Power Capability Research Articles

Adaptive search for broad attention based vision transformers

Vision Transformer (ViT) has prevailed among computer vision tasks for its powerful capability of image representation recently. Frustratingly, the manual design of efficient architectures for ViTs can be laborious, often involving repetitive trial and error processes. Furthermore, the exploration of lightweight ViTs remains limited, resulting in inferior performance compared to convolutional neural networks. To tackle these challenges, we propose Adaptive Search for Broad attention based Vision Transformers, called ASB, which automates the design of efficient ViT architectures by utilizing the broad search space and an adaptive evolutionary algorithm. The broad search space facilitates the exploration of a novel connection paradigm, enabling more comprehensive integration of attention information to improve ViT performance. Additionally, an efficient adaptive evolutionary algorithm is developed to efficiently explore architectures by dynamically learning the probability distribution of candidate operators. Our experimental results demonstrate that the adaptive evolution in ASB efficiently learns excellent lightweight models, achieving a 55% improvement in convergence speed over traditional evolutionary algorithms. Moreover, the effectiveness of ASB is validated across several visual tasks. For instance, on ImageNet classification, the searched model attains a performance of 77.8% with 6.5M parameters and outperforms state-of-the-art models, including EfficientNet and EfficientViT networks. On mobile COCO panoptic segmentation, our approach delivers 43.7% PQ. On mobile ADE20K semantic segmentation, our method attains 40.9% mIoU. The code and pre-trained models will be available soon in ASB-Code.

Flexibility potential quantification of electric vehicle charging clusters

A significant obstacle to providing flexibility services with electric vehicles (EVs) is the uncertainty surrounding the profitability and flexibility potential of charging clusters when utilised as a flexible load. Currently, there is a lack of comprehensive and easily applicable methods for quantifying flexibility in the literature. This paper introduces an evaluation tool and a set of flexibility indexes to assess the capability of charging clusters to deliver flexibility services. The method is designed to evaluate and quantify the flexibility potential of charging clusters in terms of short-term and long-term power adjustments and charge scheduling. Through sensitivity analysis, we examine how connection capacity, EV battery capacities, power capabilities, and the number of daily charging sessions influence the flexibility potential of charging clusters. Our findings highlight a direct relationship between the grid connection capacity of clusters and their ability to perform short-term power adjustments. Moreover, while larger batteries tend to reduce energy and time flexibility, their increased storage capability facilitates managing and scheduling a larger energy volume. Furthermore, for the days analysed, the flexibility potential showed minimal sensitivity to the number of daily charging sessions. Instead, the amount of energy requested and connection patterns emerge as key determinants of overall flexibility. In summary, this research provides valuable insights that can inform the design, monitoring, and assessment of EV charging clusters when evaluating their suitability for various flexibility services.

CLlight: Enhancing representation of multi-agent reinforcement learning with contrastive learning for cooperative traffic signal control

Multi-agent reinforcement learning has shown great potential for coordinating multi-intersection traffic signals due to its powerful adaptive capabilities, treating each intersection as an agent. However, in the real world, different intersections possess differentiating characteristics such as unique vehicle distributions and traffic patterns. Most existing methods directly add neighboring intersection states to local intersections and optimize the cooperative policy network based on synthesized global features. This indirect optimization approach makes it difficult to thoroughly explore the mutual interactions among different intersection agents, preventing agents from truly learning features with cooperative awareness. To resolve these challenges, we introduce contrastive learning as representation task to the multi-intersection traffic signal control approach named CLlight for two-stage policy network updating. In the first stage, we utilize policy-based or actor-critic-based reinforcement learning methods such as A2C, SAC, and PPO to train policy networks with certain representational capabilities. In the second stage, by extracting pre- and post-masked features and reconstructing the post-masked features, the agents are encouraged to learn the similarities and differences between different intersection policies, which in turn enhances the cooperative and individual representation capabilities of the policy network. To the best of our knowledge, this is the first application of contrastive learning in the field of traffic signal control. Experimental results demonstrate, compared to other state-of-the-art traffic signal control methods, superior average travel time and average waiting time performance under various scenarios, tested on synthetic and real-world datasets.

Identification of anoikis-related long non-coding RNA signature as a novel prognostic model in lung adenocarcinoma.

