Article
Construction & Building Technology
Chenghao Wei, Ryozo Ooka
Summary: Obtaining a detailed indoor airflow field is crucial for accurately controlling indoor environmental comfort. Traditional computational fluid dynamics (CFD) methods are time-consuming and may produce inaccurate results due to difficulties in reproducing accurate inlet boundary conditions. Artificial neural networks (ANN) can directly reconstruct the airflow field from measurement data, but may yield unphysical results. A physics-informed neural network (PINN) was introduced in this study to reconstruct the airflow field without inlet boundary conditions and showed more physical results than ANN, proving its potential in practical applications.
BUILDING AND ENVIRONMENT
(2023)
Article
Energy & Fuels
Xiaodong Gao, Pingchuan Dong, Jiawei Cui, Qichao Gao
Summary: This study aims to develop a more accurate viscosity model of diluted heavy crude based on machine learning techniques. By using a multilayer neural network to predict the viscosity of heavy oil diluted with lighter oil, it was found that the new model can predict the viscosity of diluted heavy oil with higher accuracy and outperforms other models.
Article
Multidisciplinary Sciences
Kiana Peiro Ahmady Langeroudy, Parsa Kharazi Esfahani, Mohammad Reza Khorsand Movaghar
Summary: Oil viscosity is crucial in petroleum engineering, and experimental methods and compositional methods can accurately estimate it. However, experimental data is difficult to obtain, so there is a need for convenient and fast methods to predict viscosity. This study uses machine learning methods (XGBoost, CatBoost, and GradientBoosting) based on gradient boosting decision tree to reduce the prediction error of viscosity by considering dissolved gas content, temperature, pressure, and API gravity. XGBoost outperforms other methods with higher precision and lower error, showing the effectiveness of the approach.
SCIENTIFIC REPORTS
(2023)
Article
Food Science & Technology
Hao-Hsiang Ku, Ching-Fu Lung, Ching-Ho Chi
Summary: This study designs a Sesame Oil Quality Assessment Service Platform that utilizes Internet of Things sensors to detect changes in volatile gases and color of the oil during storage. By employing deep learning mechanisms, the platform accurately evaluates the quality of sesame oil. The platform effectively addresses digitization, quality measurement, supply quality observation, and scalability issues.
Article
Environmental Sciences
Linqi Zhang, Yi Liu, Liliang Ren, Adriaan J. Teuling, Xiaoxiang Zhang, Shanhu Jiang, Xiaoli Yang, Linyong Wei, Feng Zhong, Lihong Zheng
Summary: This study proposes an ANN method to reconstruct missing surface soil moisture records with good agreement with ground-based observations, especially in regions with low-density vegetation. The ANN model outperformed the OK model in reconstructing SM with absent swaths and in densely and sparsely vegetated regions.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Article
Energy & Fuels
Uzair Ahmad, Salman Raza Naqvi, Imtiaz Ali, Faisal Saleem, Muhammad Taqi Mehran, Umair Sikandar, Dagmar Juchelkova
Summary: The study demonstrates the production of biolubricant from castor oil using Fe3O4 nanoparticles and ethylene glycol as additives. The reaction conditions were optimized, and the yield of the biolubricant was predicted using artificial neural networks. The biolubricant exhibited improved tribological properties compared to raw castor seed oil and other biolubricant samples.
Article
Polymer Science
Hossein Saberi, Ehsan Esmaeilnezhad, Hyoung Jin Choi
Summary: In this study, artificial intelligence techniques were used to evaluate the performance of polymer flooding operation, utilizing multilayer perceptron, radial basis function, and fuzzy neural networks to estimate the output EOR performance, with MLP neural network demonstrating a high ability for prediction. This proposed model can significantly assist engineers in selecting appropriate EOR methods, with API gravity, salinity, permeability, porosity, and salt concentration having the greatest impact on polymer flooding performance.
