Method of Hydrocarbons Search (Variants) and Method of Reservoir
Depth Determination
This invention relates to seismic survey, more specifically, to the search of hydrocarbons, and can be used for the search of hydrocarbons in mainland, on the shelf and in water basins, for the location of hydrocarbon/water boundary during hydrocarbon extraction and for the control of hydrocarbon storage in natural underground caverns.
At present almost the entire scope of works related to the search of hydrocarbons are in any way related to seismic survey. Conventionally, seismic survey includes the registration of the propagation of artificially generated seismic oscillations in the ground with subsequent mathematical processing of the data. Existing seismic survey methods currently use seismic oscillations with a frequency of above 10.0 Hz. Over the past period of using these frequencies for seismic survey, hardware for the generation and recording of these seismic oscillations and the mathematical formalism for the processing of the resultant data have been well elaborated. The seismic oscillations are generated with vibrators or by blasting. Blast works require drilling blastholes for the placement of explosives. This method adversely affects the environmental situation in the vicinity of the survey area. Moreover, the ratio of efficient seismic prediction with the use of the existing methods and techniques of seismic survey is within 0.5. Hence, at least every second well drilled on the basis of
conventional seismic survey results is located erroneously. Along with the loss of resources for well drilling this bears irrecoverable and unjustified damage to the environment.
Known is method of Vibroseis survey for the search of hydrocarbons (RU Patent 2045079). According to this method seismic oscillations are excited with a vibrator and the seismic signal is recorded with three-component seismic receivers and mathematically processed, wherein the seismic oscillations are generated at frequencies of 2 to 5 Hz and for at least 3 minutes, the seismic signal is recorded for at least 20 minutes before the onset of the seismic oscillations and for at least 5 minutes after the termination of the seismic oscillations, the seismic signal is the seismic background of the earth, and the presence of a reservoir is determined by an increase in the area under respective signal components in the combined spectrum recorded during seismic background and signal registration after the onset of the seismic oscillations as compared with the background and signal spectrum before the onset of the seismic oscillations.
Disadvantage of this method is its low information value which does not allow one to determine the depth of the reservoir and the complexity of the mathematical processing of the signals because noise cannot be separated from the proper signals.
Known also is hydrocarbon reservoir search method (RU Patent 2161809). According to this method seismic oscillations are excited with a vibrator at frequencies of 2 to 5 Hz, the seismic signal is recorded with three-component seismic receivers capable of recording the seismic signal in the infrasonic range and located at a distance of
not more than 500 m from one another and not more than 500 m from the vibrator, the registration being at frequencies of 2 to 5 Hz for all the three components simultaneously both before and after the generation of the seismic signal, and the presence of a reservoir is determined by the appearance of a spectral anomaly in at least one of the components in the spectrum recorded during seismic signal registration after the onset of the seismic oscillations as compared with the background and signal spectrum before the onset of the seismic oscillations.
Disadvantage of this method is its low information value which does not allow one to determine the depth of the reservoir and the complexity of the mathematical processing of the signals because noise cannot be separated from the proper signals.
Known is hydrocarbon search method (RU Patent 2217778, 2003) according to which the seismic noise of the Earth is recorded and the data are analyzed. Preliminarily, the standard waveform of the seismic signal (the energy spectrum) is determined for an a priori known reservoir location. A similar signal is determined for a presumable reservoir location. The time range of the recorded signal is split into discrete portions, and each discrete portion is analyzed for the presence of a standard seismic signal waveform and for the presence of seismic signal distortion having manmade nature. The discrete portions that do not contain the standard seismic signal waveform and the discrete portions that contain manmade nature distortions are excluded from further analysis. The remaining portions are analyzed to conclude on the presence or absence of a hydrocarbon reservoir.
Disadvantage of this method is its low accuracy.
Therefore the object of this invention is to increase the accuracy of detecting hydrocarbon reservoirs and provide for the determination of their depth in order to reduce the percentage of erroneously drilled wells, and also to provide for the control of the operation of productive wells and gas caverns in the oil and gas industry.
As one embodiment of this invention it is suggested to use the hydrocarbon search method wherein seismic oscillations in the surface layer of the Earth are generated and recorded with seismic receivers capable of recording the seismic signal in the 0.1 to 20 Hz range and located at a distance of 50 to 500 m from one another, the registration being for all the components simultaneously, the time range of the signal recorded in the presumably productive area is split into discrete portions synchronized for all the seismic receivers, the spectral characteristic is calculated for each of the portions to obtain a discrete sequence, each discrete portion is analyzed for the presence of seismic signal distortion having manmade nature and an event related to a reservoir signal, the discrete portions that do not contain event related to a reservoir signal in each of the respective component records and the discrete portions that contain manmade nature distortions are excluded from further analysis, and the remaining portions are analyzed to conclude on the presence or absence of a hydrocarbon reservoir. Preferably, the method embodiment comprises measurement of seismic oscillations in an area that is a priori known not to contain hydrocarbons, and the presence of oil or gas is determined from the presence of deviations in the spectral characteristic as compared with that of the area that is a priori known not to contain hydrocarbons. The
method can be implemented either in mainland or in a water basin, wherein the seismic receivers are located in the ground or on the basin bed, respectively, or deepened into the subsurface layer, in the water and/or installed on vessels in areas where the vessel hull is the least subject to intrinsic oscillations, the vessels being equally distant from the seismic oscillation source.
