Vibrating wire sensor frequency measuring device and method based on maximum likelihood estimation method

Abstract

The invention discloses a vibrating wire sensor frequency measuring device and method based on a maximum likelihood estimation method. The device comprises a vibration excitation module, a vibration pickup module and a main control core module. The input end of the excitation module is connected to the output end of the main control core module, and the output end of the excitation module is connected with a vibrating wire sensor; the input end of the vibration pickup module is used for connecting a vibrating wire sensor; the excitation module is used for generating excitation voltage and carrying out sweep frequency excitation on the vibrating wire sensor; the vibration pickup module is used for carrying out data acquisition on a vibration signal of the vibrating wire sensor and transmitting the data to the main control core module; the main control core module is used for controlling the vibration excitation module according to data collected by the vibration pickup module. According to the invention, the frequency measurement of the vibrating wire sensor is more accurate, and meanwhile, the excitation reliability of the vibrating wire sensor and the accuracy of measured data are greatly improved.

Description

Technical Field

[0001] The invention relates to a frequency measurement technology of a vibrating string sensor, and in particular to a frequency measurement device and method of a vibrating string sensor based on a maximum likelihood estimation method. Background Art

[0002] During the operation of pumped-storage power stations, vibrating-string sensors are widely used to monitor the safety of their building structures. With the construction of pumped-storage power stations such as Lianghekou, pumped-storage power stations face challenges such as high altitude, large total installed capacity, and harsh operating environment. Commonly used vibrating-string sensors use the time domain method to measure the output frequency of the vibrating-string sensor. When the signal is interfered by noise, these measurement methods will have large errors. At the same time, the amplitude of the output signal of the vibrating-string sensor gradually decays and the duration is limited. In order to accurately measure the output frequency, the vibrating-string sensor needs to be reliably excited. In order to ensure the long-term reliability of the vibrating-string sensor, the excitation method mostly uses low-voltage sweep frequency excitation.

[0003] There are many methods of swept frequency excitation. The full-frequency excitation method is the simplest, but the excitation quality is poor and the efficiency is low; the frequency-band excitation method can improve the measurement reliability to a certain extent, but it is impossible to predict the frequency range in which the frequency to be measured is located, thus reducing operability; the feedback low-voltage excitation method adopts low-voltage string plucking pre-excitation and swept frequency complex excitation, but due to the poor effect of low-voltage string plucking pre-excitation, the complex excitation is unreliable; the adaptive swept frequency excitation method divides the excitation process into two stages: pre-sweep frequency excitation and complex sweep frequency excitation by changing the incremental step size of the swept frequency pulse sequence. Although this excitation method can greatly improve the excitation quality of complex excitation, the equal-precision frequency measurement method used has poor anti-interference performance and still does not solve the problem of insufficient reliability of pre-excitation itself. The excitation strategy based on spectrum feedback can realize adaptive and reliable excitation of vibrating string sensors, but it increases the frequency measurement time and loses a certain amount of measurement efficiency. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a vibrating-string sensor frequency measurement device and method based on the maximum likelihood estimation method, which makes the frequency measurement of the vibrating-string sensor more accurate, and at the same time greatly improves the reliability of the vibrating-string sensor excitation and the accuracy of the measurement data.

[0005] Technical solution: A frequency measurement device of a vibrating string sensor based on a maximum likelihood estimation method of the present invention comprises an excitation module, a vibration pickup module and a main control core module; the input end of the excitation module is connected to the output end of the main control core module, and the output end thereof is used to connect to the vibrating string sensor; the input end of the vibration pickup module is used to connect to the vibrating string sensor;

[0006] The excitation module is used to generate an excitation voltage and perform sweep frequency excitation on the vibrating string sensor; the vibration pickup module is used to amplify and filter the vibration signal of the vibrating string sensor and transmit it to the main control core module; the main control core module is used to analyze and calculate the amplified and filtered vibration signal, and control the excitation module to perform complex sweep frequency excitation on the vibrating string sensor according to the analysis and calculation results.

