Article
Environmental Sciences
Beom-Jin Kim, You-Tae Lee, Byung-Hyun Kim
Summary: This study showed the process and method of selecting the optimal deep learning model for predicting dam inflow using hydrologic data over the past 20 years. The optimal models for Andong Dam and Imha Dam were selected for drought and typhoon conditions, showing closer prediction to the observed inflow than the SFM. The deep learning models were more accurate under various typhoon conditions, but the SFM showed better results under certain conditions. Comparing the inflow predictions of the SFM and deep learning models is necessary for efficient dam operation and management.
Article
Environmental Sciences
Hyeon Seok Choi, Joong Hoon Kim, Eui Hoon Lee, Sun-Kwon Yoon
Summary: This study improves data preprocessing and machine learning algorithms to increase the accuracy of dam inflow prediction. By applying deep learning models and an improved algorithm, accurate predictions of historical data from 2004 to 2021 for the Soyang Dam Basin in South Korea were achieved, resulting in improved accuracy of dam inflow prediction.
Article
Engineering, Environmental
Hajar Feizi, Halit Apaydin, Mohammad Taghi Sattari, Muslume Sevba Colak, Muhammad Sibtain
Summary: A Hybrid Deep Learning Inflow Prediction-Rolling Window (HDeepLIP-RW) framework is proposed in this study to predict reservoir inflow efficiently by utilizing hybrid alternatives of four different RNN architectures and the rolling window technique. The results on real data from Ermenek Dam in Turkey demonstrate the superiority of the HDeepLIP-RW based BiLSTM-GRU model in terms of performance and accuracy.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2022)
Article
Engineering, Electrical & Electronic
Gunwoo Park, Sihyun Ryu, Sang-Kyoon Park, Sukhoon Yoon, Seungbum Koo
Summary: This study investigates the accuracies and uncertainties associated with using a single virtual accelerometer for whole-body motion predictions. The results show that sensor position and training dataset size have an impact on the prediction errors and uncertainties. Increasing dataset size reduces error and epistemic uncertainty, while aleatoric uncertainty remains constant. Different sensor positions lead to different prediction errors and uncertainties for each gait parameter.
IEEE SENSORS JOURNAL
(2023)
Article
Computer Science, Interdisciplinary Applications
Li-Li Bao, Jiang-She Zhang, Chun-Xia Zhang, Rui Guo, Xiao-Li Wei, Zi-Lu Jiang
Summary: In seismic exploration, reservoir prediction is important to reveal reservoir characteristics. A Bayesian neural network (BNN) model is proposed to predict reservoir thickness and quantify uncertainty. The BNN combines attribute data with spatial information and uses the Monte Carlo dropout (MC-dropout) approach to capture uncertainty.
COMPUTERS & GEOSCIENCES
(2023)
Article
Nuclear Science & Technology
Lesego E. Moloko, Pavel M. Bokov, Xu Wu, Kostadin N. Ivanov
Summary: This study uses Deep Neural Networks (DNNs) to predict assembly axial neutron flux profiles in the SAFARI-1 research reactor and quantifies the uncertainties in DNN predictions. Uncertainty Quantification is done using Monte Carlo Dropout (MCD) and Bayesian Neural Networks solved by Variational Inference (BNN VI). The results show that regular DNNs, DNNs with MCD, and BNN VI all have good prediction and generalization capabilities, and the uncertainty bands produced by MCD and BNN VI accurately envelope the measurement data points.
ANNALS OF NUCLEAR ENERGY
(2023)
Article
Engineering, Marine
Chengcheng Qiu, Qiaogao Huang, Guang Pan
Summary: The study introduces a variational Bayesian convolution neural network (VB-CNN) framework for obtaining the pump-jet propulsor (PJP) transient velocity field. This method saves computing time and resources compared to the computational fluid dynamics (CFD) method. The results show that VB-CNN outperforms CNN in accurately predicting the PJP transient flow field.
