Probabilistic data self-clustering based on semi-parametric extreme value theory for structural health monitoring

Mechanical Systems and Signal Processing, Journal Year: 2022, Volume and Issue: 187, P. 109976 - 109976

Published: Nov. 28, 2022

Language: Английский

---

Buckling and ultimate load prediction models for perforated steel beams using machine learning algorithms

Journal of Building Engineering, Journal Year: 2022, Volume and Issue: 51, P. 104316 - 104316

Published: March 11, 2022

Language: Английский

---

Prediction of the load-shortening curve of CFST columns using ANN-based models

Journal of Building Engineering, Journal Year: 2022, Volume and Issue: 51, P. 104279 - 104279

Published: Feb. 26, 2022

Language: Английский

---

Rapid seismic damage state assessment of RC frames using machine learning methods

Journal of Building Engineering, Journal Year: 2022, Volume and Issue: 65, P. 105797 - 105797

Published: Dec. 29, 2022

Language: Английский

---

A Hybrid ANN-GA Model for an Automated Rapid Vulnerability Assessment of Existing RC Buildings

Applied Sciences, Journal Year: 2022, Volume and Issue: 12(10), P. 5138 - 5138

Published: May 19, 2022

Determining the risk priorities for building stock in highly seismic-prone regions and making final decisions about buildings is one of essential precautionary measures that needs to be taken before earthquake. This study aims develop an Artificial Neural Network (ANN)-based model predict reinforced-concrete (RC) constitute a large part existing stock. For this purpose, network parameters structure have been optimized by establishing hybrid with Genetic Algorithm (GA). As result, ANN can make accurate predictions maximum efficiency. The suggested feedforward back-propagation model. It 329 RC most successful way, which performance score was calculated using Turkey Rapid Evaluation Method (2013). In paper, GA-ANN implemented ANN, gene revealed model, produced results calculating score. addition, required input obtaining more efficient solving such problem need used optimized. With help operation process will eliminated. created 98% determining priority buildings.

Language: Английский

---

Assessing the compressive and splitting tensile strength of self-compacting recycled coarse aggregate concrete using machine learning and statistical techniques

Materials Today Communications, Journal Year: 2023, Volume and Issue: 38, P. 107970 - 107970

Published: Dec. 28, 2023

The construction industry is adopting high-performance materials due to technological and environmental advances. Researchers worldwide are studying the use of recycled coarse aggregates (RCA) as a partial alternative natural in concrete their sustainability benefits. This study compares predictive abilities three different machine learning techniques evaluating mechanical properties 28-day-old self-compacting (SCC) incorporating RCA better understand how design parameters affect SCC containing RCA. used range statistical methodologies algorithms, such ANN, SVM, M5P trees, examine relationship between elements accurately forecast characteristics concrete. ANN model exhibited notable superiority effectively forecasting compressive strength (CS) splitting tensile (STS) compared other models, with uncertainty bands 15.038%-21.154% for CS 15.701%-19.008% STS. Moreover, all uncertainties were under threshold 35%. Notably, water-cement ratio emerged most crucial parameter predicting SCC. Finally, parametric evaluation conducted revealed that STS inversely proportional aggregate-cement ratio, whereas, directly water-binder water-solids percentage.

Language: Английский

---

Assessment of Convolutional Neural Network Pre-Trained Models for Detection and Orientation of Cracks

Materials, Journal Year: 2023, Volume and Issue: 16(2), P. 826 - 826

Published: Jan. 14, 2023

Failure due to cracks is a major structural safety issue for engineering constructions. Human examination the most common method detecting crack failure, although it subjective and time-consuming. Inspection of civil structures must include detection categorization as key component process. Images can automatically be classified using convolutional neural networks (CNNs), subtype deep learning (DL). For image categorization, variety pre-trained CNN architectures are available. This study assesses seven networks, including GoogLeNet, MobileNet-V2, Inception-V3, ResNet18, ResNet50, ResNet101, ShuffleNet, categorization. diagonal (DC), horizontal (HC), uncracked (UC), vertical (VC). Each architecture trained with 32,000 images equally divided among each class. A total 100 from category used test models, results compared. Inception-V3 outperforms all other models accuracies 96%, 94%, 92%, 96% DC, HC, UC, VC classifications, respectively. ResNet101 has longest training time at 171 min, while ResNet18 lowest 32 min. research allows best automatic orientation selected, based on accuracy taken model.

Language: Английский

---

A systematic review of machine learning approaches in carbon capture applications

Journal of CO2 Utilization, Journal Year: 2023, Volume and Issue: 71, P. 102474 - 102474

Published: April 13, 2023

Language: Английский

---

Potential Role and Challenges of ChatGPT and Similar Generative Artificial Intelligence in Architectural Engineering

SSRN Electronic Journal, Journal Year: 2023, Volume and Issue: unknown

Published: Jan. 1, 2023

The incorporation of generative artificial intelligence (AI) systems, such as ChatGPT, holds great potential in reshaping diverse facets architectural engineering. This research investigates the profound influence AI technologies on structural engineering, HVAC (Heating, Ventilation, and Air Conditioning) electrical plumbing fire protection sustainability, net zero, green building design, information modeling (BIM), urban planning, project management. In ChatGPT's capacity to analyse extensive datasets simulate intricate structures expedites design process, ensuring integrity while optimizing materials costs. it aids devising energy-efficient systems climate control solutions, significantly contributing sustainable practices. Similarly, AI's capabilities enhance optimization both safety reliability. ChatGPT assists creating efficient layouts suppression compliance with regulations. Moreover, plays a pivotal role advancing sustainability design. By evaluating environmental factors suggesting eco-friendly designs, fosters development environmentally responsible structures. domain BIM, facilitates seamless collaboration, automates model generation, improves clash detection, streamlined execution. Nevertheless, integration engineering presents challenges. Ethical concerns, data security, necessity for skilled professionals interpret AI-generated insights are significant issues. delves into these contribution challenges effectively harness AI, paving way transformative era

Language: Английский

---

Evaluating fire resistance of timber columns using explainable machine learning models

Engineering Structures, Journal Year: 2023, Volume and Issue: 296, P. 116910 - 116910

Published: Sept. 21, 2023

Language: Английский

---