Subject: Earth Scieces Seminars
Seminar: Earth Sciences Division of Lawrence Berkeley National Laboratory,
Division Seminars
Date: Friday, July 18, 1997
Time: 11:00am to 12:00 Noon
Place: Building 90, Room 2063
===============================================
The Role of Soft Computing Techniques in Earth Sciences:
I. Mining and Fusion of Data
Masoud Nikravesh
Earth Sciences Division of Lawrence Berkeley National Laboratory and
BISC Program-Electrical Engineering and Computer Science of UC Berkeley
In the past, classical data processing tools and physical models solved many
real-world complex problems. However, this should not obscure the fact that
the world of information processing is changing rapidly. Increasingly we are
faced on the one hand with more unpredictable and complex real-world,
imprecise, chaotic, multi-dimensional and multi-domain problems with many
interacting parameters in situations where small variability in parameters can
change the solution completely. On the other hand, we are faced with profusion
and complexity of computer-generated data. Unfortunately, making sense of
these complex, imprecise and chaotic data which are very common in earth
sciences applications, is beyond the scope of human ability and understanding.
What this implies is that the classical data processing tools and physical
models that have addressed many complex problems in the past may not be
sufficient to deal effectively with present and future needs.
To solve such a complex problem, one needs to go beyond standard techniques
and silicon hardware. The model needs to use several emerging methodologies
and soft computing techniques: Expert Systems, Artificial Intelligence,
Neural Network, Fuzzy Logic, Genetic Algorithm, Probabilistic Reasoning, and
Parallel Processing techniques. Soft computing differs from conventional
(hard) computing in that it is tolerant of imprecision, uncertainty, and
partial truth. Soft Computing is also tractable, robust, efficient and
inexpensive.
In this seminar, we will reveal role of Soft Computing techniques in Earth
Sciences through several examples (from pore scale to field scale modeling,
from black box models to knowledge based models, from petroleum reservoirs to
environmental problems, from lithology to seismic data). The basis for each
example will be data and minimum knowledge about the problem (if this
knowledge exists). In each example, our model will use one or all of the
following techniques;
1. Fuzzy Logic for its linguistic nature, its tolerance for imprecision and
uncertainty in data, and low cost solution,
2. Neural Network because it is a fundamentally parallel technique to capture
the presence of fuzzy rules, and it does not require specification of
structural relationships between input-output data for multi-dimensional,
multi-domain, and complex data sets,
3. Genetic Algorithm for extraction of patterns, structure of the data, its
robustness, and to reduce the complexity of the Neuro-Fuzzy Structure, and
4. Quantum Computation as a Parallel Processing technique for integration into
the Neuro-Fuzzy-Genetic model for fast processing of the data in a super-fast
and parallel processor machine.