PROCESS CONTROL ESSENTIALS
ARIAS PROCESS
WHAT DOES STEADY-STATE MEAN?
It means no or negligible change during time. So, after every changes in the simulation, the final steady-state
condition will be calculated regardless of the transition and the path.
TRANSITION
STEADY-STATE 2
PARAMETER X
STEADY-STATE 1
TIME
ARIAS PROCESS
2
THE MAIN PROCESS CONTROL SYSTEMS
1 Feedback Closed-loop control Corrective
2 Feed-forward Open-loop control Predictive
3 Combined Feedback and Feed-forward
3
FEEDBACK CONTROL
4
FEEDBACK CONTROL
LOAD
ERROR OP MANIPULATED + PV
SP + CONTROLLER + PROCESS
- ELEMENT
PV
SENSOR
5
FEEDBACK CONTROL
PROCESS VARIABLE
SET POINT
1 Error-based control.
It only reacts on the error which is the difference between the PV and SP.
2 It does a self-correction.
If the process variable deviates form the SP, the controller continues to correct it.
3 Minimum knowledge of process is needed.
Mathematical modelling of the process is not required.
4 It is not expensive.
If the process variable deviates form the SP, the controller continues to correct it.
5 It is not predictive.
It can not do corrections before disturbance(s) have affected the process.
6 It can introduce instability in the system.
Controller tuning is critical.
7 It is not good for processes with large time constant
If a large disturbances occur, the process may operate continuously in a transient state and never reach the desired steady-state.
6
FEEDBACK CONTROL
EXAMPLE
OP TC SP
TT
PV
COOLER
PV: PROCESS VARIABLE
SP: SET POINT
OP: CONTROLLER OUTPUT
7
PID CONTROLLER
𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎𝐎 𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜𝐜 𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟𝐟
𝒕𝒕
𝟏𝟏 𝒅𝒅𝒅𝒅 𝒕𝒕
𝑲𝑲𝒄𝒄 𝒆𝒆 𝒕𝒕 + . � 𝒆𝒆 𝒕𝒕 𝒅𝒅𝒅𝒅 + 𝑻𝑻𝒅𝒅 .
𝑻𝑻𝒊𝒊 𝟎𝟎 𝒅𝒅𝒅𝒅
PID CONTROLLER
P 𝐾𝐾𝑐𝑐 . 𝑒𝑒 𝑡𝑡
I 𝐾𝐾𝑐𝑐 𝑡𝑡
𝑒𝑒 𝑡𝑡
SP (SETPOINT) + . � 𝑒𝑒 𝑡𝑡 𝑑𝑑𝑑𝑑 �. PROCESS
- 𝑇𝑇𝑖𝑖 0
D 𝐾𝐾𝑐𝑐 . 𝑇𝑇𝑑𝑑 .
𝑑𝑑𝑑𝑑 𝑡𝑡
𝑑𝑑𝑑𝑑
PV (PROCESS VARIABLE)
SENSOR
8
PID CONTROLLER – OVERALL CONTROL FUNCTION
DIFFERENTIAL TIME
ERROR
𝒕𝒕
𝟏𝟏 𝒅𝒅𝒆𝒆 𝒕𝒕
PROPORTIONAL TERM 𝑲𝑲𝒄𝒄 𝒆𝒆 𝒕𝒕 + . � 𝒆𝒆 𝒕𝒕 𝒅𝒅𝒅𝒅 + 𝑻𝑻𝒅𝒅 .
𝑻𝑻𝒊𝒊 𝟎𝟎 𝒅𝒅𝒅𝒅
INTEGRAL TIME
9
PID CONTROLLER
PID CONTROLLER
PROPORTIONAL 𝐾𝐾𝑐𝑐 . 𝑒𝑒 𝑡𝑡 Present error
𝐾𝐾𝑐𝑐 𝑡𝑡
INTEGRAL . � 𝑒𝑒 𝑡𝑡 𝑑𝑑𝑑𝑑
𝑇𝑇𝑖𝑖 0
Past errors
𝑑𝑑𝑑𝑑 𝑡𝑡
DERIVATIVE 𝐾𝐾𝑐𝑐 . 𝑇𝑇𝑑𝑑 .
𝑑𝑑𝑑𝑑
Prediction of Future errors
10
PID CONTROLLER SETTING
INITIAL PI CONTROLLER TUNING SUGGESTIONS
Kc Ti (min)
Flow 0,1 0,1
Level 2 5
Pressure 0,5 2
Temperature 2 10
11
FEEDFORWARD CONTROL
12
FEED FORWARD CONTROL
SENSOR LOAD
FEED FORWARD OP MANIPULATED +
+ PROCESS
CALCULATIONS ELEMENT PV
13
FEEDFORWARD CONTROL
1 Equation or correlation-based control.
It reacts before a disturbance effects is observed in the process.
2 It is predictive.
It takes corrective actions before disturbance(s) affect the system.
3 It does not affect the system’s stability.
4 It can’t perform self-correction.
If the adjustment fails, the system continue with it.
5 It is expensive.
Due to the efforts required for the mathematical modelling etc.
6 Extensive knowledge of process is needed.
To obtain the equations and correlations.
14
FEEDFORWARD CONTROL
OP
FC
FT
PV
COOLER
PV: PROCESS VARIABLE
SP: SET POINT
OP: CONTROLLER OUTPUT
15
