[0079] The output result of the DTW algorithm is the morphological similarity index, and its value range is usually from 0 to 1. The closer the value is to 1, the more similar the shape of the measured A(f) curve is to the preset standard template curve, and the closer the performance of the accelerometer is to the ideal state.

[0080] e. According to the calculated smoothness index and morphological similarity , as well as the preset threshold , and reference standard deviation , a fault determination is made. When > or < it is determined that the accelerometer has a fault.

[0081] > indicates that the smoothness of the measured A(f) curve is too low, exceeding times the reference standard deviation , indicating that there are obvious fluctuations or noises in the curve, and the performance of the accelerometer is interfered or abnormal. is the smoothness threshold coefficient, and its value range is 1.2 - 1.8, which can be adjusted according to the actual application scenario and the requirements for smoothness.

[0082] < indicates that the morphological similarity between the measured A(f) curve and the preset standard template curve is too low, lower than the threshold , indicating that the curve shape has obvious distortion, the frequency response characteristics of the accelerometer deviate from the normal range, and the performance fails. is the similarity threshold coefficient, and its value range is 0.9 - 0.95, which can be adjusted according to the requirements for the morphological similarity of the curve.

[0083] is the preset reference standard deviation, representing the smoothness standard of the frequency response curve of a normal accelerometer. It can be obtained through statistical analysis of the frequency response curves of a large number of normal accelerometers.

[0084] The second and third steps: Acquisition and parameter extraction of vibration time-domain signals:

[0085] This step mainly evaluates the performance of the accelerometer under normal operating conditions. By collecting the vibration time-domain signal and extracting key parameters, such as the main frequency energy, harmonic components, and signal-to-noise ratio, it further verifies whether the performance of the accelerometer meets the expectations. Among them, to obtain the vibration time-domain signal, its sampling frequency satisfies , where is the highest frequency of the oscillator, is a coefficient greater than 2, which is an engineering extension based on the Nyquist sampling theorem.

[0086] Driven by the self-test device, the accelerometer continuously collects the vibration time-domain signal. The sampling frequency is set in the range of 10 Hz - 20 kHz. The specific sampling frequency is selected according to the highest frequency of the oscillator and the requirements of the signal bandwidth. The sampling frequency needs to satisfy the Nyquist sampling theorem to ensure that the signal is not distorted.

[0087] The signal processing module performs low-pass filtering on the collected time-domain signal. The low-pass filter is used to filter out high-frequency noise and interference and improve the signal-to-noise ratio of the signal. The cut-off frequency is set to > , that is, = 1.2 × , which is 1.2 times the highest frequency of the oscillator under normal operation. If the highest frequency of the oscillator is 5 kHz, the cut-off frequency of the low-pass filter can be set to 6 kHz. The low-pass filter can use a digital filter.

[0088] The signal processing module converts the filtered time-domain signal into a frequency-domain signal using the fast Fourier transform (FFT) algorithm and extracts at least one of the main frequency energy, harmonic components, and signal-to-noise ratio parameters.

[0089] Parameter extraction:

[0090] Main frequency energy: In the frequency-domain signal, the square of the spectral amplitude corresponding to the main frequency represents the main frequency energy. The main frequency energy reflects the degree of concentration of the main energy of the vibration signal. Under normal circumstances, the main frequency energy should account for the vast majority of the total energy.

[0091] Harmonic components: Harmonic components refer to the spectral components at integer multiples of the main frequency. The amplitude of the harmonic components reflects the degree of non-linear distortion of the vibration signal. Under normal circumstances, the harmonic components should be as small as possible.

[0092] Signal-to-noise ratio (SNR): The signal-to-noise ratio (SNR) is the ratio of the main frequency energy to the noise energy. The signal-to-noise ratio reflects the purity of the signal. The higher the SNR value, the better the signal quality. The noise energy can be estimated as the sum of the energies of the spectral components other than the main frequency and its harmonic components in the frequency-domain signal.

[0093] Fourth Step: Dynamic Threshold Adjustment and Fault Alarm

[0094] This step realizes the intelligence and reliability of the self-checking device. By dynamically adjusting the judgment threshold, it adapts to the changes of environmental parameters, and adopts a mechanism of multiple repeated tests and parameter correction to reduce the misjudgment rate, and finally realizes accurate fault alarm and fault information recording.

[0095] The signal processing module dynamically adjusts the judgment threshold according to environmental parameters, such as temperature, humidity, air pressure, external vibration interference intensity, etc.

