Multi-resolution graph-based clustering

Details Multi-resolution graph-based clustering Multi-resolution graph-based clustering

2000-06-02

2001-09-25

McElheny, Jr., Donald E. (Department: 2862)

Data processing: measuring, calibrating, or testing

Measurement system in a specific environment

Earth science

C702S011000

Reexamination Certificate

active

06295504

ABSTRACT:

BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to the geological study of earth formations for the location and exploitation of mineral deposits using electrofacies analysis. More particularly, the present invention relates to a new system and method for identifying formations of mineral deposits using a user-friendly and reliable clustering technique that can extract natural clusters from sets of logged data points for improved electrofacies analysis of the formation.
2. Description of the Related Art
Mineral and hydrocarbon prospecting is based upon the geological study and observation of formations of the earth's crust. Correlations have long been established between geological phenomena and the formation of mineral and hydrocarbon deposits that are sufficiently dense to make their exploitation economically profitable.
The study of rock and soil facies encountered while prospecting for minerals takes on particular importance. As used herein, a facies is an assemblage of characteristics that distinguish a rock or stratified body from others. A facies results from the physical, chemical and biological conditions involved in the formation of a rock with respect to other rocks or soil. This set of characteristics provides information on the origin of the deposits, their distribution channels and the environment within which they were produced. For example, sedimentary deposits can be classified according to their location (continental, shoreline or marine), according to their origin (fluviatile, lacustrine, eolian) and according to the environment within which they occurred (estuaries, deltas, marshes, etc.). This information in turn makes it possible to detect, for example, zones in which the probability of hydrocarbon accumulation is high.
The set of characteristics used to define a facies depends on the situation. For example, a lithofacies may be defined by the rock's petrographic and petrophysical characteristics. These are the composition, texture and structure of the rock. Examples of mineral composition are silicate, carbonate, evaporite, and so on. A rock's texture is determined by its grain size, sorting, morphology, degree of compaction, and degree of cementation. The rock structure includes the thickness of beds, their alternation, presence of stones, lenses, fractures, degree of parallelism of laminations, etc. All of these parameters are related to the macroscopic appearance of the rock.
For extraction of hydrocarbons from geologic formations, the particularly desirable characteristics of the lithofacies are the porosity of the reservoir rocks and their permeability, as well as the fraction of the pore volume occupied by these hydrocarbons. These aid in estimating the nature, quantity, and producibility of the hydrocarbons contained in such strata.
There are various sources of information on formation lithofacies. Information may be gathered from subsurface observations such as, for example, by the study of core samples taken from rock formations during the drilling of a bore hole for an oil well. Such information can also be provided by drill cuttings sent up to the surface from the bottom of a well by means of a fluid (generally drilling mud) injected near the drilling tool. It is not normally cost-effective to identify facies using these methods. Information on geological formations traversed by a bore hole is more commonly gathered by a measurement sonde passing through the bore hole. The gathered information as a function of the sonde's position along the bore hole is then stored or “logged”.
Many downhole measurement techniques have been used in the past, including passive measurements such as measuring the natural emission of gamma rays; and active measurements such as emitting some form of energy into the formation and measuring the response. Common active measurements include using acoustic waves, electromagnetic waves, electric currents, and nuclear particles. The sonde measurements are designed to reflect the distinguishing characteristics of the rock facies. Multiple logs and sondes may be used to gather the measurements, which are then correlated and standardized so as to furnish measurements at discrete levels separated by equal depth intervals. The measurement standardization allows the automation of data interpretation in order to obtain estimates of the porosity of the rocks encountered, the pore volume occupied by hydrocarbons, and the ease of flow of hydrocarbons out of the reservoirs in the case of petroleum prospecting. The set of measured formation characteristics values that distinguish the strata in a given bore hole is herein termed the electrofacies.
Interpretation studies have demonstrated a strong correlation between the electrofacies and lithofacies, thereby making it possible to identify with confidence the compositional characteristics of the rocks traversed by bore holes based on the sonde measurements. It has been established that the sets of log measurements (i.e. sample points) which correspond to a given lithofacies form a “cluster” in “data space”. That is, when the measured characteristic values of a formation are graphed, the points generally fall into a continuous region which is distinguishable from the regions where points for other formations would fall.
Various systems and method that use the correlation or the observed clusters to identify lithofacies from electrofacies have been created. These systems take the logged measurements and convert them to a graph that furnishes, as a function of position along the bore hole, an image of a succession of facies. The graph typically also provides some indication of the measured formation characteristic values alongside the image. An example of one such system and its output is described in U.S. patent application No. 4646240, which is hereby incorporated herein by reference. However, before these systems can do the conversion, they must be tailored to the drilling region.
The most accurate existing systems and methods require a substantial amount of user participation to set up, and conversely, those existing systems which are highly automated tend to perform poorly. One proven approach uses a two-step methodology to correlate different measured characteristic values into generalized electrofacies charts for analysis. In the first step, the number of clusters is specified to an automatic clustering algorithm such as maximum likelihood algorithm, hierarchical clustering method, dynamic clustering or neural network. The number of clusters specified is large, creating clusters containing small numbers of points. A petrophysicist or geologist then manually assigns geological characteristics from the facies to each cluster and simultaneously merges similar small clusters into electrofacies.
Another approach for creation of electrofacies charts requires that the number of clusters specified to the automatic clustering algorithm be relatively small. In this approach, the geologist often has a problem assigning specific geological facies to the clusters, which tend to be much larger than the clusters in the previous approach. The geologist may also be required to “lump” together geological facies at a coarser level of distinction than might be appropriate. A large number of clusters require much work by the geologist to match clusters to geology; too few clusters cause the geologist problems in making meaningful linkages between clusters and geology.
The electrofacies analysis systems described above suffer from various limitations and drawbacks. The automatic clustering methods require the user to provide an initial number of clusters before processing. This is a limitation because the results are very sensitive to this parameter. Furthermore, unless the number is large, the identified clusters may have shapes that are not geologically meaningful. This prevents them from being directly used for facies analysis. On the other hand, manual merging of a large number of small clusters based on similar geological characteri

Affiliated with

Also associated with

Rating

Multi-resolution graph-based clustering does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multi-resolution graph-based clustering, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-resolution graph-based clustering will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2549864