446494 research outputs found
Sort by
AI-Enabled Adaptive DVFS Controller Using LMS-Based Prediction and FSM Logic: FPGA and Real-Time MATLAB Implementation
No full textA popular method for maximising power and performance in contemporary processors and System-on-Chip (SoC) designs is dynamic voltage and frequency scaling, or DVFS. It is widely used in modern processors and System-on-Chip (SoC) designs to balance power and performance. However, conventional DVFS controllers often struggle with highly dynamic workloads, rapid temperature changes, and process uncertainties. Since they depend on fixed threshold values andpredefined lookuptables, they tend to react slowly and can lead to unnecessary power consumption and lead to poor performance. So to avoid such limitations this paper proposes an AI-enabled adaptive DVFS controller using LMS-based prediction and FSM logic. At the same time, FSM states will ensure steady transitions between performance states and a real-time MATLAB environment is created to replicate the performance of FPGA-style AXI register behaviour, enabling cycle by cycle updationsimilar to hardware execution. This design is fully compatible with HDL coder, Vivado and AXI4- lite integration for FPGA implementation. Simulation results show that the predictor produces stable forecasts, the FSM responds smoothly with hysteresis, and the MATLAB real-time imitates closely resembles hardware operation. The proposed architecture offers intelligent and efficient DVFS control with very low computational overhead, making it practical for edge devices and power-sensitive SoCs
Poxisafe Helmet: A Smart Safety Solution for Industrial Environments
No full textIn high-risk industries like mining, construction, and the chemical industry, worker safety is frequently compromised by insufficient monitoring and a delayed response to emergencies. The PoxiSafe Helmet, a wearable safety device with two communication channels and cutting-edge sensors, is discussed in this paper. Real-time monitoring is made possible by the embedded design. The PoxiSafe Helmet is based on the ESP32-S3 microcontroller. The BNO055 was used for fall detection, and the MAX30102 was used for vital signs (heart rate and SpO₂). CCS811 was used to test air quality and gas detection, while DHT22 was used to collect temperature and humidity measurements. The data was sent via Bluetooth Low Energy (BLE) for short-distance communication as well as LoRa for long-distance and power-efficient communication when signal impairments are present. The results demonstrate reliable fall detection, health monitoring accuracy within ±3 % of commercial health monitors, and stable gas detection after settling. There was reliable BLE communication up to 20 meters indoors. For the LoRa tests, there was great long-distance and power-efficient communication up to 500 meters outdoors, with packet loss of less than 5%. The findings indicate that the PoxiSafe Helmet is a viable and economical means of promoting safety in industrial occupations, enabling timely notification of supervisors who can then mitigate risks in the workplace
Design and Analysis the Performance of Ternary Logic Gates using Doping-Less FET
No full textThis paper presents the analysis of performance and design of ternary logic gates using doping- less field-effect transistors (DLFET) integrated with resistive memory (RM). The goal which we’re trying to achieve is low-power, high-speed operation suitable for multi-valued logic systems. The key ternary gates— Inverter, NAND, and NOR—are designed using DLFET-RM architecture and evaluated. Their performance is compared against conventional and emerging technologies, including Single Gate MOSFET, Double Gate MOSFET, FET, CNFET, and FINFET. Parameters such as power consumption, propagation delay, and power-delay product (PDP) are used as the basis for comparison. Significant reductions are shown for both delay and PDP in the simulation results for the proposed design. Compared to traditional logic gates, the DLFET-RM gates achieve up to 90% improvement in PDP . These improvements are hugely due to the doping-less structure, which avoids random dopant fluctuations, and the efficiency of RM elements. By eliminating the need for passive resistors, the proposed circuits also reduce area and complexity. Overall, DLFET-RM-based ternary logic is shown to be a ideal candidate for future low-power nanoelectronic systems
Towards Precision Agriculture: A Real-Time Soil Fertility Monitoring System Using IoT and Deep Learning
No full textSoil fertility is a critical determinant of agricultural productivity and sustainability. This study presents a deep learning framework for classifying soil fertility levels based on a comprehensive dataset of 880 soil samples, each characterized by 12 parameters: Nitrogen (N), Phosphorus (P), Potassium (K), pH, Electrical Conductivity (EC), Organic Carbon (OC), Sulfur (S), Zinc (Zn), Iron (Fe), Copper (Cu), Manganese (Mn), and Boron (B). The data underwent rigorous preprocessing, including handling missing values, removing duplicates, addressing class imbalance via oversampling, and eliminating outliers using the Interquartile Range (IQR) method. An Artificial Neural Network (ANN) model was developed for multi-class classification, featuring three hidden layers with ReLU activation and HeNormal initialization, and a softmax output layer. The model, trained with the Adam optimizer and sparse categorical cross-entropy loss, achieved a high validation accuracy of 93.04% with a loss of 0.2102. Furthermore, this research integrates an IoT-based system utilizing sensors such as the DS18B20 for temperature and NPK sensors for nutrient monitoring to enable real-time soil condition assessment. The synergy of machine learning and IoT technologies established in this work provides a scalable, e