[0096] Dynamic Threshold Adjustment Mechanism: A mapping relationship model between environmental parameters and judgment thresholds is established in advance. For example, when the temperature rises, the zero bias of the accelerometer increases, and at this time, the judgment threshold for the zero bias deviation should be appropriately relaxed. When the external vibration interference intensity increases, the signal-to-noise ratio decreases, and at this time, the signal-to-noise ratio threshold should be appropriately relaxed. The mapping relationship model can be obtained through experimental calibration, data analysis or machine learning algorithms and stored in the calibration database. The signal processing module dynamically adjusts the judgment threshold according to the real-time collected environmental parameters and the mapping relationship model. This dynamic threshold adjustment mechanism can improve the adaptability and accuracy of the self-checking device under different environmental conditions and reduce the influence of environmental factors on the self-checking results.

[0097] Initial Abnormality Marking and Repeated Vibration Tests: The signal processing module compares the main frequency energy, signal-to-noise ratio or harmonic components of the vibration response data collected for the first time with the preset standard parameters. If the deviation exceeds the judgment threshold, it is marked as the initial abnormality and triggers 3 - 5 repeated vibration tests, with each vibration lasting 10 - 500 ms; In order to reduce misjudgments caused by accidental errors or instantaneous interference, after the initial abnormality occurs, 3 - 5 repeated vibration tests are triggered. The number of repeated tests can be adjusted according to the requirements for the misjudgment rate. Each vibration lasts 10 - 500 ms to ensure enough time to collect stable vibration response signals. The repeated vibration tests can verify the authenticity of the initial abnormality and improve the reliability of fault judgment.

[0098] The signal processing module uses the least squares method to fit the multiple test data to correct the zero bias and sensitivity parameters of the accelerometer, and obtains the corrected calibration parameters;

[0099] Least Squares Method Fitting Parameter Correction: The signal processing module uses the least squares method to fit the vibration response data (such as accelerometer output signal, drive current signal, etc.) collected from multiple repeated vibration tests. The goal of fitting is to correct the zero bias and sensitivity parameters of the accelerometer.

[0100] Zero Bias Correction: Through the fitting algorithm, estimate the constant offset (i.e., zero bias) in the accelerometer output signal and subtract it from the accelerometer output signal to eliminate the zero bias error.

[0101] Sensitivity correction: Through the fitting algorithm, estimate the sensitivity coefficient of the accelerometer, and scale the output signal of the accelerometer according to the sensitivity coefficient to calibrate the sensitivity error.

[0102] Obtain the corrected calibration parameters: After the least squares fitting is completed, the corrected bias and sensitivity parameters are obtained, and these parameters will be used as the new calibration parameters for subsequent accelerometer data processing and self-verification.

[0103] Based on the corrected calibration parameters, re-execute the first step - the third step for verification testing. The purpose of the verification testing is to test the effect of parameter correction and confirm whether the performance of the corrected accelerometer returns to normal.

[0104] Verification test deviation threshold: The deviation threshold of the verification test is set to 80% of the determination threshold, which is more stringent than the initial abnormal determination threshold, further reducing the false positive rate and ensuring the accuracy of the non-recoverable fault alarm.

[0105] Non-recoverable fault alarm signal generation: When the signal processing module determines a permanent fault of the accelerometer, it generates a non-recoverable fault alarm signal. The alarm signal can be output through digital signal output, analog voltage output, or wireless communication module, etc., for external systems or users to perform fault prompt and processing.

[0106] Disable the self-test function: After generating the non-recoverable fault alarm signal, disable the self-test function. Disabling the self-test function can prevent the faulty accelerometer from continuously performing self-tests, causing system resource waste or incorrect information output. Disabling the self-test function can be achieved by setting software flag bits, disconnecting hardware circuits, etc.

[0107] Fault type coding and fault information storage: After generating the non-recoverable fault alarm signal, code the fault type as hardware damage or environmental interference overload, and write the fault information into the non-erasable storage area of the accelerometer through an encrypted check code.

[0108] Fault type coding: The fault type is coded as "hardware damage" or "environmental interference overload".

[0109] Hardware damage: It means that the internal hardware components of the accelerometer fail, such as the failure of the MEMS sensor, the damage of the circuit chip, the aging of components, etc. Hardware damage is usually irreversible and requires replacing the accelerometer.

[0110] Environmental interference overload: It means that the accelerometer is interfered by environmental factors beyond its working range, such as excessive external vibration, too high / low temperature, too strong electromagnetic interference, etc. Environmental interference overload is usually recoverable, and when the environmental conditions return to normal, the performance of the accelerometer returns to normal.

[0111] Encryption check code: Encrypt and verify the fault information using an encryption check code (such as CRC check code, MD5 hash value, etc.) to prevent the fault information from being tampered with or misread, and ensure the integrity and reliability of the fault information.

