Efficient System Research Articles

Using new control strategies to improve the effectiveness and efficiency of the hybrid power system based on the battery storage system

Hybrid energy systems (HESs) are integrated systems that have successfully addressed the problems of meeting the increasing demand for electrical power. Like all known power systems, the energy and stream quality are among the most important issues in addition to the durability of the HES. In this study, the battery-powered HES is presented, where this designed system consists of a wind system and a photovoltaic (PV) system. The strategy of maximum power point (MPP) tracking (MPPT) based on the adaptive neuro-fuzzy inference system (ANFIS) method is used to command the PV system and the wind system, and the MPPT based on the neural method is used. These proposed strategies do not need the mathematical model of the studied system and augment the robustness and stability, where the system performance is great. Also, the fractional-order proportional-integral regulator and the integral sliding mode control approach are combined to control the battery-based storage system, and the particle swarm optimization approach was used to estimate the gain values of the resulting controller. The HES was realized using MATLAB, where the competence is tested under different work scenarios. The results showed excellent efficacy of the designed control and were compared with conventional control. The simulation results showed that using the neural MPPT strategy in the case of the wind speed being 12 m/s, the values of rise time, response time, MPP, and steady-state error (SSE) are improved by rates estimated at 99.32%, 60%, 1.5%, and 60%, respectively compared to the perturbations and observations-based MPPT approach. Compared to the traditional strategy, the ANFIS-MPPT strategy improves the values of MPP, response time, SSE, and rise time in the case of irradiation, which takes the value of 1000 W/m2, by percentages estimated at 18%, 60%, 94.70%, and 69.23%, respectively. Also, the PSO-FOPI-ISMC strategy improves the harmonic distortion of the current value in the second test by 55.20% and 72.90% for mode 1 and mode 2, respectively, compared to the traditional approach. These results make the designed approach of great importance in the future in other industrial applications.

Exploration of the optimal deep learning model for english-Japanese machine translation of medical device adverse event terminology

BackgroundIn Japan, reporting of medical device malfunctions and related health problems is mandatory, and efforts are being made to standardize terminology through the Adverse Event Terminology Collection of the Japan Federation of Medical Device Associations (JFMDA). Internationally, the Adverse Event Terminology of the International Medical Device Regulators Forum (IMDRF-AET) provides a standardized terminology collection in English. Mapping between the JFMDA terminology collection and the IMDRF-AET is critical to international harmonization. However, the process of translating the terminology collections from English to Japanese and reconciling them is done manually, resulting in high human workloads and potential inaccuracies.ObjectiveThe purpose of this study is to investigate the optimal machine translation model for the IMDRF-AET into Japanese for the part of a function for the automatic terminology mapping system.MethodsEnglish-Japanese parallel data for IMDRF-AET published by the Ministry of Health, Labor and Welfare in Japan was obtained from 50 sentences randomly extracted from the terms and their definitions. These English sentences were fed into the following machine translation models to produce Japanese translations: mBART50, m2m-100, Google Translation, Multilingual T5, GPT-3, ChatGPT, and GPT-4. The evaluations included the quantitative metrics of BiLingual Evaluation Understudy (BLEU), Character Error Rate (CER), Word Error Rate (WER), Metric for Evaluation of Translation with Explicit ORdering (METEOR), and Bidirectional Encoder Representations from Transformers (BERT) score, as well as qualitative evaluations by four experts.ResultsGPT-4 outperformed other models in both the quantitative and qualitative evaluations, with ChatGPT showing the same capability, but with lower quantitative scores, in the qualitative evaluation. Scores of other models, including mBART50 and m2m-100, lagged behind, particularly in the CER and BERT scores.ConclusionGPT-4’s superior performance in translating medical terminology, indicates its potential utility in improving the efficiency of the terminology mapping system.

Full-Percutaneous Trans-Kambin Lumbar Interbody Fusion With a Large-Footprint Interbody Cage.

