Article
Engineering, Mechanical
Wansu Song, Jongsung Park, Hyungjo Seo, Jinsoo Choi, Jung Ju Lee, Seok Su Sohn, Ho Jang
Summary: The influence of brake pad curing conditions on brake emissions was studied. The Box-Behnken method and an artificial neural network were used to determine the optimal curing condition for minimum brake emissions and appropriate friction effectiveness. The results showed that the curing condition significantly affected the friction level, pad wear, and brake emissions. Pad wear was proportional to the mass concentration of airborne particles, while no correlation was found with the number concentration of ultrafine particles. The pad hardness, determined by the degree of crosslinking of the binder resin, correlated well with pad wear and brake emissions. The artificial neural network provided reliable results for finding the optimum curing condition with minimum brake emissions and satisfying friction effectiveness for brake performance.
Article
Multidisciplinary Sciences
Seah Yi Heng, Wanie M. Ridwan, Pavitra Kumar, Ali Najah Ahmed, Chow Ming Fai, Ahmed Hussein Birima, Ahmed El-Shafie
Summary: This paper presents a comprehensive study on the meteorological data and backpropagation algorithms used to develop the best solar radiation predicting artificial neural network (ANN) model. The findings show that temperature and relative humidity have high correlation with solar radiation, while wind speed has little influence. The Bayesian Regularization algorithm trained ANN models performed the best in terms of predictive ability.
SCIENTIFIC REPORTS
(2022)
Article
Thermodynamics
J. Garcia-Morales, M. Cervantes-Bobadilla, J. A. Hernandez-Perez, Y. Saavedra-Benitez, M. Adam-Medina, G. Guerrero-Ramirez
Summary: This work focused on a novel non-linear control approach for regulating the output cold water temperature of a double tube heat exchanger. The approach utilizes an inverse artificial neural network (ANNi) and an integral control law to achieve temperature regulation. Experimental results demonstrated that the proposed control scheme exhibits good reference tracking, fast settling time, and low overshoot.
CASE STUDIES IN THERMAL ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Xinglan Liu, Bin Xu, Yingxin Shou, Quan-Yong Fan, Yingxue Chen
Summary: This article focuses on event-based collaborative design for strict-feedback systems with uncertain nonlinearities, proposing a controller design method based on neural network weight adaptive law and updating the controller and NN weights adaptive law at triggering instants determined by a novel composite triggering threshold. By integrating state-model error, the requirements of system information and allowable range of event-triggering error are relaxed, reducing the number of triggering instants significantly while maintaining system performance. The stability of the closed-loop system is proven using the Lyapunov method at time intervals and sampling instants, with simulation results demonstrating the effectiveness of the proposed scheme.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Review
Thermodynamics
Mohammad Ghalandari, Misagh Irandoost Shahrestani, Akbar Maleki, Mostafa Safdari Shadloo, Mamdouh El Haj Assad
Summary: The performance modeling and forecasting of heat exchangers can utilize intelligent methods, with accuracy and applicability dependent on factors such as algorithm architecture, model inputs, and system complexity. Considering influential factors in the model is crucial for producing models with the greatest accuracy, while the performance of intelligent methods is influenced by various factors.
JOURNAL OF THERMAL ANALYSIS AND CALORIMETRY
(2021)
Review
Thermodynamics
Mohammad Hossein Ahmadi, Ravinder Kumar, Mamdouh El Haj Assad, Phuong Thao Thi Ngo
Summary: Research shows that using intelligent models can effectively model heat pipes and accurately estimate their thermal behavior. The accuracy and applicability of the models depend on various factors, such as input variables, algorithms, and model structures. In the future, data-driven approaches should be promoted in heat pipe modeling and optimization methods should be applied to enhance the accuracy of the models.
