Green Energy and Intelligent Transportation最新文献

{"title":"Feature-enhanced ensemble learning for accurate capacity estimation of lithium-ion batteries using partial discharging segments in initial stage based on second-order voltage derivatives","authors":"Ziheng Zhou ,&nbsp;Chaolong Zhang ,&nbsp;Shi Chen ,&nbsp;Yan Zhang ,&nbsp;Lei Wang","doi":"10.1016/j.geits.2025.100388","DOIUrl":"10.1016/j.geits.2025.100388","url":null,"abstract":"<div><div>Accurate and rapid capacity estimation is essential for efficient battery management in industrial settings particularly for cell grading, pack assembly, and second-life screening where throughput, cost, and energy efficiency are paramount. Conventional approaches require complete discharge cycles, leading to testing times of several hours per cell, which severely limits scalability and increases operational costs. To address this bottleneck, this paper proposes a fast capacity estimation method for battery capacity grading in the production process, which utilizes only the early-stage voltage measurements within the first 300–480 s of the initial discharge cycle during cell grading to accurately predict the cell's nominal capacity, enabling reliable battery capacity grading within minutes instead of hours. Although real-world grading data from production lines are often inaccessible, this first-cycle setup serves as a well-controlled surrogate that replicates key aspects of factory-based capacity labeling. The method exploits early-voltage transients that encode degradation-sensitive electrochemical signatures such as lithium inventory loss and solid-electrolyte interphase (SEI) evolution arising from microscopic changes in charge-transfer resistance and ion transport dynamics. From this short window, we extract physically interpretable health indicators (HIs) that reflect underlying aging mechanisms. A nonlinear feature enhancement strategy is then applied to amplify subtle capacity-related patterns while suppressing manufacturing-induced variability. These engineered features feed into a Multi-Decision Ensemble Learning (MDEL) architecture, which adaptively fuses multiple regression pathways to improve robustness across diverse cell chemistries and aging stages. Evaluated on both in-lab cells, the public CALCE and MIT dataset spanning fresh to end-of-life capacity conditions, the proposed approach achieves a mean absolute error (MAE) of ≤0.039,1 Ah (≤1.63% of nominal capacity), which is comparable to the methods with complete cycle data while reducing testing time by over 80%. This enables reliable capacity assessment in minutes rather than hours, offering a practical, scalable solution for high-throughput battery manufacturing, precise pack matching, and rapid second-life qualification.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 3","pages":"Article 100388"},"PeriodicalIF":16.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dynamic range compression dual-domain attention network for tunnel extreme exposure image enhancement in transportation visual systems","authors":"Bu Xu ,&nbsp;Jingyi Tang ,&nbsp;Jue Li ,&nbsp;Shuai Zhou ,&nbsp;Chen Liu","doi":"10.1016/j.geits.2025.100337","DOIUrl":"10.1016/j.geits.2025.100337","url":null,"abstract":"<div><div>The rapid expansion of road transportation infrastructure has led to an increased prevalence of tunnel scenarios, which are often characterized by extreme lighting conditions. These “black hole” and “white hole” effects, caused by stark brightness contrasts between tunnel entrances and interiors, severely compromise the image acquisition capabilities of transportation visual systems. To address these challenges, this study proposes a dynamic range compression dual-domain attention network (DRC-DFANet) for real-time and high-precision enhancement of tunnel images. The core architecture integrates a dynamic frequency-domain attention module (DFAM) and a spatial self-calibrated convolution (SCConv) module to concurrently optimize global illumination coordination and local detail restoration. The DFAM employs wavelet transform to decouple features into low-frequency and high-frequency components, enabling dynamic brightness adjustment and enhanced detail preservation. The SCConv module establishes interdependencies between channel and spatial dimensions to adaptively calibrate local contrast. Experiments on benchmark tunnel datasets demonstrate the superior performance of DRC-DFANet, with peak signal-to-noise ratio improvements of up to 8.8% and significant enhancements in high-frequency energy ratio, subband correlation, and exposure error metrics compared to state-of-the-art methods. Qualitative analyses validate the model's effectiveness in mitigating the “black hole” and “white hole” effects, preserving critical details such as vehicle contours, lane markings, and traffic signs. The transferability of DRC-DFANet is further confirmed on related transportation scenarios, underscoring its potential for wide-ranging applications in visual systems for autonomous driving, traffic monitoring, and other transportation-related tasks.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 3","pages":"Article 100337"},"PeriodicalIF":16.