Physics-Informed Drift Diagnosis for Laser-Plasma Accelerator Operations

Abstract

Laser-plasma accelerators (LPAs) sustain accelerating gradients of order 100 GV/m, but routine operation remains difficult: electron beam metrics drift over an operating shift, and the root physical cause is often invisible to the available diagnostics. We formulate LPA operation as a latent state-space model in which three effective interaction-point variables, the normalized laser amplitude a₀, the normalized plasma electron density ñ_e and the residual pulse chirp 𝒞, are inferred from routine electron beam observations by an extended Kalman filter. The emission model, which maps the latent state to the diagnostics, is kept structurally separate from the {transition} model, which describes how the latent state evolves between shots. The separation supports diagnosis in two stages, one asking which latent variable moved and one asking what moved it. The implemented emission model is a toy model, yielding an expected performance in line with current facilities and using 3D blow-out regime dependencies where relevant. We conduct synthetic sessions to test the effectiveness of the detection and attribution protocols, finding that attribution is limited by excitation rather than by shot count or diagnostic resolution. Because the construction needs only a set of physical latent variables, an emission model and a family of hardware-derived transition models, it transfers to other drift-prone subsystems. We argue that the accuracy of the whole procedure is limited by the emission model rather than by the inference method.

Experimental validation of a fast control-oriented, physics-informed surrogate model for plasma equilibrium reconstruction in the TCV tokamak
M. Grandin, A. Mele, C. Galperti +5
arXiv23 Sep 2026PreprintarXiv licenceFull textPlasma physicsarXiv
A Two-Dimensional Physics-Informed Neural Network Model of a Hall Effect Thruster Plasma: Self-Consistent Electrostatics and Physical-Consistency Diagnostics
Josh Adjedj, Subrata Roy
arXiv3 Nov 2026PreprintarXiv licenceFull textAtomic, molecular & optical physics Plasma physicsarXiv
Brilliance: toward autonomous computational research in particle accelerator physics
Jinyu Wan, Yu Zhao, Wei Bao +4
arXiv28 Oct 2026PreprintarXiv licenceFull textApplied physics & instrumentationarXiv
Reveal normal form structure for nonlinear map in accelerator beam physics with symplectic neural network
Leo He, Yue Hao
arXiv5 Oct 2026PreprintCC BY 4.0Full textApplied physics & instrumentationarXiv
Direct observation of electron shedding from a laser-plasma accelerator
Sheroy Tata, Salome Benracassa, Heychal Davidovich +6
arXiv14 Sep 2026PreprintarXiv licenceFull textApplied physics & instrumentation Plasma physicsarXiv

Cite

BibTeX RIS

BibTeX

@misc{labun2026physicsinformed,
  title = {{Physics-Informed Drift Diagnosis for Laser-Plasma Accelerator Operations}},
  author = {Ou Labun and Calin Hojbota and Mara Klebonas and Mike Downer and Rafal Zgadzaj and Phil Franke and Lance Labun},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.11874},
  eprint = {2609.11874},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.11874},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Physics-Informed Drift Diagnosis for Laser-Plasma Accelerator Operations
AU  - Ou Labun
AU  - Calin Hojbota
AU  - Mara Klebonas
AU  - Mike Downer
AU  - Rafal Zgadzaj
AU  - Phil Franke
AU  - Lance Labun
PY  - 2026
DA  - 2026/09/18
PB  - arXiv
DO  - 10.48550/arxiv.2609.11874
UR  - https://arxiv.org/abs/2609.11874
AB  - Laser-plasma accelerators (LPAs) sustain accelerating gradients of order 100 GV/m, but routine operation remains difficult: electron beam metrics drift over an operating shift, and the root physical cause is often invisible to the available diagnostics. We formulate LPA operation as a latent state-space model in which three effective interaction-point variables, the normalized laser amplitude a_0, the normalized plasma electron density ñ_e and the residual pulse chirp 𝒞, are inferred from routine electron beam observations by an extended Kalman filter. The emission model, which maps the latent state to the diagnostics, is kept structurally separate from the {transition} model, which describes how the latent state evolves between shots. The separation supports diagnosis in two stages, one asking which latent variable moved and one asking what moved it. The implemented emission model is a toy model, yielding an expected performance in line with current facilities and using 3D blow-out regime dependencies where relevant. We conduct synthetic sessions to test the effectiveness of the detection and attribution protocols, finding that attribution is limited by excitation rather than by shot count or diagnostic resolution. Because the construction needs only a set of physical latent variables, an emission model and a family of hardware-derived transition models, it transfers to other drift-prone subsystems. We argue that the accuracy of the whole procedure is limited by the emission model rather than by the inference method.
ER  -

Abstract

Related

Cite