PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
August 30, 2026Energy Conversion and EconomicsOpen Access

An edge AI‐powered power converter hardware platform for practical end‐to‐end AI‐to‐converter closed‐loop deployment

View Full Paper
Ask AI
Bookmark
Share

Authors

PSPingyang SunFZFan ZhangZQZihang Qiu

Discussion

Loading...

Member takes

Overview

Experimental study demonstrates real-time closed-loop control in interconnected power converters, indicating the feasibility of end-to-end edge artificial intelligence for adaptive grid dispatch.

Key Points

  • To design and experimentally validate an edge AI-powered hardware platform that enables direct, closed-loop execution of artificial intelligence control commands on physical power converters.
  • Interconnected physical AC/DC and DC/DC power converters with an NVIDIA Jetson Orin Nano edge processing unit via the industrial CAN protocol for bidirectional data transfer.
  • Tested single- and multi-converter systems across three AI implementations: an offline multilayer perceptron for power sharing, an offline spatio-temporal graph neural network for solar and load forecasting, and an online bidirectional LSTM for battery dispatch.
  • Successfully established an end-to-end AI-to-converter closed-loop framework capable of directly driving physical converter hardware without relying on cloud processing.
  • Achieved verified algorithm delivery across all test scenarios, confirming functional power-sharing estimation, photovoltaic generation forecasting, and real-time peak-shaving battery management.

Cite This Study

Sun et al. (2026) studied this question.

synapsesocial.com/papers/6a93f1266c1a8fb52e79de22https://doi.org/10.1049/enc2.70048
View Full Paper
Ask AI
Bookmark
Share