Article
Engineering, Multidisciplinary
Kareem Othman, Hassan Abdelwahab
Summary: The study aims to provide artificial neural network (ANN) prediction models for efficiently estimating the CBR value of subgrade soil in Egypt. Deep neural networks outperform shallow ANNs, and compared to traditional multiple linear regression (MLR) models, ANNs demonstrate better performance and produce highly accurate predictions.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Engineering, Chemical
Zafar Said, Nese Keklikcioglu Cakmak, Prabhakar Sharma, L. Syam Sundar, Abrar Inayat, Orhan Keklikcioglu, Changhe Li
Summary: A direct sol-gel technique was used to produce rGO-Fe3O4-TiO2 ternary hybrid nanocomposites, and prediction models were developed to forecast their stability and density. Machine learning techniques played a crucial role in the prediction models and proved to be more accurate than traditional analytical methods.
Article
Multidisciplinary Sciences
Mahmood Ahmad, Mohammad A. Al-Zubi, Ewa Kubinska-Jabcon, Ali Majdi, Ramez A. Al-Mansob, Mohanad Muayad Sabri Sabri, Enas Ali, Jamil Abdulrabb Naji, Ashraf Y. Elnaggar, Bakht Zamin
Summary: A new model based on Gaussian process regression (GPR) was developed to predict the CBR value of hydrated lime-activated rice husk ash (HARHA) treated soil. The GPR model showed higher accuracy compared to the artificial neural network (ANN) and gene expression programming (GEP) models in the literature.
SCIENTIFIC REPORTS
(2023)
Article
Engineering, Geological
Ashish Bastola, Xiaoqiang Gu, Kangle Zuo
Summary: This study investigates the effect of fines on the liquefaction potential of saturated silty sands through cyclic triaxial tests and centrifuge tests, emphasizing the importance of selecting the proper density index. It is found that a unique relation can be established between cyclic stress ratio, number of loading cycles required for triggering liquefaction, and equivalent granular void ratio for silty sands with fines content less than the threshold amount. The study also indicates that a finite element method using an advanced constitutive model can effectively simulate the dynamic responses of silty sands but underestimates settlement due to liquefaction, with parametric studies showing that the cyclic resistance of silty sands can be unified by the equivalent granular void ratio.
SOIL DYNAMICS AND EARTHQUAKE ENGINEERING
(2021)
Article
Engineering, Environmental
Malgorzata Kida, Sabina Ziembowicz, Kamil Pochwat, Piotr Koszelnik
Summary: The research presented in this paper investigates the environmental impact of pollutant emissions from used car tire microparticles. The study found that under different environmental conditions, these microparticles leach out various harmful chemicals and emit greenhouse gases such as carbon dioxide and methane.
JOURNAL OF HAZARDOUS MATERIALS
(2022)
Article
Chemistry, Multidisciplinary
Abolfazl Baghbani, Minh Duc Nguyen, Ali Alnedawi, Nick Milne, Thomas Baumgartl, Hossam Abuel-Naga
Summary: This study investigates the potential of artificial intelligence (AI) methods for predicting the California bearing ratio (CBR) of soils stabilized with alum sludge. Three AI methods, including two black box methods (artificial neural network and support vector machines) and one grey box method (genetic programming), were used to predict CBR, based on a database with nine input parameters. The results demonstrate the effectiveness of AI methods in predicting CBR with good accuracy.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Mechanical
Yu Hou, Xi Wang, Bihe Xu, Yangliao Geng, Qingyong Li, Di Yang
Summary: This paper proposes an experimental data-driven artificial neural network (ANN) model to accurately predict the frictional moment of cylindrical roller bearings (CRB) under various operating conditions. The complex relationship between the frictional moment and multiple operating parameters is established using the ANN model. Compared to conventional physical models, the ANN model shows a higher prediction performance for the frictional moment.
JOURNAL OF TRIBOLOGY-TRANSACTIONS OF THE ASME
(2023)
Article
Engineering, Chemical
Ashutosh Pare, Subrata Kumar Ghosh
Summary: The study utilized artificial neural networks to simulate and predict the effects of nanoparticle concentration and temperature on the thermal conductivity of nanofluids, with the final results showing good agreement with experimental data.
