摘要
研究粮食准确预测问题,由于粮食产量受到多种高维非线性、随机性和有限样本等因素的影响,单一模型不能全面描述其变化趋势,预测精度较低。为了提高粮食产量预测精度,提出一种将灰色关联支持向量机的粮食产量预测模型。首先采用灰色关联分析确定粮食产量变化主要影响因子,然后通过支持向量机学习建立粮食产量与因子之间的非线性映射关系,最后为避免人为选择参数的盲目性,采用遗传算法确定支持向量机参数并对将来粮食产量进行预测。利用1978-2011年我国粮食产量进行仿真,并将预测结果与单一机模型进行对比。结果表明,灰色关联支持向量机提高了粮食产量的预测精度,可以较好地应用于粮食产量预测中。
In order to improve the grain yield prediction accuracy, this paper presented a grain output prediction model of grey support vector machine. First, grey correlation analysis was used to identify the main factors affecting the changes in grain output, and then by support vector machine, the nonlinear mapping relation between grain crop and factors was built. Finally, to avoid the blindness of artificial selecting parameters, a genetic algorithm was used to determine the parameters of support vector machine and the future of grain output prediction. The simulation experi-ments are carried out using the grain crops of China from 1978 to 2011. The predicted results were, compared with a single machine model, and the results show that, the grey support vector machine can increase the prediction accura-cy of grain output.
出处
《计算机仿真》
CSCD
北大核心
2012年第9期220-223,227,共5页
Computer Simulation
关键词
粮食产量
支持向量机
灰色关联
预测
Grain production
Support vector machine（SVM）
Grey relation
Prediction