As another embodiment of this invention it is suggested to use the hydrocarbon search method wherein seismic oscillations in the surface layer of the Earth are generated and recorded with seismic receivers capable of recording the seismic signal in the 0.1 to 20 Hz range, the seismic oscillations being generated in the 1 to 20 Hz range, the receivers being located at a distance of 50 to 500 m from one another and 50 to 500 m from the seismic oscillation source, the registration being in the 0.1 to 20 Hz range for all the components both before and after the generation of seismic oscillations, the time range of the signal recorded in the presumably productive area is split into discrete portions synchronized for all the seismic receivers, the spectral characteristic is calculated for each of the portions to obtain a discrete sequence, each discrete portion is analyzed for the presence of seismic signal distortion having manmade nature and an event related to a reservoir signal, the discrete portions that do not contain event related to a reservoir signal in each of the respective component records and the discrete portions that contain manmade nature distortions are excluded from further analysis, and the remaining portions are analyzed to conclude on the presence or absence of a hydrocarbon reservoir. Preferably, the microseismic noise of the Earth is measured, and the presence of hydrocarbons is determined from the
presence of deviations in the spectral characteristic in at least one of the components of the signal recorded during and/or after oscillation generation as compared with the seismic signal measured before the generation. The method can be implemented either in mainland or in a water basin, therefore the seismic receivers are located in the ground or on the basin bed, respectively, and/or installed on vessels in areas where the vessel hull is the least subject to intrinsic oscillations, the vessels being equally distant from the seismic oscillation source. In either embodiment of this method the seismic oscillation receivers are usually grouped and synchronized. Moreover, during the mathematical processing of the recorded signal the latter is split into discrete portions the duration of which is at least 2-3 periods of the least frequency signal in the range considered.
Said objective can also be achieved with the method of hydrocarbon reservoir depth determination wherein at least four seismic oscillation receivers are installed that are capable of recording infrasonic range oscillations and the seismic signal is recorded, said seismic signal being the microseismic noise of the Earth, the seismic receivers capable of recording both vertical and horizontal infralow frequency oscillations are installed over the preliminarily detected source of microseismic activity of an oil/gas reservoir in the nodes of a preferably convex quadrangle for a period that is long enough to record a statistically reliable noise signal in the infralow frequency range generated by the oil/gas reservoir, seismic oscillations are generated with an oscillation source in the 1 to 10 Hz range, the spectral densities of the vertical and horizontal components are calculated, along with the spectral densities of the crosscorrelation
functions of the infralow frequency oscillations recorded and then, by solving the inverse problem of acoustic oscillation propagation from a symmetrical cylindrical source, the depth of the hydrocarbon reservoir is determined.
For the first embodiment of the invention related to hydrocarbon search in mainland, the following is suggested. Over the presumed hydrocarbon reservoir area, at least one seismic oscillation receiver is placed that is capable of recording oscillations in the infrasonic frequency range for at least one component, and the spectral characteristic of the Earth microseismic noise is calculated simultaneously for all of the receivers used at frequencies from 0.1 to 20 Hz for all the components being measured. Preferably the registration is repeated at other points over the presumed hydrocarbon reservoir area. The presence of a reservoir is determined from change in the spectral characteristic of the seismic signal or different combinations of such signals capable of reducing the contribution from the noise. For the purpose of this invention the term spectral characteristic shall mean functions or the integrity thereof obtained as a result of the spectral analysis of the useful signal wherein said analysis may include Fourier transformation (wavelet analysis) or deconvolution of the useful signal into an only asymptotically converging series. The useful signal in terms hereof shall be the modified signal of microseismic oscillations after mathematical processing to several algorithms including those described below for noise elimination and reservoir signal separation. For example, the spectral characteristic can be the spectral power of the measured signals and/or their crosscorrelation or their combinations that reduce
the noise. The presence of a reservoir can be determined from change in the spectral power JjJJ) of the useful signal at the frequencies being measured, from an increase in the correlation of the useful signal of respective components (at least one of them) in different observation points relative to the useful signal for an area that a priori does not contain any hydrocarbon resources (figs 1 and 2). The crosscorrelation of the useful signal can be characterized, for example, by the correlation coefficient kAB(f) and hence related to another spectral characteristic of the signal, i.e. the spectral density, the crosscorrelation function JΛB{f) by the following ratio:
where A and B are two observation points and/is the frequency. Combination of these spectral characteristics that reduces the effect of noise can be exemplified as follows:
(2) where f, and fb are the top and bottom useful range limits, respectively, and v and p are the recording indices before and after (and during) the excitation, respectively. The parameter /?; characterizes change in the radiation energy fluence for one of the components after the excitation of the medium with the source of seismic oscillations.
