Based on the nonlinear sweep Design of Signal method of real data spectrum signature
Technical field
The present invention relates to vibroseis seismic acquisition technology, particularly relate to a kind of nonlinear sweep Design of Signal method based on real data spectrum signature.
Background technology
Current Seismic Exploration with Vibrator acquisition technique is the trend of land seismic exploration acquisition technique development, because vibroseis exists the plurality of advantages such as frequency energy is controlled, safety and environmental protection, cost are low, operating efficiency is high, so be more and more subject to pursuing of Ge great petroleum geophysics oil company compared with explosive source.Sweep signal is the important factor in order affecting vibroseis earthquake-capturing list big gun quality, and different sweep signals is by the seismic acquisition quality of results different quality.
What adopt in common seismic collection is all linear sweep signal, the frequency spectrum of this signal is more satisfactory spectrum signature, frequency band is very wide, spectrum curve shows as linearly, each frequency content energy is all identical, but the attenuation by absorption degree of actual geologic model to different frequency composition is different, HFS attenuation by absorption is larger, the decay of medium and low frequency part relative absorbance is less, this just makes the linear sweep signal spectrum of Theoretical Design be unavailable in practice, especially in low signal-to-noise ratio area, even if improve several times to HFS in Theoretical Design, still due to strong attenuation by absorption, make the raising of HFS energy very little, cause the waste in design, this redesigns according to the attenuation by absorption degree of actual geologic model to frequency with regard to needing.We have invented a kind of nonlinear sweep Design of Signal method based on real data spectrum signature newly for this reason, solve above technical matters.
Summary of the invention
The object of this invention is to provide a kind of nonlinear sweep Design of Signal method based on real data spectrum signature of effective energy of maximum raising seismic event.
Object of the present invention realizes by following technical measures: based on the nonlinear sweep Design of Signal method of real data spectrum signature, should comprise based on the nonlinear sweep Design of Signal method of real data spectrum signature: step 1, adopts two kinds of sweep length signals to carry out linear sweep; Step 2, analyzes the single big gun frequency spectrum data after these two kinds of linear sweeps, finds the separation improving Energy Efficient frequency by extending sweep length and improve energy idling frequency; Step 3, according to this separation, is optimized the frequency spectrum of the linear sweep signal of routine production; And step 4, when the spectrum curve after optimizing meets the spectrum envelope line features of linear sweep signal list big gun, calculate time-frequency curve according to the spectrum curve after this optimization, calculate the waveform of time domain nonlinear sweep signal according to this time-frequency curve.
Object of the present invention also realizes by following technical measures:
In step 1, the two kinds of sweep length signals carrying out linear sweep are a shorter sweep length, another comparatively long scan length, and the length of these two kinds of sweep length signals adopts identical initial frequency and stops frequency, adopt identical slope and exert oneself, sweep length has a certain distance.
In step 2, obtain which frequency earth penetrating ability in actual work area by comparative analysis strong, which frequency earth penetrating ability is weak; Determine which frequency is obvious by extending sweep length energy lift, namely determine effective band scope, which frequency is not too obvious or basic with regard to no change by extending sweep length energy lift, finds and improves Energy Efficient frequency and this separation improving energy idling frequency by prolongation sweep length.
In step 2, single big gun frequency spectrum long for sweep length and the short single big gun frequency spectrum of sweep length are carried out superimposed display, there are differences to identify which frequency energy, thus determine effective band scope.
In step 3, the linear sweep signal of this conventional production is the linear sweep signal determined by system test before routine is produced, using this separation as the separatrix of revising on frequency spectrum, frequency content higher than this separation suitably reduces energy, frequency content lower than this separation suitably improves energy, the energy improved and reduce is roughly the same, and namely the sweep length of sweep signal remains unchanged, and improves and falls low-energy size by spectrum curve characteristics determined.
In step 4, this spectral enveloping line is the spectral enveloping line of the linear sweep signal list big gun of this conventional production, and envelope feature is line smoothing, and dominant frequency band is given prominence to, and HFS energy is lower slightly.
