Article
Mechanics
Abouzar Jafari, Lingyue Ma, Amir Ali Shahmansouri, Roberto Dugnani
Summary: Quantitative fractography is important in analyzing the failure of brittle materials, but its application is limited due to unknown factors. Artificial neural networks (ANNs) were used to analyze the fracture strength of glasses and ceramics. The developed ANN models outperformed empirical relations in predicting fracture strength and could be extended to a broader range of brittle materials.
ENGINEERING FRACTURE MECHANICS
(2023)
Article
Thermodynamics
P. C. Mukesh Kumar, R. Kavitha
Summary: The study predicted the dynamic viscosity ratio of nanofluids using machine learning techniques, achieving low error values with multilayer perceptron and Gaussian process regression models. This helps to reduce experimental costs and improves prediction accuracy.
JOURNAL OF THERMAL ANALYSIS AND CALORIMETRY
(2021)
Article
Chemistry, Multidisciplinary
Nemesio Fava Sopelsa Neto, Stefano Frizzo Stefenon, Luiz Henrique Meyer, Rafael Bruns, Ademir Nied, Laio Oriel Seman, Gabriel Villarrubia Gonzalez, Valderi Reis Quietinho Leithardt, Kin-Choong Yow
Summary: This study focuses on the application of machine learning to interpret ultrasound signals obtained from electrical grid insulators, using Multilayer Perceptron networks for the classification of ceramic insulators based on a restricted database of ultrasonic signals recorded in the laboratory.
APPLIED SCIENCES-BASEL
(2021)
Article
Health Care Sciences & Services
Rahman Ali, Jamil Hussain, Seung Won Lee
Summary: In this study, a feed-forward artificial neural network (ANN)-based self-care prediction methodology, called multilayer perceptron (MLP)-progressive, has been proposed to improve the early detection of self-care disabilities in children. The proposed MLP-progressive model outperforms existing methods and achieves a classification accuracy of 97.14% and 98.57% on multi-class and binary-class datasets, respectively.
Article
Energy & Fuels
Ramesh Kanthasamy, Eydhah Almatrafi, Imtiaz Ali, Hani Hussain Sait, Mohammed Zwawi, Faisal Abnisa, Leo Choe Peng, Bamidele Victor Ayodele
Summary: This study uses Bayesian optimized multilayer perceptron neural network to model the prediction of biochar and syngas from biomass-derived waste pyrolysis. The performance of the neural networks is influenced by the number of connecting layers and the size of the hidden neurons. The best-performing neural network architecture for predicting biochar yield is 3-2-10-10-1 with R2 of 0.984 and RMSE of 0.34, while for predicting syngas yield it is 3-7-10-3-1 with R2 of 0.999.
Article
Optics
Menglong Luo, Bishal Bhandari, Hongliang Li, Stuart Aberdeen, Sang-Shin Lee
Summary: Machine learning is utilized in optical applications to efficiently predict divergence and deflection angles of beams using artificial neural networks. The optimized networks greatly simplify lens design, provide multiple lens specification solutions, and significantly reduce computation time.
Article
Biodiversity Conservation
Fatemeh Panahi, Mohammad Ehteram, Ali Najah Ahmed, Yuk Feng Huang, Amir Mosavi, Ahmed El-Shafie
Summary: The study focuses on predicting streamflow in four rivers of Malaysia using the CBMA as an improved version of the BMA model. The CBMA corrects the assumption of Gaussian distortion in the BMA and uses different distribution and Copula functions for the variables. The study trains the MLP model with the AOA algorithm and benchmarks it against other optimization algorithms, showing a significant improvement in predictive performance.
ECOLOGICAL INDICATORS
(2021)
Article
Thermodynamics
Manash Jyoti Deka, Pankaj Kalita, Dudul Das, Akash Dilip Kamble, Bhaskor Jyoti Bora, Prabhakar Sharma, Bhaskar Jyoti Medhi
Summary: The study aims to train an artificial neural network to optimize input parameters and determine the most suitable PV/T system for a specific environmental situation. The rectangular spiral tube PV/T system is found to have the highest average thermal and electrical efficiency (50.7% and 12.15% respectively) among the studied systems. The artificial neural network models can predict the performance of different PV/T systems.
ENERGY CONVERSION AND MANAGEMENT
(2023)
Article
Chemistry, Physical
Yixiao Wang, Mingzhu Tang, Jiangang Ling, Yunshan Wang, Yiyang Liu, Huan Jin, Jun He, Yong Sun
Summary: In this study, three modeling techniques were used to investigate the biohydrogen process. A new effective strategy for modeling and optimizing the complex BioH2 production during the dark fermentation was proposed. The proposed strategy is a useful and practical paradigm in modeling the BioH2 production process.
