Performance Limitations Research Articles

Thrust-efficiency limits of a swimming tail with variable chordwise flexural rigidity

Emulating oscillations performed by natural swimmers can provide different functionalities than those of propeller-based underwater robots. Yet, to successfully accomplish specific missions under limited power, there is a need to design efficient bio-inspired robots. Adding an appropriate level of flexibility to flapping caudal fins (tails) of robots emulating the thunniform swimming mode has been shown to enhance the thrust generation over a finite range of the flapping frequency. Still, in many cases, adding flexibility to increase thrust generation may require increased input power, which may cause a significant reduction in the efficiency. These observations lead to the concept of enhanced performance by varying the stiffness of the tail as in the case of natural swimmers. This study is concerned with assessing the impact of varying the chordwise stiffness on the tail deflection and flow dynamics, including contributions of added mass and circulation forces to thrust generation and their impact on efficiency. The simulation data are used to identify specific flow dynamics and tail deflections associated with the enhanced thrust generation and/or efficiency, and to define a performance limit expressed as the maximum efficiency as a function of the thrust coefficient.

Thick Hydrogel Membrane With Macro‐Channel for Rapid Uranium Extraction From Seawater

AbstractThe growing demand for uranium driven by nuclear power development necessitates the creation of macroscale, cost‐effective, and efficient adsorbents for uranium extraction from seawater, a challenge yet to be fully addressed. Amidoxime‐based hydrogel adsorbents are promising candidates but suffer from performance limitations due to bound water obstructing pore channels and impeding uranyl ion migration. Herein, a thick macro‐channel poly(amidoxime)‐polyvinyl alcohol hydrogel membrane (MCHM) with a double cross‐linking structure is introduced, designed to enhance uranium adsorption. Unlike traditional powders and microfibers, the macroscale thick membrane facilitates deployment and recovery in marine environments, while its directional channel structure improves mass transfer properties and accessibility to amidoxime groups. Using COMSOL Multiphysics, design parameters such as channel size, membrane thickness, and arrangement are optimized, significantly enhancing performance while reducing costs. In natural seawater, the optimized MCHM with 2.0 mm channels, 20 times thicker than ultra‐thin hydrogel membranes demonstrated improvements of 99% in adsorption capacity and 179% in adsorption rate compared to non‐channel counterparts. This method uniquely combines material innovation with computational modeling, providing valuable insights into how the internal channel structure of adsorbents influences their performance. The findings provide a scalable design solution for hydrogel adsorbents, enabling sustainable uranium extraction from seawater.

Mobility‐Lifetime Products in Organic Infrared Photodiodes with Peak Absorption at 1550 nm

AbstractInfrared photodiodes based on organic semiconductors are promising for low‐cost sensors that operate at room temperature. However, their realization remains hampered by poor device efficiency. Here, performance limitations are analyzed by evaluating the mobility‐lifetime products and charge collection efficiency of devices operating in the shortwave infrared with a peak absorption at 1550 nm. Through complementary impedance and current‐voltage measurements on devices with different donor‐to‐acceptor semiconductor ratios, a trade‐off between mobility and recombination time and the need to balance between transport and interfacial charge transfer are observed. Thus, this study revisits the mobility‐lifetime metric to shed new light on charge collection constraints in organic infrared photodiodes.

SBCP-YOLO-R3D: Student Behavior Recognition and Visualization Framework Using Improved YOLO and R3D for Class Video

Real-time recognition and visualization of students' behaviors in face-to-face classrooms serve as pivotal indicators of learning engagement. However, current methods exhibit limitations in both real-time performance and accuracy. Additionally, in-depth studies have not been extensively conducted to evaluate learning status more conveniently by utilizing computer vision techniques. To address these issues, a novel Student Behavior Recognition and Dynamic Class Portraits Construction framework, named SBCP-YOLO-R3D, incorporating the StB-YOLO and R3D methods, has been proposed to detect student behaviors and construct class portraits. The developed framework comprises two layers: the StB-YOLO detection layer and the R3D classification layer. In the StB-YOLO detection layer, the Lightweight-SEAM (LW-SEAM) is incorporated into YOLOv5 to enhance the recognition of occluded students, by capturing contextual information and enhancing occlusion-related features. Moreover, a Double-SlideLoss function is devised, employing adaptive weighting mechanisms to strike an optimal balance between simple and challenging samples. In the R3D classification layer, the results generated by StB-YOLO are then processed using R3D to produce class portraits. Experiments conducted on the StuAct and SCB-DATASET3-S datasets demonstrate the effectiveness of the StB-YOLO. Compared with the baseline model, StB-YOLO increases the mAP by 3.1%.

