arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.23141physics.plasm-phcs.LG

基于FPGA加速的量化递归概率神经网络的实时等离子体状态预测

Real-Time Plasma State Prediction via FPGA-Accelerated Quantized Recurrent Probabilistic Neural Networks

Daniel Gaytan-Villarreal, Aiken Xie, Tu Pham, Rohit Sonker, Chiara Amendola, Matteo Cremonesi, Cong Hao, Jeff Schneider

首次发表
浏览论文内容

中文总结 AI 辅助

针对托卡马克等离子体控制系统的实时延迟挑战,提出基于FPGA的量化递归概率神经网络端到端部署流程,实现亚10微秒延迟并满足资源预算。

中文摘要 AI 辅助

托卡马克装置等离子体控制的实时状态估计因等离子体控制系统(PCS)的严格延迟要求而具有挑战性。我们提出了一种在FPGA硬件上部署递归概率神经网络(RPNN)的端到端工作流程。我们将架构规模缩减与通过QKeras进行的量化感知训练相结合。随后,使用hls4ml对模型进行综合,目标设备为Xilinx Alveo U50。我们报告了一种设计,该设计在确定性亚10微秒单时间步延迟下,舒适地满足所有四个资源预算(DSP、LUT、FF、BRAM),满足模型预测控制式等离子体控制回路内实时推理的要求。

英文摘要

Real time plasma state estimation for control of Tokamak devices are challenging due to the stringent latency requirements of the plasma control system (PCS). We present an end-to-end workflow for deploying a recurrent probabilistic neural network (RPNN) on FPGA hardware. We combine architecture size reduction with quantization-aware training via QKeras. The model is then synthesized using hls4ml, targeting a Xilinx Alveo U50 device. We report a design that fits comfortably within all four resource budgets (DSP, LUT, FF, BRAM) at deterministic sub-10~$μ$s single-timestep latency, meeting the requirements for real-time inference inside a model-predictive-control-style plasma control loop.

发表机构

  • Carnegie Mellon University(卡内基梅隆大学)
  • Columbia University(哥伦比亚大学)
  • Georgia Institute of Technology(佐治亚理工学院)

机构由 AI 辅助整理,请以论文原文为准。

↑