YOKOGAWA digitalYEWFLO Series, Руководство пользователя

Страница: 131 / 169

<APPENDIX 8. PID BLOCK>
A8-4
IM 01F06F00-01EN
A8.4 PID Computation Details
For PID control, the PID block in a digitalYEWFLO
employs the PV-proportional and -derivative type
PID control algorithm (referred to as the I-PD
control algorithm), or the PV-derivative type PID
control algorithm (referred to as the PI-D control
algorithm) depending on the mode, as described
below.
• PV-proportional and -derivative Type PID (I-
PD) Control Algorithm versus PV-derivative
Type PID (PI-D) Control Algorithm
The I-PD control algorithm, which is expressed
by the equation below, ensures control stability
against sudden changes in the setpoint, such as
when the user enters a new setpoint value. The
I-PD algorithm also ensures excellent controllability
by performing proportional, integral, and derivative
control actions in response to changes of
characteristics in the controlled process, changes in
load, and occurrences of disturbances.
When the PID block is in Auto or RCas mode,
this I-PD algorithm is used for control. In Cas
mode, however, the PV-derivative type PID (PI-D)
algorithm takes over since the response to setpoint
changes is more important. The control algorithm in
use thus switches over automatically in line with the
mode transitions. The following shows the basic
computation formulas of these algorithms.
PV-proportional and -derivative (I-PD) control
algorithm:
∆ MVn = K ∆ PVn + (PVn – SPn) + ∆ ( ∆ PVn)
∆ T
Ti
T
d
∆ T
FA0802.ai
PV-derivative (PI-D) control algorithm:
∆ MVn = K ∆ (PVn – SPn) + (PVn – SPn) + ∆ ( ∆ PVn)
T
d
∆ T
∆ T
Ti
FA0803.ai
Where,
∆MVn = change in control output
∆PVn = change in measured (controlled)
value = PVn - PVn-1
∆T = control period = period_of_execution
in Block Header
K = proportional gain = GAIN (= 100/
proportional band)
TI = integral time = RESET
T
D
= derivative time = RATE
The subscripts, n and n-1, represent the time
of sampling such that PVn and PVn-1 denote
the PV value sampled most recently and the PV
value sampled at the preceding control period,
respectively.
• PID Control Parameters
The table below shows the PID control parameters.
Parameter Description Valid Range
GAIN Proportional gain 0.05 to 20
RESET Integral time 0.1 to 10,000 (seconds)
RATE Derivative time 0 to infinity (seconds)
A8.5 Control Output
The final control output value, MV, is computed
based on the change in control output ∆MVn, which
is calculated at each control period in accordance
with the aforementioned algorithm. The PID block
in a digitalYEWFLO performs the velocity type
output action for the control output.
• Velocity Type Output Action
The PID block determines the control output (OUT)
value by adding the change in control output
calculated in the current control period, ∆ MVn, to
the value read back from the output destination,
BKCAL_IN. This velocity type output action can be
expressed as:
OUT = BKCAL_IN – ∆MVn’
where ∆MVn’ is ∆MVn scaled based on PV_SCALE
and OUT_SCALE.
Note: MV indicates the PID computation result.
A8.6 Direction of Control Action
The direction of the control action is determined by
the Direct Acting setting in CONTROL_OPTS.
Value of Direct Acting Resulting Action
True The output increases when the input
PV is greater than the setpoint SP.
False The output decreases when the
input PV is greater than the setpoint
SP.