Albin Heimerson
Albin Heimerson
Recently spent a while very confused by the `Flux.normalise` function behaving inconsistently, but upon closer inspection it was just that I wrote `Flux.normalize` some times and `Flux.normalise` some. `Flux.normalise` is...
Noticed that plotting by sending in a vector of plots does not seem to work when there is a mix of discrete and continuous, but when sending them in one...
Have two systems H and H2 to compare. Running ```julia bodeplot([H, H2]) ``` I get a nice plot with legend G1 and G2 so I know which is which. Running...
``` h = 0.1 z = tf("z", h) sp = 2/h * (z-1)/(z+1) sys = (1+sp)/(1+sp/10)^2 bodeplot(sys) ``` This does not manage to plot correctly, while it does in matlab....
For time domain plots it seems like we get one plot per output, and all inputs for each output are sharing that plot. For freq domain it seems like it...
Saw that the nyquist plot in the updated [ControlExamplePlots](https://github.com/JuliaControl/ControlExamplePlots.jl/pull/20) looked a bit odd. It writes "u(1) to y(1)" on all four subplots, and the xrange does not seem to scale....
Have not looked into how stable is decided, but it links to 0.5.3 which seems old and should probably be updated. Latest docs seems okay.
See https://discourse.julialang.org/t/controlsystems-transfer-functions-with-delay/58003/7 * Inconsistency between normal and delayed systems in how `u` is defined, `u(x, t)` vs `u(t)`. Can easily lead to defining a zero input since delay system also...
From #441, add pole zero mapping for c2d on zpk system. Maybe need to rethink how to handle c2d a bit more general.
Noticed that some methods in `pid_design.jl` still used non-positive Ts to denote continuous system. Seems rather harmless since default if no Ts is supplied it returns continuous and if a...