摘要
基于分数阶反向累加生成构建一种新的GM(1,2)模型,为使所构建模型能更好贴近和反映两个累加生成序列指标之间的真实关联关系,提出了基于不同序列采用不同累加阶数的GOM^((p,q))(1,2)模型.首先通过灰关联模型识别并筛选与特征序列关联度最大的相关因素序列,然后建立不同累加阶数的灰色模型,通过带压缩因子的粒子群优化算法求解模型最优阶数p和q,最后运用BP神经网络修正GOM^((p,q))(1,2)的模型值,构建GOM^((p,q))(1,2)-BP神经网络组合模型.模型应用于武汉市空气质量指数的预测,结果表明与单一模型相比,组合模型具有更好的性能和建模精度.
In this paper,a new GM(1,2)model is constructed based on the fractional order reverse cumulative generation.In order to make the constructed model better approach and reflect the true correlation between the two cumulative generation sequence indicators,a GOM^((p,q))(1,2)model with different cumulative orders is proposed based on different sequences.Firstly,the sequence of related factors with the highest correlation degree with the feature sequence is identified and screened through the grey correlation model,and then grey models with different cumulative orders are established.The particle swarm optimization algorithm with compression factor is used to solve the optimal order P and Q of the model.Finally,the BP neural network is used to modify the model value of GOM^((p,q))(1,2)and build the GOM^((p,q))(1,2)-BP neural network combination model.The model is applied to the prediction of air quality index in Wuhan,and the results show that the combination model has better performance and modeling accuracy.
作者
邹旺
刘军
柳福祥
ZOU Wang;LIU Jun;LIU Fu-xiang(School of Science,China Three Gorges University,Yichang 443002,China;Three Gorges Mathematical Research Center,China Three Gorges University,Yichang 443002,China)
出处
《数学的实践与认识》
2023年第4期151-161,共11页
Mathematics in Practice and Theory