Exploratory prospective observational cohort. Aim of this study was to evaluate clinical and radiologic outcome, and surgical efficiency of a novel universal delivery system that allows full-percutaneous extraforaminal, trans-Kambin lumbar interbody fusion with a large-footprint lumbar interbody cage using only fluoroscopic imaging and open-surgery instrumentation. We prospectively evaluated patients that underwent elective trans-Kambin TLIF surgery with a large-footprint interbody cage using a novel universal delivery system. Clinical follow-up was evaluated pre-and post-operatively with Visual Analogic Scale (VAS) and Oswestry Disability Index (ODI) scores, while radiologic follow-up was performed with a computed tomography scan and standing films post-operatively at hospital discharge and 1year follow-up. A total of 47 patients were evaluated. Clinically, post-operative VAS and ODI scores significantly (P < 0.001) improved compared to pre-operative scores and 29.4months mean follow-up. Radiologic evaluation yielded an intervertebral fusion rate of 90% of the operated levels and a significant increase in segmental lordosis by 3.7°. Median surgical time for interbody cage insertion per level was 28minutes. Complications included 14 (29%) cases with transitory post-operative radiculitis, 4 (8%) cases with partial muscle weakness and 2 (4%) cases that required revision surgery. Post-operative ambulation started at a median 5hours and median hospital length of stay was 28hours. A new universal delivery system allows overcoming most limitations of current full-endoscopic trans-Kambin fusion as it allows a time- efficient full-percutaneous insertion of a large-footprint interbody cage under fluoroscopy imaging only, with standard open-surgery instruments and optional endoscopic visualization.

Enhancing Agile Responsiveness in Flexible Supply Chains: A Case Study of Automobile Companies

As the leading industry in China's current development, the automobile industry is seeking international development under the background of globalization. In order to gain a competitive advantage, the management model of the flexible supply chain is introduced to quickly adapt to market changes and improve the agile responsiveness of the supply chain. This paper first gives an overview of the research, introduces the development status faced by automobile companies, and explains the basic concepts of flexible supply chains. Then the research content is expanded, and a multi-angle analysis is adopted. From the front end to the back end of the supply chain, the four links of the procurement system and supplier management, production system and design development, inventory system and logistics efficiency, and the role of the audit system in the supply chain are analyzed and studied to improve the agile responsiveness of the flexible supply chain of automobile companies. Finally, for the four links, the multi-point supply strategy, "three-ization" strategy, digital strategy and strategy to improve audit transparency are summarized. In the future, automobile companies will also use a variety of strategies to improve the agile responsiveness of the flexible supply chain to adapt to the ever-changing market and new challenges.

Observation of Sinosteel's Effective Ecologisation and the Direction of Green Metallurgy Development

The metallurgical industry raw material industrial sector, providing metal materials for all sectors of the national economy, is also the material basis of economic development, but its dependence on natural resources, consumption intensity, and has a greater impact on the quality of the environment, so the development of metallurgical industry ecology has become an important topic. The study examines Sinosteels practices in areas such as low-carbon mining, energy-efficient engineering, employee environmental training, and international cooperation, highlighting its efforts to promote industrial ecology. Key initiatives include the adoption of advanced technologies for waste heat recovery, hydrogen-rich carbon cycle processes, and digital solutions to optimize resource utilization and minimize emissions. Furthermore, this paper discusses the broader development direction of Chinas metallurgical industry, emphasizing low-carbon innovation, digital transformation, talent cultivation, and enhanced pollution management. By addressing systemic inefficiencies and fostering green practices, Sinosteel exemplifies the potential for metallurgical enterprises to achieve sustainable growth. The findings underscore the importance of integrating technological innovation, policy reforms, and corporate responsibility to ensure the industrys green transformation. This research offers valuable insights for global metallurgical enterprises aiming to balance economic growth with environmental preservation.