[0007] Furthermore, the output end of the vibration pickup module is used to connect to a storage module, which is connected to a main control core module; the main control core module is also connected to a communication module, which is used to realize external transmission of data.

[0008] Furthermore, the vibration pickup module includes a filter amplifier circuit for amplifying and filtering the voltage excited by the vibrating string sensor; the main control core module adopts the maximum likelihood estimation method as the frequency domain measurement method to perform vibration pickup measurement on the output frequency of the vibrating string sensor.

[0009] Furthermore, when the maximum likelihood estimation method is used to perform vibration pickup measurement on the output frequency of the vibrating string sensor, different measurement requirements are met by superimposing different window functions.

[0010] Furthermore, the main control core module uses a power spectrum weighted estimation method to accurately measure the output frequency of the vibrating string sensor.

[0011] Furthermore, it also includes a power supply module, which is connected to the excitation module and the main control core module and is used to supply power to each module of the device.

[0012] Based on the same inventive concept, a method for measuring the frequency of a vibrating wire sensor based on maximum likelihood estimation of the present invention is applied to the above-mentioned device for measuring the frequency of a vibrating wire sensor based on maximum likelihood estimation, comprising:

[0013] Setting sampling frequency parameters of a vibrating wire sensor frequency measurement device based on maximum likelihood estimation;

[0014] The excitation module performs frequency sweep excitation on the vibrating string sensor;

[0015] The vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor;

[0016] The main control core module collects data and measures the frequency of the vibration signal;

[0017] The main control core module calculates the amplitude of the vibrating string and determines whether the vibrating string is reliably excited based on the amplitude of the vibrating string;

[0018] The main control core module uses the frequency of the vibration signal measured by the vibration pickup module as a point estimate According to the asymptotic normal distribution characteristics of the maximum likelihood estimate, the point estimate variance is calculated, the confidence level is determined, and the confidence interval is calculated based on the obtained variance;

[0019] The main control core module uses the confidence interval as the sweep frequency complex excitation interval of the vibrating string sensor, and controls the excitation module to perform complex sweep frequency excitation on the vibrating string sensor;

[0020] The vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor;

[0021] The main control core module performs digital sampling and frequency measurement on the vibrating string sensor to obtain the output frequency of the vibrating string sensor.

[0022] Furthermore, the main control core module collects data on the vibration signal of the vibrating string sensor and measures the frequency of the vibration signal, including:

[0023] The main control core module adopts the maximum likelihood estimation method to pick up the vibration signal of the vibrating string sensor and measure the frequency, and then adopts the power spectrum weighted estimation method to accurately measure the measured frequency.

[0024] Furthermore, when the power spectrum weighted estimation method is adopted, the number of weighted estimation points is determined according to the measurement requirements.

[0025] Furthermore, the main control core module uses a maximum likelihood estimation method to calculate the vibration string amplitude.

[0026] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are:

[0027] The present invention adopts the maximum likelihood estimation method as the spectrum measurement method, which greatly improves the anti-interference ability of the measurement and can make the frequency measurement of the vibrating string sensor more accurate.

[0028] The present invention adopts the confidence interval under 99% confidence level as the swept frequency complex excitation interval. The swept frequency interval is shorter, which shortens the complex excitation time. For the same number of swept frequency steps, the swept frequency step length can reach 0.01 Hz level, which greatly improves the reliability of the swept frequency excitation of the vibrating string sensor and the measurement accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram of the structure of a frequency measurement device of a vibrating-wire sensor based on a maximum likelihood estimation method disclosed in an embodiment of the present invention;

[0030] Figure 2 The present invention discloses a flow chart of a method for measuring the frequency of a vibrating-string sensor based on a maximum likelihood estimation method. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is described in detail below in conjunction with specific implementation methods and the accompanying drawings of the specification.