[0112] Non-erasable storage area: Write the fault type code and the encryption check code into the non-erasable storage area of the accelerometer (such as a specific area of OTPROM or FlashROM). The non-erasable storage area has the characteristics of data not being lost when powered off and cannot be modified by the user, and can permanently record the fault information for subsequent fault tracing and quality analysis.

[0113] A self-checking device for an accelerometer, which includes a first cavity and a second cavity divided inside a rigid encapsulation housing; a vibrator and a drive circuit board are fixed in the first cavity; the second cavity is isolated and installed with an accelerometer and a signal processing module through a metal shielding cover.

[0114] Composition of the signal processing module: The signal processing module integrates an FFT analysis unit and a calibration database.

[0115] FFT analysis unit: A hardware FFT accelerator chip or a software FFT algorithm library (such as the CMSIS-FFT library or the FFTW library). The FFT analysis unit is responsible for converting the time-domain signal into a frequency-domain signal, and performing spectrum analysis and parameter extraction.

[0116] Calibration database: A non-volatile memory (such as FlashROM or EEPROM) is used to store calibration data, preset standard parameters, decision threshold models, fault type code tables, encryption check algorithms and other data. The calibration database can be pre-written with factory calibration data and dynamically update calibration parameters and fault information during the self-check process.

[0117] Wireless communication module: Integrate a wireless communication module (such as a Wi-Fi module, a Bluetooth module, a LoRa module or an NB-IoT module), which supports receiving external trigger instructions and uploading the self-check results to a cloud server.

[0118] Receiving external trigger instructions: The wireless communication module listens for trigger instructions from an external system or a cloud server. The trigger instructions can be periodic self-check instructions, remote manual trigger instructions or event trigger instructions (such as device startup, abnormal environmental parameters, etc.). After the received instructions are decrypted and verified, they are passed to the signal processing module to start the self-check program.

[0119] Upload the self - test results to the cloud server: After the self - test is completed, the signal processing module uploads the self - test results (including performance parameters, fault status, fault type codes, calibration parameters, etc.) to the cloud server through the wireless communication module. The uploaded data uses an encrypted transmission protocol (such as TLS / SSL) to ensure the security of data transmission. The cloud server can store, analyze, visualize, and alarm the received self - test results, facilitating users to remotely monitor the operating status and health of the accelerometer.

[0120] Multi - axis coordination interface: Configure a multi - axis coordination interface (such as an SPI interface, an I2C interface, or a synchronous trigger signal line), configured to synchronously trigger the pairwise self - test of the three - axis accelerometer, and generate a global fault alarm signal when the self - test results of at least two axes are abnormal.

[0121] Synchronous trigger of pairwise self - test of the three - axis accelerometer: The self - test device supports self - testing any two axes of the three - axis accelerometer simultaneously. For example, it can simultaneously self - test the X - axis and Y - axis, Y - axis and Z - axis, and Z - axis and X - axis. Synchronous self - testing can improve the self - test efficiency and reduce the influence of inter - axis interference. Through the multi - axis coordination interface, an external system can synchronously trigger multiple accelerometers for self - testing, realizing distributed and multi - point health monitoring of accelerometers.

[0122] Generation of global fault alarm signal: The signal processing module comprehensively analyzes the self - test results of multiple axes. If the self - test results of at least two axes are abnormal (for example, the smoothness index exceeds the standard or the morphological similarity is lower than the threshold, or the parameter deviation exceeds the judgment threshold), a global fault alarm signal is generated. The global fault alarm signal can be output through digital signal output, analog voltage output, or the wireless communication module, indicating to the external system that there is a fault in the overall accelerometer system and maintenance or replacement is required.

[0123] When the sensor performs the self - test function, it gives a self - test signal. After receiving this signal, the signal processing module first performs data pre - sampling of the accelerometer for a duration, generally 10 - 10000 ms, as Figure 2 shown: The signal processing module starts the oscillator and waits for about 15 milliseconds.

[0124] As Figure 3 and Figure 4 shown, the collected accelerometer data is compared with the previously pre - collected data. At this time, the signal processing module calculates through an algorithm that the accelerometer is working properly and issues a correct self - test signal.

[0125] As Figure 5 and Figure 6As shown, the signal processing module calculates through an algorithm that the accelerometer is not working properly and issues an incorrect self-check signal. Through this self-check signal, the quality of the accelerometer can be truly judged, making it more worry-free and reassuring for customers to use.

