阅读说明：本技术 一种mems电容式陀螺仪敏感结构品质因数q值的获取方法 (Method for acquiring quality factor Q value of sensitive structure of MEMS capacitive gyroscope ) 是由 李淼 赵建颖 李严峰 杨宇生 于 2019-09-02 设计创作，主要内容包括：本发明公开了一种MEMS电容式陀螺仪敏感结构品质因数Q值的获取方法,设定一个采样时间阈值长度,采样时间阈值长度大于设定时间长度,计算输出信号衰减时间常数：第一步：以所述设定时间长度结束为起点对顺序获得的采样点信号幅值进行判断获得衰减过程连续有效点,直至达到采样时间阈值长度后结束、获得有效点的集合,第二步：建立时间常数计算坐标,将采样点信号幅值数组作为纵坐标赋值衰减过程连续有效点集合,将采样时间间隔数组作为横坐标,做有效点集合线性回归,并由此计算出时间常数。通过对数据的过滤避免传统方式对整个包络线全部数据的处理,减少了数据的处理量,提高了测试效率,解决大数据量计算慢和一致性度不够高的问题。(The invention discloses a method for acquiring a quality factor Q value of a sensitive structure of an MEMS capacitive gyroscope, which comprises the following steps of setting a sampling time threshold length, calculating an output signal attenuation time constant, wherein the sampling time threshold length is greater than the set time length: the first step is as follows: judging the sequentially obtained sampling point signal amplitudes by taking the end of the set time length as a starting point to obtain continuous effective points in the attenuation process, ending the judgment until the sampling time threshold length is reached to obtain a set of the effective points, and the second step of: and establishing a time constant calculation coordinate, taking the sampling point signal amplitude array as a continuous effective point set in the evaluation and attenuation process of the ordinate, taking the sampling time interval array as the abscissa, performing effective point set linear regression, and calculating the time constant. The data are filtered to avoid the traditional method for processing all the data of the whole envelope curve, so that the data processing amount is reduced, the testing efficiency is improved, and the problems of slow calculation of large data amount and low consistency degree are solved.)

1. A method for obtaining quality factor Q value of a sensitive structure of an MEMS capacitive gyroscope comprises the steps of sampling a sampling frequency to obtain an output signal of the gyroscope, calculating an attenuation time constant of the output signal, multiplying the attenuation time constant by the inherent resonant frequency of the MEMS capacitive gyroscope and then multiplying by the inherent resonant frequency of the MEMS capacitive gyroscopeAnd taking an absolute value to obtain a quality factor Q value, wherein: the output signal is an output signal corresponding to a test signal of the inherent resonance frequency of the MEMS capacitive gyroscope, the test signal is a vibration signal applied to the MEMS capacitive gyroscope according to a set time length, the method is characterized in that a sampling time threshold length is set, the sampling time threshold length is greater than the set time length, and the process of calculating the attenuation time constant of the output signal is as follows:

the first step is as follows: and judging the sequentially obtained sampling point signal amplitudes by taking the end of the set time length as a starting point to obtain continuous effective points in the attenuation process, and ending the judgment until the sampling time threshold length is reached to obtain a set of the effective points, wherein the effective point judgment process is as follows: comparing the signal amplitude of the next sampling point with the signal amplitude of the previous sampling point, if the signal amplitude of the next sampling point is smaller than the signal amplitude of the previous sampling point, the next sampling point is an effective point, otherwise, the next sampling point is an invalid point;

The second step is that: and establishing a time constant calculation coordinate, taking the sampling point signal amplitude array as a continuous effective point set in the evaluation and attenuation process of the ordinate, taking the sampling time interval array as the abscissa, performing effective point set linear regression, and calculating the time constant.

2. the acquisition method according to claim 1, wherein the sampling frequency is selected from 100KHz to 1 MHz.

3. The acquisition method according to claim 1, characterized in that the method further comprises: and setting a sampling point signal amplitude auxiliary finishing judgment threshold, and finishing sampling when the sampling time does not reach the length of the sampling time threshold and the acquired sampling point signal amplitude is equal to the sampling point signal amplitude auxiliary finishing judgment threshold.

4. The acquisition method according to claim 1, wherein the sampling point signal amplitude auxiliary ending judgment threshold is 5% of the sampling point signal amplitude acquired from the starting point, and when the sampling time does not reach the length of the sampling time threshold and the acquired sampling point signal amplitude is equal to 5% of the sampling point signal amplitude acquired from the starting point, the sampling is ended.

5. The acquisition method according to claim 1, wherein the effective point judgment process is performed in real time during the sampling process.

Technical Field

The invention relates to a method for acquiring a quality factor Q value of a sensitive structure of an MEMS capacitive gyroscope.

