Cited by Fast Prediction of Car Driving Direction Velocity Field Based on Convolutional Neural Network with Data of Flow Simulation Nodes after Feature Enhancement

AI-infused characteristics prediction and multi-objective design of ultra-high performance concrete (UHPC): From pore structures to macro-performance DOI

Wangyang Xu,

Lingyan Zhang,

Dingqiang Fan

et al.

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 111170 - 111170

Published: Oct. 1, 2024

Language: Английский

Citations

Considering integrated information on environmental features and neighborhood deformation: A missing value filling framework for arch dam deformation sequence DOI

Xudong Chen, Wenhao Sun, Yajian Liu

et al.

Advanced Engineering Informatics, Journal Year: 2024, Volume and Issue: 63, P. 102959 - 102959

Published: Dec. 2, 2024

Language: Английский

Citations

A Combined Noise Reduction Method for Floodgate Vibration Signals Based on Adaptive Singular Value Decomposition and Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise DOI

Wentao Wang,

Huiqi Zhu,

Y. Cheng

et al.

Water, Journal Year: 2023, Volume and Issue: 15(24), P. 4287 - 4287

Published: Dec. 15, 2023

To address the issue of vibration characteristic signals floodgates being affected by background white noise and low-frequency water flow noise, a reduction method combining improved adaptive singular value decomposition algorithm (ASVD) complete ensemble EMD with (ICEEMDAN) is proposed. Firstly, Hankel matrix constructed based on collected discrete time signals. After performing SVD matrix, ASVD used to automatically select effective values filter out most retain useful frequency components similar energy in signal. Then, ICEEMDAN combined Spearman correlation coefficient further residual flows. The performance this verified through simulation experiments. Filtered ASVD-ICEEMDAN method, signal-to-noise ratio signal (50% level) increased from 4.417 16.237, root mean square error reduced 2.286 0.586. Based practically measured floodgate at large hydropower station, result shows that exhibits good feature information extraction abilities for signals, can provide support operational mode analysis damage identification practical structures under complex interference conditions.

Language: Английский

Citations

Arch dam point cloud segmentation based on deep feature learning and normal vector data optimization DOI

Huokun Li, Yuekang Li,

yijing li

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: April 5, 2024

Abstract Separating the dam body, spillway, and other structures from point cloud in area is an important step deformation monitoring. Manual segmentation time consuming inaccurate. This study proposes a neural network model based on normal vector optimization suitable for environment: 1) utilizes voxel uniform sampling method of equal length cubes to solve problem uneven density caused by wide range long distance measurement during areas. 2) Designed block input combined output modules model, achieving efficient large volume eliminating impact interpolation points offset seq2seq decoding process. 3) In response diverse characteristics vectors presented vegetation, rock mass, complex area, this paper adaptive radius plane fitting estimation eigenvalue improve accuracy segmentation. Experiments prototype arch show that proposed improves classification PointNet + original 96.26–98.27%. Compared with three methods (2-jets, Hough CNN, iterative PCA), overall has improved 0.82%, 1.22%, 0.22%, mean intersection over union 0.0293, 0.0325, 0.0104. provides high-precision scheme applications such as detection cloud.

Language: Английский

Citations

Fast Prediction of Car Driving Direction Velocity Field Based on Convolutional Neural Network with Data of Flow Simulation Nodes after Feature Enhancement DOI

Shengrong Shen, Tian Han, Jiachen Pang

et al.

Published: Jan. 1, 2024

Language: Английский

Citations