Whether you’re interested in renewable energy forecasting, metamaterial design, or the latest in robotics, there’s something here for everyone.
Top Articles
BAOTING China Hosts Climate and Health International Conference in Dec 2024
The BAOTING conference brought together experts on climate change and health for a two-day event (Dec 15-16, 2024), focusing on implementation and action to mitigate the effects of climate change on human health. Co-hosted by the Ningyuan Institute and Baoting County Government, the event aimed to foster global cooperation and knowledge sharing in sustainable development.
EDF Group Contracts Envision Energy for Battery Energy Storage Systems in South Africa
EDF Group selects Envision Energy for battery energy storage systems, supporting sustainable power infrastructure development in South Africa’s Oasis 1 portfolio.
JA Solar Achieves Record-Breaking Open-Circuit Voltage with TOPCon Technology
JA Solar achieves record-breaking open-circuit voltage using tunnel oxide passivated contact (TOPCon) technology, surpassing previous records and validated by ISFH’s independent testing.
Community-Centered Solar Energy Solutions for a Sustainable Future
Renewable energy projects can thrive by addressing local needs and concerns, as demonstrated by the Ranch Sereno project’s innovative approach to community engagement and sustainable development.
Automated Grape Bunch Detection and Counting for Agricultural Robots Optimization
Scientists develop an automated grape bunch detection system using YOLOv3 network and local tracking algorithm, enabling accurate 3D positioning and counting in agricultural robots.
Transformers in Robotics: Recent Advances and Applications Overview
Recent breakthroughs in Deep Learning have enabled the use of Transformers architecture in robotics, outperforming traditional neural networks with their self-attention mechanism and scalability. In this paper, we explore recent advances in Transformer architectures integrated into robotic perception, planning, and control for autonomous systems, as well as their applications in pre-trained foundation models and Deep Reinforcement Learning (DRL) for reliable planning, human-robot interaction, long-horizon decision-making, and generalization.
Metamaterial Design Optimization Using Neuroevolution Algorithm Techniques Effectively
**Metamaterial Design Optimization Using Neuroevolution Algorithm Techniques Effectively** Discover how neuroevolution algorithm techniques efficiently optimize metamaterial design, balancing multiple competing properties such as stiffness, toughness, weight, and energy-absorbing capacity, revealing empirical bounds of elastic properties for diverse applications in robotics, biomedicine, thermal engineering, and photonics.
Super Resolution Climate Data Generation for Renewable Energy Forecasting
Scientists develop a super-resolution recurrent diffusion model (SRDM) to enhance temporal resolution of climate data and simulate short-term uncertainty of renewable energy resources. The SRDM incorporates a pre-trained decoder and denoising network, generating high-resolution climate data for wind and photovoltaic power forecasting. Case studies demonstrate the SRDM’s superiority in producing accurate super-resolution climate data, with implications for sustainable power system development and climate change mitigation.
Two-Level Battery Charger Design for Electric Vehicles Optimized Performance Efficiency
This article presents a two-level battery charger design optimized for electric vehicles, focusing on modeling, simulation, and performance evaluation. Key features include: * Complementary switch operation * Detailed steady-state and dynamic modeling * Precise efficiency calculations and controller synthesis * Robust proportional-integral compensator design The study provides valuable insights for optimizing charger design and control, enhancing electric vehicle performance, and reliability.
Predictive Plant Growth Modeling Techniques in Simulated Environments Explained
This review explores cutting-edge predictive modeling techniques for simulating plant growth patterns in controlled environments, combining deterministic, probabilistic, and generative approaches with dynamic environmental interactions to improve high-throughput phenotyping and forecasting accuracy.
Renewable Energy Uncertainty in Electricity Markets and Efficient Forecasting Solutions
Optimizing electricity markets with large renewable energy sources requires accurate forecasting to minimize operating costs. This study proposes a value-oriented forecasting approach that improves entering quantities for renewable energy sources, reducing overall costs by 15% compared to traditional methods.
FPGA Accelerator for Lightweight Neural Convolutional Networks with Balanced Dataflow
A novel FPGA accelerator is proposed, leveraging multi-Computing-Engine (CE) architecture and balanced dataflow to efficiently accelerate lightweight neural convolutional networks (LWCNNs). The design minimizes on-chip/off-chip memory overhead, enhances computational efficiency, and achieves state-of-the-art performance with up to 68.3% reduced memory size and 2092.4 FPS processing speed.
LEO Satellite-Assisted MIMO Systems with Spatial Modulation and SSK Designs
Improve global connectivity through Low Earth Orbit (LEO) satellites, leveraging spatial modulation and space shift keying techniques to enhance spectral efficiency and reduce bit-error rate in 6G satellite-assisted wireless communication systems.
Solar Panel Efficiency Breakthrough Revolutionizes Green Energy Production
Scientists achieve a significant breakthrough in solar panel technology, marking a major step towards higher efficiency and reduced waste in green energy production with fewer replacements needed.