CN104122582B - The method that accurately seismic velocity is asked for using stack velocity - Google Patents

Abstract

The present invention is the method for asking for accurately seismic velocity in geophysical prospecting technology using stack velocity, propose in the area of earth's surface, subsurface geology condition, all the having an impact to stack velocity such as localized variation, the offset distance of datum level, inaccurate static correction, formation lithology and thickness, establish the relation of complex area stack velocity and root mean sequare velocity.The present invention is realized by following steps：1) it is the stack velocity under near surface to ask for datum level；2) with inaccurate and formation heterogeneity the influence of the average elimination static correction of stack velocity in limited bin or superposition interval velocity；3) just drilled with stratiform rate pattern time curve, estimation offset distance forms stack velocity error.Through excessive mouth VSP log well check, depth error within 10 meters, result of calculation reliability, high precision.

Description

The method that accurately seismic velocity is asked for using stack velocity

Technical field

The present invention relates to geophysical exploration technology, belong to seism processing and the technology category explained, be a kind of profit The method that accurately seismic velocity is asked for stack velocity

Technical background

In seismic prospecting, seimic wave velocity is a critically important parameter, particularly seism processing, seismic data Explanation, migration imaging, AVO oil and gas detections etc. are required for seimic wave velocity parameter.In seism processing, obtain substantial amounts of Normal-moveout spectrum data, the stack velocity that treatment person explains is mainly used in dynamic correction, can be the reflection line-ups school of CMP road collections It is flat, but when calculating seimic wave velocity with it, the error with VSP well loggings, synthetic seismic record calibration result is larger；Adjacent superposition speed The change of degree is also very big, and difference in change can reach 350m/s, does not also meet geologic rule, all the error of explanation stack velocity compared with Greatly.To generally believe that ask for seimic wave velocity error with stack velocity larger in seismic prospecting circle, it is impossible to be applied to accurate speed Research, can only do the variation tendency of speed, informative effect.High precision seismic velocity information mainly uses well-log information and VSP Well logging is obtained.

For layered medium, stack velocity (Va) is with the relation of root mean sequare velocity (Vr)：Vr=Va*cos φ, φ are anti- Penetrate inclination layer (referring to《PRINCIPLE OF SEISMIC PROSPECTING WITH》, continental rise Meng chief editor, publishing house of University of Petroleum, in April, 2001, the 1st edition the 3rd time print Brush, the first volume, page 217), in the treatment of existing domestic and international seismic prospecting and interpretation software all be application this formula.It is right Preferable layered medium, this relation is to set up, but in the seismic prospecting of reality, earth's surface is complicated, subsurface formations also have It is localized variation, heterogeneous, therefore have many errors.

To land seismic exploration, surface conditions are more complicated, it is necessary to can just carry out velocity analysis after doing static correction, in benchmark On the P of face (Fig. 1), stack velocity Va, p and the root mean sequare velocity of scanning do not have the relation of above-mentioned formula, because changing ray biography Path (Fig. 2) is broadcast, it is necessary to ask for stack velocity of the datum level under near surface G；Other static correction is inaccurate, subsurface lithologic is locally non- The influence that homogenieity changes and spread length has to stack velocity, eliminating these influences can just obtain high-precision wave speed of the earthquake Degree.

The content of the invention

The purpose of invention is to provide a kind of method that utilization stack velocity asks for accurately seismic velocity

The method that utilization stack velocity of the invention asks for accurately seismic velocity, establishes in earth's surface, subsurface geology The relation of complicated condition area stack velocity and root mean sequare velocity, it is proposed that eliminate datum level, inaccurate static correction, formation lithology and The method that localized variation, spread length of thickness etc. influence on stack velocity, using following steps：

1) stack velocity (Va) and vertical incidence TWT when using below equation to ask for datum level near surface Between (To)：

To=t_{0, p}-Δτ

In above-mentioned formula：X is geophone offset, is distance of the shot point to geophone station, can use either constant, and such as 1000 meters, Δ τ is quiet Correcting value, is in earthquake-capturing arrangement, from ground to the static correction value of datum level P (doing the datum level of stack velocity time spectrum)；V_{A, p} On the basis of face P when stack velocity, t_{O, p}On the basis of face P when vertical incidence two-way travel time；Its advantage is to eliminate base Influence of the standard in face of stack velocity.

