An official website of the United States government
This paper compares three different dc-dc topologies, i.e. boost converter, three-level flying capacitor multilevel converter (FCMC) and one-cell switching tank converter (STC) for a 100 kW electric vehicle power electronic system. This bidirectional dc-dc converter targets 300 V - 600 V voltage conversion. Total semiconductor loss index (TSLI) has been proposed to evaluate topologies and device technologies. The boost converter and one-cell STC have been fairly compared by utilizing this index. The simulation results of a 100 kW one-cell STC working at zero current switching (ZCS) mode have been provided. A 100 kW hardware prototype using 1200 V 600 A SiC power module has been built. The estimated efficiency is about 99.2% at 30 kW, 99.13% at half load, and 98.64% at full load. The power density of the main circuits is about 42 kW/L
more »
« less
2024 IEEE International Communications Energy Conference (INTELEC)
This work introduces a machine learning approach for developing Digital Twins (DTs) for DC-DC converters, focusing on in-situ implementation in real-world operational conditions. A system based on a boost converter has been developed in MATLAB Simulink. To mirror real-world scenarios, commercial datasheets along with a range of input parameters, health degradation elements, temperature influence, and random noises have been considered. The study employs Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN) for predicting critical circuit responses of the boost converter, including inductor current, output voltage, and efficiency. Investigations show that MLP performs relatively poorly in the presence of noise. The CNN and RNN outperform the MLP under various noise levels, with the RNN exhibiting the best performance. This work advances DTs technology in power electronics, aiming to improve converter system optimization and enable predictive maintenance.
more »
« less
- Award ID(s):
- 2239966
- PAR ID:
- 10577639
- Publisher / Repository:
- IEEE
- Date Published:
- ISBN:
- 979-8-3503-7057-7
- Format(s):
- Medium: X
- Location:
- Bengaluru, India
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This paper investigates integration of distributed energy resources (DERs) in microgrids (MGs) through two-stage power conversion structures consisting of DC-DC boost converter and DC-AC voltage source converter (VSC) subsystems. In contrast to existing investigations that treated DC-link voltage as an ideal constant voltage, this paper considers the non-ideal dynamic coupling between both subsystems for completeness and higher accuracy, which introduces additional DC-side dynamics to the VSC. The analysis shows parameters of the boost converter's power model that impact stability through the DC-link. Carefully selecting these parameters can mitigate this effect on stability and improve dynamic performance across the DC-link. Hence, an optimization framework is developed to facilitate in selecting adequate boost converter parameters in designing a stable voltage source converter-based microgrid (VSC-MG). The developed optimization framework, based on particle swarm optimization, considers dynamic coupling between both subsystems and is also effective in avoiding inadequate boost converter parameters capable of propagating instability through the DC-link to the VSC. Simulations are performed with MATLAB/Simulink to validate theoretical analyses.more » « less
-
null (Ed.)This paper presents a reverse electrowetting-on-dielectric (REWOD) energy harvester integrated with rectifier, boost converter, and charge amplifier that is, without bias voltage, capable of powering wearable sensors for monitoring human health in real-time. REWOD has been demonstrated to effectively generate electrical current at a low frequency range (< 3 Hz), which is the frequency range for various human activities such as walking, running, etc. However, the current generated from the REWOD without external bias source is insufficient to power such motion sensors. In this work, to eventually implement a fully self-powered motion sensor, we demonstrate a novel bias-free REWOD AC generation and then rectify, boost, and amplify the signal using commercial components. The unconditioned REWOD output of 95–240 mV AC is generated using a 50 μL droplet of 0.5M NaCl electrolyte and 2.5 mm of electrode displacement from an oscillation frequency range of 1–3 Hz. A seven-stage rectifier using Schottky diodes having a forward voltage drop of 135–240 mV and a forward current of 1 mA converts the generated AC signal to DC voltage. ∼3 V DC is measured at the boost converter output, proving the system could function as a self-powered motion sensor. Additionally, a linear relationship of output DC voltage with respect to frequency and displacement demonstrates the potential of this REWOD energy harvester to function as a self-powered wearable motion sensor.more » « less
-
This paper presents a reverse electrowetting-on-dielectric (REWOD) energy harvester integrated with rectifier, boost converter, and charge amplifier that is, without bias voltage, capable of powering wearable sensors for monitoring human health in real-time. REWOD has been demonstrated to effectively generate electrical current at a low frequency range (<3 Hz), which is the frequency range for various human activities such as walking, running, etc. However, the current generated from the REWOD without external bias source is insufficient to power such motion sensors. In this work, to eventually implement a fully self-powered motion sensor, we demonstrate a novel bias-free REWOD AC generation and then rectify, boost, and amplify the signal using commercial components. The unconditioned REWOD output of 95-240 mV AC is generated using a 50 μL droplet of 0.5M NaCl electrolyte and 2.5 mm of electrode displacement from an oscillation frequency range of 1-3 Hz. A seven-stage rectifier using Schottky diodes having a forward voltage drop of 135-240 mV and a forward current of 1 mA converts the generated AC signal to DC voltage. ~3 V DC is measured at the boost converter output, proving the system could function as a self-powered motion sensor. Additionally, a linear relationship of output DC voltage with respect to frequency and displacement demonstrates the potential of this REWOD energy harvester to function as a self-powered wearable motion sensor.more » « less
-
Federated Learning (FL) enables collaborative model training across distributed devices while safeguarding data and user privacy. However, FL remains susceptible to privacy threats that can compromise data via direct means. That said, indirectly compromising the confidentiality of the FL model architecture (e.g., a convolutional neural network (CNN) or a recurrent neural network (RNN)) on a client device by an outsider remains unexplored. If leaked, this information can enable next-level attacks tailored to the architecture. This paper proposes a novel side-channel fingerprinting attack, leveraging flow-level and packet-level statistics of encrypted wireless traffic from an FL client to infer its deep learning model architecture. We name it FLARE, a fingerprinting framework based on FL Architecture REconnaissance. Evaluation across various CNN and RNN variants—including pre-trained and custom models trained over IEEE 802.11 Wi-Fi—shows that FLARE achieves over 98% F1-score in closed-world and up to 91\% in open-world scenarios. These results reveal that CNN and RNN models leak distinguishable traffic patterns, enabling architecture fingerprinting even under realistic FL settings with hardware, software, and data heterogeneity. To our knowledge, this is the first work to fingerprint FL model architectures by sniffing encrypted wireless traffic, exposing a critical side-channel vulnerability in current FL systems.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Proceeding:
- https://doi.org/10.1109/INTELEC60315.2024
