Bayseian microseismic source inversion

Patent metadata

Field	Value
Publication number	US-10634803-B2
Application number	US-201615268047-A
Country	US
Kind code	B2
Filing date	Sep 16, 2016
Priority date	Sep 16, 2015
Publication date	Apr 28, 2020
Grant date	Apr 28, 2020

Abstract

A method for using microseismic data during an injection or perforation event includes injecting fluid or perforating a well to create cracks in the formation. Microseismic data is obtained from the formation and forward modelling source parameter estimations are performed using a full moment tensor space source model and a double-couple source model. Likelihoods of the microseismic data are calculated for each model type by forward modelling synthetic data from a sampled source parameter probability distribution derived from each estimation, and by comparing the synthetic data with the microseismic data. The likelihoods are marginalized over prior probabilities for the source models, and Bayesian inference converts the likelihoods and prior probabilities to posterior probabilities. The posterior probabilities for the full tensor space and double-couple source models are compared to reveal whether an event is a fracture opening, fracture closing, or a slip on a fault plane.

First claim

What is claimed is:

1. A method for performing a hydraulic fracturing process, comprising:
injecting fluid into a well to create cracks in downhole rock formations;
obtaining microseismic data from the downhole rock formations at least while injecting the fluid into the well;
performing at least two forward modelling source parameter estimations on the microseismic data, the at least two forward modelling source parameter estimations including a first estimation using a full moment tensor space source model and a second estimation constrained to one or more source model types that include at least a double-couple source model;
calculating likelihoods of the microseismic data for the full moment tensor space source model and each of the one or more source model types by forward-modelling synthetic data from a sampled source parameter probability distribution derived from each of the at least two forward modelling source parameter estimations, and by comparing the synthetic data with the microseismic data;
marginalizing the calculated likelihoods over prior probabilities for model parameters for the full moment tensor space source model and the one or more source model types to give respective likelihoods for the full moment tensor space source model and the one or more source model types;
using Bayesian inference to convert the full moment tensor space source model likelihoods, the respective source model type likelihoods, and the prior probabilities to posterior probabilities for the full moment tensor space source model and the one or more source model types;
comparing the posterior probabilities for the full tensor space source model and the one or more source model types; and
using the comparison of the posterior probabilities to reveal probabilities an event triggering the microseismic data is a fracture opening event, a fracture closing event, and a slip on a fault plane.

2. The method according to claim 1 , wherein the constrained source parameter estimation is performed for one or more source model types selected from the group consisting of: a volumetric opening source model type, a volumetric closing source model type, and a combined tensile crack and double-couple source model type.

3. The method according to claim 1 , wherein the sampled source parameter probability distribution includes a probability distribution of seismic wave first-arrival polarity.

4. The method according to claim 1 , wherein the sampled source parameter probability distribution includes a probability distribution of seismic wave amplitude ratio.

5. The method according to claim 1 , including a preliminary step of performing seismic testing using one or more hydrophones, geophones, accelerometers, and/or distributed acoustic sensing to obtain the microseismic data.

6. The method according to claim 1 , wherein using the comparison of the posterior probabilities includes:
revealing the extent of cracks created in response to the injection of the fluid.

7. The method according to claim 1 , wherein using the comparison of the posterior probabilities includes quantifying the fracture opening or closing.

8. The method according to claim 1 , further comprising:
supplying proppant to the cracks in the downhole rock formations, an amount of the proppant being based on the quantified fracture opening.

9. The method according to claim 1 , wherein comparing the posterior probabilities for the unconstrained source model and the one or more source model types includes normalizing the posterior probabilities for a particular prior probability together for the full moment tensor space source model and the double-couple source model.

10. The method according to claim 1 , wherein comparing the posterior probabilities includes comparing a best fitting double-couple and full moment tensor solution to the event triggering the microseismic data.

11. A method for performing a perforation operation, comprising:
perforating a well to initiate cracks in a downhole rock formation;
obtain microseismic data from the rock formation at least while perforating the well;
performing at least two forward modelling source parameter estimations on the microseismic data, the at least two forward modelling source parameter estimations including a first estimation with an unconstrained source model and a second estimation that is constrained to one or more selected source model types that include at least a double-couple source model;
calculating likelihoods of the microseismic data for the unconstrained source model and each of the one or more selected source model types by forward-modelling synthetic data from a sampled parameter probability distribution derived from each of the at least two forward modelling source parameter estimations, and comparing the synthetic data against the microseismic data;
marginalizing over prior probabilities for model parameters for the unconstrained source model and each of the one or more selected source model types; and
using Bayesian inference to convert the calculated likelihoods and the prior probabilities to posterior probabilities for the unconstrained source model and each of the one or more source model types;
comparing the posterior probabilities for the unconstrained source model and the one or more source model types; and
using the comparison of the posterior probabilities to reveal probabilities an event triggering the microseismic data is a fracture opening event, a fracture closing event, and a slip on a fault plane.

Assignees

• Schlumberger Technology Corp
• Cambridge Entpr Ltd

Inventors

• Pugh David J
• White Robert S
• Christie Philip Andrew Felton

Classifications

• G01V1/288 Primary

  Event detection in seismic signals, e.g. microseismics (G01V1/36 takes precedence) · CPC title

• G01V1/282

  Application of seismic models, synthetic seismograms · CPC title

• E21B49/00

  Testing the nature of borehole walls; Formation testing; Methods or apparatus for obtaining samples of soil or well fluids, specially adapted to earth drilling or wells · CPC title

• G01V2210/646

  Fractures · CPC title

• G01V2210/663

  Modeling production-induced effects · CPC title