INTERNATIONAL JOURNAL OF HYDROGEN ENERGY
(2021)
Article
Computer Science, Artificial Intelligence
Sajad Ahmadian, Seyed Mohammad Jafar Jalali, Saeid Raziani, Abdolah Chalechale
Summary: Cardiovascular diseases are the leading cause of death in recent decades, and predicting and treating these diseases is crucial for improving life quality and reducing treatment costs and death risks. The multilayer perceptron neural network and moth-flame optimization algorithm can be used effectively to detect cardiovascular diseases and provide more accurate predictions.
Article
Construction & Building Technology
Can Jiang, Xiong Liang, Yu-cheng Zhou, Yong Tian, Shengli Xu, Jia-Rui Lin, Zhiliang Ma, Shiji Yang, Hao Zhou
Summary: In Chinese building codes, it is necessary for residential buildings to receive a minimum amount of direct sunlight on a specified winter day. This requirement is essential for obtaining a building permit during the conceptual design phase, and official software is commonly used to assess sunlight performance. The paper proposes a real-time shading time interval prediction method based on multilayer perceptron, which reduces computation time significantly with high accuracy. A residential neighborhood layout planning plug-in for Rhino 7/Grasshopper is also developed based on the proposed model. The study demonstrates the potential of using deep learning techniques for accelerating sunlight hour simulations in the conceptual design phase.
BUILDING AND ENVIRONMENT
(2023)
Article
Energy & Fuels
Ramesh Kanthasamy, Imtiaz Ali, Bamidele Victor Ayodele, Hisham A. Maddah
Summary: This study models biohydrogen production from photocatalytic conversion of wastewater using a data-driven approach and neural network models. The best network architectures were obtained for different algorithms, and all models demonstrated good predictability with high R2 values and low RMSE values. The MLPNN-BR model displayed the best performance with an R2 of 0.999 and RMSE of 0.138. The independent variable analysis showed that all factors significantly influenced the predicted biohydrogen production, with catalyst size having the most significant effect.
Article
Chemistry, Applied
May Ali Alsaffar, Mohamed Abdel Rahman Abdel Ghany, Jamal Manee Ali, Bamidele Victor Ayodele, Siti Indati Mustapa
Summary: The study involved predictive modeling of hydrogen production by thermo-catalytic methane decomposition using neural networks trained with Bayesian regularization and Levenberg-Marquardt algorithms. The Levenberg-Marquardt trained neural network with a topology of 7-16-1 demonstrated the best performance, achieving a strong agreement between predicted and observed hydrogen yields. Sensitivity analysis revealed that the intrinsic properties of catalysts, specifically the specific surface area and pore volume, have the most significant influence on predicted hydrogen yield.
TOPICS IN CATALYSIS
(2021)
Article
Medical Informatics
Md Ashikur Rahman Khan, Jony Akter, Ishtiaq Ahammad, Sabbir Ejaz, Tanvir Jaman Khan
Summary: Dengue fever is a fatal disease that has been spreading worldwide, with Bangladesh being heavily affected. This research aims to predict Dengue outbreaks by analyzing real-time data and utilizing techniques such as neural networks and decision trees.
HEALTH INFORMATION SCIENCE AND SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Ruiyuan Lin, Zhiruo Zhou, Suya You, Raghuveer Rao, C. -C. Jay Kuo
Summary: This work interprets the multilayer perceptron (MLP) neural network from a geometrical viewpoint and proposes a new three-layer feedforward MLP (FF-MLP) architecture for its implementation. Experiments show that the FF-MLP outperforms the traditional backpropagation-based MLP (BP-MLP) in terms of design time, training time, and classification performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Review
Energy & Fuels
Adrian Knapczyk, Slawomir Francik, Marcin Jewiarz, Agnieszka Zawislak, Renata Francik
Summary: This paper aims to summarize and discuss current research trends in biomass thermal treatment, revealing a clear upward trend in publications after 2010. The main countries contributing to the research are China, USA, Canada, and South Korea, with research topics focusing on torrefaction process, hydrothermal carbonization process, pyrolysis process, and gasification and co-combustion process. The study also highlights the use of various raw materials in biomass thermal treatment, including energy crops, wood, waste from agri-food industry, sewage sludge, and microalgae.
Article
Chemistry, Physical
Slawomir Francik, Pawel Knapik, Boguslawa Lapczynska-Kordon, Renata Francik, Zbigniew Slipek
Summary: The aim of this study was to investigate the influence of internode number and water content on the biomechanical parameters of giant miscanthus stalks. The results showed that both water content and internode number had a significant impact on the modulus of elasticity and maximum stress.
