Production logs from distributed acoustic sensors
AssigneeConocoPhillips Company
InventorsHerbert W. Swan, Ge Jin, Kyle R. Krueger, Baishali Roy
Life of this patent
Estimating the term…
Abstract
A system and method for monitoring oil flow rates along a producing oil or gas well using a Distributed Acoustic Sensing fiber is described. This system uses the low-frequency component of the acoustic signal as a measurement of temperature variations within the well. The relative flow contributions can then be inferred from these temperature fluctuations.
Claims15
A method of monitoring oil flow rates along a hydrocarbon reservoir comprising:
- a) installing one or more fiber optic cables along a wellbore in a hydrocarbon formation;
- b) installing one or more interrogators on at least one of said fiber optic cables;
- c) interrogating at least one of said fiber optic cables with an interrogation signal during production;
- d) obtaining one or more datasets of data from at least one of said interrogators;
- e) converting one or more of said datasets from at least one of said interrogators into a continuous record;
- f) transforming the continuous record with a low-pass filter to 1-100 milliHz while down sampling the data into a transformed well signal;
- g) approximating flow velocities at a given depth by calculating the slope of temperature variation as a function of depth from the transformed well signal;
- and
- h) estimating from the slope of temperature variation as a function of depth a cumulative production rate of hydrocarbons from said hydrocarbon formation.
A method of monitoring oil flow rates along a hydrocarbon reservoir comprising:
- a) installing one or more fiber optic cables along a wellbore in a hydrocarbon formation;
- b) installing one or more interrogators on at least one of said fiber optic cables;
- c) interrogating at least one of said fiber optic cables with an interrogation signal during production;
- d) obtaining one or more datasets of data from at least one of said interrogators;
- e) converting one or more of said datasets from at least one of said interrogators into a continuous record;
- f) transforming the continuous record with a low-pass filter to 1-100 milliHz while down sampling the data into a transformed well signal;
- g) quantitatively measuring flow velocities by removing the effect of constant-temperature produced fluids from the transformed well signal to generate a differentiated distributed acoustic sensing (DAS) signal;
- h) computing a semblance function for the differentiated DAS signal;
- and
- i) determining from the semblance function for the differentiated DAS signal a cumulative production rate of hydrocarbons from said hydrocarbon formation.
A method of monitoring oil flow rates along a hydrocarbon reservoir comprising:
- a) installing one or more fiber optic cables along a wellbore in a hydrocarbon formation;
- b) installing one or more interrogators on at least one of said fiber optic cables;
- c) interrogating at least one of said fiber optic cables with an interrogation signal during production;
- d) obtaining one or more datasets of data from at least one of said interrogators;
- e) converting one or more of said datasets from at least one of said interrogators into a continuous record;
- f) transforming the continuous record with a low-pass filter to 1-100 milliHz while down sampling the data into a transformed well signal;
- g) quantitatively measuring flow velocities from the transformed well signal comprising:
- i) use and initial estimate of fluid velocity to compute a traveltime function f(x),
- ii) numerically form an inverse function through piece-wise interpolation,
- iii) form a series of traveltime functions and their inverses,
- iv) compute a semblance over a range of changes in time (δτ),
- v) set a new traveltime function f(x) at a trial function g(x, δτ), and
- vi) repeat (i)-(v) until convergence is achieved;
- and
- h) determining from the measured flow velocities a cumulative production rate of hydrocarbons from said hydrocarbon formation.
Dependencies are inferred from the claim text.