Anoikis, as a specific form of programmed cell death, involves in tumor metastasis. However, there is still lacking of anoikis-related long non-coding RNA (lncRNA) risk signature in the diagnosis and prognosis of lung adenocarcinoma (LUAD). This study constructed a prognostic risk model by comprehensively analyzing anoikis-related lncRNAs which could effectively diagnose and predict the outcomes of LUAD patients. A list of anoikis-related genes (ARGs) was retrieved from literatures. Anoikis-related lncRNAs were selected using co-expression analysis from The Cancer Genome Atlas (TCGA) database. Univariate and multivariate regression analyses were used to construct a prognostic model. The performance of the risk signature in predicting the prognosis and clinical significance were determined by Kaplan-Meier survival analysis, receiver operating characteristic (ROC) curves, univariate and multivariate regression analyses. Moreover, the differences of tumor immune microenvironment between the high- and low-risk groups were explored. Finally, a novel nomogram was developed by combining the signature and clinicopathological factors, and the association between lncRNAs and differential N6-methyladenosine (m6A) genes was analyzed by Spearman's analysis. A total of 1,694 anoikis-related lncRNAs were identified from 479 cases of LUAD. According to the univariate and multivariate Cox analyses, we established a prognostic risk model consisting of seven lncRNAs (AC026355.2, AL606489.1, AL031667.3, LINC02802, LINC01116, AC018529.1, and AP000844.2). This prognostic risk model could efficiently classify low- and high-risk patients. The area under the curve (AUC) value was 0.717, which indicated more powerful predictive capability than commonly used clinicopathological factors. The high- and low-risk groups demonstrated different immune microenvironment. Moreover, the nomogram also demonstrated good performance in predicting the prognosis. Twelve differential m6A regulators were identified, and RBM15 was found to be correlated positively with the hub lncRNA AL606489.1. Our study constructed a prognostic risk model based on anoikis-related lncRNAs, which could provide novel perspective on the prognosis of LUAD patients.

Retinal fundus image super-resolution based on generative adversarial network guided with vascular structure prior

Many ophthalmic and systemic diseases can be screened by analyzing retinal fundus images. The clarity and resolution of retinal fundus images directly determine the effectiveness of clinical diagnosis. Deep learning methods based on generative adversarial networks are used in various research fields due to their powerful generative capabilities, especially image super-resolution. Although Real-ESRGAN is a recently proposed method that excels in processing real-world degraded images, it suffers from structural distortions when super-resolving retinal fundus images are rich in structural information. To address this shortcoming, we first process the input image using a pre-trained U-Net model to obtain a structural segmentation map of the retinal vessels and use the segmentation map as the structural prior. The spatial feature transform layer is then used to better integrate the structural prior into the generation process of the generator. In addition, we introduce channel and spatial attention modules into the skip connections of the discriminator to emphasize meaningful features and accordingly enhance the discriminative power of the discriminator. Based on the original loss functions, we introduce the L1 loss function to measure the pixel-level differences between the segmentation maps of retinal vascular structures in the high-resolution images and the super-resolution images to further constrain the super-resolution images. Simulation results on retinal image datasets show that our improved algorithm results have a better visual performance by suppressing structural distortions in the super-resolution images.

Deciphering laser shock peening quality monitoring: Wavelet-driven network with interpretability

Quality monitoring of laser shock peening based on acoustic emission technology is a topical multidisciplinary issue that has received much attention in recent years. To acquire complex time-varying information in acoustic emission signals, convolutional neural networks with powerful learning capabilities have shown potential for a wide range of applications. However, the black-box property of the network imposes limitations on its further development and decision credibility. Therefore, this study proposes a wavelet-driven network with theoretical basis and physical significance. This network can cyclically utilize discrete wavelet packet transform to map input features to the wavelet domain during the learning process, thereby obtaining more robust and valuable information. This paper also constructs a novel wavelet attention mechanism that takes into account the difference between low-frequency and high-frequency information, and is able to allocate resources in both the decomposition component and the time-domain dimension. The proposed method can be seen as a multiresolution analysis technique that combines existing physical knowledge with nonlinear feature processing and feature selective enhancement. The results of the two laser shock peening cases show that the proposed method not only outperforms current state-of-the-art models in terms of monitoring performance, but also has better physical interpretability. Importantly, the proposed method has the potential to be further extended to other interpretable structural health monitoring.

