Article
Engineering, Civil
Mahesh Shelke, S. N. Londhe, P. R. Dixit, Pravin Kolhe
Summary: This study compared the performance of a conceptual semi distributed HEC-HMS model and an ANN-based model in predicting reservoir inflow in the Koyna reservoir catchment area. The results showed that the semi distributed HEC-HMS model performed slightly better than the ANN model. This research is significant for the planning of reservoir operations in the Koyna reservoir catchment area.
WATER RESOURCES MANAGEMENT
(2023)
Article
Thermodynamics
Vinicius David Fonseca, Willian Moreira Duarte, Raphael Nunes de Oliveira, Luiz Machado, Antonio Augusto Torres Maia
Summary: This study proposes a method of using an Artificial Neural Network model to predict the mass flow rate in refrigeration systems, and validates the accuracy of the model through optimization and experimental data.
APPLIED THERMAL ENGINEERING
(2022)
Article
Geosciences, Multidisciplinary
Xiaoying Zhang, Fan Dong, Guangquan Chen, Zhenxue Dai
Summary: In this study, artificial intelligence techniques were used to develop time convolutional network (TCN) and long short-term memory (LSTM) models for predicting groundwater levels with different leading periods in coastal aquifers. The TCN-based models showed higher prediction accuracy and lower root mean square error (RMSE) values in a shorter time compared to the LSTM-based model. Both models demonstrated the ability to learn complex patterns from historical data with different leading periods, making them valuable tools for localized groundwater level prediction in subsurface environments.
HYDROLOGY AND EARTH SYSTEM SCIENCES
(2023)
Article
Physics, Particles & Fields
Laurits Tani, Diana Rand, Christian Veelken, Mario Kadastik
Summary: Analyzing vast amounts of data in modern high energy physics experiments is a challenge, often requiring the use of machine learning methods trained on simulated data. Choosing the right parameters for the machine learning algorithm is crucial for performance optimization.
EUROPEAN PHYSICAL JOURNAL C
(2021)
Article
Multidisciplinary Sciences
Chengxu Zhuang, Siming Yan, Aran Nayebi, Martin Schrimpf, Michael C. Frank, James J. DiCarlo, Daniel L. K. Yamins
Summary: Recent advancements in unsupervised learning have narrowed the gap in using deep neural networks to model the response patterns of neurons in the primate ventral visual stream, achieving neural prediction accuracy comparable or superior to current supervised methods. These methods even produce brainlike representations when trained solely on real human child data, demonstrating potential for a biologically plausible computational theory of primate sensory learning.
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2021)
Article
Computer Science, Artificial Intelligence
Xu Huang, Bowen Zhang, Shanshan Feng, Yunming Ye, Xutao Li
Summary: In this paper, an interpretable local flow attention (LFA) mechanism is proposed for traffic flow prediction (TFP), which has the advantages of flow-awareness, interpretability, and efficiency. Based on LFA, a novel spatiotemporal cell called LFA-ConvLSTM is developed to capture the complex dynamics in traffic data. Experimental results demonstrate that our method outperforms previous approaches in prediction performance and is also faster by 32% than global self-attention ConvLSTM.
Article
Energy & Fuels
Krystian Gora, Mateusz Kujawinski, Damian Wronski, Grzegorz Granosik
Summary: Artificial neural networks are rarely used for power consumption estimation in the field of mobile robotics, as researchers prefer to develop analytical models of the robots being studied. A comparison of mathematical models and non-complex artificial neural networks for energy prediction tasks shows that both methods can be used interchangeably, with AI methods being more universal and tolerant of designers lacking complex knowledge about the system.
Article
Multidisciplinary Sciences
Zhengcai Li, Xinmin Hu, Chun Chen, Chenyang Liu, Yalu Han, Yuanfeng Yu, Lizhi Du
Summary: This paper investigates the optimization algorithms based on machine learning for settlement prediction. By comparing the performance of different algorithms, the study finds that Sparrow Search Algorithm (SSA) significantly improves the optimization effect of the gradient descent model and enhances its stability to a certain degree.