[0126] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any ordinary changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A self-checking method for an accelerometer, characterized in that, It includes the following steps: S1. Generate a preset vibration signal through the vibration signal module inside the sensor module. The vibration signal module includes a drive circuit and an oscillator. S2. Collect the vibration response data excited by the oscillator by the accelerometer to obtain a vibration time-domain signal, and its sampling frequency satisfies , where is the highest frequency of the oscillator, is a coefficient greater than 2; S3. Process the vibration time-domain signal to obtain characteristic parameters, including: S31. Perform low-pass filtering on the time-domain signal, with a cut-off frequency > ; S32. Convert the filtered time-domain signal into a frequency-domain signal, and extract at least one of the main frequency energy, harmonic components, and signal-to-noise ratio as a characteristic parameter. S4. Based on the comparison result between the characteristic parameter and the preset standard parameter, judge the working state of the accelerometer.

2. The self-checking method of an accelerometer according to claim 1, characterized in that, The specific steps of S4 include: S41. Dynamically adjust the determination threshold according to the environmental parameters. S42. When the characteristic parameter first exceeds the determination threshold, it is marked as a primary anomaly, and 3 - 5 repeated vibration tests are triggered, with each vibration lasting for 10 - 500 ms. S43. Use the least squares method to fit the multiple test data, correct the zero bias and sensitivity parameters of the accelerometer, and obtain the corrected calibration parameters. S44. Based on the calibration parameters, re - execute steps S1 - S3 for verification testing.

3. The self-checking method of an accelerometer according to claim 2, characterized in that The sampling frequency ranges from 10 Hz to 20 kHz; The cut-off frequency = 1.2 × ; When the deviation of the characteristic parameter in the verification test exceeds 80% of the determination threshold, an irrecoverable fault alarm signal is generated, and the self - test function is disabled.

4. The self-checking method of an accelerometer according to claim 2, characterized in that, In step S1, it also includes detecting the actual vibration amplitude and frequency of the oscillator through a feedback loop, and real - time adjusting the PWM duty cycle or voltage amplitude output by the drive circuit to make the frequency error of the actual vibration signal less than 3% and the amplitude error less than 5%.

5. The self-checking method of an accelerometer according to claim 4, wherein In step S1, it also includes: Scanning the frequency of the oscillator in the range of 50 Hz - 5 kHz, detecting the peak value of the response amplitude of the accelerometer to determine the resonance frequency, and using this resonance frequency as the main frequency of the subsequent vibration signal.

6. The self-checking method of an accelerometer according to claim 4, characterized in that, The feedback loop includes: Detect the phase of the drive current through the current sensor in series in the oscillator drive circuit, combine the vibration time - domain signal output by the accelerometer, calculate the actual vibration frequency and compare it with the preset frequency, and dynamically adjust the duty cycle of the drive signal.

7. A self-checking method for an accelerometer according to claim 5, characterized in that, After the frequency scanning, it further includes: a. With the resonance frequency as the center, use logarithmically spaced frequency points to perform sinusoidal sweep excitation in the range of 0.8 -1.2 . The amplitude of the excitation signal is dynamically adjusted according to , where is the amplitude of the excitation signal; is the resonance frequency corresponding to the peak driving voltage; is the frequency of the excitation signal; b. Synchronously collect the differential output signal of the accelerometer and the phase difference of the drive current, and generate a response amplitude curve A(f) after removing the carrier interference. c. Perform a Hartley transform on the A(f) curve and calculate the standard deviation of the energy distribution in the frequency domain as the smoothness index; d. Calculate the morphological similarity between the measured A(f) curve and the preset standard template curve using the dynamic time warping algorithm ; e. When > or < , it is determined that there is an abnormality in the damping material of the accelerometer, where is 1.2 - 1.8; is 0.9 - 0.95; is the preset reference standard deviation.

8. The self-checking method of an accelerometer according to claim 3, characterized in that, 9. A self-checking device for an accelerometer for implementing the method according to any one of claims 1-8, characterized in that, After generating the irrecoverable fault alarm signal, encode the fault type as hardware damage or environmental interference overload, and write the fault information into the non - erasable storage area of the accelerometer through an encrypted check code. It includes: A rigid encapsulation housing, which is internally divided into a first cavity and a second cavity. The oscillator and the drive circuit board are fixed inside the first cavity. The second cavity is isolated and installed with the accelerometer and the signal processing module through a metal shielding cover.

10. The self-checking device of an accelerometer according to claim 9, characterized in that, The drive circuit board includes an H - bridge circuit and a closed - loop control chip, and the signal processing module integrates an FFT analysis unit and a calibration database. It also includes: A wireless communication module, which supports receiving external trigger instructions and uploading the self - test results to the cloud server. A multi - axis collaboration interface, which is configured to synchronously trigger the pairwise self - test of the three - axis accelerometer, and generate a global fault alarm signal when the self - test results of at least two axes are abnormal.

Images