Domain generalization for image classification based on simplified self ensemble learning. by Zhenkai Qin, Xinlu Guo, Jun Li, Yue Chen

“…Experimental results on three public benchmark datasets (OfficeHome, PACS, and VLCS) demonstrate that our proposed algorithm achieves an improvement of up to 3 . 38% over existing methods in terms of generalization performance, particularly in complex and diverse real-world scenarios, such as autonomous driving and medical image analysis. …”
Get full text

Polymer electrolyte membrane fuel cell performance Revolutionized: Artificial intelligence-validated asymmetric flow channels enhance mass transport via hybrid analytical-numerical... by Nima Ahmadi, Ghader Rezazadeh

“…Experimental validation confirms a current density improvement of up to 5.6 % over conventional designs, while Artificial Intelligence (AI)-driven optimization via a hybrid Convolutional Neural Network and Genetic Algorithm independently identifies the same optimal configuration. …”
Get full text

Industrial-scale prediction of cement clinker phases using machine learning by Sheikh Junaid Fayaz, Néstor Montiel-Bohórquez, Shashank Bishnoi, Matteo Romano, Manuele Gatti, N. M. Anoop Krishnan

“…The framework can potentially enable real-time optimization of cement production, thereby providing a route toward reducing material waste and ensuring quality while reducing the associated emissions under real-world conditions.…”
Get full text

Weak Connection Regulation Technology for Source-Storage-Load Collaborative Microgrids Based on Direct Power Control by Peng REN, Zhigang ZHAO, Hongchao GAO, Wu WEN, Yingyi JIANG

“…The paper also studied the regional DC power supply and flexible regulation technology, analyzed the flexible characteristics of air conditioners, energy storage, photovoltaic, etc., and verified the power direct-control flexible regulation algorithm and evaluation indexes. From the effect of testing and verification, the whole system enables real-time response and rapid adjustment, thus improving the energy utilization efficiency, reducing the grid-side quota and realizing the local consumption of PV.…”
Get full text

Numerical Solution of the Inverse Thermoacoustics Problem Using QFT and Gradient Method by Syrym E. Kasenov, Aigerim M. Tleulesova, Almas N. Temirbekov, Zholaman M. Bektemessov, Rysbike A. Asanova

“…The inverse problem was reduced to an optimization problem, where the objective function was minimized using gradient methods, including the accelerated Nesterov algorithm. …”
Get full text

Research on the policy inconsistency, network motifs and low carbon effects for municipal solid waste management by Bo Lv, Tianxu Cui, Daiheng Li, Weiyue Yao

“…It offers strategies for enhancing carbon reduction through network optimization and phased policy design, aiding low - carbon MSWM policy improvement.…”
Get full text

Parametric Sensitivity Analysis of Safe Train Interval Model Based on Relative Braking Distance by QIAN Hua, LYU Haojiong

“…This study sought to analyze the sensitivity of key parameters affecting these safe intervals, to harness the potential of virtual coupling technology in improving efficiency while ensuring operational safety. …”
Get full text

Advanced artificial intelligence with federated learning framework for privacy-preserving cyberthreat detection in IoT-assisted sustainable smart cities by Mahmoud Ragab, Ehab Bahaudien Ashary, Bandar M. Alghamdi, Rania Aboalela, Naif Alsaadi, Louai A. Maghrabi, Khalid H. Allehaibi

“…Eventually, the walrus optimization algorithm (WOA) is used for hyperparameter tuning to improve the parameters of the SSAE approach and achieve optimal performance. …”
Get full text

The Current Development Status of Agricultural Machinery Chassis in Hilly and Mountainous Regions by Renkai Ding, Xiangyuan Qi, Xuwen Chen, Yixin Mei, Anze Li

“…In alignment with China’s “Dual Carbon” strategy and the unique operational requirements of hilly–mountainous areas (such as high gradients, uneven terrain, and small field sizes), this study proposes three key technological directions for the development of intelligent agricultural machinery chassis: (1) Multi-mode traveling mechanism design: Aimed at improving terrain traversability (ground clearance ≥400 mm, obstacle-crossing height ≥ 250 mm) and traction stability (slip ratio < 15%) across diverse landscapes. (2) Coordinated control algorithm optimization: Designed to ensure stable torque output (fluctuation rate < ±10%) and maintain gradient operation efficiency (e.g., less than 15% efficiency loss on 25° slopes) through power–drive synergy while also optimizing energy management strategies. (3) Intelligent perception system integration: Facilitating high-precision adaptive leveling (accuracy ± 0.5°, response time < 3 s) and enabling terrain-adaptive mechanism optimization to enhance platform stability and operational safety. …”
Get full text

Local Alignment of DNA Sequence Based on Deep Reinforcement Learning by Yong-Joon Song, Dong-Ho Cho

