Mitsubishi Electric MELSEC iQ-R Series, Руководство по программированию

Страница: 1392 / 2110

1390
11 PROCESS CONTROL INSTRUCTIONS
11.7 Auto Tuning
11.7
Auto Tuning
Auto tuning is designed to make initial setting of PID constants.
Auto tuning can be used for processes that can be approximated with the "primary lag plus dead time" represented by the
following expression.
Ex.
Process with relatively slow response such as temperature control
K: Gain, T: Time constant, L: Dead time, s: Laplace operator
Auto tuning can be used for the loop that uses S.PID or S.2PID instruction.
Auto tuning is performed in the ZN method: step response method of Ziegler and Nichols.
Outline of step response method
With no control operation being performed, change the manipulated value (MV) in a stepwise manner and look how the
process value (PV) varies.
• When MV is changed in a stepwise manner, PV begins to change slowly. Soon, the change speed becomes faster and then
becomes slow again, and finally is settled as a fixed value.
• Draw a tangent line at the place where PV varies fastest, and find the points of intersection A and B where the tangent line
crosses the horizontal axis corresponding to the first process value (

0
) and last process value (

1
). This obtains the
equivalent dead time (L) and equivalent time constant (T) as shown below.
• Determine maximum ramp (response speed) R = Y/T from the equivalent time constant (T) and maximum process value
width (Y). Apply the equivalent dead time (L) and maximum ramp (R) to the Ziegler and Nichols' adjustment rule to
calculate the proportional gain KP(P), integral constant TI(I), and derivative constant TD(D).
e
1+Ts
K -Ls
L T
A
θ
0
θ
1
B
Y
T
R=
X%
[%]
Y%
AT1MV
PV
t