The Joint Accelerator Conferences Website (JACoW) is an international collaboration that publishes the proceedings of accelerator conferences held around the world.
TY - CONF AU - Hanten, J.H. AU - Arnold, M. AU - Birkhan, J. AU - Caliari, C. AU - Pietralla, N. AU - Steinhorst, M. ED - Schaa, Volker RW ED - Jansson, Andreas ED - Shea, Thomas ED - Olander, Johan TI - Enhancement of the S-DALINAC Control System with Machine Learning Methods J2 - Proc. of IBIC2019, Malmö, Sweden, 08-12 September 2019 CY - Malmö, Sweden T2 - International Beam Instrumentation Conferenc T3 - 8 LA - english AB - For the EPICS-based control system of the superconducting Darmstadt electron linear accelerator S-DALINAC**, supporting infrastructures based on machine learning are currently developed. The most important support for the operators is to assist the beam setup and controlling with reinforcement learning using artificial neural networks. A particle accelerator has a very large parameter space with often hidden relationships between them. Therefore neural networks are a suited instrument to use for approximating the needed value function which represents the value of a certain action in a certain state. Different neural network structures and their training with reinforcement learning are currently tested with simulations. Also there are different candidates for the reinforcement learning algorithms such as Deep-Q-Networks (DQN) or Deep-Deterministic-Policy-Gradient (DDPG). In this contribution the concept and first results will be presented. PB - JACoW Publishing CP - Geneva, Switzerland SP - 475 EP - 478 KW - network KW - target KW - controls KW - linac KW - electron DA - 2019/11 PY - 2019 SN - 2673-5350 SN - 978-3-95450-204-2 DO - doi:10.18429/JACoW-IBIC2019-WEBO04 UR - http://jacow.org/ibic2019/papers/webo04.pdf ER -