The Joint Accelerator Conferences Website (JACoW) is an international collaboration that publishes the proceedings of accelerator conferences held around the world.
TY - CONF AU - Mohamadian, M. AU - Afarideh, H. AU - Chai, J.-S. AU - Ghergherehchi, M. AU - Salehi, M. ED - Petit-Jean-Genaz, Christine ED - Kim, Dong Eon ED - Kim, Kyung Sook ED - Ko, In Soo ED - Schaa, Volker RW TI - Development of Optimized RF Cavity in 10 MeV Cyclotron J2 - Proc. of IPAC2016, Busan, Korea, May 8-13, 2016 C1 - Busan, Korea T2 - International Particle Accelerator Conference T3 - 7 LA - english AB - Cyclotron cavity modelled by an artificial neural net-work, which is trained by our optimized algorithm. The training samples are obtained from simulation results, which are done by MWS CST software for some defined situation and parameters, and also with the conventional BP algorithm. It is shown that the optimized FFN can estimate the cyclotron model parameters with acceptable outputs. Hence, the neural network trained by this algorithm represents the proper estimation and acceptable ability to our structure modelling. The cyclotron cavity parameter modelling illustrate that the neural network trained by this algorithm could be the acceptable method to design parameters. PB - JACoW CP - Geneva, Switzerland SP - 1250 EP - 1252 KW - network KW - cyclotron KW - cavity KW - simulation KW - resonance DA - 2016/06 PY - 2016 SN - 978-3-95450-147-2 DO - 10.18429/JACoW-IPAC2016-TUPMR011 UR - http://jacow.org/ipac2016/papers/tupmr011.pdf ER -