Hierarchical Architecture Research Articles

Hierarchical structure of poly(ε-caprolactone) crystals on its copolymer poly(l-lactide-co-ε-caprolactone) fibers

In crystallization, heterogeneous structures of copolymers are employed to establish a hierarchical architecture comprising two polymers, this typically occurs on the surface of block copolymers or blended polymers. While hierarchical structures in block copolymer fibers are well-established and widely utilized, achieving a regular hierarchical structure on random copolymer fibers is both meaningful and challenging. Here, we present a case of forming a hierarchical structure on the surface of the random copolymer poly(L-lactide-co-caprolactone) (PLCL)fibers. The (ε-caprolactone)(CL) polymer chain segments of the random copolymer are released on the surface of the modified copolymer fiber using sodium hydroxide. Subsequently, poly(ε-caprolactone)(PCL) forms a shish kebab-like crystalline structure under the guidance of the CL segments. The CL segments act as crystalline precursors to induce the formation of heterogeneous PLCL/PCL shish kebab structures. The variation of CL segments on the fiber changes with modification time, leading to the formation of crystalline lamellae of different sizes, which had been confirmed by characterization techniques such as scanning electron microscope (SEM), 1H NMR, X-ray photoelectron spectra (XPS), differential scanning calorimetry (DSC), etc. The implementation of this heterogeneous structure provides possibilities for the implementation of hierarchical structure in random copolymers.

All-Scale Hierarchical Structuring, Optimized Carrier Concentration, and Band Manipulation Lead to Ultra-High Thermoelectric Performance in Eco-Friendly MnTe.

MnTe emerges as an enormous potential for medium-temperature thermoelectric applications due to its lead-free nature, high content of Mn in the earth's crust, and superior mechanical properties. Here, it is demonstrate that multiple valence band convergence can be realized through Pb and Ag incorporations, producing large Seebeck coefficient. Furthermore, the carrier concentration can be obviously enhance by Pb and Ag codoping, contributing to significant enhancement of power factor. Moreover, microstructural characterizations reveal that PbTe nanorods can be introduced into MnTe matrix by alloying Pb. This can modify the microstructure into all-scale hierarchical architectures (including PbTe nanorods, enhances point-defect scattering, dense dislocations and stacking faults), strongly lowering lattice thermal conductivity to a record low value of 0.376W m-1 K-1 in MnTe system. As a result, an ultra-high ZT of 1.5 can be achieved in MnTe thermoelectric through all-scale hierarchical structuring, optimized carrier concentration, and valence band convergence, outperforming most of MnTe-based thermoelectric materials.

A nacre-inspired thermo conductive and healable nanocomposite captures extremely enhanced stiffness and toughness

Polymer-based nanocomposites that exhibit exceptional mechanical properties, rapid self-healing capability, high thermal conductivity, and superior electrical insulation are highly sought after for thermal management applications in modern electrical systems and electronic devices, owing to their versatility, ease of processing, and cost-effectiveness. However, traditional methods for enhancing thermal conductivity often compromise self-healing ability and mechanical and electrical properties. In this study, inspired by nacre, we propose an innovative approach to fabricating nanocomposites with a hierarchical architecture by incorporating boron nitride nanosheets (BNNSs) into a polyurethane matrix through a bottom-up assembly process and lamination technology. Leveraging the nacre-like layered structure and strong interfacial hydrogen bonding interactions, the nanocomposites were successfully transformed from exhibiting brittle fracture to displaying ductile fracture behavior, effectively reconciling the contradiction of high stiffness and toughness. The unique design of the nanocomposite leads to simultaneous enhancement in stiffness (5.3 times), strength (20.1 times), toughness (337.4 times), and fracture toughness (16.4 times) compared to those of the polyurethane matrix. Additionally, the resulting nanocomposites demonstrate excellent self-healing efficiency (∼98 %), high thermal conductivity (8.1 ± 0.3 W m−1 K−1) and superior electrical insulation property (>1012 Ω cm). Overall, our study provides a promising avenue for fabricating polymer-based nanocomposites for thermal management applications, which exhibit multifunctional properties without compromising on mechanical strength, electrical insulation, and self-healing capabilities.