Evaluating machine learning models for predictive analytics of liver disease detection using healthcare big data

Liver diseases rank among the most prevalent health issues globally, causing significant morbidity and mortality. Early detection of liver diseases allows for timely intervention, which can prevent the progression of such diseases to more severe stages such as cirrhosis or liver cancer. To this end, many machine learning models have been previously developed to early predict liver diseases among potential patients. However, each model has its accuracy and performance limitations. In this paper, we present a comprehensive comparison of three different machine learning models that can be employed to enhance the prediction and management of liver diseases. We utilize a big data set of 32,000 records to evaluate the performance of each model. First, we implement a preprocessing technique to rectify missing or corrupt data in liver disease datasets, ensuring data integrity. Afterwards, we compare the performance of three machine models: k-nearest neighbors (KNN), gaussian naive Bayes (Gaussian NB) and random forest (RF). We concluded that the RF algorithm demonstrates superior performance in our evaluation, excelling in both predictive accuracy and the ability to classify patients accurately regarding the presence of liver disease. Our results show that RF outperforms other models based on several performance metrics including accuracy: 97.3%, precision: 97%, recall: 96%, and f1-score: 95%.

Particle tracking modelling in coastal marine environments: Recommended practices and performance limitations

Particle tracking modelling in coastal marine environments: Recommended practices and performance limitations

Design Principles and Performance Limitation of InGaN Nanowire Photonic Crystal Micro-LEDs

Design Principles and Performance Limitation of InGaN Nanowire Photonic Crystal Micro-LEDs

Nanophotonic electron accelerator: A review of particle accelerator technology

Particle accelerators are indispensable tools in various industries, spanning a wide range of research fields such as nuclear and particle physics. They are particularly valuable in the medical sector for applications like medical imaging, radiotherapy and tumor treatment. Currently, the largest and most powerful particle accelerator is the Large Hadron Collider (LHC) at CERN, a 27-kilometer-long ring-shaped tunnel that accelerates particles, such as protons, to near-light speeds and collides them. While the LHC represents the pinnacle of accelerator technology, there is significant interest in developing compact particle accelerators that can fit within buildings, laboratories, or even on tabletops. However, even smaller accelerators often occupy several square meters of valuable space and may suffer from performance limitations. Not all applications require massive machines like the LHC, and more compact solutions could fulfill many needs efficiently. Furthermore, a breakthrough in this field has been achieved by a team of laser physicists from Friedrich-Alexander University Erlangen-Nürnberg (FAU), who successfully demonstrated the first nanophotonic electron accelerator. This nanoscale device has been used to accelerate electrons, achieving a remarkable energy gain. What makes this development truly exciting is the unprecedented compactness of the nanophotonic accelerator - it is so small that it fits on a one-cent coin, making it the smallest particle accelerator currently available. Often referred to as a "particle accelerator on a chip," this technology holds immense potential, including applications in advanced cancer radiation therapies.

Superstrong Lightweight Aerogel with Supercontinuous Layer by Surface Reaction.

Breaking the thermal, mechanical and lightweight performance limit of aerogels has pivotal significance on thermal protection, new energy utilization, high-temperature catalysis, structural engineering, and physics, but is severely limited by the serious discrete characteristics between grain boundary and nano-units interfaces. Herein, a thermodynamically driven surface reaction and confined crystallization process is reported to synthesize a centimeter-scale supercontinuous ZrO2 nanolayer on ZrO2-SiO2 fiber aerogel surface, which significantly improved its thermal and mechanical properties with density almost unchanged (≈26 mgcm-3). Systematic structure analysis confirms that the supercontinuous layer achieves a close connection between grains and fibers through Zr─O─Si bonds. The as-prepared aerogel exhibits record-breaking specific strength (≈84615 Nmkg-1, can support up to ≈227272 times aerogel mass) and dynamic impact resistance (withstanding impacts up to 500 times aerogel mass and up to 200 cycling stability at 80% strain). Besides, its temperature resistance has also been greatly optimized (400°C enhancement, stability at 1500°C). This work will provide a new perspective for exploring the limits of lightweight, high strength, and thermal properties of solid materials.

