一种新型自适应变权累加优化的AVW-DGM(1,1)模型

A New AVW-DGM(1,1) Model of Adaptive Variable Weight Accumulation

高等教育的发展是国家科技创新,实现创新发展的源动力。针对经典DGM(1,1)模型的原始数据序列以常数1进行累加这一不足,本文提出基于变权累加优化DGM(1,1)模型,简称AVW-DGM(1,1)模型。以中国的研究生,硕士研究生,本专科生和本科生的招生人数为数值实例,分别建立DGM(1,1)模型和AVW-DGM(1,1)模型进行模拟预测,并用粒子群算法对AVW-DGM(1,1)模型的加权系数进行优化求解。结果表明,在本文提供的四个数值实例中,AVW-DGM(1,1)模型均比经典的DGM(1,1)模型有更高的模拟和预测精度。由此可知,通过粒子群算法实现原始数据序列自适应累加,可以使得一阶累加序列更符合DGM(1,1)模型对数据特征的要求,从而提高模拟和预测精度。

The development of higher education is the source power of national innovation in science and technology. The original data sequence of the classical DGM(1,1) model is accumulated with the constant 1. Therefore, this paper proposes an optimization DGM(1,1) model based on variable weight accumulation. The DGM(1,1) model and AVW-DGM(1,1) model are established respectively to simulate and predict the enrollment of graduate students, master students, specialized subject of raw and undergraduates in China. Next, using PSO to optimize the weighting coefficient of the AVW-DGM(1,1) model. The four numerical examples provided show that the AVW-DGM(1,1) models all have higher simulation and prediction accuracy than the classical DGM(1,1) models in this paper. It can be seen that the adaptive accumulation of original data series through particle swarm optimization algorithm can make the first-order accumulation sequence more in line with the requirements of DGM(1,1) model for data characteristics, thus improving the accuracy of simulation and prediction.