论文部分内容阅读
采用支持向量机的组合预测方法,对黑龙江垦区农机装备水平进行预测。在确定单一预测模型的基础上,运用自组织神经网络方法,将权系数确定问题转化为粗糙集理论中属性重要性评价的问题;计算各单一预测方法对组合模型的依赖度、重要度和权系数;利用建立的基于支持向量机非线性农机装备水平组合预测模型,对黑龙江垦区2002—2012年农机装备水平的历史数据进行检验。误差分析表明:该模型对农机总动力、大中型拖拉机、小型拖拉机、大中型拖拉机配套机具和小型拖拉机配套机具的预测平均相对误差为0.471%、1.328%、3.738%、1.193%、3.574%,均低于各单一预测模型的平均相对误差;利用该模型对黑龙江垦区农机装备水平进行预测,到2020年拥有农机总动力999.33万kW、大中型拖拉机88 921台、小型拖拉机38 453台,大中型拖拉机与配套农机具台数比为1.51∶1,小型拖拉机与配套农机具台数比为1.68∶1。所建模型适用于黑龙江垦区农机装备水平的预测。
The combined forecasting method based on support vector machine is used to predict the level of agricultural equipment in Heilongjiang Reclamation Area. On the basis of determining the single forecasting model, the self-organizing neural network method is used to convert the problem of determining the weight coefficient into the evaluation of the importance of attributes in the rough set theory. The dependence, importance and weight of each single forecasting method on the model are calculated Coefficient; using the established non-linear agricultural machinery and equipment level combination forecasting model based on support vector machine to test historical data of farm machinery equipment level in Heilongjiang reclamation area from 2002 to 2012. The error analysis shows that the average relative error of this model is 0.471%, 1.328%, 3.738%, 1.193%, 3.574% for the total power of agricultural machinery, medium and large tractors, small tractors, medium and large tractor supporting equipment and small tractor supporting equipment Which is lower than the average relative error of each single forecast model. Using this model to predict the level of agricultural machinery and equipment in Heilongjiang Reclamation Area, it will have 9,993,300 kW of agricultural machinery, 88,921 medium and large tractors, 38,453 small tractors, medium and large tractors And the number of supporting agricultural machines is 1.51: 1, the number of small tractors and supporting agricultural machines is 1.68: 1. The model is suitable for the forecast of agricultural machinery and equipment in Heilongjiang Reclamation Area.