[0032] Example 1

[0033] like Figure 1 As shown, a frequency measurement device of a vibrating string sensor based on the maximum likelihood estimation method of the present invention includes an excitation module, a vibration pickup module, a main control core module, a storage module, a communication module, and a power supply module. The main control core module has a main control chip to realize functions such as digital sampling, frequency measurement, and data communication; the power supply module includes a power supply, a power switch control circuit, and a low-power control circuit; the excitation module includes a boost circuit and an excitation control circuit; the vibration pickup module includes an amplification circuit and a bandpass filter circuit; the storage module includes ROM, SD, FLASH, etc.; the communication module uses RS485 communication, etc. The input end of the excitation module is connected to the output end of the main control core module, and its output end is used to connect the vibrating string sensor; the input end of the vibration pickup module is used to connect the vibrating string sensor. The excitation module is used to generate an excitation voltage to perform frequency sweep excitation on the vibrating string sensor; the vibration pickup module is used to amplify and filter the vibration signal of the vibrating string sensor and transmit it to the main control core module; the main control core module is the core part of the device, and is used to control the excitation module according to the data collected by the vibration pickup module. The output end of the vibration pickup module is used to connect to the storage module, which is connected to the main control core module. The vibration pickup module amplifies and filters the collected vibration signal and transmits it to the main control core module, which calculates the frequency and then stores the calculated data in the storage module. The communication module and the storage module store and transmit the data collected and calculated by the main control core module. The power module connects the excitation module and the main control core module to power each module of the device. Due to the requirement for the vibrating string sensor to work for a long time, a low-power design must be adopted.

[0034] It is worth emphasizing that Figure 1 In the figure, the dotted box represents that although the two modules have different functions, they can be equivalent to a whole. The storage module only stores the data collected and calculated by the main control core module. The main control core module performs frequency sweep complex excitation based on the stored data. At the same time, the frequency and amplitude obtained by the secondary calculation are also stored in the storage module. The communication module is also connected to this equivalent whole. The communication module is not connected to the vibration pickup module. The communication module here refers to the transmission of the frequency amplitude collected and calculated by the main control core module for the staff to view.

[0035] The vibration pickup module includes a filter amplifier circuit. The purpose of setting the filter amplifier circuit is to limit the measurement frequency to the measurement frequency range (400-6000Hz) of the vibrating string sensor. It is necessary to design a filter circuit and a bandpass filter to filter out the frequency signal that is not within the range. At the same time, because the amplitude of the output signal of the vibrating string sensor is very low, only 2-6mV, it is necessary to amplify its amplitude for easy measurement. The filter amplifier circuit is composed of signal amplifier circuits at various levels. The main control core module uses the maximum likelihood estimation method to perform vibration pickup measurement on the output frequency of the vibrating string sensor. The maximum likelihood estimation method is a spectrum measurement method. Compared with the time domain method, it has higher measurement accuracy and better anti-interference ability. In this embodiment, when the maximum likelihood estimation method is used to perform vibration pickup measurement on the output frequency of the vibrating string sensor, different measurement requirements are met by superimposing different window functions. In this embodiment, different measurement requirements are met by superimposing different Blackman-Harris window functions. This is because the Blackman-Harris window function has better sidelobe attenuation characteristics. Superimposing different window functions means that other window functions can also be used, such as Hanning window; Chebyshev window, etc. On the basis of maximum likelihood estimation measurement, the main control core module also uses a power spectrum weighted estimation method to measure the output frequency of the vibrating string sensor more accurately, which can reduce the maximum error of the measurement frequency. As the number of sampling samples increases, the power spectrum weighted estimation method tends to be unbiased estimation. The power spectrum weighted estimation method in the main control core module can meet the measurement requirements by using three weighted estimation points. The maximum likelihood estimation method used by the main control core module can also obtain the amplitude and phase of the vibrating string. In this solution, digital sampling and frequency measurement using the maximum likelihood estimation method are completed in the main control core module, and the vibration pickup module only needs to amplify and filter.

