Control System Analysis for the Perturbed Linear Accelerator Rf System

Kwon, Sungil; Regan, A.

doi:10.2172/810037

Cited by 3 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to achieve efficient analysis and synthesis for a klystron, and for the cascade of the klystron and cavity in the linear accelerator, a linear klystron model around each operating point is required where the operating point is determined by the required power of the cavity. That is, a linear parameter varying klystron model can be obtained when the amplitude and phase saturation curves are approximated by polynomials [2], , , where is the input voltage and , , are coefficients. When the output amplitude and phase operating point of a klystron is determined by the cavity operating condition, the input amplitude operating point, , and the input phase operating point are obtained by the inverse mapping of the amplitude and phase saturation curves.…”

Section: A Klystron Modelmentioning

confidence: 99%

“…When the output amplitude and phase operating point of a klystron is determined by the cavity operating condition, the input amplitude operating point, , and the input phase operating point are obtained by the inverse mapping of the amplitude and phase saturation curves. By linearizing the amplitude and phase saturation curves around the operating voltage , the second-order linearized klystron model is obtained as (1) (2) in the baseband coordinate [2] where is the low-level RF system output. Note that the output matrix reflects the signal pickup loop gain attenuation factor.…”

Section: A Klystron Modelmentioning

confidence: 99%

“…In the context of the input uncertainty representation of a perturbed system [5], the perturbed output voltage of a klystron is expressed as the polynomial where is the perturbed input voltage around the nominal operating input voltage , is the lumped sum of amplitude ripple and amplitude droop in percentage [2]. The effect of HVPS ripple and capacitor bank droop on the output phase of the klystron is described by two terms: the first is the perturbed output of the phase saturation curve, , due to the perturbed amplitude , and the second is the direct additive phase perturbation, , which is the lumped sum of the phase ripple and the phase droop [2]. Then, the perturbed klystron model is defined as a second-order additive uncertainty model.…”

Section: B Perturbed Klystron Modelmentioning

confidence: 99%

“…In order to investigate the system characteristics, a MATLAB/SIMULINK [1] model has been constructed and an enormous number of simulations have been performed [2].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Decoupling PI controller design for a normal conducting RF cavity using a recursive LEVENBERG-MARQUARDT algorithm

Kwon¹,

Lynch²,

Prokop³

2005

IEEE Trans. Nucl. Sci.

View full text Add to dashboard Cite

This paper addresses the system identification and the decoupling PI controller design for a normal conducting RF cavity.Based on the open-loop measurement data of an SNS DTL cavity, the open-loop system's bandwidths and loop time delays are estimated by using batched least square. With the identified system, a PI controller is designed in such a way that it suppresses the time varying klystron droop and decouples the In-phase and Quadrature of the cavity field. The Levenberg-Marquardt algorithm is applied for nonlinear least squares to obtain the optimal PI controller parameters. The tuned PI controller gains are downloaded to the low-level RF system by using channel access. The experiment of the closed-loop system is performed and the performance is investigated. The proposed tuning method is running automatically in real time interface between a host computer with controller hardware through ActiveX Channel Access.Index Terms-Capacitor bank droop, high voltage power supply ripple, least square, Levenberg-Marquardt algorithm, normal conducting RF cavity, operating point, PI feedback control, spallation neutron source, system identification.

show abstract

Section: A Klystron Modelmentioning

confidence: 99%

Section: A Klystron Modelmentioning

confidence: 99%

Section: B Perturbed Klystron Modelmentioning

confidence: 99%

“…In order to investigate the system characteristics, a MATLAB/SIMULINK [1] model has been constructed and an enormous number of simulations have been performed [2].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Decoupling PI controller design for a normal conducting RF cavity using a recursive LEVENBERG-MARQUARDT algorithm

Kwon¹,

Lynch²,

Prokop³

2005

IEEE Trans. Nucl. Sci.

View full text Add to dashboard Cite

show abstract

“…As previously reported [4,5], we have performed extensive modeling of the SNS RF control system. This model has been developed in MATLAB/Sirnulink.…”

Section: Firm Warementioning

confidence: 99%

Newly designed field control module for the SNS

Regan

Kasemir

Kwon

et al.

