Optimal Design and Process of Threshing Units Based on a Genetic Algorithm. I. Algorithm
Petre I. Miu · 2001 Sacramento, CA July 29-August 1,2001 · 2001
This paper concludes with the formulation of a multiobjective genetic algorithm foroptimization of design and functional parameters of threshing units. This approach implied theblending of performant, reliable mathematical models that describe the whole process and anevolutionary search method (a genetic algorithm in relationship with fuzzy logic). Design and operation parameters in connection with physical properties of a certain crop areexpressed as variables of a complex, constrained, nonlinear, multiobjective problem. Solvingthe problem requires finding the setting of above-mentioned decision variables, so that anevaluation function based on process quality criteria is maximized. To solve the problem at hand a genetic algorithm is implemented. Improvements of thealgorithm performance are obtained using elitist selection strategy and selective weights in theevaluation function. Once this successful evolutionary computation approach has beendeveloped, it can be incrementally adapted to other threshing units in various crops.