With the growth of industrialization and the logistics industry,steel structure shelves have become increasingly vital in storage and logistics systems,making accurate deformation prediction essential.This paper presents a prediction algorithm based on Bidirectional Long Short-Term Memory Networks(BiLSTM)and Attention mechanisms.Utilizing time series data and training through deep learning models,this algorithm allows for detailed analysis and prediction of deforma-tions in steel structure shelves.The model's performance,validated by a typical application,shows high robustness and exceptional predictive accuracy.Experimental results indicate that the model can accurately predict deformations in steel structure shelves,with an average error of only 0.15%to 3.33%.These outcomes suggest potential applications of this algorithm in the automated monitoring of steel structure shelves,providing a viable solution for predicting structural de-formations.
automated monitoring technology/deep learning/time series data/Bidirectional Long Short-Term Memory Networks with an Attention mechanism(BiLSTM-Attention)
维普
万方数据