LSTM-AR Prediction Method for Bridge Foundation Settlement in Goaf Under Traffic Load
摘要
In order to ensure the driving safety of bridges in goaf areas, it is necessary to predict the foundation settlement of bridges in goaf areas and prevent potential dangerous accidents in a timely manner. Based on DB wavelet, long short-term memory neural network model, and autoregressive model, this article proposes an LSTM-AR prediction model for bridge foundation settlement in goaf considering engineering noise. The model is used to predict two bridge foundations, and the reliability of the model is verified through comparative analysis. The research results provide theoretical data support for the management and maintenance department to correctly analyze the operational safety of bridge foundations in goaf areas.