In step 4, when the spectrum curve after optimizing does not meet the spectrum envelope line features of linear sweep signal list big gun, flow process turns back to step 3.
In step 4, when calculating time-frequency curve, be S (f) according to the distribution function of linear sweep signal spectrum energy in frequency [f1, f2] scope, sweep signal gross energy be sweep length T, and on frequency spectrum energy density with corresponding sweep time density a pass be:
Sweep signal T/F distribution function can be obtained thus
Wherein, f1 is initial frequency, and f2 stops frequency, and t represents frequency distribution in time, and its value represents the sweep time that energy S (f) of corresponding instantaneous frequency f distributes.
Also should comprise based on the nonlinear sweep Design of Signal method of real data spectrum signature, after step 4, the linear sweep signal of this time domain nonlinear sweep signal and this conventional production and application is carried out field test in same place, contrast the fraction frequency of these two kinds of single big guns, frequency spectrum, judge whether obvious the overall signal to noise ratio (S/N ratio) of single big gun promotes by comparative analysis, whether this time domain nonlinear sweep signal HFS data loses larger, signal to noise ratio (S/N ratio) requirement is improved when not reaching, or high frequency loss obvious time, flow process turns back to step 3.
When being analyzed these two kinds of single big guns, when the raising of signal to noise ratio (S/N ratio) reaches requirement and HFS data loss in nonlinear sweep list big gun is little, export this time domain nonlinear sweep signal
The nonlinear sweep Design of Signal method based on real data spectrum signature in the present invention, the method is suitable for the signal to noise ratio (S/N ratio) that seismic data is improved in low signal-to-noise ratio area, which first need to obtain this area's frequency by after prolongation sweep length, its frequency energy can be significantly improved, namely which frequency is dominant frequency, even if after which frequency extends sweep length, its frequency energy still promotes not obvious, by contrasting the frequency spectrum of the linear sweep list big gun of two kinds of different scanning length, obtaining which frequency is dominant frequency, then the frequency spectrum of the linear sweep signal of routine production is optimized, holding signal sweep length is constant, suitable reduction HFS energy, compensate to dominant frequency band part, spectrum curve after optimization requires smooth, and its feature meets the spectrum envelope line features of conventional linear sweep signal list big gun, the spectrometer designed is calculated corresponding time-frequency curve, again time-frequency opisometer is calculated corresponding time domain sweep signal waveform.
The spectral enveloping line of the linear sweep signal list big gun that the method also can utilize actual acquisition to arrive is as the frequency spectrum of nonlinear sweep signal, and this frequency spectrum considers different frequency composition attenuation by absorption feature in the earth formation, designs nonlinear sweep signal by this frequency spectrum.The method has taken into full account the difference of actual geologic model to the attenuation by absorption degree of various frequency content, carry out the design of nonlinear sweep signal targetedly, reduce the deficiency in the design of conventional linear sweep signal, farthest improve the effective energy of seismic event.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of a specific embodiment of the nonlinear sweep Design of Signal method based on real data spectrum signature of the present invention;
Fig. 2 is the spectral contrast figure of the linear sweep signal list big gun of two kinds of different scanning length (6S and 30S) in a specific embodiment of the present invention;
Fig. 3 is the schematic diagram of a specific embodiment neutral line sweep signal list big gun spectral enveloping line of the present invention, what select is production that system test is determined linear sweep signal list big gun used, to the spectrum analysis that whole single big gun carries out, and according to the envelope that the distribution area of energy in frequency spectrum is drawn;
Fig. 4 is the schematic diagram of the transformation of theoretical linear sweep signal frequency spectrum in a specific embodiment of the present invention, by part A energy supplement in linear sweep signal list big gun frequency spectrum to part B, and make improved spectrum curve smooth, curvilinear characteristic meets the spectrum envelope line features shown in Fig. 3;
