Machine learning based device modeling and performance optimization for dual gate FinFET Prashanth Kumar, R Kiran Kumar, Vinod A, Ravi Tiwari, V Shalini, P Anusuya Engineering Research Express, 2026 In this research, we present a unique machine learning (ML)-based pipeline for simulating the I – V (current–voltage) characteristics of dual-gate fin field-effect transistor (FinFET) devices and improving them. The creative method builds models to effectively and precisely predict crucial outputs while maintaining the most reliable ML model. Several key input elements used in this work, including work function, fin height, temperature, and doping concentration, define the FinFET structure form and related doping profile. Another way we depict this is by splitting the notion of several variables and then the interactions between them in our dataset model. The ML models are then trained using this data to forecast the I – V behavior of FinFET devices. The current–voltage response is correctly predicted by the training model and has a strong correlation with traditional technology computer aided design (TCAD) simulations. Device model: less reliant on TCAD usage and time, more predictable. Fast and accurate DC characteristic forecasts are made possible by well-trained ML, which makes it a very dependable maintenance tool. The findings reveal that ML might be used to model the design and performance of advanced FinFET technologies and show that our framework achieves a strong optimization in the process with maintained high accuracy.
Exploring battery-powered electric vehicles: Energy sources, configurations, and management approaches Ravi Tiwari, Akilan K, Harish S, Lalit R, Prashanth Kumar Next Energy, 2026 Electric vehicles (EVs) have been widely studied as a promising option to mitigate the greenhouse effect. With improvements in drivetrain electricity, energy storage, and assistance, plug-in hybrid electric vehicles (PHEVs) offer competitive range and fuel efficiency compared to internal combustion engine vehicles. By operating with optimized control strategies or using energy management system (EMS) concepts, the efficiency of PHEVs can be significantly improved. This research study investigates advances in EV technology, with an emphasis on battery breakthroughs, charging infrastructure, and environmental implications. As the global need for sustainable transportation solutions grows, EVs have emerged as a critical component in lowering greenhouse gas emissions and reliance on fossil fuels. The study looks at current advances in lithium-ion and solid-state batteries, emphasizing improvements in energy density, charging speed, and lifecycle sustainability. Furthermore, the report examines the existing state of charging infrastructure, identifying problems and potential to improve accessibility and efficiency. This study emphasizes the importance of legislative frameworks and technology collaboration in advancing the adoption of EVs by conducting a thorough analysis of existing literature and case studies. This review paper provides a comprehensive analysis of EMSs in EVs, focusing particularly on hybrid energy storage systems (HESS). Our original contributions include: 1) A comparative analysis of various HESS configurations (e.g., Battery-Supercapacitor) under different driving cycles. 2) Identification and detailed summary of the current technological and control challenges faced by EV EMSs. 3) A summary of the most recent technological advances in battery management systems and power-split control strategies for enhanced efficiency and battery lifespan. • Comprehensive review of battery and hybrid energy storage systems for electric vehicles. • Presents comparative analysis of HESS architectures under varying driving conditions. • Details advances in energy management strategies and control methods for efficiency. • Highlights the challenges of battery recycling, charging infrastructure, and grid impact. • Offers actionable recommendations for future EV policy, design, and standardization.
Dielectrically-Modulated Stacked Nanosheet Schottky TFETs for Biosensing Applications Shalini Virumandi, Tanmoy Majumder, Prashanth Kumar Semiconductors, 2026 This work presents the design and analysis of a high-sensitivity Nanosheet Schottky Tunnel Field-Effect Transistors (NS-STFETs) for biosensing applications. The proposed device leverages Schottky source/drain contacts and stacked nanosheet geometry to enhance electrostatic control and tunneling efficiency. Device performance was investigated using Sentaurus TCAD simulations. The biosensor demonstrates promising characteristics, including an ION of 2.43 × 10–5 A, IOFF of 1.33 × 10–18 A, and ION/IOFF of 1013 for protein biomolecules with a dielectric constant of k = 8. Further analysis was carried out to evaluate the influence of biomolecules with varying dielectric constants (k = 1 to 20), different fill factors (25–100%), and positively/negatively charged biomolecules with charge densities from –1011 to +1011. The results confirm that higher fill factors and charged biomolecules have a significant impact on sensitivity, while the device maintains stable operation under various conditions. Overall, the NS-STFET biosensor exhibits high sensitivity, low power consumption, and scalability, making it a strong candidate for next-generation label-free diagnostic platforms.
