对于复杂工业系统非线性时间序列预测精度不高问题,引入了多种预测方法的预测相对误差、预测对象的变化趋势、灰色基本权重和自适应调节系数等概念,建立了模糊自适应变权重非线性组合预测模型。结果表明,此模糊自适应变权重非线性组合预测模型的精度较高,并且平均误差和预测平方根误差均较小。该组合预测模型为复杂非线性工业系统所需决策提供了有力支持。
As for the low forecasting precision problem about nonlinear time series in complex industry system, a nonlinear combined forecasting model was established by using of the conception such as the relative error, the tendency of the forecasted object, gray basic weight and adaptive control coefficient due to the method of fuzzy variable weight. The results reveal that the forecasting precision of the nonlinear combined forecasting model is higher than that of some single combined forecasting models. The combined forecasting model is very useful for requirement decision in complex industry system.