Fig. 5 is the schematic diagram of nonlinear sweep signal time-frequency curve and nonlinear sweep signal in a specific embodiment of the present invention, left figure is time-frequency curve, can find out that the dominant frequency band part time used is longer, namely energy is relatively high, and right figure is this sweep signal oscillogram;
Fig. 6 is the schematic diagram of this method signal list big gun and the fixed gain of conventional linear sweep signal list big gun Contrast on effect, left figure is this method signal list big gun, right figure is conventional linear sweep signal list big gun, and can find out that this method signal list big gun zone of interest lineups are more clear, signal to noise ratio (S/N ratio) is higher;
Fig. 7 is the schematic diagram of this method signal list big gun and conventional linear sweep signal list big gun Contrast on effect (agc), left figure is this method signal list big gun, right figure is conventional linear sweep signal list big gun, and can find out that this method signal list big gun zone of interest lineups are more clear, signal to noise ratio (S/N ratio) is higher;
Fig. 8 is the schematic diagram of this method signal list big gun and conventional linear sweep signal list big gun Contrast on effect (30-60Hz), left figure is this method signal list big gun, right figure is conventional linear sweep signal list big gun, and can find out that this method signal list big gun zone of interest lineups are more clear, signal to noise ratio (S/N ratio) is higher;
Fig. 9 is the schematic diagram of this method signal list big gun and conventional linear sweep signal list big gun spectral contrast, left figure is this method signal list big gun frequency spectrum, right figure is conventional linear sweep signal list big gun frequency spectrum, can find out that in this method signal list big gun frequency spectrum, dominant frequency band energy is improved (in figure the above area of blue line) significantly, frequency span widens (-24dB) 5Hz;
Figure 10 is the schematic diagram of this method signal section and conventional linear sweep signal section Contrast on effect, upper figure is this method signal section, figure below is conventional linear sweep signal section, can find out that in this method signal section, signal to noise ratio (S/N ratio) is higher, zone of interest imaging is more clear, reflective information is abundanter, is very favorable for instructing oil-gas exploration.
Embodiment
For making above and other object of the present invention, feature and advantage can become apparent, cited below particularly go out preferred embodiment, and coordinate institute's accompanying drawings, be described in detail below.
As shown in Figure 1, Fig. 1 is the process flow diagram of the nonlinear sweep Design of Signal method based on real data spectrum signature of the present invention.
In step 101, two kinds of sweep length signals are adopted to carry out linear sweep.In one embodiment, a shorter sweep length, another comparatively long scan length.The linear sweep signal of the different scanning length of described design, the length requirement of these two kinds of sweep signals adopts identical initial frequency and stops frequency, adopt identical slope and exert oneself, two kinds of signals are only the differences in sweep length, and the sweep length of these two kinds of signals requires a certain distance, such as 6S and 30S, being conducive to which frequency content of comparative analysis like this can better earth penetrating, obtain higher reflected energy, the ability of which frequency content earth penetrating is more weak, decays more.Flow process enters into step 102.
In step 102, analyze (as shown in Figure 2) the single big gun frequency spectrum data after these two kinds of linear sweeps, object is that to obtain which frequency earth penetrating ability in actual work area by comparative analysis strong, and which frequency earth penetrating ability is weak; Determine which frequency is obvious by extending sweep length energy lift, namely determine effective band scope, which frequency is not too obvious or basic with regard to no change by extending sweep length energy lift, finds and improves Energy Efficient frequency and the separation improving energy idling frequency by prolongation sweep length.These two kinds single big gun spectrum signatures are contrasted after collection completes, from contrast, find which frequency content energy can effectively be improved by extending sweep length, it is generally the dominant frequency part of this data, this component frequency is as " improving Energy Efficient frequency ", which frequency content energy still can not get improving by extending sweep length, it is generally HFS, this component frequency is as " improving energy idling frequency ", and find the separation of these two kinds " improving Energy Efficient frequency " and " improving energy idling frequency ", in this example, 10-50Hz is " improving Energy Efficient frequency ".