Article
Computer Science, Artificial Intelligence
Ali Hosseini Salari, Hossein Mirzaeinejad, Majid Fooladi Mahani
Summary: Tire normal forces are important for vehicle dynamic control systems, and accurate estimation of them can improve vehicle handling and safety. These forces change based on static parameters, road grade, and dynamic states of the vehicle. The proposed algorithm consists of two parts: an ANN-based estimation algorithm for vehicle mass and CG position, and a deep-learning-based algorithm and integrated hardware-software method for estimating dynamic normal forces.
APPLIED SOFT COMPUTING
(2023)
Article
Energy & Fuels
Houssain Zitouni, Alae Azouzoute, Charaf Hajjaj, Massaab El Ydrissi, Mohammed Regragui, Jesus Polo, Ayoub Oufadel, Abdellatif Bouaichi, Abdellatif Ghennioui
Summary: The study found that the energy drop per day reached 0.43 kWh/day during dry periods and 0.03 kWh/day during rainy periods, with an expected energy production of 5.59 kWh/day. The daily performance ratio dropped by an average of 6.1%/day and 1.6%/day during dry and rainy periods respectively. The soiling ratio during the dry period reached an average of 0.35%/day.
SOLAR ENERGY MATERIALS AND SOLAR CELLS
(2021)
Article
Engineering, Geological
Christian Carow, Frank Rackwitz
Summary: This paper focuses on the irrecoverable deformation caused by cyclic changes of mean effective stress in sands. Triaxial tests on Toyoura Sand are conducted to study the effects of mean effective stress variations. New constitutive functions for an existing Bounding Surface Plasticity model are proposed to improve the simulation results. The experimental data from the triaxial tests validate the effectiveness of the new constitutive functions.
Article
Computer Science, Artificial Intelligence
Junjie Shi, Jiang Bian, Jakob Richter, Kuan-Hsun Chen, Jorg Rahnenfuhrer, Haoyi Xiong, Jian-Jia Chen
Summary: The predictive performance of a machine learning model depends on hyper-parameter setting, making hyper-parameter tuning crucial. In distributed machine learning, collecting all data is challenging, thus the MODES framework is proposed to deploy MBO on resource-constrained distributed embedded systems to optimize combined prediction accuracy.
Article
Multidisciplinary Sciences
Lingling Fan, Kai Wang, Heming Wang, Avik Dutt, Shanhui Fan
Summary: Photonic convolution, a crucial operation in signal and image processing, can overcome computational bottlenecks and outperform electronic implementations. This study demonstrates the realization of convolution operations in the synthetic frequency dimension using a modulated ring resonator. By synthesizing arbitrary convolution kernels with high accuracy, we showcase the computation between input frequency combs and synthesized kernels. Our work highlights the efficient data encoding and computation capabilities of the synthetic frequency dimension, paving the way for compact and scalable photonic computation architecture.
Article
Energy & Fuels
Omer Boyukdipi, Gokhan Tuccar, Hakan Serhad Soyhan
Summary: The experimental study investigated the effects of NH3 as a fuel additive on engine vibration parameters, revealing that increasing levels of NH3 additive led to increased engine vibration and had a negative impact on engine vibration when blended with sunflower biodiesel. High accuracy rates were achieved in predicting vibration data through artificial neural networks models.
Article
Computer Science, Artificial Intelligence
Dang Khoa Nguyen, Trong Phuoc Nguyen, Chayut Ngamkhanong, Suraparb Keawsawasvong, Van Qui Lai
Summary: This paper presents a novel investigation on the bearing capacity of ring footings embedded in undrained anisotropic clays using a hybrid soft computation technique that combines finite element limit analysis (FELA) and artificial neural networks (ANNs). The study considers the anisotropic behaviour of the undrained clays by employing an anisotropic undrained shear strength (AUS) model. The results are presented in dimensionless design charts and tables, providing practical insights on the relationships between the bearing capacity factor and various parameters. Additionally, the application of ANNs enables sensitivity analysis and the development of an empirical equation, which proves to be efficient with high predictive accuracy.
NEURAL COMPUTING & APPLICATIONS
(2023)