GB2253708A

GB2253708A - Method for determining underground stresses

Info

Publication number: GB2253708A
Application number: GB9204861A
Authority: GB
Inventors: Timothy Richard Harper
Original assignee: BP PLC
Current assignee: BP PLC
Priority date: 1991-03-14
Filing date: 1992-03-06
Publication date: 1992-09-16
Anticipated expiration: 2012-03-06
Also published as: GB2253708B; GB9105405D0; GB9204861D0

Abstract

A method for the determination of the direction of stresses on a subterranean rock formation comprises the steps of: (a) taking a symmetrical core sample of the rock, (b) successively heating the core sample to progressively higher temperatures and cooling the core sample after each heating step back to a reference temperature, in a plurality of heating-cooling cycles, or successively cooling the core sample to progressively lower temperatures and heating the core sample after each cooling step back to a reference temperature, in a plurality of cooling-heating cycles, (c) measuring the strain relief after each cycle along at least three measurement base lines in the same plane of the core sample which exceed a critical minimum length, and (d) determining successive increments of time and temperature dependent extensional strains on the core sample. <IMAGE>

Description

METHOD FOR DETERMINING UNDERGROUND STRESSES This invention relates to a method for the determination of the direction of stresses in subterranean rock formations.

Rock formations in situ are subject to stresses in all dimensions which give rise to strains. When the formations are disturbed, for example, by drilling, tunnelling or mining, the stresses are relieved in the vicinity of free surfaces and, in the absence of support from the surrounding formations, the rock may fracture where the stresses are greatest or along planes of weakness. Such fractures are generally undesirable and a knowledge of where they are most likely to occur is of use in planning to avoid them or take remedial action.

A particular form of disturbance of importance to the oil industry is the technique known as hydraulic fracturing. In this technique a fracturing fluid is pumped down an oil well under considerable pressure and is forced into the surrounding formation to fracture it and thereby improve communication between the fluids in the formation and the well. The fractures generally form in a plane normal to the least stress. In this case the fractures are deliberately induced and are beneficial. Again, a nowledge of where they are most likely to occur is of value.

In both these and other instances it would clearly be desirable to know the directions of the stresses acting on the rocks in situ.

One proposal for determining the direction of fracturing involves loading discs of a core sample and measuring the orientation of the resulting cracks in the laboratory. This method is of limited accuracy, however, and relies on a macroscopic failure mechanism, i.e. the formation of actual cracks in the disc and requires application of load to the sample.

A previous method for determining the directions of the stresses is the so-called "anelastic strain relief" method which reportedly measures viscoelasticity. It is based on the theory that when the subterranean stresses are removed, the core will expand with the greatest expansion occurring in the areas where it was subjected to greatest stress and the least in the areas where it was subjected to the least stress. Viscoelasticity is defined in the book "Composite Materials" (Academic Press, 1984) edited by G.P. Sendeckys in Chapter 4, pages 87 - 88, by R.A. Schapery.

However, this method suffers from several disadvantages. It is essentially a field test and must be carried out irnmediately the core has been drilled. This frequently involves a race against time, working in adverse conditions. The cores are subjected to changes in temperature, pressure and moisture, all of which affect the results and it is difficult, if not impossible, to introduce compensating factors for these. Furthermore, although it is possible to carry out multiple measurements on the sample at the same time, each sample can only be used once, which makes a statistically sound analysis of the sample impractical.

We have now discovered that more accurate determinations of the direction of the stresses involved, which are susceptible to statistical analysis, can be made without undue haste in the laboratory using the phenomenon of 1,residual stra n". By residual strain is meant self-equilibrating strain existing with or without the presence of boundary tractions.

These strains persist in the rock cores after they have been removed from the formation, except in the immediate vicinity of cut surfaces, and can be relieved and determined time and again by inducing thermal perturbations in the cores since finite extensional strains develop in response to minor temperature changes.

Rock cores are generally anisotropic and therefore when the stress on a core is relieved a core which was circular in cross-section when under stress will tend to form an ellipse.

From the differences in lengths of the axes of the ellipse it is possible to determine the direction of the strain relief and hence the direction of the stress.

Similar phenomena are shown by cores of other symmetrical shapes.

According to the present invention there is provided a method for the determination of the direction of stresses on a subterranean rock formation which method comprises the steps of (a) taking a symmetrical core sample of the rock (b) successively heating the core sample to progressively higher temperatures and cooling the core sample after each heating step back to a reference temperature, in a plurality of heating-cooling cycles, or successively cooling the core sample to progressively lower temperatures and heating the core sample after each cooling step back to a reference temperature, in a plurality of cooling-heating cycles, (c) measuring the strain relief after each cycle along at least three measurement base lines in the same plane of the core sample which exceed a critical minimum length, and (d) determining successive increments of time and temperature dependent extensional strains on the core sample.

