论文部分内容阅读
灰色预测方法和人工神经网络,在建筑物变形预测中有各自的优势和不足。为了提高预测精度,该文结合灰色GM(1,1)模型和BP神经网络模型的特点,构造并联型灰色神经网络模型(PGNN)对南京地铁隧道某监测点的沉降量进行预测。结果显示,PGNN的预测精度明显高于单一的灰色GM(1,1)模型和BP神经网络模型,证明了PGNN组合方法在地铁隧道沉降量预测中的有效性。
Gray prediction method and artificial neural network have their own advantages and disadvantages in the prediction of building deformation. In order to improve the prediction accuracy, this paper constructs a parallel gray neural network model (PGNN) to predict the settlement of a monitoring point in Nanjing Subway Tunnel by combining the characteristics of gray GM (1,1) model and BP neural network model. The results show that the prediction accuracy of PGNN is obviously higher than that of a single gray GM (1,1) model and BP neural network model, which proves the effectiveness of PGNN combination method in predicting subsidence of subway tunnels.