首页|基于时序序列分解和IBAS-LSTM的滑坡数据预测模型

基于时序序列分解和IBAS-LSTM的滑坡数据预测模型

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
针对传统静态机器学习模型在周期项位移预测中的缺陷和动态神经网络超参数人工选择困难的问题,在时序序列分解的基础上,提出一种新的滑坡预测耦合模型.首先,用最大相关最小冗余算法对周期项位移筛选合适的环境特征,作为长短期记忆人工神经网络的输入.然后,在天牛须搜索算法搜索过程中引入反馈机制,以避免原算法中出现远离最优解的问题;在算法迭代过程中将固定的递减因子改为动态递减因子,以提升前期全局和后期局部的寻优能力;利用改进的天牛须搜索算法对长短期记忆人工神经网络超参数进行寻优,以获得最佳的网络参数组合.最后,重构趋势项和周期项预测结果,得到最终预测位移.以发耳滑坡为例进行分析,结果表明:相较于其他方法,所提模型在平均绝对误差、均方根误差以及拟合度等方面更具优势.
Landslide Displacement Prediction Model Based on Time Series Decomposition and IBAS-LSTM
In order to solve the problems of the defects of the traditional static machine learning model in the displace-ment prediction of periodic terms and the difficulty of manual selection of hyperparameters of dynamic neural net-works,a new coupling model for landslide prediction was proposed on the basis of time series sequence decomposi-tion.Firstly,the maximum correlation and minimum redundancy algorithm was used to screen suitable environmental features for the displacement of periodic items,which were used as the input of long short-term memory artificial neural network.Then,a feedback mechanism was introduced into the search process of the beetle antennae search al-gorithm to avoid the problem of deviating from the optimal solution in the original algorithm.In the process of algo-rithm iteration,the fixed decreasing factor is changed to a dynamic decreasing factor to improve the optimization abil-ity of the early global and late local optimization.The improved long-term memory artificial neural network hyperpa-rameters were optimized by using the improved beetle antennae search algorithm to obtain the best combination of network parameters.Finally,the prediction results of trend term and period term are reconstructed to obtain the final predicted displacement.Taking the Fa'er landslide as an example,the results show that compared with other meth-ods,the proposed model has more advantages in terms of mean absolute error,root mean square error and fit.

dynamic neural network modeltemporal sequence decompositiongrey modellong short-term memorybeetle antennae search algorithm

荆严飞、党建武、王阳萍、岳彪

展开 >

兰州交通大学 电子与信息工程学院,兰州 730070

兰州交通大学 轨道交通信息与控制国家级虚拟仿真实验教学中心,兰州 730070

动态神经网络模型 时序序列分解 灰色模型 长短期记忆人工神经网络 天牛须搜索算法

国家自然科学基金甘肃省教育科技创新项目甘肃省军民融合专项甘肃省重点研发计划甘肃省知识产权计划

617630252021jyjbgs-052020JG0121YF5GA15821ZSCQ013

2024

10.3969/j.issn.2096-9066.2024.02.008

兰州交通大学学报
兰州交通大学

兰州交通大学学报

影响因子：0.532
ISSN：1001-4373
年,卷(期)：2024.43(2)
  • 23