METHOD FOR THE ESTIMATION OF FLUIDS MOVED IN COMPART- MENTED AREAS OF THE SUBSOIL

The present invention relates to method for estimating volumes of fluids moved in compartmented areas of the subsoil, such as, for example, natural deposits, also called reservoirs, it can be applied industrially in oil fields, and additionally, also for monitoring the production and storage of hydrocarbons and reducing mining risks during the explorative phase and development of onshore and off-shore reservoirs.

In the exploration of subsoil resources, it is customary to rely on the measurement of the vertical component of the gravity field and the vertical gradient of the same field. From an analysis of these data, it is possible to deduce information on the density distribution of the subsoil which characterizes a particular site.

On the basis of this methodology, it is also possible to obtain the mass variation of the hydrocarbons present inside a reservoir.

The idea at the basis of this technique is that as the

movements of hydrocarbons inside a reservoir are correlated to density variations, they can be appreciated by means of vertical gravity gradient measurements .

It is since the thirties', in fact, that measurements of the gravimetric field gradient have been successfully used in the exploration of resources of the subsoil. Since 1936 the importance has been known of the use of the vertical gradient which, as it has a better resolution and is relatively insensitive to regional effects, often has particular structures which cannot be easily obtained from gravimetric field data.

The measurement of the vertical gravimetric field gradient can be effected by means of specific instruments called gradiometers . Alternatively, the vertical gradient of the gravity field of a point can be measured, with good approximation, by means of the almost contemporaneous acquisition of two gravimetric measurements referring to different heights. In this second case, before interpreting the data acquired in the field in geological terms, their reduction in terms of Bouguer anomaly is frequent, from which the undesired effects are removed and the calculation and analysis of the vertical gradient is subsequently ef- fected.

The most important corrections to be made are the following:

- Instrumental drift

- Tidal correction - Latitude correction

- Free Air correction

- Bouguer correction

- Topographic correction

The description of these corrections is treated hereunder.

Instrumental drift: the readings of data with a gravime- ter undergo time variations due to the elastic characteristics of the materials which form the instrument itself. The instrumental drift can be easily determined by re- peating the measurement in the same station in different times, typically every 1-2 hours. The representation referring to Cartesian axes gives the drift curve which, for many gravimeters is of the linear type.

A definite value is subtracted with each measurement effected in subsequent stations, on the basis of the measurement time .

Tidal correction: the drift measured in reality contains the further contribution of an effect of the sea type due to moon-sun attraction (tide) . The correction to be made is calculated on a theoretical basis by means of formulae

which allow the quantification of this effect, such as, for example, the Longman formula.

Latitude correction: both the Earth's rotation and its equatorial swelling produce an increase in gravity with the latitude, and this must be considered when reducing the gravity data observed.

Free air correction: this is a correction used in order to consider the altitude of the measuring station. Bouguer correction: this correction is used to consider the attraction due to the interposed masses between the measuring station and the reference surface. In 1749, Bouguer suggested that this additional attraction could have been calculated like that due to the action of an infinite horizontal plate having a thickness equal to the elevation from sea level of the measuring station.

Topographic correction: The approximation of the plate may be unsatisfactory in an area with an articulated topographic trend.

Under these conditions, it is appropriate to add a correction in order to consider the masses above the plate and those whose contribution has been erroneously subtracted in the Bouguer correction.

After the reductions listed, the vertical gradient of the field is calculated as described hereunder. Variations in density in the subsoil with time can be

measured and monitored from the measurements of the vertical gravity field gradient.

This method is already in use for measuring and monitoring water layers and geothermic fields. The objective of the present invention is to provide a method for the estimation of fluid volumes, such as, for example, liquid and/or gaseous hydrocarbons moved in compartmented areas of the underground, following, for example, production, injection, and/or storing. Within the above mentioned objective, it is necessary to draw a calibration curve characteristic for each single field reservoir.

These and other objectives can be achieved, according to the present invention, by means of an estimation method of the volumes of fluids moved in compartmented areas, as specified in claim 1.

Further characteristics of the method are the object of the dependent claims.

The characteristics and advantages of an estimation method of the fluid volumes in compartmented areas, according to the present invention, will appear more evident from the following illustrative and non- imitating description.

The estimation method of fluid volumes in the sub- soil, according to the invention, comprises a first gra-

diometric measurement phase "to frame/calibrate the area" on the basis of which a model is determined, by statistical approximation, for quantifying the movement of the fluid volumes, such as, for example, hydrocarbons, pro- duced, injected and/or stored in the subsoil.

More specifically, a series of measurement campaigns are effected, repeated over a certain period of time, with the aim of quantifying the amount of the fluid movements, by means of a model of their production, injection and/or storage.

The degree of accuracy of this model is then evaluated through the statistical analysis of the data obtained.

The measurement operation campaigns are effected on a series of stations suitably dislocated in the survey area.

The measurement is effected, for each single station, using a precision gravimeter and a tripod suitably positioned to allow measurement at a certain distance and constant from the ground.

For the geophysical applications mentioned above, it is advisable to use gravimeters having a precision not lower than μGal (microgal) .