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981469","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated bidirectional electric vehicle battery network for sustainable communities: A planning framework","authors":"Srisanthosh Sekar ,&nbsp;Alampratap Singh Tiwana ,&nbsp;Kuljeet Singh Grewal","doi":"10.1016/j.geits.2025.100333","DOIUrl":"10.1016/j.geits.2025.100333","url":null,"abstract":"<div><div>The transition towards sustainable and net-zero energy communities has become imperative in addressing the challenges of climate change and ensuring a resilient energy future. This work proposes an innovative planning framework through the development of an integrated bidirectional electric vehicle (EV) battery storage network for net-zero communities. The proposed framework is intended for neighborhood planning and integrates a bidirectional charging infrastructure that allows EV batteries to seamlessly contribute to the grid during periods of high demand or store excess renewable energy during off-peak hours. To analyze various EV microgrid integration scenarios, a combined Matlab-Simulink and EnergyPlus simulation environment is proposed to simulate EV battery networks in neighborhood settings. This study examines state of charge (SoC), and energy exchange characteristics based on specific user behaviors and charging scenarios. A neighborhood archetype of 48 single-family detached houses is considered along with five EV use profiles (EVPs) for the demonstration of the proposed method. For the considered neighborhood, in winter, EVPs have eliminated the peak loads during early morning hours (1 am–6 am) by discharging stored energy. In spring, loads exceeding the base load are observed from 1 am to 10 am, with all EVPs discharging energy until 9 am and then recharging during off-peak hours. Summer required strategic charging management, with EVPs supporting peak loads from 7 am to 6 pm. In the fall, EVPs discharged energy from 12:01 am–6 am and recharged from 10 am to 6 pm. The study introduces the EVP peak support index facilitating real-time charging adjustments and incentivizing greater participation. By leveraging this index, smart charging systems can develop algorithms to control charging times based on grid needs, ensuring efficient energy distribution and enhanced grid stability. This framework offers a robust approach to scenario generation for energy and urban planners during the neighborhood planning stages predicting energy performance and management.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 3","pages":"Article 100333"},"PeriodicalIF":16.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079600","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Toward smart railway maintenance: AI-enhanced Non-Destructive Evaluation using Vision Transformers and CNNs for fastener defect detection","authors":"Samira Mohammadi ,&nbsp;Sasan Sattarpanah Karganroudi ,&nbsp;Mehdi Adda ,&nbsp;Hussein Ibrahim","doi":"10.1016/j.geits.2025.100332","DOIUrl":"10.1016/j.geits.2025.100332","url":null,"abstract":"<div><div>Predictive health management and maintenance of transport infrastructure are critical for preventing accidents and service disruptions. Applying Non-Destructive Evaluation (NDE) and imaging techniques is essential for identifying irregularities without causing harm. This research utilizes pre-trained models and incorporates transfer learning concepts to overcome dataset constraints. This study assesses the effectiveness of various machine learning models, including the Vision Transformer (ViT), Data-efficient Image Transformer (DeiT), VGG19, VGG16, and ResNet50, in enhancing NDE for railway track fasteners. ViT and DeiT, both transformer-based models, emerged as the top performers due to their superior learning efficiencies and generalization capabilities, augmented by precise hyperparameter tuning. VGG models are a reliable alternative, while ResNet50 is better suited for applications prioritizing computational efficiency over accuracy.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 3","pages":"Article 100332"},"PeriodicalIF":16.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncertainty quantification-based framework for predicting degradation trends of proton exchange membrane fuel cell","authors":"Bingxin Guo ,&nbsp;Changjun Xie ,&nbsp;Wenchao Zhu ,&nbsp;Yang Yang ,&nbsp;Hao Li ,&nbsp;Yang Li ,&nbsp;Hangyu Wu","doi":"10.1016/j.geits.2025.100297","DOIUrl":"10.1016/j.geits.2025.100297","url":null,"abstract":"<div><div>Accurately predicting the degradation trends of proton exchange membrane fuel cells (PEMFCs) can provide a solid basis for optimizing the control of vehicles and stations based on PEMFCs. However, most prediction methods do not consider factors such as measurement errors from experimental environments and the inherent cognitive uncertainty of the model. These methods can only offer point estimates, lacking credibility. This paper introduces a deep learning prediction framework that combines a bidirectional gated recurrent unit (BiGRU) model with a truncated Bayes by backpropagation through time (TB) algorithm. The TB algorithm reconstructs fixed parameters in the model into probability density distributions, transforming the output from point estimation to interval estimation with probability density distributions. Under dynamic conditions, the TB-BiGRU (truncated Bayes-based bidirectional gated recurrent unit) improves the mean absolute error (MAE) and root mean square error (RMSE) by 37.28% and 36.09%, respectively, compared to the TB-GRU (truncated Bayes-based gated recurrent unit). Compared with TB-GRU and B-GRU (Bayesian gated recurrent unit), TB-BiGRU has significantly improved uncertainty quantification ability. Under different working conditions and noise levels, the prediction accuracy of TB-BiGRU is superior to that of the other seven models, and it exhibits better noise resistance and stability. This method holds greater practical significance compared to other prediction approaches. Additionally, the paper proposes four effective evaluation metrics for uncertainty quantification, providing higher reference value in effectively characterizing the model's prediction accuracy and uncertainty quantification capability.