JOURNAL OF THERMAL ANALYSIS AND CALORIMETRY
(2021)
Article
Green & Sustainable Science & Technology
Salma Zaim, Mohamed El Ibrahimi, Asmae Arbaoui, Abderrahim Samaouali, Mouhaydine Tlemcani, Abdelfettah Barhdadi
Summary: This study proposes the use of artificial neural networks and XGBoost algorithm for modeling hourly global solar radiation in a humid climate. Important meteorological data are selected and validated, and two ANN models and one XGBoost model are chosen with similar performances, with coefficient of determination values of 98% and 97% respectively. Statistical indicators prove to be effective in assessing the accuracy and fidelity of each model. The intent of the modeling in terms of accuracy, simplicity, and fidelity is a crucial factor in selecting the model algorithm to adopt.
Article
Environmental Sciences
Saeid Fallahizadeh, Majid Kermani, Ali Esrafili, Zahra Asadgol, Mitra Gholami
Summary: The study on PM10 health effects and air quality forecasting revealed the association between exposure to PM10 and increased symptoms of asthma and bronchitis, with the use of ANN modeling for prediction and managerial planning in tackling air pollution.
Article
Construction & Building Technology
Renner de Assis Garcia Sobrinho, Franklin Piauhy Neto, Henrique Fernandes
Summary: The research aimed to create a public database of images of cracks in mortar coating, considering different types of surface finish. The training accuracy varied based on surface finish and data balancing, with the scrapped type showing the lowest accuracy.
Article
Thermodynamics
Akbar Maleki, Arman Haghighi, Misagh Irandoost Shahrestani, Zahra Abdelmalek
Summary: In this study, GMDH and multilayer perceptron artificial neural networks were used to determine the thermal conductivity of nanofluids containing silica particles and different base fluids. The models showed good agreement with experimental results, with R2 values reaching up to 0.9998.
JOURNAL OF THERMAL ANALYSIS AND CALORIMETRY
(2021)
Article
Engineering, Mechanical
S. Gregori, M. Tur, J. Gil, F. J. Fuenmayor
Summary: This paper proposes a new method to predict the current collection quality by measuring the vertical acceleration of the pantograph collector head. The method involves preprocessing the acceleration signals and training Artificial Neural Networks (ANNs) to predict the standard deviation of the interaction force. The proposed approach is proven to be accurate and effective through academic examples and experimental data.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Automation & Control Systems
Sercan Yalcin, Musa Esit, Onder Coban
Summary: This study focused on drought parameter estimation using the standard precipitation evapotranspiration index (SPEI) in four different stations in Turkiye. A hybrid deep structure of convolutional neural network (CNN) and long short-term memory (LSTM) was proposed and compared with other methods. The results showed that better estimation results were achieved with a 12-month time scale for all methods. The proposed method outperformed existing methods in terms of performance metrics and was robust to changes in input time series data from different stations. The analysis also revealed that the highest occurrence percentages of the wet category were observed at different time scales.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Rutuja Rajendra Patil, Sumit Kumar
Summary: By utilizing a plant disease forecasting model, the emergence of rice diseases can be predicted in advance, allowing for proactive measures to be taken to reduce losses. This paper focuses on the use of artificial neural network (ANN) to detect, classify, and predict the occurrence of rice diseases based on agro-meteorological conditions, achieving a high accuracy rate of 92.15% using the softmax activation function on a 70-30% dataset split.