Article
Chemistry, Physical
Boguslawa Lapczynska-Kordon, Zbigniew Slipek, Karolina Slomka-Polonis, Jakub Styks, Tomasz Hebda, Slawomir Francik
Summary: This study analyzed the feasibility of using goldenrod for biochar production. The results showed that torrefaction process increased the ash content, calorific value, and heat of combustion of biochar while decreasing the volatile matter content. The plant species and sampled parts were found to have a significant impact on the physicochemical properties of both raw biomass and biochar.
Article
Chemistry, Physical
Agnieszka Zawislak, Renata Francik, Slawomir Francik, Adrian Knapczyk
Summary: This study evaluated the antioxidant properties of red clover, sweet violet, and elderflower extracts obtained through different drying methods. The results showed that ethanolic extracts of red clover had the highest antioxidant activity, while sweet violet extracts had the lowest. Freeze drying was found to be the most effective preservation method for retaining the antioxidant properties of the flowers and their compounds.
Article
Green & Sustainable Science & Technology
Slawomir Francik, Boguslawa Lapczynska-Kordon, Norbert Pedryc, Wojciech Szewczyk, Renata Francik, Zbigniew Slipek
Summary: The aim of this study is to develop neural models for determining the biomechanical parameters of giant miscanthus stems. The static three-point bending test is used to determine the bending strength parameters. Four neural models have been developed which demonstrate high accuracy in determining the modulus of elasticity in bending and the maximum stress in bending.
Article
Nutrition & Dietetics
Pawel Pasko, Krzysztof Okon, Ewelina Prochownik, Miroslaw Krosniak, Renata Francik, Jadwiga Kryczyk-Koziol, Marta Grudzinska, Malgorzata Tyszka-Czochara, Mateusz Malinowski, Jakub Sikora, Agnieszka Galanty, Pawel Zagrodzki
Summary: This study examined the effects of kohlrabi sprouts diet on rat thyroid function. The results showed that there were no significant changes in TSH, thyroid hormones, antioxidant enzymes, and histopathology parameters after kohlrabi sprouts ingestion, except for an increase in thioredoxin reductase activity. The sprouts diet did not prevent thyroid damage in hypothyroid animals. Additionally, the sprouts diet decreased TNF-alpha level, but did not negatively affect red blood cell parameters, glucose and uric acid concentrations, or kidney function. However, it resulted in reduced white blood cell levels and adverse interference with liver function, likely due to a higher dietary intake of glucosinolates. Furthermore, the possible impact of the rat breed on the evaluated parameters was indicated.
Article
Biochemistry & Molecular Biology
Piotr Hydzik, Renata Francik, Slawomir Francik, Ewa Gomolka, Ebru Derici Eker, Miroslaw Krosniak, Maciej Noga, Kamil Jurowski
Summary: In conventional clinical toxicology practice, the blood level of carboxyhemoglobin is commonly used as a biomarker for carbon monoxide poisoning. However, this does not reflect the complete clinical picture and severity of the poisoning. This article aims to explore the potential use of oxidative-stress-related parameters as biomarkers for evaluating CO poisoning severity and guiding treatment decisions. Preliminary studies involving CO-poisoned patients showed significant changes in antioxidative parameters, such as catalase and glutathione levels, but further research is needed to fully understand their role as biomarkers for CO poisoning severity.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Chemistry, Multidisciplinary
Urszula Sadowska, Klaudia Jewiarz, Magdalena Kopak, Kinga Dziadek, Renata Francik, Aneta Kopec
Summary: White mustard plant is an easy-to-grow species commonly sown as a catch crop in Northern Europe. Mustard seeds have high nutritional value and are widely used in food production. However, little is known about the nutritional composition of young mustard plants. This research aimed to determine the composition and antioxidant activity of young green plants of Polish cultivars of white mustard.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Physical
Renata Francik, Jadwiga Kryczyk-Koziol, Miroslaw Krosniak, Slawomir Francik, Tomasz Hebda, Norbert Pedryc, Adrian Knapczyk, Mehmet Berkoez, Zbigniew Slipek
Summary: This study examined the effects of organic vanadium complexes on metabolic rate and antioxidant activity of adipose tissue, finding that the BM complex significantly increased ferric reducing antioxidant power while the V complex increased glutathione concentration and reduced weight gain in rats on a high-fat diet. Additionally, all tested vanadium complexes led to a significant increase in basal metabolic rate, regardless of the diet applied. Further research is needed to understand the mechanisms by which these organic vanadium complexes affect adipose tissue.