Lecture 3 Flashcards
(12 cards)
Benefits of Cascade (over Direct)
- Faster Disturbance Rejection
- Improved control
- More precise
- More complex, requires additional sensors
Benefits of Direct (over Cascade Control)
- Slower reaction to disturbances- corrections are only made when the deviation from set point has already occured
- Less Precise
- Simpler design
On-Off Control
Most simple control - control valve is either on or off according to the measurement of the process variable
Deadband (On-Off control)
Deadband is introduced to avoid oscillations around the set point in on off control where deadband provides a range around the set point in which the control valve will not change - avoids oscillations
Proportional Control
Continuous feedback control syastem
Error based correction - Calculates the error betweemn the setpoint and the measured variable and the controller output is adjusted proportionally
Offset means that the process variable may not reach the exact setpoint espcially if there are sustained disturbances
Integral Control/Action
A control loop that eliminates steady state errors
Eliminates offset by continuouisly summing the error signal over time
Considers past errors and applies corrective action
Time of correction is faster for bigger errors
Derivative Action/Control
A control loop that responds to the rate of change of the error signal
Predicts future errors by analyzing current rate of change of errors - applies corrective action proportional to the rate of change
Dampens or stabilises the sytem to prevent overshooting and oscillations - Slows down the control action when the process variable changes rapidly
PID Actiopn
P - responds to current error, corrective action is proportional to error magnitude
I - Eliminates steady state errors by acccumulating error over time, drives process towards SP
D - responds to rate of change of the error and predictive action dampens oscillations and stabilizes
Derivative Spike
When the error signal changes rapidly, causes a large sudden response from derivative action
- The derivative calculates the rate of change, its inherently sensitive to noise, even small fluctuations can produce a large spike in derivative signal
Load Change
External disturbances which affect the PV
- Unintended variations in the system
- Can lead to deviation from SP
- Feed forward control can mitigate the impact of load change through predictive action
Tuning
Essential part of the process - basically adjusting the settings so that the control is optimal
- Need a quantitative measure of controller performance
Penalty Function (Control)
Quantitative measure used to evaluate the performance of a controllers response - helps assess how well the controller is achieving the desired control objectives
Control system tuning then becomes an optimization problem where the controller parameters are adjusted to minimize this penalty function