2) root mean sequare velocity (Vr) when datum level is near surface is asked for below equation：

Vr=(Va- Δ Vs- Δ Vd- Δ Ve) * Cos φ

In above-mentioned formula：Stack velocity when face is near surface on the basis of Va, φ is the inclination angle in reflecting layer, and Δ Vs is by quiet school Just forbidden to cause stack velocity error near, that the time difference is formed a long way, Δ Vd be by lithology, the change of the locality of thickness cause it is near, The stack velocity error that the time difference is formed a long way, Δ Ve be layered medium in large offseting distance, superposition that non-double curve line causes speed Degree error；Its advantage is established in earth's surface subsurface geology complex area, the relation of stack velocity and root mean sequare velocity.

3) in limited bin (certain areal extent is used as 1 bin), respectively to the superposition interval velocity of same layer (by folding Acceleration Va is calculated through DIX formula) average, elimination Δ Vs, the Random Effect of Δ Vd are carried out, calculate the interval velocity of each layer；Have Limit bin domestic demand has sufficient amount of normal-moveout spectrum, and more precision are higher, at least to there is 4 normal-moveout spectrums；Its advantage is to eliminate Static correction is inaccurate,

The localized variation of lithology and thickness causes stack velocity error.

4) according to earthquake interval velocity model and maximum offset length, the ray time curve just drilled by interval velocity model Overlapped near, remote offset distance away from hyperbola during with speed for Vr+ Δ Vm, ask for speed worst error Δ Vm, estimate Δ Ve values, Its advantage is the error for eliminating Δ Ve.

5) implementing 2), 3) in step, wherein when stratum level of approximation, declination angle ≈ 0, speed is superimposed with limited bin Degree is carried out averagely, eliminating Δ Vs, the Random Effect of Δ Vd by the identical To times, calculates root mean sequare velocity；Its advantage is Quickly, do not require that layer position is carried out on seismic profile explains.

Brief description of the drawings

Fig. 1 is earthquake reflected wave time curve schematic diagram, and X is offset distance, and T is the time, and G is ground or near-earth surface, and P is Datum level during normal-moveout spectrum is done, Δ τ is static correction value of the ground G to datum level P.

Fig. 2 is homocentric seismic propagation path schematic diagram of HORIZONTAL LAYERED MEDIUM WITH HIGH ACCURACY, V₁、h₁It is respectively the seismic wave of ground floor Speed and thickness, V₂、h₂Respectively the seimic wave velocity and thickness of the second layer, Vi, hi be respectively i-th layer seimic wave velocity and Thickness, Vn, hn are respectively the seimic wave velocity and thickness of n-th layer, have n-layer, and G is ground, and P is unified base when doing normal-moveout spectrum Quasi- face or floating datum, Vo, ho are respectively the filling speed and datum level P of static correction to the depth displacement on ground.

Fig. 3 is the average speed comparison diagram of the B1 well seismic wave average speeds that the present invention is asked for and VSP actual measurements, and V is earthquake Ripple average speed, To is vertical incidence two-way travel time.Solid line is the B1 well seismic wave average speeds that the present invention is asked in figure, Dotted line is the average speed of B1 wells VSP actual measurements.

Fig. 4 is the average speed comparison diagram of the A1 well seismic wave average speeds that the present invention is asked for and VSP actual measurements, and V is earthquake Ripple average speed, To is vertical incidence two-way travel time.Solid line is the B1 well seismic wave average speeds that the present invention is asked in figure, Dotted line is the seismic wave average speed of A1 wells VSP actual measurements.

Specific embodiment

The present invention is realized by following technical step：

1) stack velocity spectrum is explained, deriving static correction values are collected in reflecting layer on explanation of seismic section.

2) stack velocity Va and vertical incidence two-way travel time when using below equation to ask for datum level near surface To：

To=t_{0, p}-Δτ

In above-mentioned formula：X is geophone offset, is distance of the shot point to geophone station, can use either constant, such as 1000 meters；Δ τ is quiet Correcting value, is in an arrangement, from ground to the static correction value of datum level P (doing the datum level of stack velocity time spectrum)；V_{A, p}It is base Stack velocity during quasi- face P, t_{O, p}On the basis of face P when vertical incidence two-way travel time.

3) root mean sequare velocity Vr when using below equation to ask for datum level near surface：

Vr=(Va- Δ Vs- Δ Vd- Δ Ve) * Cos φ

In above-mentioned formula：Stack velocity when face is near surface on the basis of Va；φ is reflection inclination layer；Δ Vs is static correction speed Degree is forbidden to cause near, the time difference, the stack velocity error of formation a long way；Δ Vd is lithology, the change of the locality of thickness causes near, remote The road time difference, the stack velocity error of formation；Δ Ve be layered medium in large offseting distance, the stack velocity that non-double curve line causes Error.

