I may be asking the impossible, but here it is.
I have a set of closed orbit measurements and I fit
quadrupole displacements to these measurements.
The problem is that the BPMs move with the quadrupoles and I am
trying to generate a self-consistent algorithm. So far I have the following
in the .ele file:
&alter_elements name=*,type=MONI,item=CO_FITPOINT,value=1 &end
.
.
.
&optimization_setup
mode = minimize
target=0.0005
n_evaluations = 500, n_passes = 4,n_restarts=8
restart_worst_term_factor=1
verbose = 1
log_file = "optimizationx"
&end
.
.
.
&optimization_term term = "M1#1.xco 6.7e-3 - abs" &end
(6.7 e-3 is the measured orbit at BPM M1. Ideally it should be this number + the BPM displacement
-which in turn depends on what I am trying to fit, the quadrupole displacement)
&optimization_term term = "M2#1.xco 3.5e-3 - abs" &end
&optimization_term term = "M3#1.xco -3.8e-3 - abs" &end
.
.
etc.
&optimization_variable name="QRF2", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF3", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF4", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF5", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF6", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF7", item="DX" step_size=0.0002 &end
&optimization_variable name="QRF8", item="DX" step_size=0.0002 &end
.
.
etc.
Any suggestion ? An iterative procedure perhaps?
Thanks, Max
Fitting quads and bpms to measured orbit
Moderators: cyao, michael_borland
-
- Posts: 1951
- Joined: 19 May 2008, 09:33
- Location: Argonne National Laboratory
- Contact:
Re: Fitting quads and bpms to measured orbit
Max,
I think this should work, but you need to link the BPM offset to the quadrupole offset. E.g.,
&optimization_covariable name=M1, item=DX, equation="Q1.DX" &end
where I assume the related quadrupole is "Q1".
--Michael
I think this should work, but you need to link the BPM offset to the quadrupole offset. E.g.,
&optimization_covariable name=M1, item=DX, equation="Q1.DX" &end
where I assume the related quadrupole is "Q1".
--Michael
Re: Fitting quads and bpms to measured orbit
Thanks a lot, Michael. What you suggest works.
Max
Max