An extensible genetic algorithm framework for problem solving in a common environment

摘要

An object-oriented framework is described for solving mathematical programs using genetic algorithms (GA). The advantages of this framework are its extensibility, modular design, and accessibility to existing programming code. The framework also incorporates a graphical user's interface that may be used to build new GA's as well as run GA simulations. Two power system problems are solved by implementing genetic algorithms using the framework. The first is a continuous optimization problem and the second an integer programming problem. We illustrate the flexibility of the framework as well as its other features on our test problems.