中文摘要：

针对具有跳跃性的中长时数据预测,提出一种改进加权灰色GM（1,1）模型对高速公路收费站交通量进行预测。将原始交通量数据经过1阶弱化和1-AGO处理后,利用灰色关联度对初始值的取值进行加权优化,同时对背景值采取光滑优化处理,从而组合成新型灰色GM（1,1）模型。应用某收费站实际交通量统计数据来验证新型灰色GM（1,1）模型算法预测准确性,结果表明：改进加权灰色GM（1,1）模型具有更好的适用性和准确性。

英文摘要：

Targeting at leaping mid/long-term data prediction,this paper put forwards an optimized weight gray GM（1,1） model to forecast traffic volume of expressway toll station. After the first-order weakening and 1-AGO processing of the original traffic volume data,using grey correction to assign weight optimization to the value of initial value,in the same time,the background value is smoothened to make a new grey GM（1,1） model. In this paper,actual traffic statistics of a toll station is used to verify the algorithm predication accuracy of the new gray GM（1,1） model. The results show that the optimized gray GM（1,1） model has better applicability and accuracy.