Article
Engineering, Geological
Pin Zhang, Zhen-Yu Yin, Yin-Fu Jin
Summary: This study proposes a modeling strategy for developing prediction models for soil properties in geotechnical engineering using the Bayesian neural network (BNN) with a strong non-linear fining capability and uncertainty. The results show that BNN can accurately predict compression index and undrained shear strength, but its reliability is low in sparse datasets. Additionally, a novel parametric analysis method is proposed to capture the relationship between input parameters and soil properties.
CANADIAN GEOTECHNICAL JOURNAL
(2022)
Article
Engineering, Industrial
Xiaoyan Shao, Baoping Cai, Yonghong Liu, Junyan Zhang, Zhongfei Sui, Qiang Feng
Summary: A novel hybrid model-data-driven RUL prediction method based on a fusion of Kalman filter and dynamic Bayesian network is proposed in this paper. The method improves accuracy by enhancing the performance of observation values through DBN and considering estimation error and observation error. The uncertainty distribution of degradation parameters and environmental parameters is integrated into the state estimation model. Numerical simulation of a subsea Christmas tree valves demonstrates the advantages of the proposed RUL prediction method.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2023)
Article
Engineering, Civil
Pedram Pishgah Hadiyan, Ramtin Moeini, Eghbal Ehsanzadeh, Monire Karvanpour
Summary: This study investigates the performance of dynamic artificial neural network models in predicting the water inflow into the Sefidruod dam reservoir in Iran. The study analyzes the discharge time series of tributaries and tests for trend and homogeneity. The study also simulates and predicts the inflow discharges using dynamic Nonlinear Auto-Regressive models, and finds a significant decreasing trend in both rivers. The study concludes that the NAR model performs better than the NARX model in this case.
WATER RESOURCES MANAGEMENT
(2022)
Article
Automation & Control Systems
Boxuan Zhong, Rafael Luiz da Silva, Minhan Li, He Huang, Edgar Lobaton
Summary: This article introduces a vision-based context prediction framework using Bayesian neural networks to quantify uncertainty and compare different wearable camera positions. The framework is used for online decision-making and fusion, demonstrating a practical procedure to interpret and improve the performance of deep neural networks.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Fernando Rodriguez-Sanchez, Concha Bielza, Pedro Larranaga
Summary: This paper introduces a multipartition clustering method for mixed data, which efficiently handles multifaceted data with several reasonable interpretations by utilizing Bayesian network factorization and the variational Bayes framework.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Green & Sustainable Science & Technology
Sarmad Dashti Latif, Ali Najah Ahmed
Summary: Correct inflow prediction is crucial for flood control and water supply efficiency. A machine learning model using SVR was generalized for reservoir inflow forecasting. Hydrological parameters, such as daily, weekly, and monthly inflow and rainfall, were used for input selection.
ENVIRONMENT DEVELOPMENT AND SUSTAINABILITY
(2023)
Article
Environmental Sciences
Jose-Luis Molina, Jose-Luis Garcia-Arostegui
Summary: This research aims to analyze and model the relationship between binomial rainfall and groundwater levels. It uses Bayesian Causal Reasoning (BCR) based on Bayesian Theorem to capture the inherent causality in the data. The methodology includes classic regression analysis and Bayesian Causal Modelling Translation (BCMT) with iterative steps. This innovative methodology has been successfully applied to aquifer management in the Campo de Cartagena groundwater body in Spain.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Construction & Building Technology
Bongjun Ji, Sushmit Sharma Bhattarai, Il-Ho Na, Hyunhwan Kim
Summary: A Bayesian deep learning framework is proposed to accurately predict the rheological properties of asphalt binders and quantify uncertainties. Atomic Force Microscopy images are used as input values, and the Dynamic Shear Rheometer test is used for measurement. The Bayesian deep learning model offers improved prediction accuracy and reduced testing time compared to conventional methods, and the quantification of uncertainties enables informed decision-making and risk assessment in asphalt binder selection and pavement design.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Environmental Sciences
Saman Javadi, Hamid Kardan Moghaddam, Aminreza Neshat
Summary: A new approach for coastal aquifers involves determining the saltwater-freshwater interface and mapping groundwater vulnerability using methods like index method and numerical modelling. Past vulnerability methods did not account for coastal regions, but this study shows the vulnerability of coastal aquifers in Iran can be very high due to saltwater intrusion from coastline reaching depths of up to 720 meters. The correlation between the DRASTICSea method and qualitative parameter was found to be 68%.
