Article
Thermodynamics
Donggyun Ku, Minje Choi, Nakyoung Yoo, Seungheon Shin, Seungjae Lee
Summary: This study investigates the optimal routing of electric vehicles by utilizing 3D spatial information data and considering the slope of each link in the route, resulting in improved energy efficiency. By assigning routes to optimize battery efficiency, the proposed approach achieved an energy efficiency improvement of 7.84 km/kWh at an average speed of 70 km/h, demonstrating its effectiveness in achieving the goal of green transportation.
Article
Computer Science, Information Systems
Huina Li, Yuan Ping, Bin Hao, Chun Guo, Yujian Liu
Summary: This article proposes an improved boundary support vector clustering (IBSVC) method that achieves reasonable boundaries and comfortable parameters through self-adaptive support. The method enhances the accuracy and efficiency of clustering through movable edge selection and flexible parameter selection.
Article
Computer Science, Artificial Intelligence
Xiaoxi Zhao, Saiji Fu, Yingjie Tian, Kun Zhao
Summary: The QTLS method proposed in this paper combines QTSELF with LSSVM, imposes different penalties on samples based on their locations, and enhances model robustness. Its generalization capacity is investigated using Rademacher complexity theory, and extensive experiments confirm its effectiveness.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Biophysics
Wenyi Zeng, Peng Chen, Shunji Li, Qiuyue Sha, Pengjie Li, Xuemei Zeng, Xiaojun Feng, Wei Du, Bi-Feng Liu
Summary: The article introduces a hand-powered vacuum-driven microfluidic device that can rapidly generate concentration gradients of different antibiotics for antibiotic susceptibility testing. The device is easy to operate, with advantages such as high throughput, cost efficiency, and shortened detection time.
BIOSENSORS & BIOELECTRONICS
(2022)
Article
Computer Science, Artificial Intelligence
Geonseok Lee, Pilwon Woo, Kichun Lee
Summary: This paper proposes a new data-generation method using geometrical edge probability for OCSVM hyperparameter selection, which improves the limitations of existing methods and determines the distribution of anomaly data. The proposed method is evaluated on datasets with 16 different dimensions and demonstrates performance improvement in the classification of target and anomaly data.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Feixia Ji, Jian Wu, Francisco Chiclana, Sha Wang, Hamido Fujita, Enrique Herrera-Viedma
Summary: This study proposes an overlapping community-driven feedback mechanism to improve consensus in social network group decision making. By guiding inconsistent subgroups to interact with each other and selecting personalized feedback parameters, this mechanism helps achieve higher levels of consensus.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Interdisciplinary Applications
Zhang-peng Tian, He-ming Liang, Ru-xin Nie, Xiao-kang Wang, Jian-qiang Wang
Summary: This paper proposes a sentiment analysis-based multi-criteria decision-making method to help consumers make EV purchase choices. The sentiment analysis results are transformed into hesitant intuitionistic fuzzy elements to derive the group opinion for each alternative. A comprehensive weighting method is developed to determine the weights of criteria. The ranking of candidate EV series can be obtained through the extended ORESTE method based on hesitant intuitionistic fuzzy Chebyshev distance. The results of sentiment analysis can also be useful for companies to explore consumers' demand for EVs.
COMPUTERS & INDUSTRIAL ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
You-Gan Wang, Jinran Wu, Zhi-Hua Hu, Geoffrey J. McLachlan
Summary: The choice of hyperparameters in support vector regression has always been challenging. This paper proposes an extended primal objective function based on probability regularization, which automatically selects appropriate parameters and has a close connection to v-support vector regression.
PATTERN RECOGNITION
(2023)
Article
Biodiversity Conservation
Nicole Miller-Struttmann, Zachary Miller, Candace Galen
Summary: Pollinators at high elevations face multiple threats from climate change, including phenological mismatches with floral resources and community disruption. Warmer conditions are decreasing abundances of range-stable alpine bumble bees and increasing abundance of range-expanding species, leading to a more diverse bumble bee community. However, the precise mechanisms accounting for these changes are not yet known.
GLOBAL CHANGE BIOLOGY
(2022)
Article
Geochemistry & Geophysics
Chen Chen, Sirui Tian, Zhiyong Xu
Summary: In this letter, an efficient ISAR motion compensation method is proposed based on fast parameter estimation. The image entropy is derived using a compensation matrix and sinc function interpolation to eliminate the high-order phase term, transforming the parameter estimation into an optimization problem. The gradient descent algorithm (GDA) is employed to accelerate the computation speed. Our method outperforms other recently proposed methods in terms of robustness and computing cost.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Economics
Luya Wang
Summary: We address the problem of selecting the smoothing parameter in adaptive testing of a parametric model against a nonparametric alternative model using a data-driven method. Simulation results demonstrate the effectiveness of our proposed procedure, which outperforms existing approaches. We also discuss the extension of our method to more general model specification testing problems, including testing a parametric quantile function and testing nonparametric significance.