fficient framework for precision agriculture, with the potential to enhance crop yield, optimize resource use, and promote sustainable farming practices
Frequency Decoupling-Based Energy Management System for Fuel Cell Hybrid Electric Vehicles Using State Machine Control Strategy
No full textFuel cell hybrid electric vehicles (FCHEVs) encounter significant challenges in energy management due to the distinct dynamic characteristics of fuel cell systems, batteries, and supercapacitors. Standard methods of managing energy use can lead to excessive hydrogen production, shorten the lifespan of fuel cells, and fail to maintain battery charge effectively when driving conditions change. This paper presents a novel frequency decoupling-based energy management strategy (FDB-EMS) integrated with state machine control to address these limitations. The suggested method uses two low-pass filters to divide power demand into three frequency bands. The battery receives the medium-frequency parts, the fuel cell receives the low-frequency parts, and the supercapacitor receives the high-frequency transients. The state machine controller adjusts power distribution in real-time based on load and SOC limits. Simulations in MATLAB/Simulink demonstrate that the system operates effectively with both constant and variable load profiles. The system uses a 12.875 kW proton exchange membrane fuel cell, a 40 Ah lithium-ion battery, and a 15.6 F supercapacitor. The results show that FDB-EMS consumes 0.060 g/s of fuel, which is 7.7% more efficient than reinforcement learning methods and 16.7% more efficient than rule-based strategies. The system maintains the battery SOC between 62% and 78%, which means that the changes are only 1.8% instead of 4.5% as in fuzzy logic controllers. The transient response time is 140 milliseconds, resulting in power losses of 3.6%. The frequency decomposition does a good job of breaking up changes in the fuel cell that happen at high frequencies. This reduces stress and extends the device’s lifespan. The proposed FDB-EMS is a simple and efficient way to control energy in real-time, which makes the system more reliable and saves fuel
BA-ANFIS: An Efficient Heart Disease Prediction Model Using Adaptive Neuro-Fuzzy Inference with Bat Algorithm
No full textOne of the means to reliably foretell it is the timely receipt of the correct medical treatment in the initial phases of heart disease. One of the most prevalent causes of death in the world is still heart disease. Traditional diagnostic methodologies are often inadequate to accommodate the complexity and ambiguity baked into clinical datasets. This study employs the Adaptive Neuro Fuzzy Inference System (ANFIS) and the Bat Algorithm (BA) to efficiently and precisely identify cardiac issues. ANFIS is a fusion of fuzzy logic and artificial neural networks has difficulties with medical data due to its non-linear nature. But this requires proper tuning of its parameters to work its best. The specific idea is that the Bat Algorithm which imitates the echolocation behaviour of bats, optimizes these parameters to improve prediction accuracy of the ANFIS model. The global search features of BA provide an optimal solution for the ANFIS membership functions together with the ANFIS rule parameters bypassing the limitation of conventional optimization methods. We validate the proposed system with characteristics derived from a clinical dataset of heart disease, such as age, blood pressure, and cholesterol level. Experimental results show that BA-ANFIS achieves an Accuracy of 98.07%, Sensitivity of 97.67%, and Specificity of 98.23%, outperforming baseline models including SVM and standard ANFIS by 86.28% and 94.12%, respectively. This method shows high efficiency in predicting heart disease and can provide support for diagnosis for practitioners, which helps to improve the prognosis of patients and to decrease health care costs due to BA global optimization and ANFIS adaptive reasoning capabilities
Efficiency Analysis of Coagulant and Energy Consumption of Surabaya Water Treatment Plant using Data Envelopment Analysis Method
No full textThe high demand for clean water in Surabaya has driven the optimisation of the water treatment process at three water treatment plants (WTPs). The primary challenge is determining the optimal dosage for coagulation and flocculation, which play a crucial role. DEA was used to assess the optimum coagulant dosage. The DEA calculates the coagulant dosage efficiency score and energy consumption in every WTP unit, with a threshold score of 1. The DEAP 2.1 software performs the assessment simulation. When the WTP score is 1, it is considered efficient in saving coagulants, and an energy A score below 1 indicates inefficiency in conserving those resources. The inputs are energy requirements and coagulant doses between 2018 and 2022. The outputs are water quality parameters such as turbidity, total dissolved solids (TDS), and colour. The highest efficiency performance was obtained by WTP 2 (46%), followed by WTP 3 (43%) and WTP 1 (29%). Overall, the recommendations for energy efficiency and coagulant dosage efficiency vary. The energy efficiency improvement recommendations are pump monitoring, maintenance of the electromotor, and supervision of the main assets. On the other hand, maintaining optimal coagulant requirements, selecting the right coagulant, and administering flocculants effectively are key