From an operative point of view, the campaign meas- urements consist of the following phases.

A first ground gravimetric measurement is acquired G _bot followed by a second gravimetric measurement G _top by positioning the gravimeter at a distance dh from the ground using a tripod. The distance dh from the ground is preferably kept constant in each single station and during the whole survey measurement. The Applicant has in fact observed that, by maintaining the distance from the ground constant, in each single station and for the whole survey, the meas- urements and definitions of gradiometric anomalies are more accurate .

On the basis of the G _bot and G _top values measured, the gravity values G ^* _bot / G ^* _t op are then determined, corrected with respect to the effects previously discussed. The vertical gradient value of the gravimetric field is obtained from the correct gravity values, by means of the following formula:

The determination of the difference in level dh between two relative measurements in each single station is effected by means of a laser distantiometer with millimet- ric precision.

This acquisition and elaboration procedure is then repeated with time on the same area, insisting on the

same measurement stations.

A time lapse signal is then obtained, i.e. a signal linked to the gravity variations in the subsoil over a period of time. The time lapse signal is calculated as the difference between two gradiometric surveys effected in the same measurement station, at different times:

In this way it is possible to evaluate which zones of the area examined have undergone a density variation, thus revealing where there have been relative movements of fluids.

In this way, a qualitative result is obtained, relating to the volume variations of fluids in the subsoil, connected, from case to case, with production, injection and/or storage operations.

The Applicant subsequently identified a calibration curve characteristic of each single storage or production reservoir, which correlates the variations in the gradi- ometric values detected over a certain time lapse, with the corresponding volumes of the fluids moved, or produced, injected and/or stored, within the same time period.

The Applicant has also developed, tested and validated an assessment method of said calibration curve.

As far as qualitative evaluations of the movements of masses with time are concerned, the Applicant deemed it suitable to find a calibration curve which would relate volumetric variations of fluids with time lapse data and not with the gravity gradient data of a single assessment .

For this purpose, it was necessary to assess a characteristic parameter of the fluid volume variation associated with the production, injection and/or storage opera- tions.

The use of the integral of time lapse values on the whole definition dominium of the area examined, proved to be adequate for this purpose: TMx.yJdxdy (3a)

The value so obtained is preferably normalized with respect to the integral of the first time lapse TLi ₂ , thus assuming the first assessment as reference point.

An adimensional parameter Pu is thus obtained, bound to the quantitative global variation of the volumes, defined as

JJ TLiifx.yJdxdy

Pu ≡ -r. . (4a)

I JJ TL ₁₂ (x,y)dxdy

The parameter Pu can be calculated on the basis of

the values acquired with the measurement campaign number i .

In the same way, it is also suitable to use as parameter Pu the summation of the time lapse values measured in the n measurement stations distributed on the whole survey area : n

Also in this case, the value thus obtained is preferably normalized with respect to the summation of the first time lapse measurements TL ₁₂ in order to obtain an adi- mensional parameter:

The Applicant then constructed the calibration curve on the basis of at least three acquisition campaigns, on known volumes of fluids moved (produced, injected and/or stored) .

On the basis of the parameters Pu associated with the time lapse whose moved ' volumes δVi are known, the relationship between the parameters Pu and the corresponding known volumes produced, injected and/or stored δVi was reconstructed by approximation.

Various controls of the results obtained demonstrated and confirmed that the variation law thus determined allows the volume variation of moved fluids (produced, injected and/or stored) to be estimated starting from gra- diometric measurement campaigns subsequent to those of the first phase "for framing/calibrating the area" .

In a purely illustrative manner, a method is described herein for the determination of a possible law which correlates the gradiometric values measured with the volume of fluids, such as, for example, hydrocarbons, moved i.e. produced, injected and/or stored.

Assuming a linear relationship between the parameter

Pii and the volume of hydrocarbons moved δVi , of the type :

P = a + b δV (5) a line can be constructed to be used for estimating δV starting from the fourth survey onwards.

The Pu parameters associated with the first three measurement campaigns and a knowledge of the relative volumes moved are necessary for defining the parameters a and b.

For the evaluation of the parameters a and b, with the relative uncertainties δa and δb, it is possible, for example, to use the approximation method of the square minima, which minimizes the differences between the theo- retical values of the ideal straight line and the data

observed, i.e. the following expression:

Once a and b have been determined in this way, it is possible, knowing the value of the parameter Pn, to evaluate the corresponding value of δv.

The uncertainty and consequently the sensitivity of the method depend on the uncertainty of the parameters a and b and the uncertainty δP on the parameter P which comes from the gravity measurements .

The equation for estimating the moved volumes is:

(P-a)

δV= (7) b

Therefore, the associated error δδV is given by:

In the same way it is possible to assume a polynomial variability law of a suitable degree and use, as an approximation method, the square minima method or interpolation.

The characteristics of the method object of the present invention, as also the relative advantages, are evident from the above description.

The Applicant has added an algorithm to the determina- tion of the vertical gravity gradient and relative time lapse processings, for estimating the fluid volumes moved in reservoirs over a period of time.

Finally, it is evident that the method thus conceived can undergo further modifications and variations, all in- eluded in the scope of the invention.