Graph Pooling Inference Network for Text-based VQA

Effectively leveraging objects and optical character recognition (OCR) tokens to reason out pivotal scene text is critical for the challenging Text-based Visual Question Answering (TextVQA) task. Graph-based models can effectively capture the semantic relationship among visual entities (objects and tokens) and report remarkable performance in TextVQA. However, previous efforts usually leverage all visual entities and ignore the negative effect of superfluous entities. This article presents a Graph Pooling Inference Network (GPIN), which is an evolutionary graph learning method to purify the visual entities and capture the core semantics. It is observed that the dense distribution of reduplicative objects and the crowd of semantically dependent OCR tokens usually co-exist in the image. Motivated by this, GPIN adopts an adaptive node dropping strategy to dynamically downscale semantically closed nodes for graph evolution and update. To deepen the comprehension of scene text, GPIN is a dual-path hierarchical graph architecture that progressively aggregates the evolved object graph and the evolved token graph semantics into a graph vector that serves as visual cues to facilitate the answer reasoning. It can effectively eliminate object redundancy and enhance the association of semantically continuous tokens. Experiments conducted on TextVQA and ST-VQA datasets show that GPIN achieves promising performance compared with state-of-the-art methods.

Synthetic Data and Hierarchical Object Detection in Overhead Imagery

The performance of neural network models is often limited by the availability of big datasets. To treat this problem, we survey and develop novel synthetic data generation and augmentation techniques for enhancing low/zero-sample learning in satellite imagery. In addition to extending synthetic data generation approaches, we propose a hierarchical detection architecture to improve the utility of synthetic training samples. We consider existing techniques for producing synthetic imagery–3D models and neural style transfer–as well as introducing our own adversarially trained reskinning network, the GAN-Reskinner, to blend 3D models. Additionally, we test the value of synthetic data in a two-stage, hierarchical detection/classification model of our own construction. To test the effectiveness of synthetic imagery, we employ it in the training of detection models and our two stage model, and evaluate the resulting models on real satellite images. All modalities of synthetic data are tested extensively on practical, geospatial analysis problems. Our experiments show that synthetic data developed using our approach can often enhance detection performance, particularly when combined with some real training images. When the only source of data is synthetic, our GAN-Reskinner often boosts performance over conventionally rendered 3D models and in all cases, the hierarchical model outperforms the baseline end-to-end detection architecture.

Common direct current (DC) bus integration of DC fast chargers, grid‐scale energy storage, and solar photovoltaic: New York City case study

AbstractThe mass deployment of distributed energy resources (DERs) to achieve clean energy objectives has become a major goal across several states in the U.S. However, the viability and reality of achieving these goals in dense urban areas, such as New York City, are challenged by several ‘Techno‐Economic’ barriers associated with available land space and the number of AC/direct current (DC) conversion stages that requires multiple electrical balance of plant (BOP) equipment for pairing/interconnecting these resources to the grid. The fundamental issue of interconnection is addressed by assessing the use of a common DC bus in a one‐of‐a‐kind configuration (to pair grid‐connected energy storage, photovoltaic, and electric vehicle chargers (EVC) systems) and reduce the number of BOP equipment needed for deployment. Building on similar work that has touched on distribution‐level DC interconnection, this paper will also address the intricacies of interconnecting third‐party and Utility DERs to a DC‐based point of common coupling. It will examine the requisite site controller configuration (control architecture) and requirements to coordinate the energy storage system's use between managing Utility and Third‐Party EVC demand while prioritising dispatch. The result shows that the DC‐coupled system is technologically feasible and hierarchical control architecture is recommended to maintain stability during various use cases proposed. This will inform a lab demonstration of this system that aims to test DC integration of the DERs with recommendations for the microgrid (MG) controllers and reduction in the BOP equipment. These learnings will then be applied to practical grid‐scale deployment of the systems at Con Edison's Cedar Street Substation. This system, if proven successful, has the potential to change the way community distributed generation and MGs are interconnected to the Utility System.

