摘要
基于灰色系统理论和马尔可夫链原理,应用系统云灰色模型拟合时序数据的总体趋势,所得拟合精度指标是随机波动的,而马尔科夫链原理适合处理波动性大的系统过程,选用能更好解决随机波动性的加权马尔可夫链预测方法,提出一种用于农作物干旱受灾面积预测的加权马尔可夫SCGM(1,1)c模型,适用时间短、数据量少且随机波动大的动态过程预测.以我国农作物干旱受灾面积预测为例,表明加权马尔可夫SCGM(1,1)c模型对于农业旱灾预测具有较高的精度.
The prediction of drought crop area is the basis of agricultural development in our country. According to grey system theory and Markov chain principle, applying a single gene system cloud grey SCGM(1,1)c model to fit the development tendency of the few time series, its error index is stochastic fluctuate. Markov chain method is suitable for forecasting stochastic fluctuating dynamic process, selecting weight Markov chain to predict the error index. Combining tow the advantages of two models, found a weighted SCGM(1,1)c model for drought crop area prediction, the new model is suitable for forecasting such kinds of system with short time, few data and stochastic fluctuating dynamic process. The example shows that the weighted SCGM(1,1)c model can have high prediction precision for drought crop area.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2009年第9期179-185,共7页
Systems Engineering-Theory & Practice
基金
国家公益性行业科研专项(200801039)