PEERJ COMPUTER SCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Fan Bailin, Zhang Yi, Chen Ye, Meng Lingbei
Summary: Based on the dynamic model of the intelligent firefighting vehicle, a linear 2-DOF lateral dynamic model and a preview error model are established. A Radial Basis Function neural network sliding mode controller is designed to solve the problems of non-linearity, time-varying parameters, output chattering, and poor robustness. Simulation results show that the controller has high accuracy in tracking the desired path and has good robustness to speed changes of the vehicle.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
N. J. Sairamya, M. S. P. Subathra, S. Thomas George
Summary: This study proposes a novel method for diagnosing epileptic seizure types, using a relaxed local neighbour difference pattern (RLNDiP) domain and artificial neural network for automatic classification. The results validate the robustness and accuracy of the proposed method.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Environmental Sciences
Youjung Jang, Hyejung Hu, Bomi Kim, Younha Kim, Seung-Jick Yoo, Kyungae Jang, Yoon-Kwan Kim, Hyungah Jin, Jung-Hun Woo
Summary: This study quantitatively analyzed the effects of climate and air pollutant reduction policies in Korea, demonstrating that these policies can lead to reductions in greenhouse gas emissions and atmospheric pollutants. The integrated model used in the study provides advantages for evaluating climate and air quality policies, and the findings offer valuable insights and data for policy development and assessment.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Giuseppe Piras, Fabrizio Pini, Paolo Di Girolamo
Summary: This study assesses the contribution of tires to atmospheric PM10 pollution and finds that tire emissions of PM10 are larger than those from exhaust gases. It suggests the need for specific strategies to reduce tire emissions, such as producing lighter vehicles, using narrower wheels, and promoting public transportation.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Laura Vallecillos, Rosa Maria Marce, Francesc Borrull
Summary: This study focuses on the implementation of a gas chromatograph-photoionization detection (GC-PID) analyzer for the continuous monitoring of 1,3-butadiene (1,3-BD) levels in urban and industrial atmospheres. The study found that the concentrations of 1,3-BD recorded by the GC-PID analyzer were comparable to those obtained by active sampling, and the concentration peaks showed consistency in values and time slots. In the test of urban atmospheres, the results showed that the concentrations of 1,3-BD were related to prevailing wind direction and activities in the petrochemical zone, while other factors had minor effects on the distribution of this pollutant.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Ajay Kumar, Arun K. Attri
Summary: This study investigated the temporal profile and composition of PM10 over a 14-month period, and found significant variations between different seasons. The highest concentrations of PM10 were observed in summer and winter, exceeding the national limits. Water-soluble ionic species and n-alkanes contributed to the PM10 mass, with the highest concentration in winter and the lowest in the monsoon season. The ion balance study revealed a strong correlation between anion and cation charge equivalents, indicating their main contribution to PM10. The main sources of PM10 components were identified using statistical correlation, regression, and principal component analysis.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Jenny Martinez, Yris Olaya Morales, Prashant Kumar
Summary: The impact of bicycle lane designs on cyclist exposure to air pollution is a significant concern. This study found that in the city of Medellin, Colombia, the sections without dedicated bicycle lanes had the highest PM2.5 exposure and inhaled dose. Cyclists had higher PM2.5 exposure and inhaled dose during morning peak hours compared to evening peak and off-peak hours. Segregated cycling lanes on the sidewalk can considerably lower PM2.5 exposure and inhaled doses for cyclists.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Ying Xu, Qingyang Liu, James J. Schauer
Summary: In this study, a quantitative method using dual-wavelength ultraviolet-visible spectroscopy and Raman spectroscopy was developed to analyze carbon black with amorphous structures and ordering in a graphene sheet. Water extracts of carbon black showed high oxidative potential, and the presence of water-soluble ions enhanced its oxidative potential. These findings can help mitigate health risks associated with nano-carbon black emissions.
ATMOSPHERIC POLLUTION RESEARCH
(2024)
Article
Environmental Sciences
Zhongmin Zhu, Hui Li, Shumin Fan, Wenfa Xu, Ruimin Fang, Boming Liu, Wei Gong
Summary: This study investigates the relationship between temperature inversions (TI) and aerosol vertical distribution in China. The results show that TI frequency, inversion strength (Delta T), and TI height (TIH) exhibit similar seasonal patterns across different regions in China. NC has a significantly higher TI frequency during summer, possibly due to the heating effect of black carbon aerosol. Aerosol optical depth (AOD) above the TIH is higher in spring and summer, indicating the presence of aerosol high-level transport over mainland China during these seasons. The study also finds that a strong inversion can suppress surface aerosols below the TI, but in regions with strong atmospheric stability, aerosols tend to accumulate above the TIH. These findings are valuable for understanding aerosol transport.
ATMOSPHERIC POLLUTION RESEARCH
(2024)