Soft Numerical Computing in Uncertain Dynamic Systems is intended for
system specialists interested in dynamic systems that operate at
different time scales. The book discusses several types of errors and
their propagation, covering numerical methods--including convergence and
consistence properties and characteristics--and proving of related
theorems within the setting of soft computing. Several types of
uncertainty representation like interval, fuzzy, type 2 fuzzy, granular,
and combined uncertain sets are discussed in detail. The book can be
used by engineering students in control and finite element fields, as
well as all engineering, applied mathematics, economics, and computer
science students.
One of the important topics in applied science is dynamic systems and
their applications. The authors develop these models and deliver
solutions with the aid of numerical methods. Since they are inherently
uncertain, soft computations are of high relevance here. This is the
reason behind investigating soft numerical computing in dynamic systems.
If these systems are involved with complex-uncertain data, they will be
more practical and important. Real-life problems work with this type of
data and most of them cannot be solved exactly and easily--sometimes
they are impossible to solve.
Clearly, all the numerical methods need to consider error of
approximation. Other important applied topics involving uncertain
dynamic systems include image processing and pattern recognition, which
can benefit from uncertain dynamic systems as well. In fact, the main
objective is to determine the coefficients of a matrix that acts as the
frame in the image. One of the effective methods exhibiting high
accuracy is to use finite differences to fill the cells of the matrix.
