摘要
为了提高灰色预测模型GM(1,1)在复杂系统模型中的预测精度,从原始数据和预测值两个方面对灰色GM(1,1)模型进行改进。根据原始数据的信息特点对模型作补充定义;预测值改进则利用背景值重构和粒子群优化算法对传统GM(1,1)模型的预测值进行改进,求出最佳预测值。结果表明:改进GM(1,1)模型的平均残差和相对残差都远远小于传统模型,其预测效能和可信度都有大幅提高。
In order to improve the prediction accuracy of grey forecasting model GM（1,1） in a complex system,the improvement to the gray GM（1,1） model from the original data and the predicted value was made.The improvement of predicted value using background value restructuring and particle swarm optimization to improve the predicted value for the traditional GM（1,1） model was studied,and the best predicted value was obtained.The computation of the example proved that the residual mean and the relative residual of improved GM（1,1） model was far smaller than the conventional model.Its prediction efficiency and the confidence level have improved largely.
出处
《沈阳农业大学学报》
CAS
CSCD
北大核心
2012年第2期253-256,共4页
Journal of Shenyang Agricultural University
关键词
GM(1
1)模型
背景值重构
粒子群优化算法
GM（1,1） model
background value restructuring
particle swarm optimizes algorithm
