446494 research outputs found
Sort by
Research on an intelligent monitoring system for the entire tunnel construction process based on IoT and digital twin
No full textThere are high-risk problems, such as peripheral rock instability and palm face collapse, during tunnel construction, and the traditional monitoring methods are difficult to meet the safety management needs due to sparse data and lagging response. This paper proposes an intelligent monitoring system for the whole process of tunnel construction based on Internet of Things (IoT) and digital twin, which integrates a multi-source sensor network, BIM dynamic modeling, and risk intelligent analysis. The system realizes an all-around perception of environment, equipment, and surrounding rock status through real-time fusion of heterogeneous data, and uses digital twin technology for 3D visualization and risk trend prediction. It adopts the improved D-S evidence theory for multi-source risk assessment, and improves the early warning accuracy through the effectiveness factor and conflict weakening strategy. The actual engineering experiments show that the system achieves 100% monitoring coverage and 100% warning accuracy, and successfully captures the whole process of palm face collapse and the time-sequence evolution of enclosing rock deformation, which significantly improves the safety and management efficiency of tunnel construction. The study verifies the high accuracy and stability of the proposed system, which provides an intelligent solution for the safety of complex underground projects
High-Speed AXI4-Lite Dynamic MAC Accelerator for RISC-V SoC Designs
No full textThis design and realization of a high-speed dynamic-precision Multiply–Accumulate (DPMAC) accelerator targets embedded and RISC-V-based System-on-Chip (SoC) platforms. Unlike conventional MAC units that operate with a fixed precision, this design supports runtime configurable 8-, 16-, and 32-bit precision modes. These modes allow accuracy, energy efficiency, and operational behaviour to be tuned according to workload requirements, while the pipeline maintains the same latency for all modes. The dynamic-precision MAC’s data-path integrates Radix-4 Booth multiplier and Brent– Kung adder within a two-stage pipeline, ensuring deterministic and low-latency operation regardless of precision selection. The AXI4-Lite interface enables communication with processor subsystems through memory-mapped registers, providing flexible configuration and control of precision, start/enable logic, and accumulator behaviour. Through FPGA synthesis and testing, the proposed AXI-MAC unit demonstrates competitive timing and resource utilization while offering improved numerical flexibility compared to traditional fixed-precision MAC units. The absence of DSP slice requirements further enhances portability across devices. As such, the architecture provides an efficient, high-performance, and scalable solution suitable for next-generation embedded computing and edge-processing applications
Intelligent Data-Driven Modeling of SARS-CoV-2 Interactions in BP–MXene–BP Heterostructure SPR Biosensors using Ridge Regression Model
No full textThe global SARS-CoV-2 pandemic has emphasized the urgent need for rapid, accurate, and scalable diagnostic technologies suitable for widespread screening. Conventional laboratory methods such as RT-PCR and ELISA, although reliable, suffer from long turnaround times, high operational cost, and dependence on specialized personnel, limiting their applicability in resource-constrained environments. Surface Plasmon Resonance (SPR) biosensors have emerged as promising alternatives, offering real-time, label-free molecular detection with high sensitivity and specificity. Recent advances highlight that integrating 2D nanomaterials—particularly BP/MXene multilayer heterostructures—significantly enhances plasmonic field confinement, signal strength, sensitivity, and detection accuracy compared to traditional metal-only configurations. However, modeling and optimizing such advanced SPR architectures typically depend on computationally intensive analytical methods, such as the Trans- fer Matrix Method and Fresnel formulations, which rely on idealized material parameters and are difficult to scale for real-time optimization. To address these limitations, this work introduces an intelligent machine- learning-based prediction framework for CaF2/Ag/BP/MXene/BP SPR biosensors using regression models to learn nonlinear relationships between structural parameters and performance metrics, including sensitivity and resonance wavelength. The proposed data-driven approach enables faster and more accurate performance estimation without exhaustive simulations, supporting rapid optimization across diverse operating scenarios. By combining plasmonic nanostructures with AI- assisted predictive modeling, this study establishes a foundation for intelligent, self-optimizing SPR diagnostic platforms suitable for next-generation biomedical applications