One-step pyrolysis derived Fe and heteroatom co-doped biochar composites for efficient microwave absorption

Electromagnetic (EM) pollution frequently disrupts the regular operation of sophisticated electrical devices, necessitating the urgent development of lightweight EM wave absorbers that possess powerful absorption capability. Herein, we report a simple method for the preparation of Fe and heteroatom co-doped biochar composites (FeX-BC, where X = N, S) by directly carbonizing the precursors of anhydrous FeCl3, cherry kernel powder, and heteroatom dopants (melamine or Na2S2O3·5H2O). By adding different amounts of dopants during the synthesis of the precursors, it is possible to adjust the heteroatom content of the FeX-BC samples with precision. It is worth mentioning that the FeN0.1-BC composite delivers the best EM wave absorption performance with an effective absorption bandwidth of 4.8 GHz and a minimal reflection loss of −63.2 dB. Furthermore, the significant attenuation of EM wave can be attributed to the synergistic interplay of magnetic loss, dielectric loss and the enhanced impedance matching. This study introduces a simple methodology for the fabrication of EM wave absorbers, a significant contribution to the synthesis, advancement, and functional applications of biomass-derived materials.

Research on the Recognition and Characterization of Fractures in Coal Rock Based on CT Scanning

Abstract Coal rock fractures play a vital role as key channels for gas migration and storage in the exploration and development of coalbed methane. The characteristics of these fractures, such as porosity distribution, specific surface area, and connectivity, directly impact the adsorption, diffusion, and permeability behaviors of coal reservoirs. Therefore, a detailed analysis of coal rock fracture characteristics is vital in assessing the extractability of coalbed methane and during its exploration and development process. Currently, techniques like imaging logging, seismic detection, and micro-CT are employed for characterizing and studying fractures at various scales. Among these, high-resolution non-destructive CT scanning technology, by constructing three-dimensional data models of core samples, enables a quantitative analysis of fracture distribution features and size parameters. Traditional digital image processing methods, such as threshold segmentation and binary segmentation, though advantageous in computational speed, have limitations in the accuracy of fracture identification. With the advancement of artificial intelligence technology, deep learning-based image processing techniques have increasingly become significant tools for fracture detection due to their powerful feature extraction capabilities. This study conducted micrometer-level CT scanning experiments on coal rock samples, using CT image processing technology to construct a digitalized coal rock core model. Deep learning methods were applied for precise fracture identification, establishing an appropriate deep learning architecture and optimized parameters for coal rock fracture recognition. Furthermore, this paper provides a detailed quantitative characterization of the precisely identified fractures, offering robust technical support for the exploration and development of coalbed methane. The high-resolution non-destructive CT scanning technology and three-dimensional fracture network reconstruction method used in this study have been proven to be effective tools for investigating the micro-features of rocks, enabling the three-dimensional visualization and detailed characterization of rock microstructures.

Exploiting the determinant factors on the available flexibility area of ADNs at TSO‐DSO interface

AbstractActive distribution networks (ADNs) are consistently being developed as a result of increasing penetration of distributed energy resources (DERs) and energy transition from fossil‐fuel‐based to zero carbon era. This penetration poses technical challenges for the operation of both transmission and distribution networks. The determination of the active/reactive power capability of ADNs will provide useful information at the transmission and distribution systems interface. For instance, the transmission system operator (TSO) can benefit from reactive power and reserve services which are readily available by the DERs embedded within the downstream ADNs, which are managed by the distribution system operator (DSO). This article investigates the important factors affecting the active/reactive power flexibility area of ADNs such as the joint active and reactive power dispatch of DERs, dependency of the ADN's load to voltage, parallel distribution networks, and upstream network parameters. A two‐step optimization model is developed which can capture the P/Q flexibility area, by considering the above factors and grid technical constraints such as its detailed power flow model. The numerical results from the IEEE 69‐bus standard distribution feeder underscore the critical importance of considering various factors to characterize the ADN's P/Q flexibility area. Ignoring these factors can significantly impact the shape and size of Active Distribution Networks (ADN) P/Q flexibility maps. Specifically, the Constant Power load model exhibits the smallest flexibility area; connecting to a weak upstream network diminishes P/Q flexibility, and reactive power redispatch improves active power flexibility margins. Furthermore, the collaborative support of reactive power from a neighboring distribution feeder, connected in parallel with the studied ADN, expands the achievable P/Q flexibility. These observations highlight the significance of accurately characterizing transmission and distribution network parameters. Such precision is fundamental for ensuring a smooth energy transition and successful integration of hybrid renewable energy technologies into ADNs.