Article
Green & Sustainable Science & Technology
Yashvir Singh, Deepak Singh, Nishant Kumar Singh, Abhishek Sharma, Erween Abd Rahim, Arunkumar Ranganathan, Pandiarajan Palanichamy, Arkom Palamanit, Sanjeev Kumar
Summary: The goal of this study is to maximize the yield of pyrolysis oil using lychee-based biomass for a quick pyrolysis process. Response surface methodology and artificial neural network with generalized neural network technique were used to optimize the bio-oil yield. A new prediction model combining the benefits of an auto-adaptive management technique with the rapid reaction of a generalized neural network was proposed. The results showed that the generalized neural network model provided superior accuracy in predicting bio-oil yield.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Thermodynamics
Naman Parashar, Navid Aslfattahi, Syed Mohd. Yahya, R. Saidur
Summary: The dynamic viscosity of MXene-palm oil nanofluid was investigated, showing a strong dependence on temperature and decreasing with increasing temperature. The effect of MXene nanoflakes concentration on dynamic viscosity was more pronounced at lower temperatures.
JOURNAL OF THERMAL ANALYSIS AND CALORIMETRY
(2021)
Article
Chemistry, Applied
Aurelie Monie, Sophie Franceschi, Stephane Balayssac, Myriam Malet-Martino, Mathieu Delample, Emile Perez, Jean-Christophe Garrigues
Summary: Commercial oleogelators containing monoglycerides (MGs) are complex mixtures with gelling properties affected by the oil used and concentration. A chemometric approach was developed to identify key parameters in the gelling process. Results showed that specific isomers and unsaturated/saturated fatty acid ratios have different effects on G' at low and high oleogelator concentrations.
Article
Energy & Fuels
Biswajit Saha, Sundaramurthy Vedachalam, Atanu Kumar Paul, Ajay K. Dalai, Saumitra Saxena, William L. Roberts, Frederick L. Dryer
Summary: As petroleum recovery has increased the proportion of heavier crudes and refining process residues, the challenge of processing these crudes has risen. The presence of asphaltenes in heavy crudes leads to reduction in combustion efficiency, clogging of refinery pipes, and emissions of particulate matter. This study investigates the deasphalting of heavy fuel oil using different methods, and finds that microwave-assisted deasphalting is more efficient in removing asphaltenes. The optimization of process parameters further improved the quality of the fuel oil. The outcomes of this study are important for the petrochemical industry as they enable improved crude oil processing in a more effective and economical manner.
Article
Meteorology & Atmospheric Sciences
Ali Danandeh Mehr
Summary: This paper introduces an ensemble evolutionary model to improve the accuracy of seasonal rainfall hindcasting. By combining gene expression programming and multi-stage genetic programming techniques, the model increased the forecasting accuracy by 30% during testing, demonstrating its effectiveness in comparison with the state-of-the-art gradient boosted decision tree model.
THEORETICAL AND APPLIED CLIMATOLOGY
(2021)
Article
Engineering, Multidisciplinary
Cesar Marques Salgado, William Luna Salgado, Roos Sophia de Freitas Dam, Claudio Carvalho Conti
Summary: This study investigates a methodology using gamma-ray scattering to study barium sulfate scales in the oil industry. An artificial neural network was trained to predict maximum scale thickness values with over 90% accuracy.
Article
Engineering, Electrical & Electronic
Shanshan Liu, Xiaochen Tang, Farzad Niknia, Pedro Reviriego, Weiqiang Liu, Ahmed Louri, Fabrizio Lombardi
Summary: Stochastic computing is a popular choice for implementing Artificial Neural Networks due to its low complexity in arithmetic unit design, but the conventional dividers suffer from high computation latency. This paper introduces a fast stochastic divider design to reduce latency, demonstrating its effectiveness in improving computation accuracy and performance for SC-based MLPs.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2021)
Article
Thermodynamics
Chunyang Leng, Mingxing Jia, Haijin Zheng, Jibin Deng, Dapeng Niu
Summary: This paper proposes a method based on AM-LSTM, ANN, and WOA for predicting the dynamic liquid level of multiple wells. By extracting dynamic and static features and using neural networks for prediction, it can achieve accurate and efficient operation of oil wells.