Design of Piezoelectric Energy Harvester with an Adaptive MPPT for Efficient Power Extraction
No full textThis study presents a piezo electric energy harvester that includes a dedicated maximum power point tracking (MPPT) control scheme will be introduced to improve efficiency of energy harvesting using mechanical vibrations in the atmosphere because of the reduction of the impedance mismatch between piezo transducers and electronic devices. The general structure of the proposed harvester is a cantilever piezo transducer that is manufactured using lead zirconate titanate (PZT) ceramic and that has been optimized to operate at a resonant frequency of 60 Hz and a power moderator circuit that operates on less than 10 W and that includes a step-up voltage converter and a Schottky diode bridge rectifier. a microcontroller-based MPPT controller that is based on an adapted perturb-and-observe algorithm with adaptable step size; and an energy-store interface MPPT algorithm is continuously measuring the system voltage and current by using high-precision sensing circuits and dynamically controls the load impedance of the piezo transducer to the electrical load by using a digital potentiometer. Keyword -- Energy harvesting technique, maximum power point tracking for improved power efficiency, piezoelectric transducers, mechanical stress, wireless sensor networks, power management, renewable energy source
Virtual reality interview simulator: To enhance professional skills
No full textTraditional job interview practice methods are severely deficient in focusing on performance anxiety and realistic non-verbal communication skill development owing to an absence of intrinsic environmental and psychological immersion This paper outlines the architecture, design, and development of a new Virtual Reality (VR) Interview Simulator using Unity 3D and powered by the Neocortex AI Platform. The underlying intention of the simulator is to provide a safe, standardized, and a realistic simulation environment that repeatedly exposes job applicants to the whole sensory and psychological stress of a professional interview. The technology applies a premium six-step conversational cycle, merging Speech-to-Text (STT), Natural Language Understanding (NLU), and Text-to-Speech (TTS) to facilitate dynamic, AI-powered conversations with a personalized interviewer avatar. This research validates the effectiveness of conversational AI deployment in a high-fidelity immersive VR environment for soft-skills training of the utmost quality and presents a scalable framework to enable future content development, aiming to democratize access to high-quality interview practice
Artificial neural network based dynamic dump load control for integrated standalone microgrid
No full textANN control approach for DC-link stabilization and power flow coordination in a standalone hybrid microgrid formed by a 10 kWp photovoltaic array, a 20 kW wind turbine generator, and a 60 kWh battery storage system operating on a 600 V bus. A controllable dump-load branch is integrated into the design to dissipate surplus energy when the battery reaches its upper state-of-charge limit. The ANN dynamically generates converter duty ratios based on its inputs of DC-link error, its rate of change, and real-time renewable power. Simulation studies conducted in MATLAB demonstrate that the proposed controller maintains DC-link variations within ±2 %, reduces voltage ripple to approximately 3.2 V, and limits current distortion to 2.64 %, while a PI controller exhibits slower recovery, higher overshoot, and post-filter harmonics of about 3.82 %. The results confirm that ANN-based dump-load coordination ensures stable operation, prevents converter stress due to power surplus, and improves the overall utilization of renewable energy in standalone microgrid environments
Deep Learning Strategies for Diabetic Foot Ulcer Diagnosis: An Empirical Comparison of Transformer and CNN Architectures
No full textDiabetic Foot Ulcers (DFU) are among the most serious complications faced by diabetic patients. What usually starts as a small injury can worsen and in many cases lead to lower limb amputation. However with modern technologies like deep learning early screening and identification of DFU has become more achievable. This helps doctors and make their job easy. This study presents a practical comparison between ResNet18 a CNN and DeiT-Small a Vision Transformer. Both these models were tuned using transfer learning on datasets containing images of both DFU affected feet and healthy feet. There was a clear performance gap between both these models. DeiT-Small reached an impressive test accuracy of 99.17% while ResNet18 achieved 85.40% accuracy. To understand their performance better visualization tools like Grad-CAM and transformer attention maps were used. For future work, the study aims to create more consistent data splits, perform deeper architectural ablation and integrate IOT based sensors to support real time DFU monitoring