Hierarchical nano-/micro-architecture phonon scattering of p-type Bismuth telluride bulk composites with Ag-TiO2 nano particles synthesized by fluidized bed spray coating method

We optimize the thermoelectric (TE) properties by minimizing thermal conductivity utilizing by hierarchical phonon scattering from long to mid wavelength of phonon in p-type Bi0.4Sb1.6Te3.4 (BST) composite with nano Ag-coated TiO2 (Ag/TiO2). We employed an Ag nano particle (20 nm) modulation into TiO2 (500 nm) particles by spray coating method and dispersed in a p-type Bi0.4Sb1.6Te3.4 (BST), giving rise to a hierarchical architecture from nano to micro scale. The nano particle modulation of Ag/TiO2 particles decreases grain sizes of p-type BST matrix. The hierarchical structure of modulating nano particles and small grain sizes of the matrix significantly decreases lattice thermal conductivity due to wide range of wavelength phonon scattering, resulting in the enhancement of thermoelectric performance over a large temperature range. The maximum thermoelectric figure-of-merit zT value reaches to 1.28 for 0.9 wt% Ag/TiO2 modulation doping, which is superior to the other reported ones. The optimized modulating Ag/TiO2 nano-particle dispersion effectively maintains the valence band structure while concurrently mitigating the scattering of charge carriers. Therefore, the nano particle modulation by spray coating method is an effective way to manifest hierarchical architecture from nano to micro-scale phonon scattering, which can be applied in thermoelectric materials design and process.

A 0.55-mm2 8-bit 32-GS/s TI-SAR ADC with optimized hierarchical sampling architecture

This paper analyzes the bandwidth of the time-interleaved analog-to-digital converter (TI-ADC) with hierarchical sampling and presents an 8-bit 32-GS/s TI-ADC in 28-nm CMOS. The front-end sampler is implemented by a two-stage hierarchical sampling architecture to extend the analog input bandwidth. The 32-way sub-ADCs are realized with 2b/cycle successive approximation register (SAR) architecture and non-binary search strategy, which meet the requirements of high speed, medium resolution, and strong robustness. Multi-phase low-jitter clock tree is designed based on the delay-locked loop (DLL). Calibration methods for timing skew, offset and gain mismatches are introduced to improve accuracy. The proposed TI-SAR ADC achieves an analog input bandwidth of 23.7 GHz. It occupies an active area of 0.55 mm2 and consumes 356 mW at 1.8/1 V supply.

Crushing performance of bioinspired hierarchical tapered structures

This paper introduces a new type of energy-absorbing hierarchical tapered structure, mimicking the hierarchical architecture of barnacle. The proposed structures are designed by iteratively incorporating sub-tapered tubes at the junctions of primary ribs aiming to enhance the crashworthiness performance. The finite element models of the proposed structures are constructed in Abaqus software and validated using experimental testing. The effects of the geometrical parameters including the number of substructures and the external-to-internal wall thickness ratio on the energy absorption characteristics of the proposed structures are investigated. The results demonstrate that as the number of substructures increases, the specific energy absorption (SEA) and mean crushing force of the proposed design show a significant improvement. Specially, the SEA of the proposed structures with four substructures can reach 32.78 kJ/kg, which is 85.8 % higher than the conventional tapered tube. Additionally, decreasing the ratio of external to internal wall thickness leads to enhanced performance. After optimizing the wall thickness ratio, the maximum SEA can reach 40.87 kJ/kg, which is 26.0 % higher than that before optimization. To complement the findings, a theoretical study is presented, which exhibits excellent agreement with the numerical results, further validating the effectiveness of the proposed design. This study highlights the potential of incorporating hierarchical and tapered features into tapered structures, offering promising prospects for advancements in energy absorption technology across diverse industries.

SwinHR: Hemodynamic-powered hierarchical vision transformer for breast tumor segmentation

Accurate and automated segmentation of breast tumors in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) plays a critical role in computer-aided diagnosis and treatment of breast cancer. However, this task is challenging, due to random variation in tumor sizes, shapes, appearances, and blurred boundaries of tumors caused by inherent heterogeneity of breast cancer. Moreover, the presence of ill-posed artifacts in DCE-MRI further complicate the process of tumor region annotation. To address the challenges above, we propose a scheme (named SwinHR) integrating prior DCE-MRI knowledge and temporal–spatial information of breast tumors. The prior DCE-MRI knowledge refers to hemodynamic information extracted from multiple DCE-MRI phases, which can provide pharmacokinetics information to describe metabolic changes of the tumor cells over the scanning time. The Swin Transformer with hierarchical re-parameterization large kernel architecture (H-RLK) can capture long-range dependencies within DCE-MRI while maintaining computational efficiency by a shifted window-based self-attention mechanism. The use of H-RLK can extract high-level features with a wider receptive field, which can make the model capture contextual information at different levels of abstraction. Extensive experiments are conducted in large-scale datasets to validate the effectiveness of our proposed SwinHR scheme, demonstrating its superiority over recent state-of-the-art segmentation methods. Also, a subgroup analysis split by MRI scanners, field strength, and tumor size is conducted to verify its generalization. The source code is released on (https://github.com/GDPHMediaLab/SwinHR).