The core sample is preferably circular in cross-section and the strain relief may be measured across a face of the core.

The minimum length of a base line may readily be determined by increasing the length until consistent results are obtained regardless of the length.

A core from a vertical well is suitable for use as drilled, after trimming, if necessary. If the core is from a deviated well, then a vertical plug should be taken from the core.

Strain measurements may be made by strain gauges. Preferably these are arranged in the form of an equilateral triangle or rosette across the face of the core.

The core is preferably subjected to a heating-cooling cycle (although as previously mentioned, a cooling-heating cycle may alternatively be employed.) The core may be successively heated in steps of 10-30 Celsius degrees above the maximum temperature of the previous cycle, most preferably in steps of about 15 degrees.

Preferably at least three cycles are involved.

The method thus provides for multiple measurements of strain, both by multiple measurement locations and also by multiple perturbations of the core, thus providing a greater number of tests on each core sample.

Furthermore, the method has the great advantage that it is non-destructive in the sense that no part of the core is removed, apart from trimming the ends, if necessary. Thus the core remains available for further and different tests, if desired.

The invention is illustrated with reference to Figures 1-5 of the accompanying drawings wherein Figure 1 is a diagram of the equipment used, Figure 2 is a plot of time against temperature of a core sample, Figure 3 is a plot of time against strain for the same core sample, Figure 4 is a plot of time against azimuth of maximum strain for the same core sample, and Figure 5 by way of comparison, is a plot of time against strain for a metal cylinder undergoing the same temperature cycle as the core sample.

With reference to Figure 1, a temperature controlled oil bath 1 contains two 100 mm diameter cylindrical core samples 2 and 3 and a similar sized metal cylinder 4. The core sample 2 is fitted with 3 strain gauges 5, 6, 7, each of gauge length 50 mm, in a triangular configuration and a thermocouple. Core sample 3 and metal cylinder 4 are similarly fitted with strain gauges 8, 9, 10 and 11, 12 and 13 respectively, and thermocouples.

Core samples 2 and 3 and metal cylinder 4 are placed within plastic bags to protect the contents and the leads of the strain gauges and thermocouples from direct contact with the oil of the bath.

The leads are connected to strain gauge amplifiers, resistance bridges and data logging equipment.

Figure 2 shows the heating and cooling cycles to which the cores samples 2 and 3 and the metal cylinder 4 were subjected.

Traces 14, 15 and 16 of Figure 3 represent the measurements of strain gauges 5, 6 and 7.

The strain gauges form a standard strain gauge configuration for detecting maximum and minimum strain.

The azimuth of strain can be computed from the traces 14, 15 and 16 by standard techniques and is shown in Figure 4.

By way of comparison, trace 17 of Figure 5 represents the measurement of strain gauges 11, 12 and 13. A single trace which returns to the same level after heating and cooling is produced.

This indicates that the metal cylinder 14 is isotropic and was not under stress.

The three traces of Figure 3 which do not return to the same level indicate that the core 2 is anisotropic and was under residual stress.

Claims

1. A method for the determination of the direction of stresses on a subterranean rock formation which method comprises the steps of (a) taking a symmetrical core sample of the rock (b) successively heating the core sample to progressively higher temperatures and cooling the core sample after each heating step back to a reference temperature, in a plurality of heating-cooling cycles, or successively cooling the core sample to progressively lower temperatures and heating the core sample after each cooling step back to a reference temperature, in a plurality of cooling-heating cycles, (c) measuring the strain relief after each cycle along at least three measurement base lines in the same plane of the core sample which exceed a critical minimum length, and (d) determining successive increments of time and temperature dependent extensional strains on the core sample.

2. A method according to claim 1 wherein the core sample is circular in cross-section and the strain relief is measured across a face of the core.

3. A method according to either of the preceding claims wherein the strain measurements are made by strain gauges.

4. A method according to claims 2 and 3 wherein the strain gauges are arranged in the form of an equilateral triangle or a rosette across the face of the core.

5. A method according to any of the preceding claims wherein the core is successively heated in steps of 10-30 Celsius degrees above the maximum temperature of the previous cycle.

6. A method according to claim 5 wherein the core is successively heated in steps of 15 Celsius degrees above the maximum temperature of each cycle.

7. A method according to any of the previous claims wherein at least three cycles are involved.

8. A method according to claim 1 as hereinbefore described with reference to the accompanying drawings.