After recording of the Earth microseismic noise the receivers are transferred to new measurement points observing the same receiver installation requirements, and the Earth microseismic noise recording procedure is repeated.
Figure 2 shows mathematically processed results of Earth microseismic noise recording in 4 observation points over a potentially gas producing area in the South-Eastern section of the pre- Urals flexure. This figure illustrates the parameter that characterizes change in the spectral characteristic in the measurement points relative to the spectral characteristic in a point that is a priori beyond the hydrocarbon reservoir region (near a dry well). Drilling in the vicinity of the point A has confirmed the presence of hydrocarbons.
To obtain the values in the right-hand part of Eqs. (1) and (2) the time range of useful signal registration recorded over the presumable hydrocarbon reservoir area is split into discrete portions synchronized for all the seismic receivers, the spectral characteristics of each discrete portion are calculated to obtain a discrete sequence, and each discrete portion is analyzed for the seismic signal distortion having manmade nature and an event related to a reservoir signal. The discrete portions that do not contain event related to a reservoir signal in each of the respective component records and the discrete portions that contain manmade nature distortions are excluded from further analysis, and the remaining portions are analyzed to conclude on the presence or absence of a hydrocarbon reservoir. An event related to signal coming from a reservoir is determined as to a high extent of likelihood definitively treated ratio of the spectral characteristic of the seismic oscillations in the useful frequency range. For example, such
an event can be reduction in the angle between the normal to the surface and the shifting velocity vector of the oscillations being measured in each discrete relative to the respective angle averaged over all the splitting portions in the useful frequency range. This filtering allows reduction of the effect of noise on calculation results and increases the accuracy of the calculations. Moreover, the very fact of occurrence or non-occurrence of events related to signal coming from a reservoir at a specific measurement point allows concluding on the presence or absence of a reservoir based on analysis of the spectral characteristics of the discrete portions without using additional information.
During hydrocarbons search in a water basin using one of the embodiments considered here, at least one seismic oscillation receiver capable of recording at least one component of infrasonic oscillations is either installed on the bed of the water basin or deepened in the water or installed on board a vessel, preferably, self-propelled one, and the Earth microseismic noise is measured simultaneously for all the components. If the seismic oscillation receiver is installed on board a vessel one should choose vessels that produce the least noise in the useful frequency range. Preferably, signal is recorded for at least 30 minutes. The receivers are installed on the bed of the water basin (or on board vessels, or by deepening into the water) by grouping them at distances 50 to 500 m from one another. In this case the registration point is preferably on the surface of water at an approximately equal distance from all the seismic oscillation receivers used. The spectral characteristics of the Earth microseismic noise useful signal obtained beyond the hydrocarbon reservoir region and
above said region are basically identical to the characteristics shown in Figs. 2 and 4.
For the first embodiment of the invention related to hydrocarbon search in mainland, seismic oscillation receivers are placed over the presumed hydrocarbon reservoir location that are capable of recording oscillations in the infrasonic frequency range for at least one component according to the first embodiment, but, additionally, seismic oscillations are generated with a seismic oscillation source in the 1 to 10 Hz range. The receivers are located at distances 50 to 500 m from the seismic oscillation source, the Earth seismic background is recorded for preferably 20 minutes, and the seismic oscillation source is activated to generate seismic oscillations for approximately 3 minutes, the Earth microseismic noise registration being not interrupted. Earth microseismic noise registration can be continued after the termination of oscillation generation (Fig. 3). The seismic oscillations generated are processed in accordance with the first embodiment, but, additionally, the presence of a reservoir can be determined from changes in the spectral characteristics of at least one of the components of the signal recorded during the generation of the oscillations and/or after the generation of the oscillations as compared with the useful signal measured before the generation, or from analysis of the spectral characteristics of discrete portions in the Earth microseismic noise during or after the excitation. This embodiment provides for a more authentic detection of hydrocarbons (Fig. 4). Figure 4 shows mathematical processing results of Earth microseismic noise in the same 4 observation points over a potentially gas producing region, but using the second embodiment. It can be well
seen that the two groups of points (A and B vs C and D) are now separated more clearly.