[0036] The measuring device of the present invention adopts the maximum likelihood estimation theory to carry out vibration pickup measurement on the frequency of the vibrating string sensor, and simultaneously adopts the power spectrum weighted estimation to carry out accurate measurement on the frequency.

[0037] Example 2

[0038] like Figure 2 As shown, a frequency measurement method of a vibrating string sensor based on a maximum likelihood estimation method of the present invention is applied to a frequency measurement device of a vibrating string sensor based on a maximum likelihood estimation method in Example 1, and comprises the following steps:

[0039] S1. Setting sampling frequency parameters of a vibrating-wire sensor frequency measurement device based on maximum likelihood estimation.

[0040] S2. The excitation module performs frequency sweep excitation on the vibrating string sensor.

[0041] S3, the vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor.

[0042] S4. The main control core module collects data on the vibration signal of the vibrating string sensor and measures the frequency of the vibration signal. Specifically, it includes:

[0043] The main control core module uses the maximum likelihood estimation method to pick up the vibration signal of the vibrating string sensor and measure the frequency. Then, the power spectrum weighted estimation method is used to accurately measure the measured frequency.

[0044] Wherein, when the power spectrum weighted estimation method is adopted, the number of weighted estimation points is determined according to the measurement requirements.

[0045] S5. The main control core module calculates the vibration string amplitude, and determines whether the vibration string is reliably excited based on the vibration string amplitude.

[0046] In this embodiment, the main control core module uses the maximum likelihood estimation method to calculate the vibration string amplitude, and can determine whether the vibration string is reliably excited by judging the vibration string amplitude.

[0047] S6, the main control core module uses the frequency of the vibration signal obtained in step S4 as a point estimate According to the asymptotic normal distribution characteristics of the maximum likelihood estimate, the point estimate variance is calculated, the confidence level is determined, and the confidence interval is calculated based on the obtained variance.

[0048] The frequency of the vibration signal measured by power spectrum weighted amplification in step S4 is used as the point

[0049] estimate The pre-excitation effect is good and the re-excitation is more reliable.

[0050] S7. The main control core module uses the confidence interval as the swept frequency complex excitation interval of the vibrating string sensor, and controls the excitation module to perform complex swept frequency excitation on the vibrating string sensor.

[0051] Since the swept-frequency complex excitation interval is short, only a few Hz, the response speed of the swept-frequency complex excitation is fast, which reduces the time the string needs to be excited.

[0052] The vibrating string sensor usually works in the range of [400Hz~6000Hz]. The sweeping frequency excitation methods such as segmented excitation often have large sweeping steps and many sweeping steps. Under the same number of sweeping steps, the sweeping frequency step length can reach 0.01Hz level by using the swept frequency complex excitation method based on confidence interval, which greatly improves the reliability of swept frequency complex excitation and the accuracy of measurement.

[0053] S8, the vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor;

[0054] S9, the main control core module performs digital sampling on the vibrating string sensor to obtain a more accurate output frequency of the vibrating string sensor again.

[0055] The measuring method of the present invention uses a single frequency measurement value as a point estimate and a confidence interval at a 99% confidence level as a sweep frequency complex excitation interval after pre-excitation of the vibrating string sensor according to the asymptotic normal distribution characteristics of maximum likelihood estimation.

[0056] The present invention adopts a sweep frequency excitation method of a vibrating string sensor based on maximum likelihood estimation, shortens the complex excitation interval, enables the string to be rapidly excited, improves the complex excitation efficiency, and greatly improves the reliability of the vibrating string sensor excitation and the accuracy of the measurement data.