Temperature-Induced Changes in Multifin-Schottky Barrier FinFETs: An Analog/RF Linearity Investigation V Shalini, Prashanth Kumar Advanced Theory and Simulations, 2025 In this script, a Gallium Nitride (GaN)‐based FinFET structure is proposed with a multi‐channel device that is designed and simulated. Here, the 3D‐Sentaures TCAD simulator is used to investigate the analog/radio frequency performance and linearity of the MultiFin‐Schottky Barrier FinFET with different temperatures of 100–400 K. The proposed device underwent a temperature analysis, where critical parameters include drain current, ION/IOFF ratio, Transconductance (gm), higher‐order terms (gm2 and gm3), Gain Bandwidth Product (GBP), Cut‐off Frequency (fT), Transit Time (τ), Transconductance Generation Factor (TGF), Transconductance Frequency Product (TFP), Voltage Input Intercept Point (VIP2, VIP3), Input Intercept Point (IIP3), and Third Order Intermodulation Distortion (IMD3) is thoroughly examined. Thus, the proposed GaN‐based FinFET validates as a strong potential contender for GaN‐based analog/RF applications.
Performance and Reliability Assessment of Schottky Complementary Multi-FinFET Inverter for Advanced Scaling Nodes Shalini Virumandi, Prashanth Kumar IEEE Access, 2025 This study investigates the performance of Schottky Complementary Multi-FinFET inverter for advanced technology nodes of 7 nm, 5 nm, and 3 nm using Sentaurus TCAD simulations. The impact of scaling on key device parameters, including threshold voltage (V<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sub>), subthreshold swing (SS), drain-induced barrier lowering (DIBL), ON-state current (ION), and OFF-state current (IOFF), is systematically investigated. The results indicate that the 5 nm node exhibit superior performance compared to other nodes, with improved electrostatic control, higher drive current (0.00212 for n-FinFET, -1.16x10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-4</sup> for p-FinFET), and lower leakage current, making them suitable for low-power applications. Further, Voltage Transfer Characteristics (VTC) analysis was performed for the 5 nm node, demonstrating sharp transitions with well-defined logic levels and strong voltage gain. Additionally, transient analysis confirms high-speed switching operation with minimal propagation delay, highlighting the feasibility of Schottky-based FinFET inverters for next-generation ultra-scaled digital circuits. Furthermore, a comprehensive reliability assessment was carried out for the 5 nm node, focusing on analog/RF performance and VTC characteristics by varying temperatures from 100 K to 250 K. The study reveals that as temperature increases, there are notable shifts in key parameters, affecting the overall stability and efficiency of the device. The temperature-dependent analysis highlights the robustness of the Schottky-contacted FinFET inverter and provides critical insights into its applicability for future high-performance and energy-efficient logic applications under varying thermal conditions. These findings make the proposed device highly suitable for applications in microprocessors, data converters, IoT systems, and cryogenic computing platforms.
Design and Implementation of Adaptive EV Battery Management System for Dual-Mode Charging Based on Temperature Estimation of IoT Monitoring Sanjai. M, Arivarsi A, Prashanth Kumar, Visves N M, Sethupathy R Proceedings of IEEE International Conference on Modelling Simulation and Intelligent Computing Mosicom 2025, 2025 Today, batteries find a wide range of applications in portable electronics and electric vehicles for the growing demand of the world for efficient and reliable energy storage. Controlled charging/discharging is an essential way to ensure good performance and prolong battery life. However, most of the traditional charging schemes do not consider thermal factors affecting the electrochemical processes of the battery, which often causes overheating, reduced cycle life, and safety hazards. Most of the research done on BMS has been related to the regulation of voltage and current, without considering adaptive control for temperature fluctuations. Though certain works incorporate temperature monitoring at a basic level, dynamic adaptability with predictive temperature estimation to optimize charging performance is missing in them. This forms the key void in the design research of state-of-the-art BMS. In view of these shortcomings, the present study proposes an intelligent adaptive BMS that will be capable of dynamically switching between fast and slow charging modes based on realtime internal temperature estimation. It uses a dual-adapter configuration <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(12 \mathrm{V}, 3 \mathrm{A}$</tex> for fast charging and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$12 \mathrm{V}, 1 \mathrm{A}$</tex> for slow charging) and estimates internal temperature from surface temperature, current, and voltage readings using the Kalman Filter algorithm. Unlike most conventional systems, this predictive model eliminates sensor noise and provides accurate temperature estimation without invasive measurement. The ESP32 microcontroller will act as the core controller in running the Kalman Filter algorithm, managing the switching by relays, and allowing IoT-based runtime monitoring of the battery parameters. The key novelty of this work consists in an adaptive, temperature-aware charging mechanism combined with IoTenabled monitoring and Kalman Filter-based estimation of the temperature, thus providing a reliable and efficient solution for next-generation electric vehicle battery systems.
Study on recent trends of smart wearable N Anusha, B Prashanth Kumar, Lucky Agarwal 2022 2nd International Conference on Artificial Intelligence and Signal Processing Aisp 2022, 2022
