SLAC Public Lecture Series
Super-Human Operator: Controlling Accelerators with Machine Learning
Particle accelerators are used every day in a wide range of scientific, medical and industrial applications. But did you know that the task of operating these machines is far from mundane? For example, for every experiment at SLAC’s X-ray laser, the Linac Coherent Light Source, human operators regularly adjust several dozen variables to carefully shape the beam, bring it to the correct energy for that particular experiment and maintain stable operation. This is no small feat for a beam that has to travel about a mile and go through plenty of nonlinear “beam gymnastics” along the way! Accelerator researchers are starting to turn to machine learning to see if we can make it easier to create new types of challenging beam setups and to speed up routine tuning. This public lecture will take you on a journey through accelerator tuning and discuss some of the ways accelerator researchers are starting to use machine learning to help out with this challenging task.