The frequency spectrum of described comparative analysis two kinds of sweep signal list big guns, single big gun frequency spectrum long for sweep length and the short single big gun frequency spectrum of sweep length are carried out superimposed display, can clearly find out which frequency energy there are differences, thus determine the frequency band range that capacity volume variance is large, be generally the dominant frequency band of this area's seismic data.Flow process enters into step 103.
In step 103, according to this separation, based on the linear sweep signal of conventional production and application, this linear sweep signal is the linear sweep signal for the production of use by determining after system test, is optimized the frequency spectrum of this signal.As shown in Figure 4, the frequency spectrum separation of surface analysis was as the separatrix of revising on frequency spectrum in the past, frequency content higher than separation suitably reduces energy (in Fig. 4 frequency spectrum part A), frequency content lower than separation suitably improves energy (in Fig. 4 frequency spectrum part B), the energy improved and reduce should be roughly the same, namely the sweep length of sweep signal remains unchanged, and improves and falls low-energy size by spectrum curve characteristics determined.
The linear sweep signal of described routine production is the linear sweep signal determined by system test before routine is produced, in frequency spectrum optimization process, suitable reduction HFS, be not the too many of reduction, even if after must guaranteeing that HFS reduces, the HFS impact of actual single big gun is little, and this just needs to be tested further by step 106 to determine.The HFS energy reduced will compensate to dominant frequency band part, to strengthen the energy of dominant frequency band, and requires that the spectrum curve after optimizing keeps smooth.Flow process enters into step 104.
In step 104, judge through amended spectrum curve whether similar with the spectrum envelope line features of actual linear sweep list big gun, similar criterion is that the spectrum curve feature request after optimizing meets the spectrum envelope line features (Fig. 3) of linear sweep signal list big gun as far as possible, and spectrum curve must be smooth.Described spectral enveloping line is the spectral enveloping line of the linear sweep signal list big gun of conventional production in step 103, and envelope feature is line smoothing, and dominant frequency band is given prominence to, and HFS energy is lower slightly.When dissmilarity, flow process turns back to step 103.When similar, flow process enters into step 105.
In step 105, the spectrum curve obtained after this optimization is calculated corresponding time-frequency curve, method is: according to sweep signal spectrum energy at frequency [f1, f2] (f1 is initial frequency, f2 stops frequency) distribution function in scope is S (f), sweep signal gross energy be exactly sweep length T, and on frequency spectrum energy density with corresponding sweep time density a pass be:
Sweep signal T/F distribution function can be obtained thus
T represents frequency distribution in time, and its value represents the sweep time that energy S (f) of corresponding instantaneous frequency f distributes.According to this formula, calculate the time-frequency curve (in Fig. 5 left figure) of this modelled signal.Just can calculate the waveform (in Fig. 5 right figure) of time domain nonlinear sweep signal according to the time-frequency curve obtained, the txt form of output waveform, be loaded into vibroseis control casing and test.Flow process enters into step 106.
In step 106, the linear sweep signal of this nonlinear sweep signal and conventional production and application is carried out field test in same place.Be analyzed these two kinds of single big guns, can this nonlinear sweep signal effectively improve the signal ratio of seismic data, widens frequency span, judges whether the HFS in nonlinear sweep list big gun loses larger.In one embodiment, contrast the fraction frequency of these two kinds of single big guns, frequency spectrum (Fig. 6--Fig. 9), whether effectively improved by comparative analysis dominant frequency band energy, it is larger whether HFS loses, whether the overall signal to noise ratio (S/N ratio) of single big gun promotes obvious, if signal to noise ratio (S/N ratio) does not reach requirement or HFS data loss significant need returns step 103 and again optimizes frequency spectrum, until obtain satisfied data.If the raising of signal to noise ratio (S/N ratio) reaches requirement and HFS data loss in nonlinear sweep list big gun is little, flow process enters into step 107.
In step 107, export final nonlinear sweep signal.Flow process terminates.