Cost Minimized Immunoaffinity Purification of EPO and Its Analogs in Doping Control-A Step-by-Step Protocol for Human Urine and Blood.

A cost minimized immunoaffinity protocol was developed, which allows the direct purification of ERAs (urinary and recombinant human EPO, Darbepoetin, EPO-Fc, CERA) from human urine. The method applies magnetic beads and needs no covalent immobilization of the capture antibody. It requires only 10 mL of urine, 1 μg of anti-EPO antibody, and 25 μL of bead slurry. The beads are coated with the capture antibody in advance and can be stored in the refrigerator for months without any loss of functionality. The protocol was fully validated in combination with SAR-PAGE and Western blotting using the biotinylated clone AE7A5 anti-EPO antibody. It is compliant with the criteria of TD2024EPO of the World Anti-Doping Agency (WADA) for the gel electrophoretic detection of ERAs. For each ERA, the achieved limit of detection (LOD) is at least one tenth of the minimum required performance limit (MRPL) of WADA, that is, 0.1 IU/L, 0.1 pg/mL, 0.5 pg/mL, and 0.5 pg/mL for rEPO, Darbepoetin, EPO-Fc, and CERA, respectively. After slight modification, the protocol is also applicable to serum and plasma and also fulfils the corresponding MRPLs of WADA for these matrices. Compared to commercial immunoaffinity purification kits for EPO, the material costs are significantly lower but with identical results.

Regulation performance limitation of networked control systems under dual‐channel noise constraints

AbstractIn this paper, we examine the regulation performance of single‐input single‐output networked control systems (NCSs). Network limitations such as time delay, noise, and insufficient bandwidth significantly challenge the regulation performance of NCSs. We derive the regulation performance limits under dual‐channel Gaussian noise, encoding–decoding processes, and bandwidth constraints using coprime factorization and spectral techniques. Our findings reveal that the regulation performance of NCSs is closely associated with the nonminimum phase zeros, unstable poles, and the communication constraints of the network. We also present a numerical simulation to validate our findings.

Discrete Carbon Nanotubes vs. Carbon Fiber: Nano-Uniform Surface Resistance for Next Gen ESD-Safe Fixtures

Elect Nano has successfully developed next-generation ESD-safe injection molding compounds leveraging discrete carbon nanotube (dCNT) technology for use in semiconductor media handling trays. Traditional molding compounds are not suitable for the next generation of chip and fab processes where cleanliness, ESD uniformity, dimensional tolerances and sustainability are being pushed to the incumbent material tolerance and performance limits. The dCNT injection molding compounds were compared directly with carbon fiber filled molding compounds and showed outstanding benefits for surface resistance uniformity, surface finish quality, laser marking contrast and isotropic mold shrinkage.

Frequency-Modulated Antipodal Chaos Shift Keying Chaotic Communication on Field Program Gate Array: Prototype Design and Performance Insights

Using chaos for communication can provide more robust channel security, covert transmission, and inherent support for spread-spectrum modulation. Although numerous studies have explored this technology, its practical deployment remains limited due to substantial hardware demands, complex signal processing, and a lack of efficient modulation methods for chaotic signals. In this study, a novel chaotic digital communication system is proposed and studied. A prototype of a frequency-modulated antipodal chaos shift keying (FM-ACSK) system is implemented on an Intel Cyclone V field-programmable gate array (FPGA) along with a complete mathematical model using Matlab R2022a Simulink software. Using FPGAs to implement chaotic oscillators avoids analog system problems such as component drift and high thermal instability while providing determined system parameters, rapid prototyping, and high throughput. The employment of FM over a chaotic modulation layer provides a passband operation (currently at an intermediate frequency of 10.7 MHz) while adding the benefits of carrier frequency offset robustness and constant signal envelope. Within this study, the robustness of FM-ACSK to white noise in the channel was evaluated using bit error rate, which was tested through hardware experiments and simulations. The results show the feasibility and potential performance limitations of this approach to chaotic communication system design.

