灰色神经网络在粮食产量预测中的应用

研究粮食准确预测优化问题,粮食产量受到多种因素影响,同时具有复杂的非线性和随机性特点,传统单一模型难准确对其变化规律进行准确描述,预测精度较低。为提高粮食产量预测精度,提出一种将灰色理论和BP神经网络相结合的粮食产量预测模型。首先采用灰色GM(1,1)预测模型动态预测粮食产量变化趋势,然后运用BP神经网络对灰色GM(1,1)模型预测结果进行修正,以提高粮食产量预测精度。采用1978-2008年我国粮食产量数据对预测模型性能进行仿真测试,仿真结果表明,组合预测模型提高了粮食产量的预测精度,更能描述粮食产量变化规律,为粮食产量准确预测提供了一种有效研究方法。

Application of Grey Model and Neural Network in Grain Production Prediction

Study the problems of food safety.Food production is affected by many factors and has the characteristics of fluctuation and stochastic,so that single model can not accurately describe the change rule.In the paper,the gray theory and BP neural network were combined to establish a combination forecasting model of food production.First,it used the gray GM(1,1) prediction model to predict the grain yield chan grain yield ging trend.Then,it used the BP neural network to modify the predicting results by gray GM(1,1) model to improve the prediction precision.Using 1978-2008 years crop of our country grain for the performance of the predicting model was tested with the grain yields of China from 1978 to2008.The results show that the combined forecasting model increases the prediction accuracy and can describe the changing rule of the grain yields.