An important stage of seismic receiver installation in either embodiment is their grouping as it allows one to reduce the effect of noise and to utilize the proper useful signal separation algorithms during further signal processing.
During hydrocarbons search in a water basin using the second embodiment the seismic oscillation receivers are installed as for the first embodiment. The measurements are men accomplished by analogy with mainland measurements according to the second embodiment.
The problem of controlling the operation of a hydrocarbon reservoir can also be solved within the first and second embodiments of this invention. For this purpose, control points are selected over the reservoir, preferably near the operated wells. In these points, seismic oscillation receivers are installed that are capable of recording seismic oscillations in the infrasonic frequency range for at least one of the components. The Earth miscoseismic noise is recorded periodically. The presence of a water/hydrocarbon contact under the measurement point is determined from the elimination of spectral characteristic anomaly at 0.1 to 20 Hz. The anomalous behavior of the spectral characteristics is determined using any of the embodiments described herein, i.e. without applying any external force and by analyzing the behaviors of spectral characteristics of each discrete portion in the split signal time range, or by determining the ratio between the spectral characteristics of the useful signals within and beyond the reservoir area, as well as using the embodiment wherein external force
is applied and the same processing algorithms are used, but this is now for the signal recorded during or after the excitation with the seismic oscillation source; otherwise, the presence of a water/hydrocarbons boundary can be determined from the occurrence of changes in the spectral characteristics of at least one of the components in the signal recorded during oscillation generation and/or after the oscillation generation as compared with the spectral characteristics of the useful signal measured before the generation. Preferably the spectral characteristic of the Earth miscoseismic noise is recorded during 40 to 60 minutes for each point.
For the control of underground natural gas cavern filling, points on the Earth surface are selected that approximately mark different extents of gas cavern filling, seismic oscillation receivers are installed in these points that are capable of recording infrasonic oscillation for at least one components, and the spectral characteristic of the Earth miscoseismic noise is recorded periodically, wherein the absence of an anomalous change in the spectral characteristic of the useful signal at 0.1 to 20 Hz suggests the absence of natural gas under the measurement point. For comparison, the Earth miscoseismic noise is recorded with the same receiver over a place a priori located beyond the gas cavern area. Preferably, the measurement points are selected during the initial filling of the gas cavern by determining the points under which the presence of natural gas is detected in the gas cavern at different quantities of gas supplied to the cavern. However, the measurement points are anyway determined in an experimental manner. Seismic oscillations can be generated also during the
recording, and in this case recording should be made both before and during the generation of the oscillations.
One can determine the depth of hydrocarbon reservoir, for example, using the second embodiment (the one which implies oscillation generation). For this purpose at least 4 seismic oscillation receivers are used that are capable of recording infrasonic oscillations for 3 mutually orthogonal components, these receivers being located in the quadrangle nodes. For the spectral power of the useful signal, there are the following energy spectrum deconvolutions of the vertical and horizontal components with regard to the homogeneous and isotropic noise in the epicenter area of each observation point:
( r, d)
l components of the useful signal spectral power, respectively, <Nz
1> and <N/> are the vertical and horizontal components of the noise spectral power, respectively, r is the distance from the epicenter to the measurement point and d is the source depth. Following this procedure the calculated spectral power of the useful signal recorded after oscillation generation and the spectral density of the crosscorrelation functions in the useful frequency range are used to determine the depth of the hydrocarbon reservoir by determining the type of radiation source.
This method has been used for the calculation of hydrocarbon reservoir depth near a productive well in southern Orenburg Region. The calculated value was about 2800 m, the hydrocarbon reservoir depth being 3222 m.
For all ol the above embodiments of the invention disclosed herein, the critical and important stage is filtering of the time- dependent series and the surface noise to separate the useful signal. For this purpose the seismic oscillation receivers are grouped (distributed) in a special manner and the recorded signal is processed using a crosscorrelation method.
The above embodiments of the invention disclosed herein can be implemented with any type of seismic oscillation receiver capable of recording infrasonic range oscillations and containing at least one seismic oscillation sensor capable of recording infrasonic oscillations, wherein all the sensors being used should be installed on a rigid support such that the sensitivity axes of the sensors are at specific angles relative to the flat rigid support and to one another, each sensor being connected to the recording unit and the support with the sensors installed thereon being placed in a sealed rigid casing. Angular and/or linear sensors can be used that are capable of recording infrasonic range oscillations. Preferably the recording unit of each sensor contains a series connected preliminary signal amplifier and an amplitude-frequency characteristic generator, wherein each terminal amplifier provides for the connection to the common recording unit.
The invention disclosed herein will allow increase the accuracy and reliability of hydrocarbon reservoir detection.