Analysis and Synthesis of the SNS Superconducting RF Control System

Y.M. Wang,S.I. Kwon,A.H. Regan

DOI: https://doi.org/10.48550/arXiv.physics/0008163

2000-08-19

Abstract:The RF system for the SNS superconducting linac consists of a superconducting cavity, a klystron, and a low-level RF (LLRF) control system. For a proton linac like SNS, the field in each individual cavity needs to be controlled to meet the overall system requirements. The purpose of the LLRF control system is to maintain the RF cavity field to a desired magnitude and phase by controlling the klystron driver signal. The Lorentz force detuning causes the shift of the resonant frequency during the normal operation in the order of a few hundreds hertz. In order to compensate the Lorentz force detuning effects, the cavity is pre-tuned into the middle of the expected frequency shift caused by the Lorentz force detuning. Meanwhile, to reduce the overshoot in the transient response, a feed-forward algorithm, a linear parameter varying gain scheduling (LPV-GS) controller, is proposed to get away a repetitive noised caused by the pulsed operation as well as the Lorentz force detuning effects.

Accelerator Physics,Instrumentation and Detectors

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: How to design and implement an effective low - level radio frequency (Low - Level RF, LLRF) control system to ensure that the radio - frequency field of each superconducting cavity in the superconducting linac of the Spallation Neutron Source (SNS) in the United States can be stabilized at the required amplitude and phase. Specifically, the research mainly focuses on the following aspects:

1. **Compensation for Lorentz - force detuning effects**:
- The Lorentz force will cause a change in the resonant frequency of the superconducting cavity, and this change can reach several hundred hertz during normal operation. To compensate for this effect, the paper proposes a pre - tuning method and uses a Linear Parameter Varying Gain Scheduling (LPV - GS) controller to reduce overshoot in the transient response.

2. **Repetitive noise suppression caused by pulsed operation**:
- Since the working mode of the SNS superconducting linac is pulsed operation, this will introduce repetitive noise. The paper proposes a feed - forward algorithm combined with an LPV - GS controller to suppress this noise.

3. **System modeling and control algorithm design**:
- The paper establishes MATLAB models of the superconducting cavity, klystron, and PID feedback controller and feed - forward controller. By modeling these functional modules, the dynamic behavior of the system is studied, and the corresponding control algorithms are designed.

4. **Implementation of a fully - digital control system**:
- To meet the system performance requirements and considering the delay problem of the digital control system, the paper proposes a fully - digital control system architecture that combines a Complex Programmable Logic Device (CPLD) and a Digital Signal Processor (DSP). The CPLD is used to handle the critical path to minimize the time delay, while the DSP is used to execute the complex LPV - GS control algorithm.

### Formula summary

- **Superconducting cavity state - space equation**:
\[
\dot{x}=A(\omega)x + Bu + B_I I
\]
\[
y = Cx
\]
where,
\[
A(\omega)=\begin{bmatrix}
-\frac{\tau}{L}&1\\
-\Delta\omega_L-\Delta\omega_m-\frac{\omega_0^2}{L}&-\frac{\tau}{L}
\end{bmatrix}
\]

- **Lorentz - force detuning frequency equation**:
\[
\Delta\omega_L = K\left(\frac{I^2}{Q}-\frac{V^2}{Q}\right)
\]

- **LPV - GS controller design**:
\[
u = F(V, I,\Delta\omega_L)(r - y)
\]
where \(F(V, I,\Delta\omega_L)\) is the parameter - varying feedback gain matrix.

Through the above methods, the paper aims to improve the stability and performance of the radio - frequency control system of the SNS superconducting linac and ensure that it can meet strict requirements in actual operation.