Claims

1. A frequency measurement device for a vibrating wire sensor based on a maximum likelihood estimation method, characterized in that: It includes a vibration excitation module, a vibration pickup module and a main control core module; the input end of the vibration excitation module is connected to the output end of the main control core module, and its output end is used to connect to the vibrating string sensor; the input end of the vibration pickup module is used to connect to the vibrating string sensor; The excitation module is used to generate an excitation voltage and perform sweep frequency excitation on the vibrating string sensor; the vibration pickup module is used to amplify and filter the vibration signal of the vibrating string sensor and transmit it to the main control core module; the main control core module is used to analyze and calculate the amplified and filtered vibration signal, and control the excitation module to perform complex sweep frequency excitation on the vibrating string sensor according to the analysis and calculation results.

2. The vibrating wire sensor frequency measurement device based on the maximum likelihood estimation method according to claim 1, characterized in that: The output end of the vibration pickup module is used to connect to the storage module, and the storage module is connected to the main control core module; the main control core module is also connected to the communication module, and the communication module is used to realize external transmission of data.

3. The frequency measurement device of a vibrating wire sensor based on the maximum likelihood estimation method according to claim 1, characterized in that: The vibration pickup module includes a filter amplifier circuit for amplifying and filtering the voltage excited by the vibrating string sensor; The main control core module adopts the maximum likelihood estimation method as the frequency domain measurement method to perform vibration pickup measurement on the output frequency of the vibrating string sensor.

4. The vibrating wire sensor frequency measurement device based on the maximum likelihood estimation method according to claim 3, characterized in that: When the maximum likelihood estimation method is used to measure the output frequency of the vibrating string sensor, different measurement requirements are met by superimposing different window functions.

5. The frequency measurement device of a vibrating wire sensor based on the maximum likelihood estimation method according to claim 1, characterized in that: The main control core module uses a power spectrum weighted estimation method to accurately measure the output frequency of the vibrating string sensor.

6. The frequency measurement device of a vibrating wire sensor based on the maximum likelihood estimation method according to claim 1, characterized in that: It also includes a power supply module, which is connected to the excitation module and the main control core module and is used to supply power to each module of the device.

7. A method for measuring the frequency of a vibrating wire sensor based on a maximum likelihood estimation method, characterized in that: The method is applied to a vibrating wire sensor frequency measurement device based on a maximum likelihood estimation method as described in claim 1, comprising: Setting sampling frequency parameters of a vibrating wire sensor frequency measurement device based on maximum likelihood estimation; The excitation module performs frequency sweep excitation on the vibrating string sensor; The vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor; The main control core module collects data on the vibration signal of the vibrating string sensor and measures the frequency of the vibration signal; The main control core module calculates the amplitude of the vibrating string and determines whether the vibrating string is reliably excited based on the amplitude of the vibrating string; The main control core module uses the frequency of the vibration signal as a point estimate According to the asymptotic normal distribution characteristics of the maximum likelihood estimate, the point estimate variance is calculated, the confidence level is determined, and the confidence interval is calculated based on the obtained variance; The main control core module uses the confidence interval as the sweep frequency complex excitation interval of the vibrating string sensor, and controls the excitation module to perform complex sweep frequency excitation on the vibrating string sensor; The vibration pickup module amplifies and filters the vibration signal of the vibrating string sensor; The main control core module performs digital sampling and frequency measurement on the vibrating string sensor to obtain the output frequency of the vibrating string sensor.

8. The method for measuring frequency of a vibrating wire sensor based on maximum likelihood estimation method according to claim 7, characterized in that: The main control core module collects data on the vibration signal of the vibrating string sensor and measures the frequency of the vibration signal, including: The main control core module adopts the maximum likelihood estimation method to pick up the vibration signal of the vibrating string sensor and measure the frequency, and then adopts the power spectrum weighted estimation method to accurately measure the measured frequency.

9. The method for measuring frequency of a vibrating wire sensor based on maximum likelihood estimation method according to claim 8, characterized in that: When using the power spectrum weighted estimation method, the number of weighted estimation points is determined according to the measurement requirements.

10. The frequency measurement method of a vibrating wire sensor based on maximum likelihood estimation method according to claim 7, characterized in that: The main control core module uses the maximum likelihood estimation method to calculate the vibration string amplitude.

Images