In Situ Engineering of Conducting Polymer Nanocomposites at Liquid/Liquid Interfaces: A Perspective on Fundamentals to Technological Significance.

The conducting polymers have continuously been hybridized with their counterparts to overcome the intrinsic functional limitations compared to the metallic or inorganic analogs. Remarkably, the liquid/liquid interface-assisted methods represent an efficient and facile route for developing fully tunable metamaterials for various applications. The spontaneous adsorption of nanostructures at a quasi-two-dimensional interface is energetically favorable due to the reduction in interfacial tension, interfacial area, and interfacial energy (Helmholtz free energy). This Perspective highlights the fundamentals of nanostructure adsorption leading to hierarchical architecture generation at the interface from an experimentalist's point of view. Thereafter, the essential applications of the conducting polymer/nanocomposites synthesized at the interface emphasize the capability of the interface to tune functional materials. This Perspective also summarizes the future challenges and the use of the known fundamental aspects in overcoming the functional limitations of polymer/nanomaterial composites and also provides some future research directions.

Porous tremella-like NiMoP/CoP network electrodes as an efficient electrocatalyst.

The design of efficient, stable and cost-effective electrocatalysts for the hydrogen evolution reaction holds substantial significance in water electrolysis, but it remains challenging. Tremella-like nickel-molybdenum bimetal phosphide encapsulated cobalt phosphide (NiMoP/CoP) with hierarchical architectures has been effectively synthesized on nickel foam (NF) via a straightforward hydrothermal followed by low-temperature phosphating method. Based on the unique structural benefits, it significantly increases the number of redox active centers, enhances the electrical conductivity of materials, and diminishes the ion diffusion path lengths, thereby promoting efficient electrolyte penetration and reducing the inherent resistance. The as-obtained NiMoP/CoP/NF electrocatalyst exhibited remarkable catalytic activity with an ultralow overpotential of 38 mV (10 mA cm-2) and low Tafel slope of 83 mV dec-1. The straightforward synthesis process and exceptional electrocatalytic properties of NiMoP/CoP/NF demonstrate great potential for the HER to replace the precious metal catalyst.

Parallel Byzantine Consensus Based on Hierarchical Architecture and Trusted Hardware

Parallel Byzantine Consensus Based on Hierarchical Architecture and Trusted Hardware

Morphology- and defect-coordinated prominent microwave absorption, thermal exhaustion, and electrical insulation in SnO2@SnP2O7@Sn2P2O7 hierarchical architectures

A series of SnO2@SnP2O7@Sn2P2O7 hierarchical architectures synthesized via a facile hydrothermal–annealing route exhibit morphology- and defect-co-dominated microwave absorption, thermal exhaustion, and electrical insulation.

A Flexible Digital Twin Framework for ATMP Production – Towards an efficient CAR T Cell Manufacturing

This paper presents a flexible digital twin framework tailored for the biomanufacturing of advanced therapeutic medicinal products (ATMPs), particularly focusing on chimeric antigen receptor T cell (CAR T cell) therapy. CAR T cell therapies face significant challenges in the management of their personalized and complex biomanufacturing processes. To tackle these issues, we propose a novel software framework for a digital twin system aimed at digitalizing, monitoring, and managing the physical production processes. The framework has a hierarchical architecture that methodically organizes production into distinct abstraction levels: processes, tasks and skills, and microservices. This layered architecture simplifies the modeling of complex biomanufacturing production and ensures that each component can be individually updated without disrupting the overall system. Furthermore, the digital twin framework integrates both manual and automated operations within a unified system, accommodating varied requirements of diverse tasks in production. The framework’s flexibility enables easier adaptation to dynamic technological advancements and allows for swift modifications in production processes, making it a sustainable and resilient digitalization infrastructure for ATMP manufacturing.

Cold denaturation of DNA origami nanostructures.

