Article
Green & Sustainable Science & Technology
Andrew Allee, Nathaniel J. Williams, Alexander Davis, Paulina Jaramillo
Summary: Mini-grids are the lowest-cost solutions for electrifying rural communities with low energy access. Survey-based demand estimates for unelectrified customers are typically unreliable, overpredicting demand in the absence of historical data. Thorough inventories of currently-owned appliances should be prioritized in surveys to improve estimates of initial electricity demand.
ENERGY FOR SUSTAINABLE DEVELOPMENT
(2021)
Article
Energy & Fuels
Jinxi Dong, Zhaosheng Yu, Xikui Zhang, Lixi Chen, Qihong Zou, Wolin Cai, Musong Lin, Xiaoqian Ma
Summary: This study designs an algorithm structure that combines neural networks and gradient boosting decision trees to predict the safety performance of electric vehicle batteries, leading to improved prediction accuracy.
JOURNAL OF ENERGY STORAGE
(2024)
Article
Thermodynamics
Xinlin Wang, Hao Wang, Sung-Hoon Ahn
Summary: This study proposes a novel energy development strategy that combines power consumption type analysis and anomaly detection, reducing training costs and enabling the application of machine learning technologies in underdeveloped areas. This method effectively utilizes local renewable energy and improves residents' electricity usage experience.
Article
Energy & Fuels
Saeed Zamani, Mohsen Hamzeh
Summary: This study investigates the accurate determination of battery capacity in off-grid solar home systems and explores the impact of different depths of discharge batteries and pulse charging methods on required battery capacity.
Article
Chemistry, Multidisciplinary
Jose D. Hernandez-Betancur, Gerardo J. Ruiz-Mercado, Mariano Martin
Summary: Analyzing chemicals and their effects on the environment from a life cycle viewpoint can help predict and understand potential end-of-life management activities and recycling loops. This work uses quantitative structure-transfer relationship (QSTR) models based on chemical structure-based machine learning (ML) models to predict industrial end-of-life activities, chemical flow allocation, environmental releases, and exposure routes. These models assist stakeholders in making environmental decisions and assessing end-of-life exposure for chemicals.
ACS SUSTAINABLE CHEMISTRY & ENGINEERING
(2023)
Article
Energy & Fuels
Thomas Kroeger, Alexander Boes, Sven Maisel, Sara Luciani, Markus Schreiber, Markus Lienkamp
Summary: This paper proposes a machine learning-enhanced testing procedure that selects and predicts battery subsets to reduce testing costs and improve prediction accuracy.
JOURNAL OF ENERGY STORAGE
(2023)
Article
Computer Science, Artificial Intelligence
Yingqi Lu, Maede Maftouni, Tairan Yang, Panni Zheng, David Young, Zhenyu James Kong, Zheng Li
Summary: An effective lithium-ion battery recycling infrastructure is crucial for addressing the concerns of waste battery disposal and the sustainability of critical elements. Automation in the disassembly process is challenging due to the varying sizes and shapes of end-of-life batteries, as well as the hazardous materials they contain. This work presents an automatic battery disassembly platform enhanced by online sensing and machine learning technologies, enabling real-time diagnosis and control to optimize cutting quality and improve safety.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Green & Sustainable Science & Technology
Chayoung Kim, Taejung Park
Summary: This study aims to identify the key factors that influence the actual learning intention leading to participation in adult education. Using longitudinal big data from Korean adults (2017-2020), a predictive model was developed using tree-based machine learning. The results revealed that self-pay education expenses and the highest level of education completed were the most influential variables in predicting the likelihood of lifelong education participation. After grid search, the importance of these variables as well as overall figures, including the false positive rate, improved. Future studies could further enhance the performance of the machine learning model by adjusting hyperparameters using less computational methods.
Article
Biochemistry & Molecular Biology
Ana -Maria Raicu, Justin C. Fay, Nicolas Rohner, Julia Zeitlinger, David N. Arnosti
Summary: The fifth biennial symposium on Evolution and Core Processes in Gene Regulation, sponsored by ASBMB, was held at the Stowers Institute in Kansas City, Missouri from July 21 to 24, 2022. The symposium brought together experts in gene regulation and evolutionary biology to discuss topics such as enhancer evolution, the cis-regulatory code, and regulatory variation, with a focus on utilizing deep learning to decipher DNA sequence information.
JOURNAL OF BIOLOGICAL CHEMISTRY
(2023)
Review
Biology
Abel Chandra, Laura Tunnermann, Tommy Lofstedt, Regina Gratz
Summary: Recent developments in deep learning and increased protein sequencing have revolutionized life science applications, particularly in protein property prediction. Deep learning, especially using language models like the Transformer model, has shown promising results in predicting protein characteristics and post-translational modifications. These models learn multipurpose representations from large protein sequence repositories, bridging the gap between the number of sequenced proteins and proteins with known properties.
Article
Energy & Fuels
Elif Ceren Gok, Murat Onur Yildirim, Muhammed P. U. Haris, Esin Eren, Meenakshi Pegu, Naveen Harindu Hemasiri, Peng Huang, Samrana Kazim, Aysegul Uygun Oksuz, Shahzada Ahmad
Summary: The study successfully predicted the performance parameters of solar cells using machine learning technology, confirming the experimental data and indicating that this technology can help advance the commercialization of perovskite solar cells.
Article
Energy & Fuels
Faiza Mehmood, Muhammad Usman Ghani, Hina Ghafoor, Rehab Shahzadi, Muhammad Nabeel Asim, Waqar Mahmood
Summary: Load forecasting is important to avoid energy wastage by accurately estimating future energy generation and demand. However, existing approaches lack the potential of feature selection and dimensionality reduction, which can improve machine learning regressors' performance. This research introduces an end-to-end framework named EGD-SNet that predicts energy generation, demand, and temperature in multiple regions.
Editorial Material
Multidisciplinary Sciences
Simon Makin
Summary: Machine learning could assist in identifying viruses with a high potential for spillover from animals to humans, leading to future pandemics.
Article
Engineering, Multidisciplinary
Amjad Alsirhani, Mohammed Mujib Alshahrani, Abdulwahab Abukwaik, Ahmed I. Taloba, Rasha M. Abd El-Aziz, Mostafa Salem
Summary: Evaluating and forecasting stability in smart grid design is essential in order to mitigate unintended instability. This study presents a novel approach, MLP-ELM, for predicting smart grid sustainability using machine learning frameworks and principal component analysis (PCA). Simulation results show that the MLP-ELM approach outperforms traditional techniques, with high accuracy, precision, recall, and F-measure.
ALEXANDRIA ENGINEERING JOURNAL
(2023)
Article
Chemistry, Analytical
Ademir Goulart, Alex Sandro Roschildt Pinto, Adao Boava, Kalinka R. L. J. Castelo Branco
Summary: This project aims to manage various utilities and collect relevant information in an off-grid environment without commercial electricity and internet, using IoT off-grid technology. Machine learning is utilized for data selection, which is safely collected using a drone.