High-Precision Estimation Model of Water Gate Structure Strain
ZHAO Guochao
Abstract：Strain is one of the key factors affecting the stability of water gate structures,making a high-precision strain estimation model essential for safe operation management.To obtain such a model,based on multi-year in-situ strain data,a hybrid deep learning model(TB)was constructed using a Temporal Convolutional Network(TCN)and a Bidirectional Long Short-Term Memory network(BiLSTM).The Artificial Lemming Algorithm(ALA),Tornado Optimization Algorithm(TOC),and Sparrow Search Algorithm(SSA)were employed to optimize the TB model,resulting in the ATB,TTB,and STB models for strain estimation.Results showed that the ATB model′s estimated track the measured values most closely,with a fitted slope of 0.981 and superior fitting performance.The estimation error of this model was only 4.493%and 3.322%,with a consistency exceeding 0.96.The ATB model demonstrated the highest accuracy among all models,enabling high-precision estimation of water gate structure strain.
Keywords：strain estimationTemporal Convolutional Network(TCN)Bidirectional Long Short-Term Memory(BiLSTM)Artificial Lemming Algorithm(ALA)
Publication Date：2025-12-20
Online Publishing Date：2026-03-04(First online date of this platform, not the publication date of the document)
Pages：4( 98-101 )