GEOCARTO INTERNATIONAL
(2022)
Article
Environmental Sciences
Hamid Kardan Moghaddam, Sami Ghordoyee Milan, Zahra Kayhomayoon, Zahra Rahimzadeh kivi, Naser Arya Azar
Summary: Water resources management requires proper understanding of available water sources, with simulation models like GMDH proving to be effective in predicting groundwater levels. In this study, it was found that the GMDH model performed the best in predicting groundwater levels in the Birjand aquifer, which is facing a critical groundwater level decline.
ENVIRONMENTAL MONITORING AND ASSESSMENT
(2021)
Article
Engineering, Civil
Massoud Tabesh, Abbas Roozbahani, Farhad Hadigol, Elham Ghaemi
Summary: Urban water supply systems are crucial for community infrastructure, but water shortages can lead to lack of access to safe water. Water treatment plants play a vital role in providing high quality water even in emergencies, highlighting the importance of assessing and improving their reliability. A comprehensive approach to evaluating and reducing vulnerabilities in water treatment plants can lead to cost savings and more effective recovery planning.
IRANIAN JOURNAL OF SCIENCE AND TECHNOLOGY-TRANSACTIONS OF CIVIL ENGINEERING
(2022)
Article
Environmental Sciences
Saeideh Samani, Hamid Kardan Moghaddam, Mohammad Javad Zareian
Summary: By introducing groundwater sustainability indicators, this study assessed the stability of aquifers in the Salt Lake catchment in Iran and found that overall sustainability decreases over time. Aquifers in the western part of the basin show higher sustainability. The groundwater resources sustainability plan in this catchment has not been successful, indicating a need for an urgent update to the management plan.
ENVIRONMENTAL EARTH SCIENCES
(2021)
Article
Engineering, Civil
Sami Ghordoyee Milan, Abbas Roozbahani, Naser Arya Azar, Saman Javadi
Summary: The study developed a predictive model based on machine learning, and combining ANFIS with HHO significantly improved the prediction accuracy of groundwater extraction amount. The results indicated that ANFIS-HHO performed well on test data and had better predictive accuracy compared to other algorithms.
JOURNAL OF HYDROLOGY
(2021)
Article
Engineering, Civil
Atiyeh Bozorgi, Abbas Roozbahani, Seied Mehdy Hashemy Shahdany, Rouzbeh Abbassi
Summary: The study developed a novel multi-hazard risk assessment model for evaluating the risks associated with agricultural water supply systems.
WATER RESOURCES MANAGEMENT
(2021)
Article
Engineering, Environmental
Sara Azargashb Lord, Mojtaba Hamze Ghasabsarai, Maryam Movahedinia, Seied Mehdy Hashemy Shahdany, Abbas Roozbahani
Summary: This study focused on conducting local rehabilitation in stormwater collection systems in the eastern Tehran metropolis by detecting critical nodes and redesigning critical canal reaches for maximum flood discharge capacity with minimum reconstruction cost. The optimal parameters for canal bed width and depth were obtained using Ant Colony Optimization (ACO) and considering flood probability exceedance as a design constraint. The results showed that widening and deepening the canals were necessary for sufficient flow capacity in various return periods (RPs).
WATER SCIENCE AND TECHNOLOGY
(2021)
Article
Geosciences, Multidisciplinary
Zahra Kayhomayoon, Sami Ghordoyee Milan, Naser Arya Azar, Hamid Kardan Moghaddam
Summary: This study proposed a new 3-stage approach for simulating groundwater levels in an arid region of eastern Iran, utilizing clustering, simulation, and optimization stages. By dividing the study aquifer into clusters and applying various input variables to an artificial neural network (ANN), optimal models were identified and validated using optimization methods such as particle swarm optimization (PSO) and whale optimization algorithm (WOA). The results showed improved simulation accuracy and highlighted the importance of choosing the appropriate optimization method based on the clustering type.