Number Recognition Through Color Distortion Using Convolutional Neural Networks

Machine learning applied to image-based number recognition has made significant strides in recent years. Recent use of Large Language Models (LLMs) in natural language search and generation of text have improved performance for general images, yet performance limitations still exist for data subsets related to color blindness. In this paper, we replicated the training of six distinct neural networks (MNIST, LeNet5, VGG16, AlexNet, and two AlexNet modifications) using deep learning techniques with the MNIST dataset and the Ishihara-Like MNIST dataset. While many prior works have dealt with MNIST, the Ishihara adaption addresses red-green combinations of color blindness, allowing for further research in color distortion. Through this research, we applied pre-processing to accentuate the effects of red-green and monochrome colorblindness and hyper-parameterized the existing architectures, ultimately achieving better overall performance than currently published in known works.

Sargassum Nanocellulose-Based Fully Ingestible Supercapacitor.

Small high-performance energy modules have significant practical value in the biomedical field, such as painless diagnosis, alleviation of gastrointestinal discomfort, and electrical stimulation therapy. However, due to performance limitations and safety concerns, it is a formidable challenge to design a small, emerging ingestible power supply. Here, a fully ingestible supercapacitor (FISC) constructed of sargassum cellulose nanofiber is presented. FISCs exhibit an electrode areal capacitance of 2.29 F cm-2 and a high energy density of 307 µWh cm-2. Furthermore, over 90% of the antibacterial activity against Escherichia coli is achieved during the self-discharge process. Therefore, following insertion into an enteric capsule, this device can enable a disposable power supply and electrostimulation for bacteriostasis in the intestine after being swallowed by a human, which offers new possibilities for scientific and simple therapy.

Depth-Oriented Gray Image for Unseen Pig Detection in Real Time

With the increasing demand for pork, improving pig health and welfare management productivity has become a priority. However, it is impractical for humans to manually monitor all pigsties in commercial-scale pig farms, highlighting the need for automated health monitoring systems. In such systems, object detection is essential. However, challenges such as insufficient training data, low computational performance, and generalization issues in diverse environments make achieving high accuracy in unseen environments difficult. Conventional RGB-based object detection models face performance limitations due to brightness similarity between objects and backgrounds, new facility installations, and varying lighting conditions. To address these challenges, this study proposes a DOG (Depth-Oriented Gray) image generation method using various foundation models (SAM, LaMa, Depth Anything). Without additional sensors or retraining, the proposed method utilizes depth information from the testing environment to distinguish between foreground and background, generating depth background images and establishing an approach to define the Region of Interest (RoI) and Region of Uninterest (RoU). By converting RGB input images into the HSV color space and combining HSV-Value, inverted HSV-Saturation, and the generated depth background images, DOG images are created to enhance foreground object features while effectively suppressing background information. Experimental results using low-cost CPU and GPU systems demonstrated that DOG images improved detection accuracy (AP50) by up to 6.4% compared to conventional gray images. Moreover, DOG image generation achieved real-time processing speeds, taking 3.6 ms on a CPU, approximately 53.8 times faster than the GPU-based depth image generation time of Depth Anything, which requires 193.7 ms.

CXR-LLaVA: a multimodal large language model for interpreting chest X-ray images.