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 3","pages":"Article 100297"},"PeriodicalIF":16.4,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145948039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing LiFePO4 battery SOC estimation: Electrochemical impedance spectroscopy with short-period sine-wave pulses","authors":"Yizhao Gao,&nbsp;Simona Onori","doi":"10.1016/j.geits.2025.100386","DOIUrl":"10.1016/j.geits.2025.100386","url":null,"abstract":"<div><div>State-of-charge (SOC) estimation for LiFePO₄ (LFP) batteries presents challenges due to their flat open-circuit voltage. Recent studies suggest that electrochemical impedance spectroscopy (EIS) offers a promising approach for SOC estimation in LFP cells. This work investigates a practical SOC estimation method based on EIS data obtained from short-duration sinusoidal current pulses. First, the EIS of LFP cells is characterized across a broad frequency range [0.01 Hz, 1000 Hz] and SOC range [0, 1]. The EIS magnitude and phase at 0.01 Hz exhibit the highest signal-to-noise ratio and are thus selected as features for SOC estimation. An EIS identification algorithm is then developed and validated to reconstruct EIS at 0.01 Hz. This method utilizes Fourier series expansion to approximate the voltage response to small sine-wave current perturbations. SOC estimation is subsequently performed by mapping the reconstructed EIS to experimental EIS data. Finally, the proposed SOC estimation approach is validated using sine-wave currents of varying amplitudes (0.05A and 0.1A) and different cell operation modes (discharge and charge). The results demonstrate rapid and accurate initialization of LFP cell SOC using this estimation algorithm.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 2","pages":"Article 100386"},"PeriodicalIF":16.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145842474","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Feature selection-based irradiance forecast for efficient operation of a stand-alone PV system","authors":"Vijay Muniyandi ,&nbsp;V Kumar Reddy Majji ,&nbsp;Manam Ravindra ,&nbsp;Ramesh Adireddy ,&nbsp;Ashok Kumar Balasubramanian","doi":"10.1016/j.geits.2025.100308","DOIUrl":"10.1016/j.geits.2025.100308","url":null,"abstract":"<div><div>Solar irradiance (SI) forecasting and determination of optimum tilt angle (OTA) of photovoltaic (PV) panels are the key strategies for improving the power output of PV systems. Precise SI forecasting offer valuable information regarding the predictable accessibility of solar energy, empowering PV system operators to make informed decisions for PV system optimization. This research uses a bi-directional long short-term memory (Bi-LSTM) hybrid network to forecast SI. Then, the OTA of the PV module is estimated by applying the forecasted SI data to the ASHRAE (American Society of Heating, Refrigerating and Air-conditioning Engineers) SI model. The performance of the Bi-LSTM hybrid network to estimate SI is compared with the observed data and the other existing forecasting models in the literature. The impact of OTA in improving PV power output is evaluated by comparing the solar irradiance received on both tilted and horizontal surfaces. This work has been experimentally implemented using the PV module setup at Thiagarajar College of Engineering, Madurai, Tamil Nadu, India. The OTA obtained by the proposed method yielded an increased output PV power compared to all other tilt angle approaches in the literature.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 2","pages":"Article 100308"},"PeriodicalIF":16.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145979166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Power management using an improved EMS algorithm in a stand-alone hybrid PV-PEMFC microgrid with reduced converter count","authors":"Kalpana Bijayeeni Samal,&nbsp;Swagat Pati,&nbsp;Renu Sharma","doi":"10.1016/j.geits.2025.100302","DOIUrl":"10.1016/j.geits.2025.100302","url":null,"abstract":"<div><div>Stand-alone microgrids (SAMG) often encounter power quality and stability issues due to the volatile green energy sources and loads. So, the reliability of SAMG is comparably lower. To address this problem, battery energy storage systems (BESSs) are often employed. This article aims to propose an integration topology for an FC- PV- BESS to facilitate the use of smaller batteries along with a reduced count of converters. A novel energy management system (EMS) has also been developed for minimum FC involvement without compromising system reliability. The system relies on control structures and converter techniques to generate maximum power from the PV unit using a maximum power point tracking (MPPT) algorithm. The PV and FC are designed to meet load demand while supplying extra power tocharge the BESSs. The EMS functions are improved based on the day and night circumstances. In both day and night, the entire EMS function is separated into two modes: power surplus and deficiency. The two modes are then classified into six categories based on the PV-generating situation. Switching between various modes of operation is done automatically based on specified set values to achieve minimum FC involvement due to the lowest possible hydrogen consumption, while simultaneously maintaining the battery state of charge (SoC) within prescribed limits. The proposed system is developed with a minimum number of converter reducing system complexity and improving performance. The EMS proposed in this work ensures minimum FC involvement and manages the system power to operate the FC with an efficiency range of 40%–60%. The system performance is validated on the real-time platform OPAL-RT 4510 under each mode to prove the efficacy of the proposed EMS.