Scholarlink

<p>We present the first step towards an intelligent thin-disk multi-pass amplifier (I-TDMPA), by simulatively proving the ability of a neural network (NN) for the detection of misalignment. A digital twin of one of our experimental systems was implemented using a specifically developed ray tracing simulation. The digital twin was used to generate more than 200,000 sets of training data to train a neural network (NN) with the goal to be able to determine misalignment of the many optical elements of a multi-pass amplifier in a simulation model, solely by observing the beam. The predictions of the NN agree very well with the simulated misalignments, which confirms the suitability of artificial-intelligence methods to ensure a safe and stable operation of complex optical systems.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Determination of misalignment in a multi-pass amplifier using a neural network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月24日

查看全文
  • Florian Bienert,
  • Christian Schmittner,
  • Christoph Röcker,
  • Andreas Michalowski,
  • Thomas Graf,
  • Marwan Abdou Ahmed

摘要

We present the first step towards an intelligent thin-disk multi-pass amplifier (I-TDMPA), by simulatively proving the ability of a neural network (NN) for the detection of misalignment. A digital twin of one of our experimental systems was implemented using a specifically developed ray tracing simulation. The digital twin was used to generate more than 200,000 sets of training data to train a neural network (NN) with the goal to be able to determine misalignment of the many optical elements of a multi-pass amplifier in a simulation model, solely by observing the beam. The predictions of the NN agree very well with the simulated misalignments, which confirms the suitability of artificial-intelligence methods to ensure a safe and stable operation of complex optical systems.