Indian Women’s Struggle Against Domestic Violence and the Challenges They Face

This research explores various issues of domestic violence faced by Indian women, which further discusses physical, emotional, and economic abuse. Physical violence remains one of the prevalent forms, in which patriarchal norms that normalize the control over women through force, and it is how it’s perpetuated until now. Although legal reforms have been implemented, societal acceptance of such violence continues to oppress women across various social classes. Emotional abuse is less visible, but it is as equally harmful as the physical violence’s harm, which will damage women’s mental health and reinforce dependency on abusers. Economic violence is generally associated with traditional practices like dowry, that puts women in financial insecurity, and limits their ability to escape from abusive environments. Grassroots movements and public campaigns have played an important role in improving women’s awareness and empowering them to resist violence. However, there are gaps in legal enforcement, especially in rural areas, that negatively impacts the effectiveness of progressive legislation like the Protection of Women from Domestic Violence Act (PWDVA). Government initiatives have the potential, but they are underutilized because of systemic inefficiencies. The study concludes that Indian women still face significant obstacles in resisting domestic violence, that include societal stigma, limited awareness of rights, and insufficient social and economic support systems. To achieve change, the institutional support should be enhanced and have stronger attitudes in cultural shift toward gender-based violence.

Redesign Considerations for a Person-Centered Nursing Home System

The nursing home sector globally faces significant challenges, including meeting residents’ diverse needs, adhering to regulatory demands, and addressing financial imperatives. These competing priorities often lead to fragmented care, preventable hospitalizations, workforce shortages, and systemic inefficiencies. This commentary examines the nursing home system through the lens of systems and complexity science, proposing a redesign framework that emphazises adaptive leadership, outcomes-focused governance, and alignment of organizational purpose, goals, and values. We identify systemic barriers across organizational levels and recommend principles for achieving holistic, person-centered care. Key strategies include fostering stakeholder collaboration, integrating feedback mechanisms, and redefining regulatory priorities to focus on resident well-being. This approach highlights the critical role of adaptive leadership in enabling systems that are flexible, sustainable, and capable of meeting the evolving needs of aging populations.

Efficient photocatalytic hydrogen production by employing a graphdiyne/NH2-MIL-88B(Fe) composite.

The electron transfer rate in photocatalysts is one of the factors determining their hydrogen production activity. In this work, NH2-MIL-88B(Fe) (NFM) was synthesized using a one-step hydrothermal synthesis method and NFM/GDY-25 (NFMG-25) was successfully synthesized by loading graphdiyne (GDY) onto the surface of NFM. The hydrogen production performance and the mechanism of the prepared photocatalysts were systematically investigated using XRD, SEM, FT-IR, XPS, UV-Vis, PL and photoelectrochemical tests. The results showed that the hydrogen production of NFMG-25 reached 61.7 μmol in 5 hours. Photoelectrochemical and Mott-Schottky tests demonstrated that the composite catalyst exhibited high photogenerated carrier separation efficiency and single catalysts were n-type semiconductors. The conduction bands of NFM and GDY were -0.36 V and -0.56 V, respectively, while the valence bands were 1.90 V and 1.12 V. NFM acted as an electron acceptor and donor, which accelerated the transfer of the electrons, and enhanced the photocatalytic hydrogen production efficiency of the composite system. This study provides an effective method for using NFM in photocatalytic hydrogen production.

Flexible material handling system for multi-load autonomous mobile robots in manufacturing environments: a hierarchical reinforcement learning approach

The increasing complexity of customer demands has led to the implementation of flexible job-shop scheduling and automated material handling systems across manufacturing sectors. In particular, advances in robotic technology have made autonomous mobile robots (AMRs) essential for material handling tasks within these sectors. The capability of free movement, path planning, and loading multiple work-in-processes (WIPs) can significantly enhance the efficiency of material handling operations. However, the full flexibility of AMRs cannot be utilised when their decisions regarding the sequence of loading and unloading multiple WIPs are made by specific rule-based operations, resulting in inefficiencies in the throughput of WIPs in manufacturing environments. To address this inefficiency, we introduce a hierarchical reinforcement learning algorithm to optimise material handling with AMRs, thereby maximising the throughput of WIPs. In this approach, a graph attention network (GAT) serves as an encoder for the hierarchical reinforcement learning (HRL) input, effectively capturing the complex relationships between different nodes. Computational experiments demonstrate that our approach enhances the efficiency of the material handling system more effectively than existing rule-based methods.