Background

the examination and test of the MEMS capacitive gyroscope device comprises the test of a quality factor Q value of the sensitive structure, wherein the quality factor Q value is a physical quantity representing the damping property of the oscillator, and the quality factor Q value of the sensitive structure of the MEMS capacitive gyroscope device is obtained by multiplying the resonance frequency of the MEMS capacitive gyroscope by an output signal attenuation time constant and then multiplying the resonance frequency by the quality factor Q valueTaking absolute values, wherein: the traditional method for obtaining the decay time constant is to find the envelope curve of the output signal curve, find the average value of the amplitude array of each frequency point of the envelope curve, subtract the average value from each item of the amplitude array to obtain a new array, then carry out square operation on the new array through Hilbert transform, then subtract the array which is not directly squared through Hilbert transform, then carry out root-opening on the array, create an array of time intervals dt corresponding to the array, such as 0, dt, 2dt, 3dt …, the array of time intervals is taken as X, the array obtained by calculating the root-opening number is taken as Y, carry out exponential fitting through the minimum absolute residual error method, calculate the decay value, namely the time constant, the calculation method has the problems of slow calculation and insufficient accuracy when a large amount of data occurs, the efficiency and accuracy of the test are affected, and because of the diversity of the test results, this leads to a particularly poor finding of the envelope, which leads to inaccuracies in the results.

Therefore, a new algorithm is urgently needed to replace the traditional mode, the calculation in the algorithm is reduced, and the problems of slow calculation of large data volume and low consistency are solved.

Disclosure of Invention

The invention aims to provide a method for acquiring a quality factor Q value of a sensitive structure of an MEMS capacitive gyroscope, which avoids the processing of all data of the whole envelope line in a traditional mode by filtering the data, reduces the data processing amount, improves the test efficiency, simplifies the calculation in an algorithm and solves the problems of slow calculation of large data amount and low consistency degree.

In order to achieve the purpose, the technical scheme of the invention is as follows:

A method for obtaining quality factor Q value of a sensitive structure of an MEMS capacitive gyroscope comprises the steps of sampling a sampling frequency to obtain an output signal of the gyroscope, calculating an attenuation time constant of the output signal, multiplying the attenuation time constant by the inherent resonant frequency of the MEMS capacitive gyroscope and then multiplying by the inherent resonant frequency of the MEMS capacitive gyroscopeAnd taking an absolute value to obtain a quality factor Q value, wherein: the output signal is an output signal corresponding to a test signal of the inherent resonant frequency of the MEMS capacitive gyroscope, the test signal is a vibration signal applied to the MEMS capacitive gyroscope according to a set time length, wherein a sampling time threshold length is set, the sampling time threshold length is greater than the set time length, and the process of calculating the attenuation time constant of the output signal is as follows:

The first step is as follows: and judging the sequentially obtained sampling point signal amplitudes by taking the end of the set time length as a starting point to obtain continuous effective points in the attenuation process, and ending the judgment until the sampling time threshold length is reached to obtain a set of the effective points, wherein the effective point judgment process is as follows: comparing the signal amplitude of the next sampling point with the signal amplitude of the previous sampling point, if the signal amplitude of the next sampling point is smaller than the signal amplitude of the previous sampling point, the next sampling point is an effective point, otherwise, the next sampling point is an invalid point;

The second step is that: and establishing a time constant calculation coordinate, taking the sampling point signal amplitude array as a continuous effective point set in the evaluation and attenuation process of the ordinate, taking the sampling time interval array as the abscissa, performing effective point set linear regression, and calculating the time constant.

The scheme is further as follows: the sampling frequency is 100KHz to 1 MHz.

The scheme is further as follows: the method further comprises: and setting a sampling point signal amplitude auxiliary finishing judgment threshold, and finishing sampling when the sampling time does not reach the length of the sampling time threshold and the acquired sampling point signal amplitude is equal to the sampling point signal amplitude auxiliary finishing judgment threshold.

the scheme is further as follows: and when the sampling time does not reach the length of the sampling time threshold value and the obtained sampling point signal amplitude is equal to 5% of the sampling point signal amplitude obtained from the starting point, the sampling is finished.

The scheme is further as follows: the effective point judgment process is carried out in real time in the sampling process.

The invention has the beneficial effects that: the data are filtered to avoid the processing of the whole envelope line data in the traditional mode, so that the data processing amount is reduced, the testing efficiency is improved, the calculation in the algorithm is reduced, and the problems of slow calculation of large data amount and low consistency degree are solved.

The invention is described in detail below with reference to the figures and examples.

drawings

FIG. 1 is a schematic diagram of time constant calculation coordinates;

fig. 2 is a schematic diagram of three attenuation signal image trends.