Utilization of Landsat 8 Level 2 Satellite Imagery for Identification of Tea Plant Health Using the Weighted Overlay Method (Case Study: Kertamanah Unit, Malabar Plantation, Bandung)
No full textMonitoring tea plant health is important to maintain plantation productivity. This study aims to map the health status of tea plants in the Cinyiruan and Kertamanah divisions, Kertamanah Unit, Malabar Plantation owned by PT Perkebunan Nusantara 1 Regional 2 using Landsat 8 Level 2 satellite imagery. The parameters used were NDVI, LST, and slope from DEM. These parameters were integrated using the weighted overlay method with weights of 50% for NDVI, 30% for LST, and 20% for DEM in the Yielding Plant area. NDVI and LST processing were conducted in Google Earth Engine, while weighting and visualization were carried out in ArcMap 10.8.2. The resulting tea plant health map was classified into three classes: very healthy, quite healthy, and unhealthy. Validation using wet tea crop production data per-block for 2020–2024 produced an accuracy of 46.6%, influenced by the 30-meter image resolution and field variations such as crop age, rainfall, pests, and production records. Spatially, very healthy class dominated the central to southern areas, while unhealthy class was generally found in peripheral and steep slopes areas. These results indicate that high NDVI values, low surface temperatures, and gentle topography are closely related to better tea plant conditions
Spatial Analysis of Forest Fire Hotspots in Kalimantan Using Inhomogeneous Neyman-Scott Cluster Area
No full textForest and land fires in Kalimantan pose a serious environmental challenge with significant impacts on ecosystems, public health, and the achievement of the Sustainable Development Goals (SDGs). In recent years, fire intensity has increased due to the combined effects of human activities, peatland characteristics, and extreme climate conditions associated with El Niño. Spatial analysis shows that fire hotspots are unevenly distributed, with a strong concentration in peatland areas. This study applies the Inhomogeneous Neyman–Scott Cluster Process (NSCP) to model the spatial clustering of hotspots under heterogeneous conditions by incorporating covariate information. The objective is to identify hotspot distribution patterns in Kalimantan during 2022–2024 and to generate fire risk maps to support targeted mitigation and environmental management policies. Model comparison using the Akaike Information Criterion (AIC) indicates that the Thomas Cluster Process with distance to BPBD facilities as a covariate provides the best fit, with the lowest AIC value (-456,944.7). The results reveal a negative relationship between distance to BPBD facilities and hotspot intensity, indicating that fire occurrences decrease with increasing distance. These findings confirm that hotspot distribution in Kalimantan is not random and highlight the importance of spatially informed fire management strategies
Self-Presentation and Meme Culture: How Do Young People Manage Emotions on Social Media?
No full textAgainst the backdrop of global crises, social media has become a crucial platform for young generations, allowing them to self-express and alleviate anxiety. With the widespread use of social media, a growing number of young people from diverse countries have become active users. Self-expression and memes are two common reasons young people are drawn to social media. Self-presentation refers to the deliberate actions individuals take to manage their impressions on others in social situations. Memes indicate the collective creation and dissemination of modifiable and easily reproducible media units on digital platforms. Young people use selfpresentation on social media to maintain social connections and express emotions through various symbols. Therefore, this study aims to explore how self-presentation on social media can help young people alleviate social anxiety and identity confusion, and how memes can mitigate negative emotions caused by crises and help young people build their identities. This article combines dramaturgy, social identity theory, and humour relief theory to analyse the mechanisms of social media’s impact on emotion regulation at both the individual and group levels. Moreover, practical implications are revealed in terms of platforms, policies and social media users, highlighting the importance of self-portray and meme use