ETFE: Efficient transformer with feature enhancement for semantic segmentation

Abstract Despite the powerful capabilities of Transformers in relation to modeling, their effectiveness in capturing multi-scale object features in image-based tasks is often limited. This limitation arises from the diverse shapes and distances between objects in natural scenes, which pose challenges for Transformers to accurately capture object sizes and distances. To overcome this problem, we propose an effective Transformer which extracts multi-scale features at different resolutions. First, we leverage a multi-head attention mechanism to extract richer semantic features at various resolutions. The semantic features are then passed through the Feature Gate to aggregate multi-scale semantic features. Finally, Attentional Feature Enhancement further strengthens the semantic information of target features, enabling the modeling of dependencies among multi-scale features. Comprehensive experiments conducted on the Cityscapes and ADE20K datasets have shown that our model outperforms competing methods in both performance and accuracy.

A bi-level inter-phase coordinated control method for voltage regulation in unbalanced LV distribution networks using PV-battery energy storage systems

Photovoltaic systems coupled with battery energy storage (PV-BES) can help to minimize the effects of variability in PV generation including voltage problems in low voltage distribution grids (LVDNs). Therefore, a joint centralized-decentralized method consisting of two control levels based on sensitivity analysis is proposed in this paper, which employs the BES charging and discharging control in the first level and reactive power capability of inverters in the second control level. When the BES charging and discharging control is insufficient for voltage regulation, the second level of the proposed method is triggered. For full utilization of the BES capacity and to ensure adequate capacity for the following day's voltage regulation needs, an enhancement algorithm is applied to the voltage to active power sensitivity coefficients. All of the voltage sensitivity coefficients are calculated by a central controller using an offline procedure and sent to the PV-BES inverters to determine the amount of active and reactive power in real-time according to the measured local voltage. Taking into account the inter-phase voltage effects, a request for Var compensation signal flow with inter-phase coordination constraints is utilized to properly coordinate the inverters in reactive power control. The proposed method can efficiently regulate voltage by preventing over and under-voltage problems, improving the average total voltage deviation (TVD), reducing the total reactive power absorption and injection by the inverters, and avoiding reduction in PV generation. To evaluate the performance of the proposed control method, it is compared with the commercially available inverter voltage control techniques in an unbalanced residential test feeder. Moreover, a comparison is conducted between the proposed method and an adaptive decentralized control approach presented in the literature. Furthermore, a three-phase Volt-Var strategy, with and without the inter-phase coordination constraints, is simulated that indicates the necessity of the constraints for proper coordination.

Clustering of Global Magnetospheric Observations

AbstractThe use of supervised methods in space science have demonstrated powerful capability in classification tasks, but purely unsupervised methods have been less utilized for the classification of spacecraft observations. We use a combination of unsupervised methods, being principal component analysis, Self‐Organizing Maps, and hierarchical agglomerative clustering, to classify THEMIS and MMS observations as having occurred in the magnetosphere, magnetosheath, or the solar wind. The resulting classification are validated visually by analyzing the distribution of classifications and studying individual time series as well as by comparison to the labeled data set of a previous model, against which ours has an accuracy of 99.4. The model has a variety of applications beyond region classification such as deeper hierarchical analysis, magnetopause and bow shock crossing identification, and identification of bursty bulk flows, hot flow anomalies, and foreshock bubbles.

A Novel Composite Graph Neural Network.

Graph neural networks (GNNs) have achieved great success in many fields due to their powerful capabilities of processing graph-structured data. However, most GNNs can only be applied to scenarios where graphs are known, but real-world data are often noisy or even do not have available graph structures. Recently, graph learning has attracted increasing attention in dealing with these problems. In this article, we develop a novel approach to improving the robustness of the GNNs, called composite GNN. Different from existing methods, our method uses composite graphs (C-graphs) to characterize both sample and feature relations. The C-graph is a unified graph that unifies these two kinds of relations, where edges between samples represent sample similarities, and each sample has a tree-based feature graph to model feature importance and combination preference. By jointly learning multiaspect C-graphs and neural network parameters, our method improves the performance of semisupervised node classification and ensures robustness. We conduct a series of experiments to evaluate the performance of our method and the variants of our method that only learn sample relations or feature relations. Extensive experimental results on nine benchmark datasets demonstrate that our proposed method achieves the best performance on almost all the datasets and is robust to feature noises.

