Abstract
In recent years, Differential Evolution (DE) algorithm has been proposed as a technique for optimizing large and complex models. In this study, an energy consumption model for integration between beef cattle and feed production was optimized using DE. The optimum levels of 5 variables (roughage-concentrate ratio, stocking rate (animals/ha), daily body weight gain, final weight for fattening and TDN content of roughage) were determined by maximizing fossil energy efficiency. The result confirmed that fossil energy efficiency can be enhanced by supplying more high quality roughage and slaughtering high growth animals at younger age. Furthermore, the effects of DE control variables (crossover probability (CR) and weighting factor (F) applied to the mutation process) were examined under two strategies (DE/best/1/exp and DE/rand/1/bin) in Storn and Price's strategies. The criteria for evaluation were the speed of convergence and stability of solutions. As a result, it was suggested that the good choices of the control variables were 0.75 and 0.5 in DE/best/1/exp and 0.5 and 0.5 in DE/rand/1/bin, respectively.
