Measurement method and data processing

Constant micro-motion measurement can generally be carried out on the surface, underground and buildings, as shown in Figure 5.37. When measuring on the ground or in buildings, it should be ensured that the observation environment is not affected by specific vibration sources (such as traffic and engineering vibration). ) within a certain range. The measuring point should be flat to facilitate the placement and adjustment (horizontal and alignment direction) of the pickup. When measuring in a building, the measuring point should be selected on the main shaft. Underground investigation can be combined with ground investigation. When drilling, the vibration pickup can be on the bedrock surface or the bearing layer of the building.

Fig. 5.37 Schematic diagram of constant fretting measurement method

The measuring system consists of vibration sensor, amplifier, filter, recorder and waveform display. Generally, the vibration pickup adopts speed electromagnetic vibration pickup, and its natural period is1s. If you want to measure two horizontal components (north-south, east-west) and vertical components at one point, you need three vibration sensors. The vibration sensor in the well is cylindrical and has two components (horizontal) or three components (horizontal and vertical). When measuring in a high-rise building, a long-period vibration pickup is needed. The signal output from the vibration sensor is amplified by the amplifier and then input to the recorder. At the same time, there are integral circuits that convert velocity waveform into displacement waveform and differential circuits that convert displacement waveform into acceleration waveform, which can be selected according to different uses. In the data recorder, the micro-motion waveform is recorded. In the case of large short-period interference such as traffic vibration, the interference can be suppressed or eliminated by filters. When measuring, the waveform is used to display the quality of monitoring information, and the waveform with little interference is selected to be input into the recorder for recording.

5.6.2.2 data processing

The basic task of constant micro-motion data processing is to obtain the amplitude of micro-motion and characterize the site vibration characteristics in different periods. There are two main processing and analysis methods, one is periodic frequency analysis and the other is spectrum analysis. At present, spectrum analysis is widely used.

(1) periodic frequency analysis

Periodic frequency analysis is to calculate the frequency of waves with various periodic components, thus obtaining the waveform and periodic characteristics. The specific method is to select the recording segments with good quality from the observation records of about 2min, and draw a roughly symmetrical zero line according to the positive and negative changes of the waveform, and the waveform and the zero line form a series of intersections. Take twice the time difference between two adjacent points as the period of the corresponding wave (the accuracy reaches 0.0 1s). Read them in turn for statistics, and take the period as the abscissa and the number of waveforms of different periods as the ordinate to obtain the frequency curves of various periodic distributions. The period with the highest frequency is called the dominant period, the period with the largest period in recording is called the maximum period, and the period obtained by dividing the number of waves appearing on the recorded waveform by the recording length (time) is called the average period. The analysis results of this method can approximately replace spectrum analysis, and can also eliminate some high-frequency interference. For constant-time fretting with a period less than 1s, the results of the two methods in practical application are the same (Figure 5.38).

Figure 5.38 Comparison between Constant Fretting Frequency Curve and Fourier Spectrum

(2) Spectrum analysis

Because the continuous fretting wave is actually a composite wave composed of a series of frequency components, it is extremely important to know which frequency components are in this composite vibration and the energy of each frequency component. Power spectrum analysis is usually used for constant fretting, which is an irregular vibration at any time.

Let constant fretting be a function of time, expressed by x(t), and convert it into Fourier integral in frequency domain as follows

Environmental and engineering geophysics

For constant fretting with infinite duration and irregular vibration, Fourier integral cannot be obtained directly. It is necessary to divide the record into several segments and carry out Fourier integration for each time segment:

Environmental and engineering geophysics

In addition, the power spectrum P(ω) can be obtained by using x(ω) and its * * * yoke complex X (ω):

Environmental and engineering geophysics

In fact, the average power spectrum can be obtained by excluding the time periods obviously mixed with noise and using the arithmetic average of the power spectrum Pn(ω) of the waveform of each time period:

Environmental and engineering geophysics

Generally, taking 10s as a time period, and stacking it for about 20 times, the relatively stable power spectrum of the observation point can be obtained. There is no essential difference between power spectrum and Fourier spectrum, and they are generally square. It is understandable that power spectrum emphasizes the influence of structure on waves of certain frequency components.