A frequency filtering analysis method for discrete monitoring data

Abstract

This invention provides a frequency filtering analysis method for discrete monitoring data, belonging to the technical field of discrete data processing methods. The aim is to remove noise signals from monitoring data and extract the true data changes caused by variations in load, operating conditions, and other effects by designing a digital filter with a reasonable stopband frequency range. The analysis method is as follows: First, a sensor is installed at the monitoring point and connected to a temperature element. The data measured by the sensor is recorded as the original signal, and the data measured by the temperature element is recorded as the comparison signal. Then, Fourier transforms are performed on the original and comparison signals to obtain their spectra. The common dominant frequency of the original and comparison signals is taken as the stopband frequency. Finally, a digital filter is designed based on the stopband frequency, and the original signal is convolved with the digital filter to obtain the processed signal after noise removal.

Description

Technical Field

[0001] This invention belongs to the technical field of discrete data processing methods, and specifically relates to a frequency filtering analysis method for discrete monitoring data. Background Technology

[0002] Construction projects, both during construction and operation, often employ numerous sensors to monitor structural safety performance and capture changes in structural condition caused by variations in construction conditions and external loads. The monitoring data acquired by these sensors is mostly discrete digital signals, easily affected by external environmental interference. This results in the superposition of noise signals from various interfering factors, distorting the monitoring data and preventing it from directly reflecting the true changes in structural condition. Therefore, filtering the monitoring data is necessary to remove noise signals and accurately reflect the actual changes in structural condition.

[0003] Currently, most noise reduction filtering methods for discrete data employ Gaussian filtering and digital filtering. Gaussian filtering can only smooth discrete data and cannot remove noise signals at specific frequencies. Digital filtering mainly involves designing filters for signal noise reduction, which requires a given stopband frequency range. However, the stopband frequency range is currently mostly determined based on human experience, resulting in low confidence levels of the denoised data. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a frequency filtering analysis method for discrete monitoring data. By designing a digital filter with a reasonable stopband frequency range, noise signals in the monitoring data are removed, and the true data changes caused by variations in load, operating conditions, and other effects are extracted. Mechanical sensors used for building structure monitoring typically have built-in temperature elements and can simultaneously output main measurement data and temperature data. This invention utilizes the temperature data measured by the sensor as a comparison signal for frequency domain filtering of the main measurement data.

[0005] The technical solution adopted by this invention to solve its technical problem is:

[0006] A frequency filtering analysis method for discrete monitoring data includes the following steps:

[0007] Step S1: Install a sensor at the monitoring point. The sensor is connected to a temperature element. The data measured by the sensor is the monitoring data required for engineering monitoring, and it is recorded as the raw signal D. r The data measured by the temperature element is recorded as a comparison signal D. t ;

[0008] Step S2: Convert the original signal D r and contrast signal D tPerform a Fourier transform to obtain the spectrum F of the original signal and the contrast signal. r F t Take F r and F t The same main frequency is used as the stopband frequency f n f n It is a frequency range, represented as f n =[f i :f j ;f m :f n ;......];

[0009] Step S3: Based on the stopband frequency f determined in step S2 above... n Design a digital filter H to filter the original signal D. r The signal is convolved with a digital filter H to obtain the processed signal D after noise removal. f D f =H*D r .

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0011] The frequency filtering analysis method for discrete monitoring data provided by this invention first involves installing a sensor at the monitoring point, connecting the sensor to a temperature element, recording the sensor's measured data as the original signal, and recording the temperature element's measured data as the comparison signal. Then, Fourier transforms are performed on the original and comparison signals to obtain their spectra, and the common dominant frequency of the original and comparison signals is selected as the stopband frequency. Finally, a digital filter is designed using the stopband frequency, and the original signal is convolved with the digital filter to obtain the processed signal after noise removal. This invention removes noise signals from monitoring data and extracts the true data changes caused by variations in load, operating conditions, etc., by designing a digital filter with a reasonable stopband frequency range. By using the temperature element in the sensor's measurement circuit system as a reference, and using its measured temperature data as the comparison signal for filtering the monitoring data, and using the common dominant frequency of the temperature data and the sensor's monitored data as the stopband frequency of the digital filter, the noise influence caused by external interference factors can be completely removed.

[0012] Furthermore, step S2 includes: determining multiple main frequencies of the original signal based on spectral peak values. and multiple main frequencies of the comparison signal Extract the main frequency f that is the same as that of the original signal and the comparison signal. e n =[f e I f e 2f e 3 ...], that is, when At that time, stopband frequency f n The calculation formula is f n =[min(f e n ): max(f e n )).

[0013] Furthermore, the temperature element is built into the sensor.

[0014] Furthermore, the temperature element and the sensor are installed in the same measurement system.

