53 lines
959 B
Markdown
53 lines
959 B
Markdown
* create a stepped schedule
|
|
|
|
200
|
|
220
|
|
200
|
|
220
|
|
|
|
* set i=0
|
|
* set d=0
|
|
* set p=5
|
|
|
|
run schedule
|
|
if oscillation is sustained stop
|
|
otherwise, double p
|
|
|
|
once oscillation is sustained
|
|
* record P as Ku
|
|
* record period of oscillation as Pu
|
|
|
|
set p=Ku/1.7
|
|
set i=Pu/2
|
|
set d=Pu/8
|
|
|
|
-------------------------------------------------------------------------------
|
|
another method (for just PI control)
|
|
https://pdfs.semanticscholar.org/e719/f259f6c93b9a897a214824e562c886aa80b9.pdf
|
|
|
|
|
|
use step schedule
|
|
set d=0
|
|
set p=1
|
|
you should see some overshoot and barely observable undershoot
|
|
if not, change p until that's true (up or down)
|
|
|
|
next measure T between high peak and low peak
|
|
set i=T*1.5
|
|
set p=.8*lastp
|
|
|
|
p=1*.8 = .8
|
|
i=1040-315 = 725s * 1.5 = 1088
|
|
|
|
-------------------------------------------------------------------------------
|
|
https://www.eurotherm.com/pid-control-made-easy
|
|
|
|
similar to above for PI
|
|
|
|
p=2*25=50 (that's 25F overshoot times 2)
|
|
i=725*1.5 = 1088
|
|
d = 1088/5 = 217
|
|
|
|
|
|
|