Topic : Advanced control system with unified classical, modern, and AI-based approaches

  

 Scope  

                        Particle accelerators are machines used for increasing the energy of charged particles for use in various applications in many fields. The operation of these machines generally requires the monitoring and control of many system parameters. This is achieved with the distributed control system comprising different interconnected hardware and software control system layers covering different subsystems of accelerators like injector, transport lines, storage ring, and beam lines. The main requirement is that all the sub-system operations are to be performed in a synchronized and/or sequential manner. Classical and modern multivariable control algorithms have been constantly implemented in the closed-loop control schemes to control the machines. With the time-and-frequency domain designs and trade-off between disturbance rejection and noise attenuation, these control algorithms have been successfully applied to the machines. Nowadays, machine stability and timing in modern accelerators have become increasingly stringent, the signal processing and control algorithms must meet these demanding requirements. With the advancement of the computing hardware and modern software architecture, it is possible and challenging to apply complex Artificial Intelligence algorithms together with the control and signal processing algorithms to the closed-loop control of the machines. The PhD candidates will be working by designing an AI-based controller together with the new and/or existing feedback control algorithms for subsystem(s) of the accelerator machine. Both simulation and real-world implementation (hardware and software designs) are performed for practical purposes during their study.
                        Tools to be used: High-performance computer, FPGA controller boards, microcontrollers, single-board computers.
                        Skill required: Programming/coding skill. Knowledge of signals and systems/control system engineering/machine learning is preferable.

  

 Supervisor  

                        Dr. Roengrut Rujanakraikarn
                        and SLRI researchers