SCIENTIFIC REPORTS
(2022)
Article
Engineering, Civil
Panagiotis G. Asteris, Paulo B. Lourenco, Mohsen Hajihassani, Chrissy-Elpida N. Adami, Minas E. Lemonis, Athanasia D. Skentou, Rui Marques, Hoang Nguyen, Hugo Rodrigues, Humberto Varum
Summary: Masonry, a building material with a long history, remains competitive in the construction industry. The compressive strength of masonry is crucial in modern design, but estimating it accurately is still a challenge. Soft computing techniques can help identify key parameters affecting masonry compressive strength and offer better estimates than existing formulas.
ENGINEERING STRUCTURES
(2021)
Article
Mechanics
Ali Kashefi, Tapan Mukerji
Summary: The study proposes a novel deep learning framework for predicting permeability of porous media from digital images, overcoming memory restrictions of GPUs through the use of PointNet architecture. This approach allows for larger batch sizes and faster, accurate predictions of permeability of digital rocks. Comparisons with convolutional neural networks show improved performance and generalizability of the proposed deep learning strategy.
Article
Materials Science, Multidisciplinary
Sohaib Nazar, Jian Yang, Muhammad Nasir Amin, Kaffayatullah Khan, Muhammad Ashraf, Fahid Aslam, Mohammad Faisal Javed, Sayed M. Eldin
Summary: This study developed empirical models for predicting compressive strength (CS) and slump values of fly ash-based geopolymer concrete using three artificial intelligence-based algorithms - adaptive neuro-fuzzy inference system (ANFIS), artificial neural networks (ANNs), and gene expression programming (GEP). The GEP model outperformed the ANFIS and ANN models in terms of R-value, R2, and RMSE. The GEP model generated more accurate predictions for slump and CS after rigorous training and optimization of hyperparameters.
JOURNAL OF MATERIALS RESEARCH AND TECHNOLOGY-JMR&T
(2023)
Article
Mathematics
Adnan Bashir, Muhammad Ahmed Shehzad, Aamna Khan, Ayesha Niaz, Muhammad Nabeel Asghar, Ramy Aldallal, Mutua Kilai
Summary: This study investigates a new hybrid model, the wavelet bootstrap quadratic response surface, for accurate streamflow prediction. The results show that this model provides the most efficient results.
JOURNAL OF MATHEMATICS
(2023)
Article
Geosciences, Multidisciplinary
Prantik Mandal
Summary: In this study, an artificial neural network was utilized to develop a ground motion prediction model for peak ground acceleration in Kachchh, Gujarat, India. The model showed good predictability for earthquakes of M(w)5.6-7.7 in the region, with a standard deviation of PGA prediction error estimates found to be +/- 0.2554 in log10 units. The model performance was evaluated using real earthquake recordings, demonstrating its accuracy in predicting PGA values.
Article
Engineering, Civil
Thiago Victor Medeiros do Nascimento, Celso Augusto Guimaraes Santos, Camilo Allyson Simoes de Farias, Richarde Marques da Silva
Summary: This study developed self-organizing maps (SOM) for rainfall-streamflow modeling and found that the models generally performed well during the calibration phase but outcomes varied in the testing phase depending on data homogeneity pattern and SOM structure.
WATER RESOURCES MANAGEMENT
(2022)
Article
Computer Science, Interdisciplinary Applications
Farid Khosravikia, Patricia Clayton
Summary: This paper studies the advantages and disadvantages of different machine learning techniques in predicting ground-motion intensity measures given source characteristics, source-to-site distance, and local site conditions. The study quantifies event-to-event and site-to-site variability of the ground motions by implementing them as random effect terms to reduce the aleatory uncertainty. The results indicate that, when sufficient data is available, all the alternative algorithms tend to provide more accurate estimates compared to the conventional linear regression-based method, and particularly, Random Forest outperforms the other algorithms.
