遗传退火算法的作物灌溉决策优化研究

Research on Optimization of Crop Irrigation Decisions Based on Genetic Simulated Annealing Algorithm

在吸取以往灌溉决策支持系统经验的基础上,为了对系统的决策解进一步优化,将遗传算法(GA)融入其中,但遗传算法收敛速度慢,容易"早熟"。为此,针对这些缺点在遗传算法的整个操作过程中应用了一部分模拟退火技术,并详细分析了该混合算法的性能和特点。实验证明,该算法能够有效地克服GA的缺点,不仅提高了计算速度,而且在改善解的质量方面有着很好的效果,从而对作物的产量进行优化,以达到最大的收益。

On the basis of learning from the past experience in irrigation decision support system,genetic algorithm（GA） was involved in it for further optimized decision-making solution in this paper.But the convergence rate of genetic algorithm is slow and it is ofen premature convergence.For these shortcomings it applied a part of simulated annealing technique in the operating process of GA.The author analyzed the performance and characteristics of this mixed algorithm in detail.Experiments indicate that the algorithm can overcome these shortcomings of GA.It not only increased computation speed but improved the quality of solution effectively.So it optimized the production of crop to achieve maximum profit.