基于灰色预测和SVM的茶叶烘焙温湿度预测

Prediction of the Baking Temperature and Humidity of Tea Based on Grey Prediction and Support Vector Machine(SVM)

针对绿茶烘焙这一过程的复杂特性,提出了一种基于灰色预测和SVM的绿茶烘焙过程温湿度预测模型.该模型首先利用灰色预测和SVM分别对绿茶烘焙过程中的温湿度做预估,为了将两者进行有机融和,笔者选用熵值法确定各预测子模型的加权系数,进而得到了更为精准的绿茶烘焙过程温湿度预测模型结果,保证温湿度的稳定控制.仿真结果验证了该方法的有效性.

In view of the complex characteristics of the baking process of green tea,a prediction model of temperature and humidity for the baking process of green tea based on grey prediction and support vector machine（ SVM） is proposed in this paper. Firstly,the temperature and humidity during the baking process of green tea were estimated by grey prediction and SVM respectively. In order to integrate the two ones,the entropy method is used to determine the weighting coefficients of each sub-model of prediction. Therefore,a more accurate prediction model of temperature and humidity in the baking process of green tea is obtained,which ensures the stable control of baking temperature and humidity. The simulation results verify the effectiveness of the method.