An original medical logistics robot system design for the complex environment of hospitals with dynamic obstacles

Purpose COVID-19 has aggregated the need for a non-contact medical logistic system. A non-contact robot with self-navigation ability has greatly enhanced the efficiency of the medical logistic system. This paper aims to design a new medical logistics robot system for the complex environment of hospitals with dynamic obstacles. Design/methodology/approach Targeting a medical logistics robot system for a large-scale hospital environment, this study proposed a dynamic obstacle avoidance system to reduce the robot’s delay time as well as frequent route switching. In the algorithm, this study proposed a new loop closure detection with an artificial correction factor. Moreover, this study presented enhanced 3D object detection, improving detection accuracy in hospital environments. Findings Experimental results confirm that the robot can move along its global path and reach its destination without colliding with stationary or moving obstacles. Originality/value The medical logistics robot system has safe and stable performance in real hospital scenarios. The implementation verifies that the robot has effectiveness and reliability in both hardware and software design.

Examining factors influencing the adoption of enterprise resource planning systems in secondary healthcare institutions

PurposeThe healthcare system in the Philippines has insufficient patient and personnel management, inefficient inventory systems and disorganized medical record-keeping, contributing to unsatisfactory patient care. This occurrence calls for a centralized database, like enterprise resource planning (ERP) systems that manage healthcare systems efficiently. This study utilized partial least squares-based structural equation modeling and random forest classifier (RFC) to identify the factors influencing the adoption of ERP systems in secondary healthcare institutions.Design/methodology/approachThe study proposed an extended unified theory of acceptance and use of technology framework integrated with the valence model to define behavioral intention and use behavior of ERP systems. A self-administered online survey was answered by 250 secondary healthcare stakeholders in the Philippines voluntarily.FindingsEffort expectancy, performance expectancy and social influence result in a very high behavioral intention, leading to the use behavior of ERP systems, while facilitating conditions and innovativeness in information technology led to high significance. Price value and perceived benefits are significant factors. RFC was proven reliable in predicting the factors, with an accuracy of 96%, indicating that the proposed framework can be used in future research to examine enterprise system adoption in different healthcare facilities.Originality/valueThere has yet to be a study that covers ERP in the Philippines, let alone its adoption. Since the ERP system is relatively present for a certain period of time, its adoption among healthcare facilities has not been covered in the Philippines. This study is the first to decipher and understand the ERP adoption and its practical implications.

Sustainable PV-hydrogen-storage microgrid energy management using a hierarchical economic model predictive control framework

Hydrogen-based renewable microgrid is considered as a prospective technique in power generation to reduce the carbon footprint, combat climate change and promote renewable energy sources integration. The photovoltaic-hydrogen-storage (PHS) microgrid system cleverly integrates renewable clean energy and hydrogen storage, providing a sustainable solution that maximizes the solar energy utilization. However, the changeable weather conditions and fluid market make it challenging to manage energy balance of the system. Moreover, in view of the fact that the existing energy management systems often ignore the dynamic synergy of microgrids, a hierarchical economic model predictive control (HEMPC) framework is proposed to realize the optimal operation of PHS microgrid. First, a precise nonlinear model of the PHS microgrid is established and the logic variables are introduced to capture the hydrogen devices’ short-term properties, i.e., the start-up/shut-down of electrolyzers and fuel cells. Then a comprehensive economic cost function, including internal power demand tracking cost, system operation cost and contract deviation cost, is considered in the proposed two-level HEMPC framework in order to address challenges such as fluctuating weather conditions, dynamic market environments, and the often-overlooked dynamic synergy of microgrid components. Under the proposed framework, a mixed-integer nonlinear optimization problem is solved by the long-term EMPC in the upper level to regulate the start-up/shut-down of hydrogen devices and the state of charge in the battery, and the short-term EMPC in the lower level reoptimizes the power demand tracking cost while tracking the optimal reference signal from the long-term EMPC, thereby improving overall control system efficiency. The simulation results along with qualitative and quantitative analysis show that compared with rule-based control, the proposed HEMPC is effective in managing the equipment power output, realizing dynamic synergy and enhancing the economic performance.