[0015] Furthermore, the ambient temperature, electromagnetic induction, and environmental vibration experienced by the temperature element are consistent with those experienced by the sensor, thereby ensuring that the external interference experienced by the temperature element and the sensor are consistent. Attached Figure Description

[0016] Figure 1 This is an example of a time history curve of sensor monitoring data in a frequency filtering analysis method for discrete monitoring data according to an embodiment of the present invention;

[0017] Figure 2 This is an example of a frequency spectrum diagram of monitoring data in a frequency filtering analysis method for discrete monitoring data according to an embodiment of the present invention;

[0018] Figure 3 This is an example of a comparison curve of filtered monitoring data in a frequency filtering analysis method for discrete monitoring data according to an embodiment of the present invention. Detailed Implementation

[0019] The frequency filtering analysis method for discrete monitoring data proposed in this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of this invention will become clearer from the following description. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise proportions, only used to facilitate and clarify the illustration of the embodiments of this invention. For ease of description, the terms "upper" and "lower" used below are consistent with the upper and lower directions in the accompanying drawings, but this should not be construed as a limitation of the technical solution of this invention.

[0020] Example 1

[0021] The following is combined Figures 1 to 3 This invention provides a detailed description of the frequency filtering analysis method for discrete monitoring data.

[0022] Please refer to Figures 1 to 3 A frequency filtering analysis method for discrete monitoring data includes the following steps:

[0023] Step S1: Install a sensor at the monitoring point. The sensor is connected to a temperature element. The data measured by the sensor is the monitoring data required for engineering monitoring, i.e., the main measurement data, and is recorded as the raw signal D. r The data measured by the temperature element is recorded as a comparison signal D. t ;

[0024] Step S2: Convert the original signal D r and contrast signal D t Perform a Fourier transform to obtain the spectrum F of the original signal and the contrast signal. r F t Take F t and F t The same main frequency is used as the stopband frequency f n f n It is a frequency range, represented as f n =[f i :f j ;f m :f n ;......];

[0025] Step S3: Based on the stopband frequency f determined in step S2 above... n Design a digital filter H to filter the original signal D. r The signal is convolved with a digital filter H to obtain the processed signal D after noise removal. f D f =H*D r .

[0026] Specifically, the frequency filtering analysis method for discrete monitoring data provided by this invention first involves installing a sensor at the monitoring point, connecting the sensor to a temperature element, recording the sensor's measured data as the original signal, and recording the temperature element's measured data as the comparison signal. Then, Fourier transforms are performed on the original and comparison signals to obtain their spectra, and the common dominant frequency of the original and comparison signals is taken as the stopband frequency. Finally, a digital filter is designed using the stopband frequency, and the original signal is convolved with the digital filter to obtain the processed signal after noise removal. This invention removes noise signals from monitoring data and extracts the true data changes caused by variations in load, operating conditions, etc., by designing a digital filter with a reasonable stopband frequency range. By using the temperature element in the sensor measurement circuit system as a reference and its measured temperature data as a comparison signal for filtering the monitoring data, and using the common dominant frequency of the temperature data and the sensor monitoring data as the stopband frequency of the digital filter, the noise influence caused by external interference factors can be completely removed.

[0027] In this embodiment, more preferably, step S2 includes: determining multiple main frequencies of the original signal based on spectral peak values. and multiple main frequencies of the comparison signal Extract the main frequency f that is the same as that of the original signal and the comparison signal. e n =[f e 1 f e 2 f e 3 ...], that is, when At that time, stopband frequency f n The calculation formula is f n =[min(fx) n ): max(f e n )).

[0028] In this embodiment, more preferably, the temperature element is built into the sensor.

[0029] In this embodiment, more preferably, the temperature element and the sensor are installed in the same measurement system.

[0030] In this embodiment, more preferably, the ambient temperature, electromagnetic induction, and environmental vibration experienced by the temperature element are consistent with those experienced by the sensor, thereby ensuring that the external interference experienced by the temperature element and the sensor are consistent.

[0031] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure are within the scope of protection of the claims.

Claims

1. A frequency filtering analysis method for discrete monitoring data, characterized in that, Includes the following steps: Step S1: Install a sensor at the monitoring point. The sensor is connected to a temperature element. The data measured by the sensor is the monitoring data required for engineering monitoring, and it is recorded as the raw signal D. r The data measured by the temperature element is recorded as a comparison signal D. t ; Step S2: Convert the original signal D r and contrast signal D t Perform a Fourier transform to obtain the spectrum F of the original signal and the contrast signal. r F t Take F r and F t The same main frequency is used as the stopband frequency f n f n It is a frequency range, represented as f n =[f i :f j ;f m :f n ;......]; Step S3: Based on the stopband frequency f determined in step S2 above... n Design a digital filter H to filter the original signal D. r The signal is convolved with a digital filter H to obtain the processed signal D after noise removal. f D f =H*D r .

2. The analytical method according to claim 1, characterized in that, Step S2 includes: determining multiple main frequencies of the original signal based on the spectral peak values. and multiple main frequencies of the comparison signal Extract the main frequency that is the same as the original signal and the comparison signal. That is when At that time, stopband frequency f n The calculation formula is:

3. The analytical method according to claim 1, characterized in that, The temperature element is built into the sensor.

4. The analytical method according to claim 1, characterized in that, The temperature element and the sensor are installed in the same measurement system.

5. The analytical method according to claim 1, characterized in that, The temperature element is subjected to the same ambient temperature, electromagnetic induction, and environmental vibration as the sensor.

Images