Review
Food Science & Technology
G. V. S. Bhagya Raj, Kshirod K. Dash
Summary: Artificial neural network (ANN) is a simplified model of the biological nervous system and has been applied to various fields in food process engineering. It has the ability to map nonlinear relationships without prior knowledge and predicts responses even with incomplete information. The connection weights and hidden to output layer weights play a significant role in predicting the response. The parallel architecture of ANN allows for fast responses and low computational time, making it suitable for real-time systems. The predicted responses obtained by the ANN model exhibit high accuracy with low relative deviation and root mean squared error and a high correlation coefficient.
CRITICAL REVIEWS IN FOOD SCIENCE AND NUTRITION
(2022)
Article
Chemistry, Multidisciplinary
Vali Rasooli Sharabiani, Mohammad Kaveh, Ebrahim Taghinezhad, Rouzbeh Abbaszadeh, Esmail Khalife, Mariusz Szymanek, Agata Dziwulska-Hunek
Summary: In this study, the IR-HA drying kinetics of parboiled hull was modeled and predicted using three different models: ANFIS, ANN, and SVR. The results showed that higher inlet air temperature and IR power led to shorter drying time. Among the three models, SVR performed the best in terms of prediction performance.
APPLIED SCIENCES-BASEL
(2022)
Review
Agriculture, Multidisciplinary
Ran Yang, Jiajia Chen
Summary: Microwave-assisted thermal process is a high-efficiency drying method with potential applications in the food industry. Artificial neural network (ANN) models have been extensively studied for prediction, optimization, monitoring, and control of microwave drying processes. Future research can focus on testing the developed ANN models in industrial-scale processes and applying them to optimize dynamic drying processes.
JOURNAL OF THE SCIENCE OF FOOD AND AGRICULTURE
(2022)
Article
Energy & Fuels
Marcelina Ogedjo, Ashish Kapoor, P. Senthil Kumar, Gayathri Rangasamy, Muthamilselvi Ponnuchamy, Manjula Rajagopal, Protibha Nath Banerjee
Summary: This study presents the bioprocess modeling of levulinic acid synthesis from sugarcane bagasse using response surface methodology and artificial intelligence techniques. The best predictive model was determined to be the artificial neural network model with high accuracy. The utilization of waste biomass sugarcane bagasse offers a sustainable approach for the production of levulinic acid.
Article
Environmental Sciences
Vahid Nourani, Hossein Karimzadeh, Aida Hosseini Baghanam
Summary: This study developed an efficient model for predicting CO pollutant concentrations using artificial neural network (ANN) and adaptive neural-fuzzy inference system (ANFIS), demonstrating the importance of air quality monitoring and developing effective models for sustainable development goals.
ENVIRONMENTAL EARTH SCIENCES
(2021)
Article
Food Science & Technology
Taoqing Yang, Xia Zheng, Sriram K. K. Vidyarthi, Hongwei Xiao, Xuedong Yao, Yican Li, Yongzhen Zang, Jikai Zhang
Summary: This study used an artificial neural network (ANN) and a genetic algorithm (GA) to develop a model and optimize the process parameters for drying-assisted walnut breaking. The ANN-GA model successfully simulated the influence of IR temperature, air velocity, moisture content, and loading direction on response variables such as drying time, specific energy consumption, high kernel rate, whole kernel rate, and shell-breaking rate. The optimized process parameters were determined to be an IR temperature of 54.9 ?, air velocity of 3.66 m/s, moisture content of 10.9%, and vertical loading direction. The study demonstrated that the ANN model combined with GA optimization is an effective method for predicting and optimizing the process parameters of walnut breaking.
Article
Construction & Building Technology
Mohamed Noureldin, Ammad Ali, Sunghan Sim, Jinkoo Kim
Summary: This study proposes a new machine learning-based procedure for seismic design and qualitative assessment of structures. The procedure combines artificial neural network, fuzzy inference system, and ensemble bagged tree classification algorithm. The procedure provides qualitative assessment of structures based on their basic structural characteristics. The results of the procedure showed good agreement with nonlinear time history analyses and proved to be superior compared to conventional methods. The procedure also demonstrated potential in seismic design and assessment applications, with lower computational cost compared to traditional methods.
