The Joint Accelerator Conferences Website (JACoW) is an international collaboration that publishes the proceedings of accelerator conferences held around the world.
TY - CONF AU - Kong, Y.B. AU - Hur, M.G. AU - Lee, E.J. AU - Park, J.H. AU - Park, Y.D. AU - Yang, S.D. ED - Chrin, Jan ED - Schaa, Volker RW TI - Neural Network Based Generalized Predictive Control for RFT-30 Cyclotron System J2 - Proc. of Cyclotrons2016, Zurich, Switzerland, September 11-16, 2016 C1 - Zurich, Switzerland T2 - International Conference on Cyclotrons and Their Applications T3 - 21 LA - english AB - Beamline tuning is time consuming and difficult work in accelerator system. In this work, we propose a neural generalized predictive control (NGPC) approach for the RFT-30 cyclotron beamline. The proposed approach performs system identification with the NN model and finds the control parameters for the beamline. Performance results show that the proposed approach helps to predict optimal parameters without real experiments with the accelerator. PB - JACoW CP - Geneva, Switzerland SP - 212 EP - 214 KW - controls KW - cyclotron KW - network KW - simulation KW - target DA - 2017/01 PY - 2017 SN - 978-3-95450-167-0 DO - 10.18429/JACoW-Cyclotrons2016-TUP19 UR - http://jacow.org/cyclotrons2016/papers/tup19.pdf ER -