This study aimed to develop an open-source multimodal large language model (CXR-LLaVA) for interpreting chest X-ray images (CXRs), leveraging recent advances in large language models (LLMs) to potentially replicate the image interpretation skills of human radiologists. For training, we collected 592,580 publicly available CXRs, of which 374,881 had labels for certain radiographic abnormalities (Dataset 1) and 217,699 provided free-text radiology reports (Dataset 2). After pre-training a vision transformer with Dataset 1, we integrated it with an LLM influenced by the LLaVA network. Then, the model was fine-tuned, primarily using Dataset 2. The model's diagnostic performance for major pathological findings was evaluated, along with the acceptability of radiologic reports by human radiologists, to gauge its potential for autonomous reporting. The model demonstrated impressive performance in test sets, achieving an average F1 score of 0.81 for six major pathological findings in the MIMIC internal test set and 0.56 for six major pathological findings in the external test set. The model's F1 scores surpassed those of GPT-4-vision and Gemini-Pro-Vision in both test sets. In human radiologist evaluations of the external test set, the model achieved a 72.7% success rate in autonomous reporting, slightly below the 84.0% rate of ground truth reports. This study highlights the significant potential of multimodal LLMs for CXR interpretation, while also acknowledging the performance limitations. Despite these challenges, we believe that making our model open-source will catalyze further research, expanding its effectiveness and applicability in various clinical contexts. Question How can a multimodal large language model be adapted to interpret chest X-rays and generate radiologic reports? Findings The developed CXR-LLaVA model effectively detects major pathological findings in chest X-rays and generates radiologic reports with a higher accuracy compared to general-purpose models. Clinical relevance This study demonstrates the potential of multimodal large language models to support radiologists by autonomously generating chest X-ray reports, potentially reducing diagnostic workloads and improving radiologist efficiency.

The Application of Microcomputer Systems in Automatic Control Systems

With the continuous advancement and innovation of computer system technologies, the performance of microcomputers has improved significantly. The enhanced computing power of microcomputer hardware provides greater flexibility for the development of various intelligent optimization algorithms. Integrating microcomputers into automatic control systems effectively addresses the systems demands for signal stability, sensitivity, response speed, and manufacturing cost, thereby significantly advancing the practical application of automatic control systems. This paper comprehensively explores the current state and challenges of applying microcomputers in automatic control systems, while providing an in-depth analysis of development trends in this field. It systematically explains the specific improvements microcomputers bring to control systems and how to select appropriate algorithms based on different types of automatic control systems. Moreover, the review highlights the impact of hardware performance limitations of microcomputers on algorithm implementation and discusses potential solutions and directions for future technological advancements. It also examines successful applications and existing issues in representative automatic control systems employing microcomputers. Finally, the study anticipates the development trends of intelligent optimization algorithms for microcomputers in automatic control systems, as well as the emerging challenges and research directions

Utilizing convolutional neural network (CNN) for orchard irrigation decision-making

Efficient agricultural management often relies on farmers’ experiential knowledge and demands considerable labor, particularly in regions with challenging terrains. To reduce these burdens, the adoption of smart technologies has garnered increasing attention. This study proposes a convolutional neural network (CNN)-based model as a decision-support tool for smart irrigation in orchard systems, focusing on persimmon cultivation in mountainous regions. Soil moisture data and corresponding leaf RGB images were collected over two growing seasons to train, test, and validate the CNN model for determining optimal irrigation timing. The model achieved over 80% accuracy in identifying water stress levels in persimmon trees based on leaf images. These findings indicate the potential of the developed model as a key component of a remote irrigation system. However, the model’s performance limitations and challenges in adapting to diverse field conditions underscore the need for further research to enhance its robustness and applicability.

Real-time deconvolution of light fields through pixel selection in the point-spread-function and direct inversion of the image formation process

This paper explores the optimization of light field deconvolution, a key process in image processing that reconstructs a 3D object space or a 2D refocus plane from a light field. Despite the critical role of deconvolution in light field technology, existing methods are often slow, computationally intensive, and unsuitable for real-time processing. Existing algorithms, such as the Richardson-Lucy approach, while groundbreaking, still suffer performance limitations due to their iterative nature and high computational costs. Central to our approach is the strategic selection of influential pixels within the point-spread-function, reducing redundant computations by focusing only on pixels contributing to a significant portion of the point-spread-function’s total intensity. In addition, we explore the potential to directly invert the image formation model, bypass iterative computations, and further accelerate the deconvolution process. Our findings reveal notable improvements in computational efficiency, with some of our methods achieving real-time performance. The reconstruction quality, measured using metrics such as the mean squared error, remained comparable to existing approaches, indicating a favorable balance between speed and reconstruction quality.