基于灰色神经网络组合模型的城区沉降预测应用研究
Application Research of Urban Settlement Prediction Based on Grey Neural Network Combination Model
雷倩芳 1左涛 2杨晓东 2韩冲 2陈恒恒1
作者信息
- 1. 中煤航测遥感集团有限公司,陕西 西安 710199
- 2. 中煤航测遥感集团有限公司,陕西 西安 710199;中煤航测遥感集团有限公司企业技术中心,陕西 西安 710199
- 折叠
摘要
为了提高城区沉降量的预测精度,本文基于灰色预测模型(Grey Model,GM)和反向传播(Back-Propaga-tion,BP)神经网络预测模型,构建了灰色神经网络组合预测模型(Gery Neural Networks Model,GNNM).以SBAS-InSAR技术获取上海特征区的平均沉降量作为 3 种预测模型的原始序列,进行预测计算,对比分析组合预测模型预测结果与灰色模型、BP神经网络模型的预测结果.实验结果表明:相比单一的GM(1,1)和神经网络预测模型,GNNM(1,1)组合预测模型的预测精度和稳定性更高,且越接近中心城区,预测效果越好.
Abstract
In order to improve the prediction accuracy of urban subsidence,based on Grey Model(GM)and back propagation(BP)neural network prediction model,this paper constructs grey neural networks model(GNNM).Taking the average settlement of Shang-hai characteristic area obtained by SBAS-InSAR technology as the original sequence of the three prediction models,the prediction cal-culation is carried out,and the prediction results of the combined prediction model,grey model and BP neural network model are com-pared and analyzed.The experimental results show that compared with the single GM(1,1)and neural network prediction model,the GNNM(1,1)combined prediction model has higher prediction accuracy and stability,and the closer it is to the central urban area,the better the prediction effect.
关键词
灰色神经网络模型/灰色模型/BP神经网络/沉降预测Key words
grey neural network model/grey model/BP neural network/settlement prediction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据