微分方程演化建模用于工业固废产量的研究

Evolutionary Modeling of Differential Equations for Study on Year′s Output of Industrial Solid Wastes

利用演化算法的自适应、自组织、自学习的特性设计了遗传程序与遗传算法相嵌套的混合演化建模算法 ,以遗传程序设计优化模型结构 ,以遗传算法优化模型参数 ,为山东省工业固废产量随年迹变化关系自动建立微分方程演化模型 .结果表明演化模型不仅其拟合精度明显高于常规的GM(1,1)模型 。

Based on the properties of self-adaptation,self-orgainization and self-learning of evolutionary algorithm,this paper proposes a hybrid evolutionary modeling algorithm to build up models of differential equations for the year′s output of industrial solid wastes in Shandong province on annual changing automatically.The main idea is to embed genetic algorithm(GA)in genetic programming(GP)where GP was employed to optimize the structure of a model while GA was employed to optimize its parametrs.The result shows that not only the fitting accuracy but also prediction trend of evolutionary models(EM) obtained by using the algorithm is much higher and more rational than that of regular GM(1,1)models.