Gender-based violence against female patients in hospitals.

Gender-based violence (GBV) against female patients in Indian hospitals is an under-recognized but critical issue that undermines the principles of equitable healthcare. This phenomenon encompasses physical, verbal, emotional, and structural abuse, often exacerbated by deeply entrenched societal norms, gender discrimination, and systemic inefficiencies in healthcare delivery. Female patients in India frequently encounter GBV in its various forms, namely neglect, inappropriate physical contact, discriminatory attitudes, and breaches of confidentiality. Women from marginalized communities (lower socio-economic or minority groups) are particularly vulnerable. Contributing factors include insufficient privacy protocols, overburdened healthcare facilities, inadequate staff training on gender sensitivity, and power dynamics, all of which particularly disadvantage female patients. The consequences of GBV are profound, leading to psychological trauma, diminished trust in healthcare systems, or delayed or avoided medical care. Despite its prevalence, GBV in Indian hospital settings remains under-reported owing to fear of stigma, societal pressure, and lack of accessible reporting mechanisms.

Investigate Advanced Techniques for Power System Protection, Fault Detection, and Restoration to Enhance the Reliability and Resilience of Electrical Grids

Investigate Advanced Techniques for Power System Protection, Fault Detection, and Restoration is a critical area of study aimed at enhancing the reliability and resilience of electrical grids. As modern power systems face increasing complexities due to the integration of renewable energy sources, aging infrastructure, and growing demand for electricity, innovative protection strategies have become essential to mitigate risks associated with electrical faults and disturbances. The significance of this topic lies not only in improving the operational efficiency of power systems but also in minimizing economic losses and ensuring public safety. Advanced techniques in power system protection encompass a range of methodologies, including the implementation of digital protective relays, circuit breakers, and sophisticated communication systems. These technologies enable real-time monitoring and rapid fault detection, which are vital for maintaining continuous power supply. The evolution of machine learning (ML) and artificial intelligence (AI) has further revolutionized this field by facilitating self-healing grids that can autonomously detect, diagnose, and restore service after outages, significantly reducing outage durations and enhancing reliability metrics such as the System Average Interruption Duration Index (SAIDI). Despite the advancements, challenges remain, particularly concerning cyber security risks, regulatory compliance, and the adaptation of protection schemes to accommodate fluctuating energy sources. These complexities necessitate ongoing research and development to create adaptive systems capable of responding to dynamic operational conditions while safeguarding infrastructure integrity. Moreover, the integration of emerging technologies must balance innovation with practical application, ensuring that protective measures not only meet current needs but also anticipate future demands within the energy sector. This investigation into advanced power system protection techniques is pivotal for addressing the pressing issues of system reliability and resilience in the face of evolving electrical grid dynamics. It highlights the critical role of technology in modernizing utility operations and lays the groundwork for future advancements in grid management and outage restoration strategies. Key Words: Power system protection, Fault detection, Distributed Energy Resources (DERs), Islanding, Grid reconnection, Fault localization, Microgrid protection, Fault current variability, System resilience

