Опубликована: Авг. 8, 2024
Язык: Английский
Transportation Engineering, Год журнала: 2024, Номер 16, С. 100243 - 100243
Опубликована: Март 11, 2024
Pavement performance prediction is crucial for ensuring the longevity and safety of road networks. In our extensive study, we employ a diverse array techniques to enhance fatigue models in flexible pavements. The methodology begins with Random Forest feature selection, identifying top 15 critical variables that significantly impact pavement performance. These form basis subsequent model development. Our investigation into indicates superiority advanced machine learning methods such as Regression Trees (RT), Gaussian Process (GPR), Support Vector Machines (SVM), Ensemble (ET), Artificial Neural Networks (ANN) over traditional linear regression methods. This consistent outperformance underscores their potential reshape forecasting accuracy. Through optimization, reveal robust across both complete selected sets, emphasizing importance meticulous selection enhancing forecast accuracy best optimized highlighted by its Performance Measurement metrics: RMSE 22.416, MSE 502.46, R-squared 0.80848, MAE 8.9958. Additionally, comparative analysis previous empirical demonstrates outperforms existing models. work significance curation prediction, highlighting sophisticated modeling methodologies. Embracing cutting-edge technologies facilitates data-driven decisions, ultimately contributing development more networks, safety, prolonging lifespan.
Язык: Английский
Процитировано
19
Buildings, Год журнала: 2025, Номер 15(2), С. 207 - 207
Опубликована: Янв. 11, 2025
The correlation analysis between current surface cracks of structures and external loads can provide important insights into determining the structural residual bearing capacity. classical regression assessment method based on experimental data not only relies costly structure experiments; it also lacks interpretability. Therefore, a novel load estimation for RC beams, detected crack images strain contour plots calculated by FEM, is proposed. distinct discrepancies figures, coupled with stochastic nature actual distributions, pose considerable challenges tasks. new index model initially introduced to quantify two types in proposed method. Subsequently, deep neural network (DNN) trained as FEM surrogate quickly predict response considering material uncertainties. Ultimately, range optimal level its confidence interval are determined via statistical estimations under different random fields. validation results beams four-point bending show that algorithm estimate levels numerical simulation results, mean absolute percentage error (MAPE) solely single measured image 20.68%.
Язык: Английский
Процитировано
1
Automation in Construction, Год журнала: 2025, Номер 172, С. 106045 - 106045
Опубликована: Фев. 7, 2025
Язык: Английский
Процитировано
1
Sustainability, Год журнала: 2024, Номер 16(21), С. 9329 - 9329
Опубликована: Окт. 27, 2024
Currently, the environment and its natural resources face many issues related to depletion of resources, in addition increase environmental pollution resulting from uncontrolled waste disposal. Therefore, it is crucial identify practical effective ways utilize these wastes, such as transforming them into environmentally friendly concrete. Artificial lightweight aggregates (ALWAs) are gaining interest because their shift focus aggregates. Researchers have developed numerous ALWAs eliminate need for This article explores diverse applications across different industries. currently research phase due various limitations compared availability that form more durable solutions. However, researchers discovered certain artificial prioritize weight over strength, allowing use like pavements. We thoroughly studied discussed this found fly ash construction most sources primary material ALWAs. production also presents challenges terms processing optimization. article’s case study reveals ALWAs, consisting 80% ash, 5% blast-furnace slag, only 15% cement, can yield a sustainable solution. In single- double-step palletization, aggregate proved be less harmful. Additionally, has reduced carbon footprint recycling materials, including derived marble sludge, ground granulated slag. Despite limited mechanical exhibit superior performance, making suitable high-rise buildings landscapes. composition plays key role determining application-based properties discusses sustainability considerations, well future trends LWA field. Simultaneously, reduce promote construction. researches associated with
Язык: Английский
Процитировано
4
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0
Journal of Chromatography B, Год журнала: 2025, Номер 1259, С. 124599 - 124599
Опубликована: Апрель 26, 2025
Язык: Английский
Процитировано
0
Journal of Infrastructure Systems, Год журнала: 2024, Номер 30(4)
Опубликована: Сен. 28, 2024
Язык: Английский
Процитировано
3
Elsevier eBooks, Год журнала: 2025, Номер unknown, С. 255 - 285
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
0
Journal of Engineering Research, Год журнала: 2025, Номер unknown
Опубликована: Фев. 1, 2025
Язык: Английский
Процитировано
0
Signal Image and Video Processing, Год журнала: 2025, Номер 19(4)
Опубликована: Фев. 24, 2025
Язык: Английский
Процитировано
0