Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Patch-MLP-Based Predictive Control: Simulation of Upstream Pointing Stabilization for PHELIX Laser System

Jiaying Wang, Jonas Benjamin Ohland, Yen-Yu Chang, Vedhas Pandit, Stefan Bock, Andrew-Hiroaki Okukura, Udo Eisenbarth, Arie Irman, Michael Bussmann, Ulrich Schramm, Jeffrey Kelling

Abstract

High-energy laser facilities such as PHELIX at GSI require excellent beam pointing stability for reproducibility and relative independence for future experiments. Beam pointing stability has been traditionally achieved using simple proportional-integral-derivative (PID) control which removes the problem of slow drift, but is limited because of the time delay in knowing the diagnosis and the inertia in the mechanical system associated with mirrors. In this work, we introduce a predictive control strategy where the forecasting of beam pointing errors is performed by a patch-based multilayer perceptron (Patch-MLP) designed to capture local temporal patterns for more robust short-term jitter prediction. The subsequent conversion of these predicted errors into correction signals is handled by a PID controller. The neural network has been trained on diagnostic time-series data to predict beam pointing error. Using the feed-forward controller compensates for system delays. Simulations with a correction mirror placed upstream of the PHELIX pre-amplifier bridge confirm that the predictive control scheme reduces residual jitter compared to conventional PID control. Over a 10-hour dataset the controller maintained stable performance without drift, while standard pointing metrics showed consistent improvements of the order of 10 to 20 percent. The predictive controller operates without drift, and therefore may improve reproducibility and operational efficiency in high energy, low repetition rate laser experiment conditions.

Patch-MLP-Based Predictive Control: Simulation of Upstream Pointing Stabilization for PHELIX Laser System

Abstract

High-energy laser facilities such as PHELIX at GSI require excellent beam pointing stability for reproducibility and relative independence for future experiments. Beam pointing stability has been traditionally achieved using simple proportional-integral-derivative (PID) control which removes the problem of slow drift, but is limited because of the time delay in knowing the diagnosis and the inertia in the mechanical system associated with mirrors. In this work, we introduce a predictive control strategy where the forecasting of beam pointing errors is performed by a patch-based multilayer perceptron (Patch-MLP) designed to capture local temporal patterns for more robust short-term jitter prediction. The subsequent conversion of these predicted errors into correction signals is handled by a PID controller. The neural network has been trained on diagnostic time-series data to predict beam pointing error. Using the feed-forward controller compensates for system delays. Simulations with a correction mirror placed upstream of the PHELIX pre-amplifier bridge confirm that the predictive control scheme reduces residual jitter compared to conventional PID control. Over a 10-hour dataset the controller maintained stable performance without drift, while standard pointing metrics showed consistent improvements of the order of 10 to 20 percent. The predictive controller operates without drift, and therefore may improve reproducibility and operational efficiency in high energy, low repetition rate laser experiment conditions.

Paper Structure

This paper contains 15 sections, 9 equations, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methodology
  3. Workflow of the simulation
  4. Patch MLP
  5. Input and Preprocessing
  6. Patch Partitioning and Embedding
  7. Feature Concatenation and MLP
  8. Loss function
  9. Sliding-Window Prediction
  10. Simulation Results and Discussion
  11. Comparison of simulation results after compensation
  12. Model stability
  13. Result assessment
  14. Failure cases
  15. Conclusion

Figures (8)

  • Figure 1: Simplified layout of the PHELIX major2024phelix laser system highlighting the dominant source of pointing jitter (pre-amplifier bridge) and the location of the pointing measurement (Main Amplifier Sensor, MAS). These positions provide the physical background for the simulation study presented in this work.
  • Figure 2: Workflow of the Hybrid Feed-Forward MLP-Driven PID and Traditional PID Control. The inset shows the step-response of the steering mirror model used in our simulation.
  • Figure 3: Patch-based MLP Architecture for Two-Axis Beam Pointing Prediction
  • Figure 4: One-minute beam-center trajectories (X and Y): real signal vs. traditional PID (0.6 ms delay) vs. MLP-driven PID.
  • Figure 5: Comparison of beam stabilization: (a) beam-center distribution and (b) power spectral density.
  • ...and 3 more figures