Advanced informatic technologies for intelligent construction: A review

Engineering Applications of Artificial Intelligence, Год журнала: 2024, Номер 137, С. 109104 - 109104

Опубликована: Авг. 29, 2024

Язык: Английский

---

SDCGAN: A CycleGAN-Based Single-Domain Generalization Method for Mechanical Fault Diagnosis

Reliability Engineering & System Safety, Год журнала: 2025, Номер unknown, С. 110854 - 110854

Опубликована: Янв. 1, 2025

Язык: Английский

---

Smart virtual sensing for deep excavations using real-time ensemble graph neural networks

Automation in Construction, Год журнала: 2025, Номер 172, С. 106040 - 106040

Опубликована: Фев. 3, 2025

Язык: Английский

---

Unmanned aerial vehicle–human collaboration route planning for intelligent infrastructure inspection

Computer-Aided Civil and Infrastructure Engineering, Год журнала: 2024, Номер 39(14), С. 2074 - 2104

Опубликована: Фев. 27, 2024

Abstract Motivated by the strengths of unmanned aerial vehicle (UAV), UAV–human collaboration route planning (UHCRP) for intelligent infrastructure inspection is a problem worthy discussion to help reduce human costs and minimize risk noninspected infrastructures under limited resources. To facilitate UHCRP, this paper proposes novel deep reinforcement learning (DRL)‐based approach well handle multi‐source uncertain features constraints at fast speed. begin with, UHCRP mathematically described reformulated as dual interdependent (diDRL) framework reflect real‐world scenarios. Afterward, policy network named attention‐based neural (A‐DNN) introduced learn decisions combinatorial optimization problem. In particular, A‐DNN made up an encoder decoder UAV inspection, where multi‐head attention mechanism incorporated generate richer representations model performance improvement. Performance proposed (DAM) has been tested in simulations case study regarding wind farm inspection. Results indicate that DAM sampling decoding strategy can deliver high‐quality path plan show better generalizability larger scale sizes compared single‐head (SAM), (AM), two baseline models, namely OR‐Tools genetic algorithm. Moreover, trained randomly generated data be directly employed solve practical with standardization inputs. Overall, DRL integrates decision‐making method selection inspected selection, providing adaptive complex dynamic engineering environments.

Язык: Английский

---

Integration of FEM and DL for seismic performance prediction and optimization design of tunnels

Tunnelling and Underground Space Technology, Год журнала: 2025, Номер 161, С. 106535 - 106535

Опубликована: Март 4, 2025

Язык: Английский

---

Temporal-spatial-fusion-based risk assessment on the adjacent building during deep excavation

Information Fusion, Год журнала: 2024, Номер 114, С. 102653 - 102653

Опубликована: Авг. 27, 2024

Язык: Английский

---

Spatiotemporal deep learning for multi-attribute prediction of excavation-induced risk

Automation in Construction, Год журнала: 2025, Номер 171, С. 105964 - 105964

Опубликована: Янв. 17, 2025

Язык: Английский

---

Analysis of Temporal and Spatial Characteristics and Influencing Factors of Construction Deformation of Super-Large Deep Foundation Pit in Thick Sand Stratum

Applied Sciences, Год журнала: 2025, Номер 15(7), С. 3553 - 3553

Опубликована: Март 25, 2025

To the aim of this paper is to study structural and environmental deformation characteristics caused by excavation a very large deep foundation pit in sandy soil area Beijing. This based on numerical simulation field monitoring results these are compared with data soft Shanghai area. The show that influence environment surrounding super-large project studied obviously too great. With progress construction, rate amount column at side higher than middle Due “hysteresis” stress transfer sand, settlement roof north wall delayed range smaller south wall. Compared conventional pit, surface larger, reaching 4 He (He depth pit). Δvmax (the maximum settlement) between 0.2~2.3% He, relationship δvmax = 1.43% Vwm. Through orthogonal experiments simulation, it concluded structure its more sensitive unloading, precipitation amplitude, spacing. It also strong, medium, weak areas bottom uplift after construction (0~0.07) × L, (0.07~0.14) (0.14~0.5) respectively (L width When embedment ratio 1.8~2.4, displacement mode parapet T mode; when 2.4~3.4, RB mode.

Язык: Английский

---

Predicting excavation-induced lateral displacement using improved particle swarm optimization and extreme learning machine with sparse measurements

Underground Space, Год журнала: 2025, Номер unknown

Опубликована: Май 1, 2025

Язык: Английский

---

Prediction of shield tunneling attitudes: A muti-dimensional feature synthesizing and screening method

Journal of Rock Mechanics and Geotechnical Engineering, Год журнала: 2024, Номер unknown

Опубликована: Ноя. 1, 2024

Язык: Английский

---