Technical data

PID controller The parameters for a PID controller are the proportional cooefficient
K, the reset time TN and the derivative time constant TV. These in
turn determine the P, I and D actions.
Parameter changes The P action of the manipulated variable is obtained based on the
following formula:
P action = KP
*
(XW
k
- XW
k-1
)
If KP or R are changed during automatic operation, this only affects
subsequent changes of the system error XW
k
. The current value of the
manipulated variable is not affected by the parameter change. This
response allows for a bumpless change.
If, however, you do not want this response, you can eliminate it using
the following calculation, (example of a KP change). This calculation
is only made once for each parameter change:
Y
k-1
= Y
k-1
+ XW
k-1
(KP
new
- KP
old
)
If you use the following program in the case of a parameter change,
the controller responds like an analog controller.
:L KP
new
load KP
new
:L KP
old
load KP
old
:-G
:L DD38 XW
k-1
:xG
:L DD44 Y
k-1
:+G
:T DD44 = Y
k-1
Abbreviations for PID
controllers
dY
k
Calculated control increment
dZ
k
Disturbance increment
FLP Floating point representation
k k * sampling
K Proportional cooefficient
LL Lower limit (limiter)
NF Normalized fixed point representation
R R parameter
TA Sampling time
TD TV/TA
TI TA/TN
t Sampling instant = k
*
TA
TN Reset time
TV Derivative time constant
UL Upper limit (limiter)
W
k
Setpoint
X
k
Actual value
XW
k
System error
Y
k
Calculated manipulated variable
YA
k
Value of manipulated variable (control increment or
manipulated variable)
Z
k
Disturbance
6
OB 250/251: Closed-Loop Control/ PID Algorithm
CPU 928B Programming Guide
C79000-B8576-C898-01
6 - 123