An Evolutionary-Genetic Approach to Heat Transfer Analysis

In this paper, ideas from the fields of evolution and genetics are incorporated in a new approach to solve heat transfer problems. Biological metaphors have been used as problem-solving strategies in the engineering sciences for well over fifty years (Denning, 1992). Early efforts include the work of Box ( 1957 ) and Friedman (1959). A rigorous mathematical formalism was introduced by Holland and his collaborators, and that formalism was first referred to as "genetic algorithms" by his student Bagley (1967). A comprehensive review of this field can be found in the book by Goldberg (1989). Genetic algorithms have been used primarily as search and optimization strategies in the areas of design, artificial intelligence, machine learning, and control. The unquestionable ability of nature to produce solutions through an evolutionary-genetic approach and the success of "genetic algorithms" in engineering suggests that a similar approach could be also useful in analyzing situations governed by prescribed physical laws. The primary goal of this paper is to illustrate and demonstrate the feasibility of this conceptually new approach to solving problems in physics. Thus, this paper shows how concepts from evolution and genetics can be used to analyze a problem governed by the conservation of energy principle. First, the general principles will be discussed, followed by an illustration of the method through the solution of one and two-dimensional transient heat conduction problems.