Comparative Investigations on Detuning Estimator for Experimental RF Cavity

In an operational accelerator, the radio frequency (RF) cavity may operate in either the continuous wave (CW) or the pulsed mode. The value of detuning
$\Delta \omega $
needs to be determined for tracking cavity resonance frequency
$f_{r}$
as well as its tuning. However, a direct sensor for knowing the value of
$\Delta \omega $
is not available. Model/s of the RF cavity may be employed to estimate detuning value from measured quantities. One needs to carry out or make a system well-suited for secondary/inferential measurement. In this article, the main focus is on the estimation of cavity
$\Delta \omega $
from the RF cavity model using various approaches. The advantages and limitations of each approach are outlined. A Kalman filter under a stochastic state-space framework is considered for detuning estimation. This framework helps to address practical issues like uncertainties in model and measurement. The experimental setup is made using a lab prototype RF cavity of 650 MHz, tested in the pulsed mode and the frequency sweep mode. Subsequently, the KF algorithm is evaluated. However, the successful operation of KF needs tuning, which is one of the main concerns while using KF. Adaptive Kalman filter (AKF) helps to assess values of tuning parameters to an extent. One of the AKFs is based on innovation-based adaptive estimation (IAE). It is applied to the detuning estimator in this article and shown to minimize rms estimation error.