Research on risk management of ship maintenance projects based on multi agent swarm model simulation method

In recent years, the role of naval ship power has become increasingly prominent in regional conflicts and military operations worldwide, and the powerful capabilities demonstrated by naval forces have also been valued by countries around the world. While naval forces are frequently deployed to carry out various tasks, the intensity of equipment wear and tear on ships is also significantly increasing, posing a huge challenge to the maintenance and support capabilities of ship equipment. In the actual process of ship equipment maintenance, a series of risk factors such as cost, schedule, and quality will have a significant impact on the entire ship maintenance project, and even affect the success or failure of the entire project. This article first decomposes the project process of ship maintenance projects, further analyzes the various risk factors of ship maintenance projects based on the project decomposition situation, and then distinguishes the overall project, project process, and project risk factors to establish an intelligent agent group model for simulation analysis. Through the above simulation analysis, we will conduct in-depth research on the specific impact of project risks on ship maintenance projects, and analyze and verify the positive role and practical significance of project risk management in project risk control and project efficiency improvement. In the process of the above research content, methods and measures for risk management of ship maintenance projects are proposed, providing scientific and efficient risk management suggestions for such projects, improving the scientific management efficiency of ship maintenance projects, and contributing to the improvement of the quality and efficiency of project management.

Data Commons Platforms: Addressing Challenges and Opportunities in Large-Scale Biomedical Data Integration and Analysis

Abstract: Data Commons Platforms are emerging as transformative tools in the landscape of medical research, offering a centralized, accessible, and scalable infrastructure for managing, analyzing, and sharing vast amounts of biomedical data. This article provides a comprehensive examination of these platforms, exploring their core functionalities, benefits, and challenges in the context of advancing medical discovery. We discuss how Data Commons Platforms accelerate research by facilitating datadriven hypothesis generation, enhancing collaboration among researchers, improving data accessibility, and ensuring data interoperability. The article also presents a reference architecture for these platforms, detailing the key components that enable their powerful capabilities. Furthermore, we address critical challenges such as data privacy, quality standardization, and ethical considerations, while also highlighting opportunities for future development. Through case studies and an analysis of emerging trends, this work demonstrates the potential of Data Commons Platforms to revolutionize biomedical research, paving the way for more efficient, collaborative, and impactful scientific discoveries in the era of big data and precision medicine.

Neural Memory State Space Models for Medical Image Segmentation.

With the rapid advancement of deep learning, computer-aided diagnosis and treatment have become crucial in medicine. UNet is a widely used architecture for medical image segmentation, and various methods for improving UNet have been extensively explored. One popular approach is incorporating transformers, though their quadratic computational complexity poses challenges. Recently, State-Space Models (SSMs), exemplified by Mamba, have gained significant attention as a promising alternative due to their linear computational complexity. Another approach, neural memory Ordinary Differential Equations (nmODEs), exhibits similar principles and achieves good results. In this paper, we explore the respective strengths and weaknesses of nmODEs and SSMs and propose a novel architecture, the nmSSM decoder, which combines the advantages of both approaches. This architecture possesses powerful nonlinear representation capabilities while retaining the ability to preserve input and process global information. We construct nmSSM-UNet using the nmSSM decoder and conduct comprehensive experiments on the PH2, ISIC2018, and BU-COCO datasets to validate its effectiveness in medical image segmentation. The results demonstrate the promising application value of nmSSM-UNet. Additionally, we conducted ablation experiments to verify the effectiveness of our proposed improvements on SSMs and nmODEs.

Regulated charge transfer by cascade dual Z-scheme heterojunction of HCdS@ZnIn2S4-Cobalt porphyrin for efficient photocatalytic solar fuel production

Dual Z-schemed photocatalysts possess superior charge-separation efficiency and powerful redox capabilities, which have attracted significant attention in photocatalytic CO2 reduction (PCR). However, most of the dual Z-scheme systems are focused on inorganic semiconductors, which remain a challenge to control the selectivity of PCR products. Herein, we designed a cascade dual Z-scheme structured organic/inorganic heterojunction of HCdS@ZnIn2S4-Cobalt porphyrin (HCdS@ZIS-CoTPPS). HCdS@ZIS-CoTPPS demonstrated excellent PCR activity (CO rate: 93.64 μmol·g−1·h−1) and adjustable syngas ratio when compared to HCdS, ZIS, CoTPPS, and HCdS/ZIS. Experimental characterizations and density functional theory (DFT) calculations reveal that the superior PCR activity of HCdS@ZIS-CoTPPS is contributed by the broad light absorption range, short charge carrier migration path, low interface contact resistance, and strong redox ability induced by the hollow structured cascade dual Z-scheme heterojunction. This article provides valuable inspiration for the design and preparation of atomistic heterojunctions for solar-driven fuel generation.

