Systems and methods employing cooperative optimization-based dimensionality reduction

Patent metadata

Field	Value
Publication number	US-10329900-B2
Application number	US-201615348718-A
Country	US
Kind code	B2
Filing date	Nov 10, 2016
Priority date	Aug 6, 2008
Publication date	Jun 25, 2019
Grant date	Jun 25, 2019

Abstract

Dimensionality reduction systems and methods facilitate visualization, understanding, and interpretation of high-dimensionality data sets, so long as the essential information of the data set is preserved during the dimensionality reduction process. In some of the disclosed embodiments, dimensionality reduction is accomplished using clustering, evolutionary computation of low-dimensionality coordinates for cluster kernels, particle swarm optimization of kernel positions, and training of neural networks based on the kernel mapping. The fitness function chosen for the evolutionary computation and particle swarm optimization is designed to preserve kernel distances and any other information deemed useful to the current application of the disclosed techniques, such as linear correlation with a variable that is to be predicted from future measurements. Various error measures are suitable and can be used.

First claim

What is claimed is:

1. A substance fingerprinting method that comprises:
performing a compositional analysis on a sample of a substance;
obtaining a data set of compositional analysis results having a dimensionality that is to be reduced;
generating a population of chromosomes having encoded low-dimensionality coordinates for data set members;
subjecting said population of chromosomes to evolutionary computation to generate new chromosomes and corresponding low-dimensionality coordinates for data set members based on a fitness function until a threshold fitness level or predetermined number of iterations is reached, wherein the new chromosomes are used to select a dimensionality reduction mapping;
applying the dimensionality reduction mapping to the compositional analysis results to obtain a low-dimensionality representation, wherein a distance between two data points in the obtained data set to be reduced is substantially maintained in the low-dimensionality representation;
using the low-dimensionality representation to match the sample with one or more closely-related substances; and
identifying one or more characteristics of the sample based on properties of the closely-related substances.

2. The method of claim 1 , further comprising deriving the dimensionality reduction transform from the data set, wherein said deriving includes:
identifying kernels that represent clusters within the data set;
applying evolutionary computation to directly-encoded low-dimensionality coordinates for the kernels to select a bit-restricted initial encoding;
refining the initial encoding using a local search technique that is not bit-restricted; and
training at least one neural network to implement the dimensionality reduction transform based on the refined encoding.

3. The method of claim 2 , wherein the local search technique employs particle swarm optimization.

4. The method of claim 2 , wherein the substance is sedimentary rock, and wherein said one or more characteristics include one or more of the following: lithology; sedimentary facies in terms of original depositional environment; reservoir quality; and aquifer quality.

5. The method of claim 2 , wherein the substance is igneous rock, and wherein said one or more characteristics include one or more of the following: lithology; source volcano(s) of a volcanic ash bed included in the igneous rock; source location of volcanic lava from which the igneous rock originated; tectonic setting in which magma making up the igneous rock was generated; degree of fractional crystallization undergone by magma during its emplacement and cooling to form the igneous rock; degree of contamination added by surrounding rock to magma making up the igneous rock.

6. The method of claim 2 , wherein identifying one or more characteristics includes identifying a source and distribution of the substance.

7. The method of claim 6 , wherein the substance is in the set consisting of hydrocarbons, kerogens, bitumens, water, water pollutants, and soil pollutants.

8. A well-telemetry method that comprises:
obtaining a data set having a dimensionality that is to be reduced;
generating a population of chromosomes having encoded low-dimensionality coordinates for data set members;
subjecting said population of chromosomes to evolutionary computation to generate new chromosomes and corresponding low-dimensionality coordinates for data set members based on a fitness function until a threshold fitness level or predetermined number of iterations is reached, wherein the new chromosomes are used to select a dimensionality reduction mapping;
training a neural network ensemble with the dimensionality reduction mapping, wherein a distance between two data points in the obtained data set to be reduced is substantially maintained in the dimensionality reduction mapping;
configuring a downhole processor to apply the neural network ensemble to logging data to obtain reduced-dimension telemetry data for transmission uphole;
generating predictions of one or more formation properties based on the telemetry data; and
adjusting a steering mode of a downhole tool using the predictions of the one or more formation properties.

9. The method of claim 8 , wherein the evolutionary computation employs a multi-objective fitness function with a measure of kernel pair distance error and a measure of linear correlation with a prediction variable.

10. The method of claim 8 , wherein the dimensionality reduction mapping is refined by particle swarm optimization.

Assignees

• Halliburton Energy Services Inc

Inventors

• Chen Dingding
• Hamid Syed
• Dix Michael C

Classifications

• G06N3/006 Primary

  based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO] · CPC title

• E21B47/12 Primary

  Means for transmitting measuring-signals or control signals from the well to the surface, or from the surface to the well, e.g. for logging while drilling · CPC title

• G06F18/21355

  nonlinear criteria, e.g. embedding a manifold in a Euclidean space · CPC title

• G06F18/2111

  by using evolutionary computational techniques, e.g. genetic algorithms · CPC title

• G06N3/045

  Combinations of networks · CPC title