Page 285 - Design and Operation of Heat Exchangers and their Networks

Optimal design of heat exchanger networks 271

the problem and the final results. In the data base, each stream is represented
by a list. For example, a hot stream is characterized by a list of the number of
hot stream i, its input temperature and output temperatures and the heat-
capacity flowrate. Therefore, a set of hot or cold streams is defined by a
hot or cold list, labeled as H or C. All forbidden matches between hot stream
i and cold stream j are specified by a list named R. Knowledge base in the
expert system must be explicitly represented. The function of this part is to
produce rules that represent behavioral knowledge in an “if-then” form
consisting of a conditional part and an action part. The third-part controlling
strategies are used to manage and apply the available information stored in
the knowledge base to change the problem state in the dynamic database
from an initial state to the solution state through a sequence of intermediate
states. The knowledge rules are obtained from engineering experiences. For
example, some of the rules are illustrated as follows:
(1) If the input temperature of hot stream i is less than or equal to the input
temperature of cold stream j, the match between hot stream i and cold
stream j is forbidden.
(2) If the corrosiveness of hot stream i or cold stream j is serious, the match
between hot stream i and cold stream j is forbidden.
(3) If the distance between hot stream i and cold stream j is long enough,
then the match between those two streams is forbidden.
In addition, the number of HEN stages can be also decided with the expert
system. And after the expert system is developed, the rule bases can be also
replenished and perfected. From earlier, it can be seen that the expert system
controls the interaction between the various problem aspects and generates
the models to be solved. The apparent advantage of using an expert system is
the flexibility in applying or to combine several methods for subproblems. It
is easy to modify or improve the capabilities of the system by expanding the
contents of the knowledge base.

6.6 Examples of heat exchanger network synthesis

Because of the high complexity of the synthesis problems for optimal design
of heat exchanger networks, there is no available methods to prove whether
a network has reached its global minimum of the total annual cost (TAC).
A common way to check a newly developed synthesis procedure is carry out
several case studies using available examples in the literature, which have
already been investigated and optimized by other researchers with other