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 2","pages":"Article 100302"},"PeriodicalIF":16.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145927407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized longitudinal motion planning based on a combination of reinforcement learning and imitation learning","authors":"Chongpu Chen ,&nbsp;Xinbo Chen ,&nbsp;Peng Hang","doi":"10.1016/j.geits.2025.100321","DOIUrl":"10.1016/j.geits.2025.100321","url":null,"abstract":"<div><div>With advancements in autonomous driving technology, to minimize the decision-making disparities between human drivers and intelligent vehicles, the need for anthropomorphism and personalization in intelligent vehicles has become increasingly pressing. In planning longitudinal motion of intelligent vehicles, it is essential to consider multiple performance metrics as well as the driver's acceptance of the vehicle's driving style. This paper introduces a longitudinal motion planning policy that synergistically combines reinforcement learning with imitation learning. The primary framework is built on reinforcement learning, creating a foundational policy for longitudinal motion planning. Within this reinforcement learning context, this study incorporates a classic trajectory prediction method to construct an environment with prediction and deduction model (EPD). Generative Adversarial Imitation Learning (GAIL), a well-established imitation learning technique, is employed to assimilate human driver demonstration data into the reinforcement learning framework. The Deep Deterministic Policy Gradient (DDPG) algorithm, integrated with the EPD and GAIL models, is used to formulate a comprehensive personalized longitudinal motion planning policy. This policy is rigorously trained and tested on a natural driving dataset. The findings confirm that the proposed policy can adapt to the driving style of each target driver, achieving personalized driving while simultaneously meeting stringent performance indices in longitudinal motion planning compared to human drivers.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 2","pages":"Article 100321"},"PeriodicalIF":16.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Challenges and prospects in real-world battery status prediction within Industry 4.0","authors":"Xudong Qu ,&nbsp;Jingyuan Zhao ,&nbsp;Hui Pang ,&nbsp;Michael Fowler ,&nbsp;Andrew F. Burke","doi":"10.1016/j.geits.2025.100298","DOIUrl":"10.1016/j.geits.2025.100298","url":null,"abstract":"<div><div>The performance of lithium-ion batteries is critical across a range of applications, including portable devices, electric vehicles, and energy storage systems. Effective diagnostics of these battery systems require evaluating multiple factors such as charge, health, lifespan, and safety. Diagnosing batteries under real-world conditions presents notable challenges, particularly due to dynamic operating environments, inconsistent data quality, and cell-to-cell variations. These challenges complicate diagnostics further when considering the need for model integration, scalability, and managing computational costs. Industry 4.0 introduces new opportunities for intelligent, real-time battery performance evaluation, but also brings its own complexities. This review examines several real-world battery diagnostic scenarios, identifying key obstacles. We provide an in-depth analysis of the integration of intelligent diagnostic technologies in Industry 4.0, with a focus on IoT connectivity, machine learning techniques, and big data analytics. Moreover, we outline promising research directions, such as fostering interdisciplinary collaboration, improving data and model integration, utilizing diverse data patterns, and strengthening partnerships between academia and industry. Cloud-based AI solutions not only enhance diagnostics related to battery lifespan and safety but also align with the Industry 4.0 framework by facilitating automated decision-making and resource management. This review highlights recent advancements and identifies critical challenges that require further exploration. It aims to support sustainable industrial practices and drive the adoption of green technologies within smart, digital and sustainable environments. It aims to promote intelligent industrial practices and accelerate the adoption of battery technologies within smart, digital, and eco-friendly environments.</div></div>","PeriodicalId":100596,"journal":{"name":"Green Energy and Intelligent Transportation","volume":"5 2","pages":"Article 100298"},"PeriodicalIF":16.4,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145886232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}