Emerging Network Security Threats: Preparing for the Next Decade

The swift advancement of network technologies has created an unprecedented level of connectivity and efficiency in digital systems. This expansion has also led to the rise of advanced network security threats, undermining traditional defensive strategies. This research study examines the dynamics of new network security risks, assessing their core causes, impacts, and the technology improvements necessary for effective countermeasures. This study examines critical concerns including ransomware, advanced persistent threats (APTs), IoT vulnerabilities, and AI-driven cyberattacks, with the objective of offering practical insights and strategic frameworks for enterprises to anticipate the next decade of cyber threats. Furthermore, empirical case studies and examples demonstrate the urgent nature of these concerns, while suggested remedies highlight possibilities for resilience and innovation. Keywords- Network Security, Advanced Persistent Threats, Ransomware, IoT Vulnerabilities, Cybersecurity Strategies, Artificial Intelligence, Cyber Threat Mitigation, Quantum Cryptography

Continuous implicit neural representation for arbitrary super-resolution of system matrix in magnetic particle imaging

Objective. Magnetic particle imaging (MPI) is a novel imaging technique that uses magnetic fields to detect tracer materials consisting of magnetic nanoparticles. System matrix (SM) based image reconstruction is essential for achieving high image quality in MPI. However, the time-consuming SM calibrations need to be repeated whenever the magnetic field's or nanoparticle's characteristics change. Accelerating this calibration process is therefore crucial. The most common acceleration approach involves undersampling during the SM calibration procedure, followed by super-resolution methods to recover the high-resolution SM. However, these methods typically require separate training of multiple models for different undersampling ratios, leading to increased storage and training time costs.Approach. We propose an arbitrary-scale SM super-resolution method based on continuous implicit neural representation (INR). Using INR, the SM is modeled as a continuous function in space, enabling arbitrary-scale super-resolution by sampling the function at different densities. A cross-frequency encoder is implemented to share SM frequency information and analyze contextual relationships, resulting in a more intelligent and efficient sampling strategy. Convolutional neural networks (CNNs) are utilized to learn and optimize the grid sampling process in INR, leveraging the advantage of CNNs in learning local feature associations and considering surrounding information comprehensively.Main results. Experimental results on OpenMPI demonstrate that our method outperforms existing methods and enables calibration at any scale with a single model. The proposed method achieves high accuracy and efficiency in SM recovery, even at high undersampling rates.Significance. The proposed method significantly reduces the storage and training time costs associated with SM calibration, making it more practical for real-world applications. By enabling arbitrary-scale super-resolution with a single model, our approach enhances the flexibility and efficiency of MPI systems, paving the way for more widespread adoption of MPI technology.

FOPID controller design for pneumatic control valves with ultra-low overshoot, rapid response and enhanced robustness

The performance of pneumatic control valves, characterized by speed, stability, and accuracy, is critical for industrial production and energy efficiency. Traditional PID and fuzzy control methods face limitations in achieving high-precision control due to structural constraints. This study proposes a fractional-order proportional-integral-derivative (FOPID) controller optimized for pneumatic control valves, incorporating a novel overshoot-penalty objective function. To enhance optimization, the Hippopotamus Optimization (HO) is improved with Genetic Algorithm (GA). Simulation and experimental results demonstrate the proposed GAHO-based FOPID controller achieves a settling time of 5.25 s and an overshoot of 0.88%, significantly surpassing conventional methods. These results establish that the proposed FOPID controller as an effective solution for improving the stability, efficiency, and safety of pneumatic control systems in industrial applications.