The coupling of structural transitions to heat capacity changes leads to destabilization of macromolecules at both elevated and lowered temperatures. DNA origami not only exhibit this property but also provide a nanoscopic observable of cold denaturation processes by directing intramolecular strain to the most sensitive elements within their hierarchical architecture.

Engineering of a hierarchical S-scheme 2D/3D heterojunction with graphdiyne (g-CnH2n−2) coated 3D porous CoAl2O4 nanoflowers for highly efficient photocatalytic H2 evolution

In this work, a novel 2D/3D hierarchical architecture graphdiyne/CoAl2O4 (GCA) S-scheme heterojunction was successfully constructed by coupling graphdiyne (GDY) nanosheets onto porous CoAl2O4 nanoflowers.

An Efficient Matching Game Approach to Association Formation in UAV-Enabled Hierarchical Distributed Learning.

Distributed machine learning has emerged as a promising data processing technology for next-generation communication systems. It leverages the computational capabilities of local nodes to efficiently handle large datasets, creating highly accurate data-driven models for analysis and prediction purposes. However, the performance of distributed machine learning can be significantly hampered by communication bottlenecks and node dropouts. In this article, a novel unmanned aerial vehicle (UAV)-enabled hierarchical distributed learning architecture is proposed to support machine learning applications, e.g., regional monitoring. Multiple UAV receivers (URs) are introduced as wireless relays to improve the communication between the UAV transmitters (UTs) and the cloud server. Our objective is to identify the optimal UT-UR association to maximize the social welfare of the network, which is distinctly different from the existing works that focus on the unilateral profit-maximizing problem. We formulate a two-side many-to-one matching game to model the UT-UR association problem, and a two-phase many-to-one matching algorithm is designed to identify the stable matching. The validity of our proposed scheme is verified through in-depth numerical simulations.

Full Point Encoding for Local Feature Aggregation in 3-D Point Clouds.

Point cloud processing methods exploit local point features and global context through aggregation which does not explicitly model the internal correlations between local and global features. To address this problem, we propose full point encoding which is applicable to convolution and transformer architectures. Specifically, we propose full point convolution (FuPConv) and full point transformer (FPTransformer) architectures. The key idea is to adaptively learn the weights from local and global geometric connections, where the connections are established through local and global correlation functions, respectively. FuPConv and FPTransformer simultaneously model the local and global geometric relationships as well as their internal correlations, demonstrating strong generalization ability and high performance. FuPConv is incorporated in classical hierarchical network architectures to achieve local and global shape-aware learning. In FPTransformer, we introduce full point position encoding in self-attention, that hierarchically encodes each point position in the global and local receptive field. We also propose a shape-aware downsampling block that takes into account the local shape and the global context. Experimental comparison to existing methods on benchmark datasets shows the efficacy of FuPConv and FPTransformer for semantic segmentation, object detection, classification, and normal estimation tasks. In particular, we achieve state-of-the-art semantic segmentation results of 76.8% mIoU on S3DIS sixfold and 73.1% on S3DIS Area 5. Our code is available at https://github.com/hnuhyuwa/FullPointTransformer.

Cyberbullying Detection in Social Networks Using Deep Learning

Cyberbullying causes significant harm, especially among adolescents and young adults. With the growth of social media, online harassment through platforms like Facebook and Twitter has also proliferated rapidly. Though social networks have reporting mechanisms, the volume of user-generated content makes manual moderation infeasible. This necessitates automated detection systems that can accurately identify cyberbullying at scale. Recent advances in deep learning provide promising techniques for text classification tasks. This paper explores Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM), and transformer models like Bidirectional Encoder Representations from Transformers (BERT) for cyberbullying detection in social networks. The models are evaluated on a benchmark dataset containing 11,000 Facebook comments labeled as clean or cyberbullying. Extensive experiments demonstrate that BERT achieves the highest accuracy of 87.3% followed by a hierarchical CNN-LSTM architecture with 86.5% accuracy. The former benefits from bidirectional context modeling using self-attention while the latter combines the strengths of convolutional layers and LSTMs. The results verify the effectiveness of deep learning methodologies for this problem. However, enhancements in multilingual, multimodal support and adversarial robustness are required. Testing on diverse platforms and content along with user privacy considerations remain as future research directions. This empirical study provides useful insights to build robust cyberbullying detection systems.