Fabrication Time Diagrams for In-Space Manufacturing of Large Reticulated Structures

Abstract In-space manufacturing (ISM), or the on-orbit construction of structures from raw feedstock materials, is a promising approach for building large structures for future space missions. Most research and development on ISM to date has focused on ground-based or microgravity-based demonstrations of candidate fabrication processes; however, the combined design of the fabrication process and the ISM spacecraft has not been fully investigated. In this paper, we estimate the fabrication times for ISM of reticulated support structures subject to various spacecraft constraints, including the available fabrication power, the available attitude control authority, and the avoidance of undesirable control-structure interactions. We summarize the insights in fabrication time diagrams which depict the dominant constraints and estimates of the fabrication time for a wide range of structural geometries. Our results indicate that for large, dense structural geometries such as the curved gridshell and tetrahedral truss, the angular momentum storage of the ISM spacecraft's attitude control system (ACS) is the dominant constraint on fabrication time. Additionally, our results suggest the following strategies for reducing fabrication time: manufacturing with multiple spacecraft; using stiff, lightweight feedstock; maximizing fabrication power and ACS capability; and minimizing spacecraft bus mass. Altogether, our results emphasize how the design of an ISM spacecraft cannot be considered independently of the fabricated structure and motivate future work on selecting optimal fabrication paths and attitude control strategies to fabricate large structures in the shortest time.

Examining the electronic, structural, optical, and photovoltaic capabilities of Sr3AsBr3 perovskite under tensile and compressive strains with DFT and SCAPS-1D

A significant amount of interest in the field of solar energy has recently been generated by the remarkable optical, structural, and electronic attributes of inorganic perovskite-based substances. A thorough investigation was conducted on how tensile along with compressive strains influence the electronic and optical properties of halide perovskite Sr3AsBr3, utilizing FP-DFT. The PBE functional is used to determine the direct bandgap of 1.512 eV for unstrained Sr3AsBr3 at the Γ position. The bandgap redshifts (1.216 eV at -4 % strain) under compressive strain and slightly increases (1.728 eV at +4 % strain) under tensile tension, causing the absorption coefficient to blueshift. The optical properties, including the functions of dielectric, electron energy loss function, and absorption coefficient, all indicate a considerable capacity for absorption in the area of the visible spectrum, and the band characteristics of this component agree with these qualities. We used Sr3AsBr3 absorbers and CdS electron transport layers (ETL) with different thicknesses, defect densities, and doping concentrations to study solar power capabilities using the SCAPS-1D simulator. The most effective arrangement (at -4 % strain) had a fill factor (FF) of 86.76 %, a short-circuit current density (JSC) of 37.5593 mAcm−2, an open-circuit voltage (VOC) of 0.8712 V, and a power conversion efficiency (PCE) of 28.39 %. Our findings support additional research into Sr3AsBr3, a promising perovskite for optoelectronic uses.

Electrochemical Performance of Coal-Based Needle Coke Anode for Lithium Ion Batteries

Lithium-ion capacitors (LICs) with the capability of high energy and high power are considered to be attractive for advanced energy storage applications. However, the design and fabrication of suitable electrode materials with desirab-le properties by a facile approach using cost-effective precursors are still a great challenge. In this work, we have utilized needle cokes, an commercial carbon material with high carbon content and soft carbon structure, as a single carbon source for anode material. A lithium-ion battery fabricated using four kinds of needle cokes exhibits a maximum high energy density of 1357mAhg-1. Systematic characterization analysis demonstrates that unique characteristics of the green coke including large interlayer spacing, turbostratic and disordered microstructures that are composed of surface defects, nanopores or voids, and graphitic domains. That contribute synergistically to the outstanding performance of the needle coke-based LIC. More importantly, anode materials from a single source is an effective way for high value-added utilization of needle coke at the commercial level.