NATURAL RESOURCES RESEARCH
(2021)
Article
Engineering, Civil
Javad Shafiee Neyestanak, Abbas Roozbahani
Summary: The study assessed the risks of using urban treated wastewater in different sectors using a novel Hybrid Bayesian Network, finding that while risks can be predicted with acceptable performance, there are variations in risks depending on the sector.
WATER RESOURCES MANAGEMENT
(2021)
Article
Computer Science, Interdisciplinary Applications
Pezhman Mousavi-Mirkalaei, Abbas Roozbahani, Mohammad Ebrahim Banihabib, Timothy O. Randhir
Summary: Forecasting urban water consumption is crucial for efficient management in rapidly growing cities in arid regions. Comparing Bayesian Networks and Gene Expression Programming models, the study finds that Bayesian Networks provide better accuracy in predicting urban water consumption. Additionally, sunshine hours have a significant influence on water consumption.
EARTH SCIENCE INFORMATICS
(2022)
Article
Engineering, Civil
Fatemeh Bayat, Abbas Roozbahani, Seied Mehdy Hashemy Shahdany
Summary: This study quantitatively evaluates the performance of practical alternatives in modernization projects of water distribution in irrigation networks based on the water-food-energy nexus using the AHP-Entropy-WASPAS technique. The results show that the centralized automatic operation method performs more efficiently than the decentralized automatic operation method under both normal and water shortage scenarios.
WATER RESOURCES MANAGEMENT
(2022)
Article
Green & Sustainable Science & Technology
S. Dehghani, A. R. Massah Bavani, A. Roozbahani, A. Gohari, R. Berndtsson
Summary: This paper presents a framework for evaluating water scarcity in the Qazvin Plain, Iran, and assesses its impact using a system dynamics model and Bayesian averaging model. The results indicate that future water scarcity may severely affect agricultural development while having minor effects on the industry, domestic, and service sectors. Policymakers should focus on implementing adaptation strategies for the agricultural sector to prepare for unpredictable shocks.
SUSTAINABLE PRODUCTION AND CONSUMPTION
(2022)
Article
Geochemistry & Geophysics
Pejman Zarafshan, Hamed Etezadi, Saman Javadi, Abbas Roozbahani, S. Mehdi Hashemy, Payam Zarafshan
Summary: This study applies machine learning and deep learning methods to simulate groundwater levels and compares the results with a numerical model. The findings indicate that machine learning and deep learning models outperform the numerical model in terms of accuracy, and the computational capabilities and memory of all three models are similar.
Article
Engineering, Civil
Amirhossein Nazari, Abbas Roozbahani, Seied Mehdy Hashemy Shahdany
Summary: This study introduces a framework based on green infrastructure, multi-objective optimization, and decision support tools to determine the most cost-effective Low Impact Development (LID) solutions. Different combinations of LID practices were evaluated using the Storm Water Management Model (SWMM) and ranked based on technical and economic criteria using TOPSIS and COPRAS methods. Results showed that Scenario 4 had the best performance under TOPSIS while Scenario 2 performed better using the COPRAS method.
WATER RESOURCES MANAGEMENT
(2023)
Article
Environmental Sciences
Farzaneh Soltani, Saman Javadi, Abbas Roozbahani, Ali Reza Massah Bavani, Golmar Golmohammadi, Ronny Berndtsson, Sami Ghordoyee Milan, Rahimeh Maghsoudi
Summary: Assessing water resources status is crucial for long-term planning. This study focuses on evaluating the effects of climate change on water resources in the Shazand plain in Iran, which has experienced significant declines in streamflow and groundwater levels. The results predict a substantial decrease in river discharges and groundwater levels in this region under future climate conditions, emphasizing the need for sustainable management methods to mitigate these effects.