Proceedings of the 2003 Bipolar/BiCMOS Circuits and Technology Meeting (IEEE Cat. No.03CH37440)

View full text Add to dashboard Cite

Spallation Neutron Source has undergone some recent hardware changes. The intended Field and Resonance Control Module (FRCM) design has been re-vamped to minimize functionality and ease implementation. This effort spans a variety of disciplines, and requires parallel development with distinct interface controls. This paper will discuss the platform chosen, the design requirements that will be met, and the parallel development efforts ongoing.The low-level RF (LLRF) control system for the OVERVIEW OF THE NEW FCMPerformance specifications were eased late last year for the LLRF control system for SNS. Physics simulations indicated that the field control requirement could be reduced from from HS%, iA).5" to kl.O%, k1.0". In addition required functions were changed such that onboard processing for resonance control was moved to the crate controller (IOC), as well as feedforward table update and access; fewer RF channels were required (no need to provide for possible beam information via a beam diagnostics channel); no real time data link needed between the control system and the operators; and milch less memory per channel was deemed appropriate. By moving the iterative learning controls onto the IOC as well, we completely eliminated the need for DSPs on the module. A collaborative effort was established between Lawrence Berkely, Los Alamos, and Oak Ridge national labs in order to develop a control system that meets these specifications [I].Due to these relaxed system performance specifications, and a new staged approach to implementation and integration of control system functions, the original complicated, meet-all-performance-specifications-at-once FRCM de-sign has been greatly simplified. The new Field Control Module (FCM) is still a basic VXIbusbased module with multiple daughter cards. The infrastructure for a VXIbus-based RF Control System (RFCS) was already in place at the SNS: VXIbus crates had been purchased and other modules within the system are VXIbus, so it only made sense to remain consistent. However the new FCM has three new daughter cards which minimize the functionality of the FCM; and the implementation of the firmware is in a staged approach where progressive phases will be achieved as the accelerator grows from one cavity to many and functional requirements of the RFCS increase [l]. A block diagram of the module hardware is given in figure I . PRlNTED CIRCUIT CARDS'rhe imis for thc rc-designed FCM is the successful SNS Diagnostics PCI-based clcctronics, specifically the Beam Position Monitors (BPMs) 121. In that vein, we attempted to make use of a'i tnuch of the foundation provided by the l3i;ignosticc HPMs 3s possible. The BPMs have an Analog Front End daughter card nianufacturecl by 'Bergm Instrumentation of France. Likewise, we specified a very similar I L R F AFE for use on the FCM. The BPMs havc a Digital Front End (DFE) proccssing card. We, too, utilize a DFE for providing the analog-to-digital conversion of the input RF and IF signals, as well as for providing an interface to the...

show abstract

Uncertain system modeling of SNS RF control system

Kwon

Regan²,

Wang³

et al.

PACS2001. Proceedings of the 2001 Particle Accelerator Conference (Cat. No.01CH37268)

View full text Add to dashboard Cite

Abstract-This paper addresses the modeling problem of the linear accelerator RF system for SNS. The cascade of the klystron and the cavity is modeled as a nominal system. In the real world, high voltage power supply ripple, Lorentz Force Detuning, microphonics, cavity RF parameter perturbations, distortions in RF components, and loop time delay imperfection exist inevitably, which must be analyzed. The analysis is based on the accurate modeling of the disturbances and uncertainties. In this paper, a modern control theory is applied for modeling the disturbances, uncertainties, and for analyzing the closed loop system robust performance.

show abstract

Control System Analysis for the Perturbed Linear Accelerator Rf System

Cited by 3 publications

References 0 publications

Decoupling PI controller design for a normal conducting RF cavity using a recursive LEVENBERG-MARQUARDT algorithm

Decoupling PI controller design for a normal conducting RF cavity using a recursive LEVENBERG-MARQUARDT algorithm

Newly designed field control module for the SNS

Uncertain system modeling of SNS RF control system

Contact Info

Product

Resources

About