4) according to well-log information and composite seismogram calibration result, and stack velocity data, first major floor in Ding Yigegai areas Interval velocity or average speed curve, estimate each layer initial angle.

5) the superposition interval velocity Vi of each layer is calculated by stack velocity according to DiX formula.

6) in limited bin, the superposition interval velocity to same layer carries out averagely, eliminating Δ Vs, the Random Effect of Δ Vd, The interval velocity of each layer is calculated, seismic interval velocity model is obtained.Using certain areal extent as 1 bin.

7) according to earthquake interval velocity model and maximum offset length, by compare during ray that interval velocity model just drilling away from Curve and when away from hyperbola, estimation Δ Ve

8) stratigraphic dip φ, repeat step 3 are calculated according to the new seismic interval velocity model that obtains), 5), 6) and, 7), 8) and until layer Rate pattern is varied less, that is, obtain final interval velocity model.

9) such as stratum level of approximation, during declination angle ≈ 0, with limited bin, stack velocity put down by the identical To times , Δ Vs, the Random Effect of Δ Vd are eliminated, root mean sequare velocity is calculated, interval velocity is asked for according to DIX formula, be converted into average Speed, is not required to do step 4), 5), 6), 8) work.

9) combined with the rate pattern of static correction final interval velocity model is obtained, so that it may which it is unified base to obtain datum level Interval velocity model under quasi- face.

Embodiment 1

The B1 wells well point seimic wave velocity asked for the present invention, centered on the well point, with 20 stack velocities in bin Data based on spectrum, the average speed tried to achieve, the earthquake average speed that it is surveyed with B1 wells VSP contrasts (Fig. 3), itself and VSP realities Survey error and be less than 10 meters, illustrate that precision is very high.

Embodiment 2

The A1 wells well point seimic wave velocity asked for the present invention, centered on the well point, with 14 stack velocities in bin Data based on spectrum, the average speed tried to achieve contrasts (Fig. 4) with the earthquake average speed of A1 wells VSP actual measurements, and itself and VSP are surveyed and missed 10 meters or so are differed from, illustrates that precision is very high.

Claims (2)

1. the method that accurately seismic velocity is asked for using stack velocity, it is characterised in that the described method comprises the following steps：

(1) the root mean sequare velocity Vr under using below equation to ask for datum level near surface：

Vr=(Va- Δ Vs- Δ Vd- Δ Ve) × Cos φ

Wherein：Vr is that datum level is the root mean sequare velocity under near surface；Va is that datum level is the stack velocity under near surface；φ is The inclination angle in reflecting layer；Δ Vs is to be forbidden the stack velocity error for causing by static correction；Δ Vd is to be become by lithology, the locality of thickness The stack velocity error that change causes；Δ Ve be HORIZONTAL LAYERED MEDIUM WITH HIGH ACCURACY in large offseting distance, the stack velocity that non-double curve line causes Error；

(2) in limited bin, the superposition interval velocity to same layer carries out averagely, eliminating Δ Vs, the Random Effect of Δ Vd respectively, Calculate the interval velocity of each layer；Δ Vs is to be forbidden to cause stack velocity error by static correction, and Δ Vd is by lithology, the part of thickness Property the stack velocity error that causes of change；

(3) according to earthquake interval velocity model and maximum offset length, the ray time curve just drilled by interval velocity model with Stack velocity overlaps away from hyperbola when being Vr+ Δ Vm near, remote offset distance, asks for speed worst error Δ Vm, estimates by big Offset distance causes stack velocity error delta Ve, Vr to be root mean sequare velocities when datum level is near surface.

2. method according to claim 1, it is characterised in that folded under using below equation to ask for datum level near surface Acceleration Va and vertical incidence two-way travel time To：

$Va=\frac{{\mathrm{xV}}_{a,p}}{\sqrt{{(\sqrt{x^2+V_{a,p}^2{t}_{0,p}^2}-{\mathrm{\Δ\τV}}_{a,p})}^2-{(t_{0,p}-\mathrm{\Δ}\mathrm{\τ})}^2{V}_{a,p}^2}}$

To=t_{0, p}-Δτ

Wherein：Va is that datum level is the stack velocity under near surface；To is that datum level is the vertical incidence round trip trip under near surface The row time；V_{A, p}On the basis of stack velocity under the P of face, datum level P is the datum level of stack velocity spectrum；t_{0, p}On the basis of under the P of face Vertical incidence two-way travel time；X is geophone offset, is distance of the shot point to geophone station；Δ τ for be in common midpoint gather, From ground to the static correction value of datum level P.