Article
Physics, Fluids & Plasmas
Ningfei Chen, Hanyuan Hu, Xiangyu Zhang, Shizhao Wei, Zhiyong Qiu
Summary: In the presence of a given radial electric field, the zero-frequency radial electric field can significantly stabilize ITG, while the associated density perturbation has little effect on ITG stability. However, the parallel mode structure is slightly affected due to the evenly symmetric density modulation associated with the zero-frequency radial electric field.
PHYSICS OF PLASMAS
(2021)
Article
Automation & Control Systems
Chengdong Wang, Yang Ge, Jianpu Ma, Zheming Yu, Kedong Zhang, Tongshun Liu, Xuhong Guo, Shu Huang
Summary: Functionally graded materials (FGMs) are promising engineering materials, highly desirable in extreme environments. The Ni-Fe FGMs produced by 3D-printing require post-cutting treatment in nuclear industry. The machinability of the FGMs is challenging due to the gradient of mechanical property.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2023)
Article
Computer Science, Artificial Intelligence
Jinran Wu, You-Gan Wang
Summary: This article proposes a data-driven approach for support vector regression, which determines the value of the insensitivity parameter through minimizing a generalized loss function. It utilizes statistical standardization and probabilistic regularization to generate data-dependent hyperparameters. The method is demonstrated to outperform other approaches in various simulations and case studies, with lower computational costs.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Engineering, Electrical & Electronic
Rui Guo, Tianyao Huang, Maokun Li, Haiyang Zhang, Yonina C. Eldar
Summary: Electromagnetic (EM) imaging is widely used in various fields, but it is an ill-posed inverse problem. Machine learning techniques, particularly deep learning, have shown potential in fast and accurate imaging. However, the challenge lies in constructing a training set that accurately represents practical scenarios. To overcome this, recent research has focused on physics-embedded ML methods for EM imaging, which combine the benefits of big data and the theoretical constraints of physical laws.
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Article
Geochemistry & Geophysics
Wouter Deleersnyder, Benjamin Maveau, Thomas Hermans, David Dudal
Summary: The study proposes a new inversion scheme for electromagnetic induction data by leveraging the sparsity of the model in the wavelet domain, improving efficiency and accuracy. Transformation to the wavelet domain allows for exploration of the temporal and spatial characteristics of the model, simplifying it by reducing small-scale coefficients. The scheme supports various regularization methods and can choose different wavelet basis functions based on the desired conductivity profile.
GEOPHYSICAL JOURNAL INTERNATIONAL
(2021)
Article
Geosciences, Multidisciplinary
Robin Thibaut, Thomas Kremer, Annie Royen, Bun Kim Ngun, Frederic Nguyen, Thomas Hermans
Summary: The current geophysical inversion paradigm based on Occam's principle may not be suitable for complex structures in the subsurface. An alternative Minimum Gradient Support (MGS) approach is proposed to compute sharp contrasts, but its performance is highly dependent on the selection of the focusing parameter beta. A new methodology is presented to incorporate prior information and improve imaging for resistivity and chargeability structures in real case studies, showing promising results in both synthetic and field data. The challenge of automatically selecting the beta parameter remains for future developments.
JOURNAL OF APPLIED GEOPHYSICS
(2021)
Article
Computer Science, Interdisciplinary Applications
Jorge Lopez-Alvis, Eric Laloy, Frederic Nguyen, Thomas Hermans
Summary: In this study, the authors review the conceptual framework of inversion with DGMs and point out that the nonlinearity in generative mapping between latent and original representations is mainly caused by changes in topology and curvature. They identify a conflict between the accuracy of generated patterns and the feasibility of gradient-based inversion, and propose a method that achieves a tradeoff between these two goals for better inversion results.
COMPUTERS & GEOSCIENCES
(2021)
Article
Environmental Sciences
Diep Cong-Thi, Linh Pham Dieu, Robin Thibaut, Marieke Paepen, Huu Hieu Ho, Frederic Nguyen, Thomas Hermans
Summary: This study presents a methodology for the semiquantitative interpretation of Electrical Resistivity Tomography (ERT) in coastal aquifers under pressure from growing populations and climate change. The method aims to provide insights into the extent of saltwater intrusion without colocated well data, and has been successfully applied in the Luy River aquifers in Vietnam.
Article
Geochemistry & Geophysics
J. Lopez-Alvis, F. Nguyen, M. C. Looms, T. Hermans
Summary: Prior information on subsurface spatial patterns is crucial for geophysical inversion, and a VAE can assemble all possible patterns into a single prior distribution for inversion, offering lower computational cost and more realistic prior uncertainty.
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
(2022)
Article
Environmental Sciences
Marieke Paepen, Wouter Deleersnyder, Sybren De Latte, Kristine Walraevens, Thomas Hermans
Summary: This study investigates the Western Belgian coast using resistivity methods and finds that while it is difficult to quantitatively interpret FSGD (fresh submarine groundwater discharge), lateral qualitative changes can be deduced from inversion models. Field data shows that MAR (managed aquifer recharge) has a positive impact on FSGD, while groundwater extraction reduces the outflow of freshwater to the North Sea.