Article
Energy & Fuels
Celal Utku Deniz, Muzaffer Yasar, Michael T. Klein
Article
Energy & Fuels
Celal Utku Deniz, S. Hande Ozoren Yasar, Muzaffer Yasar, Michael T. Klein
Article
Materials Science, Composites
Ibrahim Bilici, Celal U. Deniz, Beytullah Oz
JOURNAL OF COMPOSITE MATERIALS
(2019)
Article
Chemistry, Physical
Ramazan Oguz Caniaz, Serhat Arca, Muzaffer Yasar, Can Erkey
JOURNAL OF SUPERCRITICAL FLUIDS
(2019)
Article
Biotechnology & Applied Microbiology
Celal Utku Deniz, Ibrahim Bilici, Mehmet Tuncer
JOURNAL OF CHEMICAL TECHNOLOGY AND BIOTECHNOLOGY
(2020)
Article
Chemistry, Physical
Humeyra Mert, Celal Utku Deniz, Cengiz Baykasoglu
MOLECULAR SIMULATION
(2020)
Article
Chemistry, Multidisciplinary
C. Utku Deniz, F. Nihal Tuzun
Article
Materials Science, Multidisciplinary
Celal Utku Deniz, Humeyra Mert, Cengiz Baykasoglu
Summary: The study investigated the hydrogen physisorption performance of lithium-doped fullerene pillared graphene nanocomposites using GCMC simulations. Results showed that the hydrogen adsorption performance could be significantly enhanced with appropriate doping ratios and types of fullerenes, especially at low temperature or pressure conditions. Furthermore, lithium doping was found to greatly improve the excess hydrogen storage capacity of FPGNs at ambient temperature.
COMPUTATIONAL MATERIALS SCIENCE
(2021)
Article
Energy & Fuels
Solmaz Akmaz, Ayse Ceyda Alpak, Mert Haktanir, Muzaffer Yasar
JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING
(2020)
Article
Energy & Fuels
Mert Haktanir, Seda Karahan, Muzaffer Yasar
Summary: The molecular modeling of fuels requires comprehensive composition data, including the names and molar fractions of components. However, for fuels like petroleum residues, the limitations of analytical techniques in detecting all molecule names and managing the vast number of components pose challenges. To address this, stochastic models and custom molecule core libraries can be used to simulate the complex compositions of fuels accurately.
Article
Chemistry, Physical
Hasan Komurcu, Kadir Yilmaz, Savas Gurdal, Muzaffer Yasar
Summary: This study investigated the hydrotreating of kerosene to obtain refined products suitable for marketing as kerosene and rocket grade fuel. Hydrogenation experiments were conducted using different amounts of crude kerosene and silica-supported nickel and kieselguhr supported nickel-sulfur catalysts. The catalysts were analyzed using various techniques, and the obtained samples were examined for aromatic fractions and paraffin structures content. The experiments showed a reduction in aromatic hydrogen structures in crude kerosene and an increase in the total H/C ratio of rocket grade hydrocarbon fuels.
INTERNATIONAL JOURNAL OF HYDROGEN ENERGY
(2023)
Article
Multidisciplinary Sciences
Serhat Arca, Savas Gurdal, Ramazan Oguz Caniaz, Kadir Yilmaz, Hasan Komurcu, Refika Cetintas, Emel Baskent Aydemir, Muzaffer Yasar
Summary: This study investigates the effects of six ionic liquids with different side-chain lengths on the self-healing properties of bitumen. The experimental results indicate that one of the ionic liquids improves the self-healing performance of asphalt by 40% at high temperatures and 100% at low temperatures, while also enhancing stripping properties by 25% and asphalt fatigue life by 20%. Therefore, different bitumen-IL modification recipes can be used for self-healing of asphalt pavements based on climatic conditions and traffic density.
ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
(2023)