Analysis of the Effects of Mesoscale Convective System on the Enhancement of Wind and Significant Wave Height in Indonesian Maritime Areas
No full textExtreme weather characterized by strong winds and high waves poses serious risks to maritime safety in the Indonesian Maritime Continent. This study comprehensively quantifies the impact of Mesoscale Convective Systems (MCS) on Significant Wave Height and Wind Speed enhancements using Himawari-8/9 imagery and multi-mission satellite altimetry. By utilizing a robust 300-km spatial filter to isolate standalone offshore events, the research employs quantile-based and multi-classification approaches to evaluate spatial and temporal anomalies. The results indicate that ocean-atmosphere coupling is highly non-linear; multilinear regression models consistently failed to predict environmental anomalies (R2<0.03). However, quantile analysis revealed critical threshold behaviors. Spatially, intense systems consistently drove higher positive anomalies. Significantly, a multi-classification interaction analysis demonstrated that the synergistic combination of the "strongest" MCS parameters (largest area, coldest temperatures, and highest top cloud) generated wave impacts eight times greater than the "weakest" systems. Conversely, temporal responses relative to a 24-hour baseline were complex and non-monotonic, suggesting the dominance of lag effects over instantaneous intensity. These findings provide quantitative evidence that MCS modulates air-sea interactions through complex non-linear mechanisms, demonstrating that operational maritime forecasting must adopt advanced probabilistic modeling rather than relying on simple linear parameterizations to capture these hazardous events
Public participation and willingness to participate in marine debris management in Batam City: A preliminary assessment using an open-ended contingent valuation approach
No full textMarine debris has become a growing environmental challenge in Batam City, driven by rapid coastal development, increasing tourism, and cross-border waste transport within the Singapore Strait. Understanding community engagement is critical for designing effective waste reduction strategies, yet empirical evidence on local participation remains limited. This preliminary study assesses public willingness to participate (WtP) in marine debris management using an open-ended contingent valuation method (CVM). A total of 70 respondents from five coastal zones Nongsa, Marina, Tanjung Pinggir, Barelang, and Batu Ampar were surveyed in September 2025 through face-to-face interviews. WtP was measured in terms of hours per month that individuals were willing to contribute to beach clean-ups, environmental monitoring, educational campaigns, and related activities. The results show an average willingness to participate of 5.68 hours/month/person, with a median and mode of 2 hours, indicating a highly skewed distribution driven by a small group of highly motivated individuals. Chi-Square analysis reveals that age is the only demographic factor significantly associated with participation levels (p = 0.008), while gender, occupation, and income show no significant relationship. The participation curve follows an exponential decline, and the marginal willingness to participate evaluated at the mean is approximately 0.80 persons per additional hour, suggesting that higher time commitments sharply reduce overall community engagement
Implementation of Preparedness Efforts for Rip Current Hazard at Pangandaran Beach
No full textRip currents are powerful current that flow from the shore through the surf zone and out to deeper waters. These currents play a critical role in coastal dynamics but they are also of significant concern for beach safety, as they pose a threat to swimmers and beachgoers. The number of fatalities due to rip current every year at Pangandaran Beach raises the question of whether the beach management is ready to face this hazard. Therefore, this study aims to assess the preparedness efforts against rip current hazard in Pangandaran Beach. The benchmarks of preparedness that will be studied include trainings, campaigns, supporting instruments, and beach closure. This qualitative research will use data from interviews with BPBD, Balawista, and Basarnas of Pangandaran Beach, supported by data from field observation. The findings indicate four criteria and eighteen metrics for evaluating rip current preparedness, applicable to beach stakeholders. The stakeholders engaged in rip current hazard management at Pangandaran Beach have undertaken preparedness initiatives that can be categorized as adequately effective. The execution of both routine and non-routine training, the distribution of campaigns via several media, the acquisition of supporting equipment, and the closure of the beach have been completed to date. Nevertheless, feedback concerning the quantity and condition of warning signs, together with the enhancement of materials and platforms for distributing information specifically about rip currents, requires further refinement