Article
Environmental Sciences
Linh Pham Dieu, Diep Cong-Thi, Tom Segers, Huu Hieu Ho, Frederic Nguyen, Thomas Hermans
Summary: The groundwater resources in Binh Thuan province, Vietnam are threatened by climate change and saltwater intrusion, leading to a significant restriction on the quality of drinking water and irrigation. Research reveals that salinization is more severe in the deeper aquifers and during the dry season, contrary to previous expectations. Freshening is the dominant process in the aquifers, especially during the rainy season.
Article
Geochemistry & Geophysics
Hadrien Michel, Thomas Hermans, Frederic Nguyen
Summary: This paper presents a new method for solving geophysical problems, which combines iterative prior resampling and rejection sampling to improve the accuracy of subsurface imaging and reduce the problem of overestimating the uncertainty range caused by prior uncertainty.
GEOPHYSICAL JOURNAL INTERNATIONAL
(2023)
Article
Environmental Sciences
Robin Thibaut, Nicolas Compaire, Nolwenn Lesparre, Maximilian Ramgraber, Eric Laloy, Thomas Hermans
Summary: Temperature logs are crucial in the geothermal industry for exploration, system design, and monitoring. However, the limited number of observations makes it necessary to optimize the number and location of boreholes. Four-dimensional temperature field monitoring using time-lapse Electrical Resistivity Tomography has been explored as a potential alternative with higher resolution and lower cost. Bayesian Evidential Learning (BEL) is used to optimize the design of the monitoring experiment by considering various combinations of data sources.
WATER RESOURCES RESEARCH
(2022)
Article
Geochemistry & Geophysics
Wouter Deleersnyder, Benjamin Maveau, Thomas Hermans, David Dudal
Summary: Regularization methods improve the stability of ill-posed inverse problems by introducing prior characteristics for the solution. In this paper, a multidimensional scale-dependent wavelet-based l(1)-regularization term is proposed to solve the ill-posed airborne electromagnetic induction inverse problem. The regularization term is flexible and can recover various inversion models based on a suitable wavelet basis function.
GEOPHYSICAL JOURNAL INTERNATIONAL
(2023)
Article
Geochemistry & Geophysics
Itzel Isunza Manrique, David Caterina, Frederic Nguyen, Thomas Hermans
Summary: The nonuniqueness of the solution to the geophysical inverse problem can cause misinterpretation of subsurface characterization. To address this, ground-truth information and probabilistic classification can be used to improve inverted models. The probabilistic approach shows good performance in classifying categories with high spatial heterogeneity and limited ground-truth data.
Article
Environmental Sciences
Wouter Deleersnyder, David Dudal, Thomas Hermans
Summary: This paper proposes an appraisal tool for evaluating the inconsistency between the inversion model and multidimensional data, using a normalized gradient calculated based on multidimensional forward modeling. Additionally, an alternative approach is suggested to account for imperfect forward modeling with low computational cost. The method is demonstrated on an AEM survey, revealing potential problematic zones in the estimated fresh-saltwater interface.
Article
Computer Science, Information Systems
Pierre Goovaerts, Thomas Hermans, Peter F. Goossens, Ellen van de Vijver
Summary: This paper addresses two common challenges in analyzing spatial epidemiological data and presents an alternative method to overcome the limitations of the Poisson kriging approach. The soft indicator kriging method shows attractive features such as no negative kriging estimates, smaller smoothing effect, and better agreement with observed municipality-level rates after aggregation.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2023)
Article
Geosciences, Multidisciplinary
Thomas Hermans, Pascal Goderniaux, Damien Jougnot, Jan H. Fleckenstein, Philip Brunner, Frederic Nguyen, Niklas Linde, Johan Alexander Huisman, Olivier Bour, Jorge Lopez Alvis, Richard Hoffmann, Andrea Palacios, Anne-Karin Cooke, Alvaro Pardo-Alvarez, Lara Blazevic, Behzad Pouladi, Peleg Haruzi, Alejandro Fernandez Visentini, Guilherme E. H. Nogueira, Joel Tirado-Conde, Majken C. Looms, Meruyert Kenshilikova, Philippe Davy, Tanguy Le Borgne
Summary: This paper discusses the interest and potential for monitoring and characterizing spatial and temporal variability in hydrogeological processes, and proposes a classification of processes and applications at different scales based on high-resolution space-time imaging. The authors call for the validation of 4D imaging techniques at highly instrumented observatories and the harmonization of open databases to share hydrogeological data sets in their 4D components.
HYDROLOGY AND EARTH SYSTEM SCIENCES
(2023)
Article
Water Resources
Md Mizanur Rahman Sarker, Thomas Hermans, Marc Van Camp, Delwar Hossain, Mazeda Islam, Nasir Ahmed, Md Abdul Quaiyum Bhuiyan, Md Masud Karim, Kristine Walraevens
Summary: People in the southwestern coastal part of Bangladesh are facing a severe freshwater crisis due to saline groundwater at a shallow depth. This study analyzed existing datasets using multivariate statistics and identified the factors affecting groundwater chemistry. Cluster analysis and factor analysis revealed the characteristics and influences of different groundwater types.