“…<italic>Goal:</italic> Over the decades, there have been improvements in the sequence alignment algorithm, with significant advances in various aspects such as complexity and accuracy. …”
Get full text

Dynamic spectrum allocation based on one-band multi-winner auction by Wen-zhu ZHANG, Ling-yun WANG

“…In order to improve the effectiveness of traditional auction in DSA (dynamic spectrum allocation) issue in cognitive radio networks,a new dynamic spectrum allocation algorithm based on one-band multi-winner auction was proposed.Compared with the original greedy algorithm,the new algorithm can achieve better spectrum allocation solutions with lower computational complexity by introducing the multiple greedy strategy.It increases the seller's revenue by improving the VCG (vickery-clarke-groves) mechanism,while retaining the dominant strategy incentive compatible property.In addition,it suppresses the occurrence of collusion effectively.Simulation results show that the algorithm proposed can get spectrum allocation efficiency close to the optimal solution and increase the auction revenue significantly.…”
Get full text

DMM-YOLO: A high efficiency soil fauna detection model based on an adaptive dynamic shuffle mechanism by Jiehui Ke, Renbo Luo, Guoliang Xu, Yuna Tan, Zhifeng Wu, Liufeng Xiao

“…Therefore, this paper proposes an improved algorithm based on You Only Look Once (YOLO) v9, which enhances feature capture capability while reducing parameters by 33.6%. …”
Get full text

An Advanced Recomposition-Based Displaying Technique: Maximizing Image Reconstruction for Virtual Museum Applications by Jingjie Zhao, Xin Shi, Olga Yezhova, Qinchuan Zhan, Xijing Zhang

“…These methods, combined with a multi-layer aggregation algorithm that encodes deep feature representations in a Gaussian Mixture Model (GMM), enable seamless scene reconstruction with improved precision. …”
Get full text

Nav2Scene: Navigation-driven fine-tuning for robot-friendly scene generation by Bowei Jiang, Tongyuan Bai, Peng Zheng, Tieru Wu, Rui Ma

“…The results demonstrate that by incorporating our Nav2Scene mechanism, the fine-tuned scene generators can produce scenes with improved navigation compatibility for home robots, while maintaining superior or comparable performance in terms of scene quality and diversity.…”
Get full text

Parts-per-quadrillion level gas molecule detection: CO-LITES sensing by Haiyue Sun, Shunda Qiao, Ying He, Xiaorong Sun, Yufei Ma

“…The artificial fish swarm algorithm auto-designed multi-pass cell (MPC) with double helix pattern, and the polymer modified round-head quartz tuning fork (QTF) with low-resonant frequency (f 0) were adopted to improve the gas absorption and QTF’s detection ability. …”
Get full text

Occlusion mapping reveals the impact of flight and sensing parameters on vertical forest structure exploration with cost-effective UAV based laser scanning by Matthias Gassilloud, Barbara Koch, Anna Göritz

“…Using sensor position reconstruction and a ray tracing algorithm to quantify occlusion, we found that: (1) A larger sensor tilt angle up to 15° increased total occlusion, enhancing exploration in the upper canopy, while decreasing it below due to reduced canopy penetration. (2) Flying multiple azimuthal directions with a linear scanning mode reduced vertical occlusion by up to 15.0% due to improved coverage from diverse perspectives. (3) Lissajous scanning patterns resulted in 10.1% less vertical occlusion compared to linear patterns, underscoring the importance of additional viewing angles. …”
Get full text

ASHM-YOLOv9: A Detection Model for Strawberry in Greenhouses at Multiple Stages by Yan Mo, Shaowei Bai, Wei Chen

“…We enhanced the YOLOv9 model by introducing the Alterable Kernel Convolution (AKConv) to improve the recognition efficiency while ensuring precision. …”
Get full text

Force-Driven Model for Automated Clear Aligner Staging Design Based on Stepwise Tooth Displacement and Rotation in 3D Space by Sensen Yang, Yumin Cheng

“…The algorithm was implemented in C++ and incorporated into the WebGL-based SmarteeCheck3.0 software for visualization. …”
Get full text

Simulation analysis of path planning for workpiece clamping robots based on digital twin technology by Xin PAN, Min LIANG, Yanchao YIN, Zhong CHEN

“…Targeting feasible regions through elliptical constraints accelerates convergence toward optimal paths, while techniques such as redundant node elimination and initial path smoothing further improve path quality, resulting in shorter, smoother routes. …”
Get full text

Artificial Intelligence-Assisted Clinical Decision-Making: A Perspective on Advancing Personalized Precision Medicine for Elderly Diabetes Patients by Hu J, Ren L, Wang T, Yao P

“…Nonetheless, several challenges persist, including issues related to data standardization, algorithm transparency, and patient privacy protection. …”
Get full text