Power and data simultaneous transmission using double-clad fibers towards 6G

To the best of our knowledge, this work represents the first report in the literature on the implementation of a radio- and power-over-fiber (RPoF) system utilizing double-clad fiber (DCF) aimed at the sixth generation of mobile communications networks (6G) operating in millimeter-waves (mm-waves). The deployed FiWi (fiber/wireless) system makes use of the DCF core and first cladding for simultaneously and optically transmitting data and power using the same fiber-optics link. For the first time, we demonstrate the simultaneous transmission of a broadband 26-GHz 5G New Radio (NR) signal in conjunction with 20 W of optical power through a 250-m DCF link. Such achievement represents a 150% reach enhancement, compared to the current commercial-available femtocells from the fourth and fifth generations of mobile communications networks (4G and 5G), which use Power-over-Ethernet (PoE) interfaces that typically provide 1 Gbps throughput limited to 100 m reach. The PoF key components of the remote antenna unit (RAU), such as the photodetector and RF amplifier, are optically powered by using PoF technology. We evaluate the PoF system efficiency and stability over extended periods and compare the analog RoF (A-RoF) system performance, by means of conventionally versus optically powered. Experimental results indicate the proposed RPoF system meets the 3rd Generation Partnership Project (3GPP) requirements, in terms of the root mean square error vector magnitude (EVMRMS), with no impairments. Overall, our FiWi system demonstrator provides a throughput of 1.6 Gbps using 16-quadrature amplitude modulation (QAM) over 5-m wireless transmission in the mm-waves band. The obtained experimental data highlight the feasibility of integrating optical and wireless technologies for supporting more reliable and scalable beyond 5G (B5G) and 6G systems.

Efficient Diabetic Foot Ulcer Classification Using Residual Linearly Clamped Layers with Detachable Convolution Layer in DCNN Xception Model

DFU (Diabetes Foot Ulcer) is caused in the foot of diabetes patients and if the disease is not diagnosed within time, it may lead to various complications and severities. The major symptoms of DFU are foot swelling, dry cracks, leg pain, varied temperature of skin and changes in skin color. When DFU is not treated properly complications tends to remove limb or foot of a person. To overcome these limitations, automatic methods are required for detecting and classifying healthy skin and DFU images. However, the classification of DFU from existing researches provides challenges like resemblance between healthy skin and DFU and has high intra-class and inter-class variations in accordance with the lighting conditions and patient’s ethnicity. The collection and labelling of DFU requires large time and it is complex to differentiate a DFU and a healthy skin, for instance when there are images with a toe, skin wrinkles and small sizes of DFU. Hence, the proposed system implemented DCNN (Deep Convolutional Neural Network) based Xception model, to involve binary classification of healthy and DFU skin images. The DCNN is employed to handle large dataset, but produces over-fitting issues. So, the proposed DCNN based Xception classifier is integrated with RLCL (Residual Linearly Clamped Layers) consisting of minimum detached convolution layers. It helps in reducing over-fitting problems and thus increases precision and computational speed of the proposed system. Further, the input images are optimized by using image enhancement techniques such as Histogram equalization, Adaptive filter and Gamma correlation. Then the efficiency of the proposed system is evaluated based on the performance metrics namely precision, F1-score, recall and accuracy to validate the performance of proposed model with existing algorithms.

Nanoparticle-Assisted Phonon Transport Modulation in Si/Ge Heterostructures Using Neuroevolution Potential Machine Learning Models

Abstract Reducing thermal boundary resistance (TBR) is critical for enhancing thermal management efficiency and optimizing the performance of electronic and thermoelectric devices. In this study, we employed non-equilibrium molecular dynamics (NEMD) simulations using Neuroevolution Potential (NEP) machine learning models to investigate how embedding nanoparticles in Si/Ge heterostructures impacts TBR. Our results show a significant reduction in TBR, attributed to enhanced phonon density of states matching via resonance, promoting more efficient elastic phonon transport across the interface. However, this approach also leads to a substantial increase in bulk thermal resistance, highlighting a trade-off where overall heat dissipation is compromised. To address this, we propose an alternative strategy of positioning a nanoparticle directly at the interface to modulate interfacial modes, improving phonon transport efficiency without adversely affecting bulk thermal properties. NEMD simulations validate this approach, showing comparable TBR reduction while mitigating the bulk thermal resistance increase observed with the resonance-based embedding method. This study offers valuable insights into resolving interfacial heat dissipation challenges, providing a balanced strategy for optimizing thermal transport efficiency in nanoscale material systems.