Information-Modeling Technology for the Environmental Monitoring - New Ecoinformatics Tools in Environmental Science: Applications and Decision-making - page 35

Fig. 1.16 Conceptual fl flow-diagram of the ith step at the kth stage of the adaptation process

In the parametric adaptation algorithm, this information is used as early as the
adaptation of the first-tier parameters, i.e. the characteristic number for each mode

of changes; multiplicity of the change mode utilization; length of the change mode
list; the distribution of change modes in the list; the distribution of k-models of A i k ,
memory volume at the stage of adaptation, the prehistory length, etc.
The endeavors to search for a building block base to synthesize network with a
variable structure in an attempt to create special flexible hardware based on the new

principles of information processing have offered up some results. Recent advances
in microelectronics have helped solve the problem of selecting components for
structures with variable

fl

fields of relations (Bukatova 1992).

cation and specialization are characteristic features of the evolutionary
software as a consequence of the minimum of a priori information, effective mech-
anisms of adaptation and the modular principle of realization. With the orientation to
up-to-date personal computer engineering and diverse range of application problems,
these speci

The uni

c features have made it possible to work out evolutionary computation
technologies, in which for an active dialogue with a user, a set of software modules is
realized, as well as the adjustment of the evolutionary facility to the speci city of the

problem being solved (Bukatova et al. 1991). Regardless of the field of application,

evolutionary technology with software support is characterized by adaptability,
fl flexibility, dynamism, self-correction. At the same time, the main distinction consists

in the high effectiveness and adjustability under the conditions of the maximum
informative uncertainty, including that of an irremovable character. This approach
permits to realize the adaptive procedure in Fig. 1.5 . Evolutionary modeling tech-
nology really does give a new structure of the AIMS (Fig. 1.17 ).
Evolutionary modeling is effective at bringing about an adaptation process within
the AIMS technology to correct the functional and parametric structure of the basic
model. Search of correlations for the description local function in framework of
common adaptation procedure other methods can use. For example, Fang and Liang
(2003) studied two LAI retrieval schemes with the neural network algorithm: