However, variations in beam pointing, known as “jitter,” caused by mechanical vibrations, hinder laser performance and prevent advances in these applications.
Dan Wang, research scientist at ATAP's Berkeley Accelerator Controls & Instrumentation program and one of the paper's lead authors, said:
However, traditional laser control systems “has struggled to keep up with rapid changes in laser position, particularly using large, slowly varying optical components used in high power, low relocation rate laser lasers,” according to Anthony Gonsalves, staff scientist and deputy director of the experiment at ATAP's Bella Center. “This will result in shot-to-shot errors that have a negative impact on the experiment.”
To overcome this limitation, the team turned to machine learning.
Unlike traditional control systems that fix laser pointing errors after they occur, “our method predicts jitter, adjusts the laser's optical components in real time, rapidly improving shot-to-shot stabilization and more accurate beam pointing,” explained Alessio Amodio, an electronics engineer in Bella Center and Engineering.
“Pilot” beam and real-time adjustment
An hour-long comparison of Freerun and ML corrected jitter (credit: Dan Wang/Berkeley Lab)
To test the effectiveness of this method, researchers adopted a low-power, high-compensation rate “pilot” laser beam as the proxy for the high-power, low-recovery rate main beam of the LPA's main research facility, Bella Petawatt laser.
“The pilot beam is fired much more frequently than the main beam, allowing you to map the movement of the beam caused by the vibrations of the mirror,” Gonsalves said. “This information can be used to predict where the beam will be when the high-power pulse arrives. Because you know the pointing errors in advance, you can adjust the mirror to correct these errors.”
They fed this position data to an ML-enabled control system and adjusted the beam pointing using a correction mirror. After testing performance, the system reduced jitter by 65% in the X direction of the beam and 47% in the Y direction.
“We plan to enhance our methods using a field-programmable gate array, electronic control circuitry that provides advanced timing and synchronization, and methods that allow for faster, more accurate real-time corrections,” Wang said. “This is expected to improve shot laser stabilization from shots as testing is planned with Verapetawatt lasers in full power and a wide range of applications.”
Research and development programs supervised by the Department of Energy's Science Bureau, the Office of High Energy Physics, and the Berkeley Institute's Institute of Laboratory support in this research.
###
The Lawrence Berkeley National Laboratory (Berkeley Lab) is committed to groundbreaking research focusing on discovery science and solutions for a rich and reliable energy supply. Lab's expertise spans materials, chemistry, physics, biology, earth and environmental science, mathematics, and computing. Researchers around the world rely on the lab's world-class scientific facilities for their own pioneering research. It was founded in 1931 with the belief that the biggest issues are best dealt with by teams. The Berkeley Lab and its scientists are recognized by 16 Nobel Prizes. Berkeley Lab is a multiprogram national laboratory managed by the Department of Energy Science and Sciences, University of California.
The DOE's Science Bureau is the sole and largest advocate for basic research in the US physical sciences and is working to address some of the most pressing challenges of our time. For more information, please visit Energy.gov/science.
