基于改进基因表达式程序设计的菠菜价格建模及预测

发布时间：2018-03-09 22:11

本文选题：GEP　切入点：IGEP　出处：《广东农业科学》2016年01期 　论文类型：期刊论文

【摘要】：基因表达式程序设计(GEP)是一种基于基因型和表现型的新型进化算法。针对传统GEP遗传算子进行建模及预测时容易受到噪声干扰,导致过早收敛,陷入局部最优等缺点,提出了一种新型GEP算法,增加了"倒串"和"基因提取"算子,该新型算法可以提高基因的有效利用率,具有更高的收敛速度和求解精度,且能更好的避免早熟现象。将该新型GEP算法用在菠菜价格预测上,通过对训练数据进行分析和进化创建数学模型,实现菠菜价格的仿真与预测,通过多组实验证明,该新型基因表达式程序设计算法在菠菜价格预测上具有更快的收敛速度和更高的精度。
[Abstract]:Gene expression programming is a new evolutionary algorithm based on genotype and phenotype. It is easy to be disturbed by noise when modeling and predicting traditional GEP genetic operator, which leads to premature convergence and local optimum. In this paper, a new GEP algorithm is proposed, in which "backward string" and "gene extraction" operators are added. The new algorithm can improve the effective utilization rate of genes, and has higher convergence speed and accuracy. The new GEP algorithm is used to predict the price of spinach. By analyzing the training data and developing the mathematical model, the simulation and prediction of spinach price can be realized. The new genetic expression programming algorithm has faster convergence speed and higher precision in predicting spinach price.
【作者单位】：华南农业大学数学与信息学院;广东省科技创新监测研究中心;广东省农科院农业经济与农村发展研究所;广东村村通科技有限公司;
【基金】：广东省科技计划项目(2015A020209119,2014A020208087,2012B040500054)
【分类号】：TP311.1;F323.7

【相似文献】