Formation Pressure Measurement

Formation Pressure
OCTOBER / 2015
When several reservoirs or a multi-layer reservoir are developed in a well, it is difficult to determine reservoir pressure in each layer or reservoir. Communication through channels, fractures or poor mechanical isolation makes pressure measurement by conventional means with packers and pressure gauges costly and unreliable.
October_2015_image_2TGT has introduced Spectral Formation Pressure, a new and easy-to-use formation pressure determination technique for flowing multi-zone wells using Spectral Noise Logging (SNL) data without affecting casing integrity or interfering with well production performance. This technique can be used rigless in both injection and production wells without shutting-in the well, thus reducing revenue losses and operating costs.
TGT’s clients have employed this technique in old and new operating injection and production wells defining proper pressure regimes for producing intervals, increasing efficiency and reducing deferred production as well as the direct and indirect costs that other measurement techniques incur.
Reservoir pressure calculation is based on the physical principle that the noise signal generated by fluid flowing from a particular reservoir unit is directly proportional to the product of differential pressure and flow rate (Q*ΔP). A noise power ratio determined by Spectral Noise Logging (SNL) at three flow rates, i.e. choke sizes, under infinite-acting radial flow conditions can be converted into a production or injection rate. Then this ratio can be used to determine the pressure for each flowing interval using pressure transient calculations.
The recorded signal has two main parameters: amplitude and frequency. The noise spectrum depends only on the type of channel through which the fluid moves, and noise amplitude depends on differential pressure. In order to determine pore pressure, the raw SNL signal is separated from other noises, such as those caused by wellbore flow or cross-flow behind casing. As acoustic waves decay only slightly in metal, reservoir noise can be recorded even through multiple metal barriers. The following cases illustrate the application of this technology in various well types.

CASE 1:  Reservoir pressure
determination for a water injector


This case illustrates the determination of reservoir pressure for a multi-reservoir injection well. SNL data were recorded at three injection rates with measurement stations positioned every one and a half feet. The reservoir flow noise is clearly identified across the A1 unit and no reservoir flow noise is identified across the A2 and A3 units, which means that water is mostly injected into the A1 unit.
The noise power logs shown in the NOISE POWER column by blue, purple and green lines were calculated for matrix flow only within a frequency range of 9 to 15 kHz. The calculated noise power increases proportionally to injection pressure at three rates within a 13-foot interval from X384.5 ft to X397.5 ft across the A1 unit. The average values of nine measurements were chosen within this interval to calculate the noise power ratio.
The calculated reservoir pressure for the A1 unit is 2979 psi. This calculated pressure was compared with shut-in pressures measured three months before and four months after a TSNL survey, as shown respectively by the red and brown curves on the PRESSURE panel. The reservoir pressure is seen to increase in this part of the oil field, and the calculated pressure lies between the two curves recorded under shut-in conditions.
As a result, the main injection unit was identified and its formation pressure was calculated.

CASE 2:  Reservoir pressure
determination for an oil producer


This case illustrates reservoir pressure determination for a well producing oil with 15% water cut at a pressure above the bubble point. SNL and wellbore pressure data were acquired at three production rates. SNL shows two streaks that correspond to rock matrix flow: in the upper half of unit A2 and the perforated part of unit A3. The noise power curves calculated for rock matrix flow are given in the NOISE POWER panel on the right.
The calculated reservoir pressure is 3598 psi for unit A2 and 3617 psi for unit A3. This result was compared with RFT data that were obtained one and a half years after the well was drilled. According to RFT measurements, the reservoir pressures were 3611 psi in unit A2 and 3648 psi in unit A3. It seems logical that the reservoir pressure decreased by 13 psi in unit A2 and by 31 psi in unit A3 after this well was put on stream.
For this reason, the formation pressure was calculated for the two flow units separately.

CASE 3:  Reservoir pressure
determination for a dual-string oil producer


This case illustrates the determination of reservoir pressure behind the tubing. Unit A1 was producing through the short string at a constant rate during the survey. Unit A2 was producing into the long string through SSD. SNL and wellbore pressure data were acquired through the long string at three flow rates in unit A2. The resulting SNL data show two main streaks that correspond to rock matrix flow: across the upper perforations of unit A1 and across unit A2.
The calculated noise power curves for unit A1 have the same values while the pressure changes at three flow rates in the long string. This means that the pressure in the short string is still stable and there is no communication between the A1 and A2 units.
The calculated noise power across unit A2 increases proportionally to production pressure at three flow rates. The calculated pressure for unit A2 is 3360 psi, i.e. only 53 psi lower than measured seven months before MDT data acquisition.
This case illustrates that formation pressure can be calculated for a flow unit even behind tubing and integrity can be checked at the same time.
Spectral Noise Logging tool specification:

Memory/Real Time*
Length: 2.68 ft (0.816 m)
Weght: 15.4 lbs (7 kg)
Titanium body
Maximum OD: 1,65” (42 mm)
H2S resistance < 25%
T max: 302°F (150 °C)
P max: 9,000 psi (60 MPa)
Frequency range, Hz: 3 – 60 000
Frequency resolution, Hz: 115
Dynamic range, dB: 90
October, 2015