Detailed Description

A method for obtaining quality factor Q value of sensitive structure of MEMS capacitive gyroscope comprises placing wafer of MEMS capacitive gyroscope on chuck of test probe station, pricking probe of the probe station onto the wafer of MEMS capacitive gyroscope, the probe station having drive circuit connected to the wafer of MEMS capacitive gyroscope via the probe, connecting a signal generator to input end of the drive circuit of the probe station, connecting output of the MEMS capacitive gyroscope to an oscilloscope via the probe, energizing the drive circuit, sending test signal by the signal generator according to resonance frequency of the MEMS capacitive gyroscope, the resonance frequency is known and inherent in the MEMS capacitive gyroscope, the oscilloscope sets a sampling frequency to sample and acquire an output signal of the gyroscope, the accuracy of test data can be ensured by higher sampling frequency, and the sampling frequency is usually 100KHz to 1 MHz;Calculating the decay time constant of the output signal, multiplying the decay time constant by the resonance frequency and then by the resonance frequencyAnd taking an absolute value to obtain a quality factor Q value, wherein: the output signal is an output signal corresponding to a test signal of the natural resonant frequency of the MEMS capacitive gyroscope, the test signal is a vibration signal applied to the MEMS capacitive gyroscope for a set time length, the set time length is usually 1 second, a decay time constant is not related to the magnitude of the vibration intensity, and for the convenience of sampling, a driving intensity, for example, a driving vibration intensity of 5V voltage output, is set.

Wherein the sampling frequency selects a set frequency of the test signal, a sampling time threshold length is set, the sampling time threshold length is greater than the set time length, and in order to fully perform the attenuation process, the sampling time threshold length is usually 3 to 5 times greater than the set time length.

The process of calculating the decay time constant of the output signal is as follows:

The first step is as follows: and judging the sequentially obtained sampling point signal amplitudes by taking the end of the set time length as a starting point to obtain continuous effective points in the attenuation process, and ending the judgment until the sampling time threshold length is reached to obtain a set of the effective points, wherein the effective point judgment process is as follows: comparing the signal amplitude of the previous sampling point with the signal amplitude of the next sampling point, if the signal amplitude of the previous sampling point is greater than the signal amplitude of the next sampling point, the previous sampling point is an effective point, otherwise, the previous sampling point is an invalid point; or vice versa according to the sampling order, namely: comparing the signal amplitude of the next sampling point with the signal amplitude of the previous sampling point, if the signal amplitude of the next sampling point is smaller than the signal amplitude of the previous sampling point, the next sampling point is an effective point, otherwise, the next sampling point is an invalid point;

The second step is that: and establishing a time constant calculation coordinate, taking the sampling point signal amplitude array as a continuous effective point set in the evaluation and attenuation process of the ordinate, taking the sampling time interval array as the abscissa, performing effective point set linear regression, and calculating the time constant.

the effective points are obtained by a difference method, the comparison result of each amplitude point and the next amplitude point is more than or equal to 1 and is effective, and less than 0 and is ineffective, then a set of effective points is obtained through circulation, the result of the previous point is multiplied by 1 and the result of the next point is subtracted, namely the highest point is 1 x (1-0) =1, the other three results are 0, the descending is 0 x (1-0) =0, the ascending is 1 x (1-1) =0, and the lowest point is 0 x (1-1) =0, thereby obtaining the set of effective points. Obtaining a time interval array from the sampling point time interval, extracting and obtaining a corresponding time array from the time interval array according to the set index of the effective point, forming a coordinate shown in figure 1 in one-to-one correspondence with the points of the effective point, namely calculating the coordinate by using a time constant, wherein: the time interval array is the abscissa t and the amplitude data array is the ordinate v. Wherein, the section a is the effective point of the test signal which is not tested, and the section b is after the test signal stops, namely: the set length of time ends with the set of decay significant points measured as the starting point. And then performing linear regression on the set of the effective points, fitting by using a least square approximation to form an attenuation simulation straight line C, and calculating the descending slope of the straight line C, namely the time constant.

To further increase the testing speed, the method further comprises: and setting a sampling point signal amplitude auxiliary finishing judgment threshold, and finishing sampling when the sampling time does not reach the length of the sampling time threshold and the acquired sampling point signal amplitude is equal to the sampling point signal amplitude auxiliary finishing judgment threshold.

Wherein: and the sampling point signal amplitude auxiliary ending judgment threshold is 5% of the sampling point signal amplitude d acquired from the starting point, and when the sampling time does not reach the length of the sampling time threshold and the acquired sampling point signal amplitude is equal to 5% of the sampling point signal amplitude acquired from the starting point, the sampling is ended.

wherein: preferably, the valid point determination process is performed in real time during the sampling process.

The method adopts the extraction of effective points and screens out useless points, thereby greatly reducing the calculation amount and improving the operation efficiency.

Fig. 2 illustrates three attenuated signal image trends, all with many spikes. In the traditional way, one million points of data are processed, and the time is more than 5 seconds. However, under the test of the improved method of the embodiment, the time is less than 1 second, the processing result is consistent with the result of the traditional test method, and the consistency degree is high. Obviously, the test efficiency is greatly improved, and the algorithm of the traditional mode can be completely replaced by the method of the snapshot envelope simulation.