COMPUTERS & GEOSCIENCES
(2021)
Article
Meteorology & Atmospheric Sciences
K. Srinivasa Raju, P. Sonali, D. Nagesh Kumar
THEORETICAL AND APPLIED CLIMATOLOGY
(2017)
Article
Environmental Sciences
Sonali Pattanayak, Ravi S. Nanjundiah, D. Nagesh Kumar
ENVIRONMENTAL RESEARCH LETTERS
(2017)
Article
Remote Sensing
Hassan Rangaswamy Shwetha, Dasika Nagesh Kumar
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2018)
Article
Geography, Physical
Subir Paul, D. Nagesh Kumar
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2018)
Article
Water Resources
H. R. Shwetha, D. Nagesh Kumar
HYDROLOGICAL SCIENCES JOURNAL-JOURNAL DES SCIENCES HYDROLOGIQUES
(2018)
Article
Engineering, Civil
Chandan Banerjee, D. Nagesh Kumar
WATER RESOURCES MANAGEMENT
(2018)
Article
Chemistry, Analytical
Lanka Karthikeyan, Ming Pan, Dasika Nagesh Kumar, Eric F. Wood
Article
Engineering, Civil
Subir Paul, Chandan Banerjee, D. Nagesh Kumar
JOURNAL OF HYDROLOGIC ENGINEERING
(2020)
Article
Geochemistry & Geophysics
Subir Paul, D. Nagesh Kumar
Summary: Hyperspectral data are more resourceful than multispectral data, but the absence of a space-borne sensor and high cost of airborne sensors restrict the use of HS data. The proposed CNNR model for MS to quasi-HS data transformation is more efficient than existing models and proven to be valuable for crop classification applications.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Remote Sensing
Subir Paul, Mamta Kumari, C. S. Murthy, D. Nagesh Kumar
Summary: This study utilizes Synthetic Aperture Radar (SAR) data and deep learning technique to conduct pre-harvest crop mapping in a large geographic area of central India. The results show that combining VH and VV data performs better, with an overall accuracy of 91.75%. The crop map can be generated as early as 45 days prior to harvesting with an accuracy of 89.15%. The 2D-CNN algorithm outperforms SVM and RF techniques, and the methodology can be applied to similar regions.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Review
Engineering, Civil
Ashok Mishra, Sourav Mukherjee, Bruno Merz, Vijay P. Singh, Daniel B. Wright, Gabriele Villarini, Subir Paul, D. Nagesh Kumar, C. Prakash Khedun, Dev Niyogi, Guy Schumann, Jery R. Stedinger
Summary: This review provides a comprehensive overview of current flood research, challenges, and future directions, emphasizing the increased flood risk in future urban systems due to continued climate change and land-use intensification. More work is needed for accurate urban flood prediction, quantifying the socioeconomic impacts of floods, and developing mitigation strategies. Integration of multiscale models, stakeholder input, and social and citizen science input is crucial to bridge the gap between model capabilities and end-user needs for flood monitoring, mapping, and dissemination. Additionally, effort is needed for downscaled, ensemble scenarios, data assimilation approaches, and enhanced capabilities for modeling compound hazards and reducing social vulnerability and impacts. Transdisciplinary research between science, policymakers, and stakeholders is essential to reduce flood risk and social vulnerability in the face of dynamic and complex interactions between climate, societal change, watershed processes, and human factors.
JOURNAL OF HYDROLOGIC ENGINEERING
(2022)
Article
Environmental Sciences
Elizabeth Baby George, Cecile Gomez, D. Nagesh Kumar, Subramanian Dharumarajan, Manickam Lalitha
Summary: Hyperspectral imaging spectroscopy is a useful tool for mapping soil properties at large scales. This study analyzed the impact of bare soil pixel identification on clay content estimation using two methods: spectral indices and spectral unmixing. The results showed that the spectral unmixing method provided slightly better performances in estimating clay content, although it reduced the spatial coverage.
GEOCARTO INTERNATIONAL
(2022)
Article
Water Resources
Rajesh Kumar Sah, D. Nagesh Kumar, Apurba Kumar Das
Summary: This study presents a novel account of avulsion records in the Himalayan foreland region from 1990 to 2019, revealing that the eastern Brahmaputra plains are among the most affected areas. The study also suggests that smaller tributaries with transitional reaches are more susceptible to avulsions.
HYDROLOGICAL SCIENCES JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Subir Paul, Vinayaraj Poliyapram, Nevrez Imamoglu, Kuniaki Uto, Ryosuke Nakamura, D. Nagesh Kumar
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2020)
Article
Engineering, Electrical & Electronic
Hassan Rangaswamy Shwetha, Dasika Nagesh Kumar
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2020)