JOURNAL OF BUILDING ENGINEERING
(2022)
Article
Engineering, Chemical
Shital B. Potdar, Bharat A. Bhanvase, Prakash Saudagar, Irina Potoroko, Shirish H. Sonawane
Summary: Encapsulation has potential in preserving flavor and health benefits of bioactive compounds. Artificial neural networks were developed to predict the effect of parameters on the encapsulation process, showing that the combined model has similar accuracy to individual models. The best prediction performance was achieved with a 5-4-3 ANN architecture.
CANADIAN JOURNAL OF CHEMICAL ENGINEERING
(2022)
Article
Materials Science, Multidisciplinary
Tong Wang, He-Ming Huang, Xiao-Xue Wang, Xin Guo
Summary: An artificial olfactory inference system based on memristive devices has been developed to classify four gases with 10 different concentrations, achieving a high accuracy of 95%. Three strategies are applied to reduce the extracted features from the reservoir computing system in order to reduce device number and power consumption.
Article
Energy & Fuels
Miguel Godinho, Rui Castro
Summary: This paper compares the performance of artificial intelligence-based methods in forecasting wind power generation 1 hour ahead, with ANFIS being the best performer and ANN and RBFN-OLS also showing strong performances. RBFN-Hybrid and RBFN-SGD performed poorly, but overall, all methods outperformed persistence.
Article
Biotechnology & Applied Microbiology
Abiola E. Taiwo, Paul Musonge
Summary: This study aims to develop and compare models that best predict the fermentation process parameters of bioethanol production using corn-steep liquor (CSL) as a media supplement. The results show that the artificial neural network (ANN) model has better predictability with statistical error indices of R-2 = 0.90; R = 0.95; SEP = 1.73.
BIOFUELS BIOPRODUCTS & BIOREFINING-BIOFPR
(2023)
Article
Engineering, Environmental
Bonface Orero, Benton Otieno, Freeman Ntuli, Tumeletso Lekgoba, Aoyi Ochieng
Summary: The study investigated the hydrodynamic characteristics of a three-phase fluidized bed reactor using a TiO2-ZnO/BAC composite catalyst for the UV-photodegradation of rhodamine 6G dye. The results indicated that a column inclination angle of 70 degrees and a gas flow rate of 0.019 ms(-1) were the optimal conditions for efficient removal of rhodamine 6G. The dominant limiting factor in the photodegradation process was found to be solid distribution.
JOURNAL OF WATER PROCESS ENGINEERING
(2023)
Article
Mechanics
Adam Subel, Ashesh Chattopadhyay, Yifei Guan, Pedram Hassanzadeh
Summary: Recent research has shown that deep neural networks trained using properly pre-conditioned data can generate stable and accurate a posteriori LES models, and transfer learning can enable accurate and stable generalization to flows with higher Reynolds numbers.
Article
Computer Science, Artificial Intelligence
Alexander J. Dyer, Lewis D. Griffin
Summary: Inferring the connectivity of biological neural networks from neural activation data is challenging, but studying the analogous problem in artificial neural networks can provide insights into the biological case. This study focuses on assigning artificial neurons to locations in the LeNet image classifier. A supervised learning approach based on features derived from the activation correlation matrix is evaluated. The experiments suggest that an image dataset needs to fully activate the network and have minimal confounding correlations for accurate localization, and perfect assignment can be achieved by combining features from multiple image datasets.
Article
Computer Science, Artificial Intelligence
P. J. Sajith, G. Nagarajan
Summary: This paper introduces the research on using Adaptive Neuro-Fuzzy Inference System (ANFIS) as a classifier for network classification. Experimental results show that ANFIS performs better compared to other models.
