Control Loops

Question 1

What is the software of a control loop? (4)

Answer 1

1. setpoint
2. manipulated variable
3. disturbance variable
4. controlled variable

Question 2

What is the setpoint?

Answer 2

The setpoint is the predetermined desired value for the controlled variable. The aim of the control loop is to ensure that the controlled variable remains at the setpoint

Question 3

What is the manipulated variable?

Answer 3

The manipulated variable has a direct influence on the controlled variable. By varying the manipulated variable in response to disturbances, the controlled variable can remain at its setpoint.

Question 4

What is a disturbance variable?

Answer 4

The disturbance variables that varies and pushes the controlled variable from the setpoint.

Question 5

What is the controlled variable?

Answer 5

It is the measured variable. It is the thing that we are trying to maintain at its setpoint.

Question 6

What is the hardware of a control loop? (3)

Answer 6

1. Sensor
2. Controller
3. Control Element

Question 7

What is the role of the sensor?

Answer 7

Measures the system variable and serves as a signal source for the control loop.

Question 8

What is the role of the controller?

Answer 8

To perform the functions of computation and comparison with the setpoint to determine how much action is required.

Question 9

What is the role of the control element?

Answer 9

It is a piece of equipment which exerts a direct influence on the process. It receives the signal from the controller and takes the appropriate action. It is generally a control valve.

Question 10

What are the characteristics of on-off control?

Answer 10

Saw-tooth pattern, cyclic, fluctuates between a maximum and minimum value. Doesn’t reach setpoint.

Question 11

When is on-off control suitable to be used?

Answer 11

For non-critical conditions such as temperature (turning on and off a heater) or level control.

Question 12

Disadvantages of on-off control.

Answer 12

Unsuitable for the control of flow, wear on the final control element if rapid opening and shutting of valve, poor control.
If the deadzone is large then there is not rapid opening and closing but the deviation from the setpoint is large. If the deadzone is small then there is a smaller deviation from setpoint but quick opening and shutting. Neither are desirable

Question 13

How are the error and the action taken in proportional only control related?

Answer 13

The action taken by the control loop is directly proportional to the error in the system.

Question 14

What are the characteristics of proportional only control?

Answer 14

There is an offset, usually oscillations at the beginning. There is an off set due to the equation for proportional only control V = K*epsilon + V0. V0 causes an offset and for the controlled variable to settle at a constant value which is different from the setpoint.

Question 15

What is the equation for the error?

Answer 15

epsilon = SP - PV (setpoint - process variable)

Question 16

What is the equation for proportional only control?

Answer 16

V = K*epsilon + V0
V is the adjustment or signal for adjustment in the control loop
K is the proportional constant or gain of the loop
epsilon is the error
V0 is the signal output with no error

Question 17

How are K and the proportional band (PB) related?

Answer 17

K = 1/PB

Question 18

What happens if there is large K ( small PB)?

Answer 18

Then there is a small offset but oscillatory (unstable) behaviour.

Question 19

What happens if there is small K ( large PB)?

Answer 19

Then there is a large offset but stable behaviour (there are no oscillations)

Question 20

What are the advantages of the proportional only control?

Answer 20

It gives a fast response, and can provide good stability.

Question 21

What are the disadvantages of proportional only control?

Answer 21

Offset and overshoot, always has to go past setpoint.

Question 22

What is the relationship between the error and the action taken in integral control?

Answer 22

The action taken is always proportional to the integral of the sum of the errors.

Question 23

Why is integral action always used with proportional control?

Answer 23

It is too slow a response and the action builds up over time.

Question 24

What is the advantage of using integral action control with proportional control?

Answer 24

It eventually removes the offset that is found in proportional only control. This is because the constant is removed and replaced by the integral of the error over time.

Question 25

What is the equation for proportional-integral action control?

Answer 25

V = K(epsilon + (1/Ti)*integral of epsilon dt

where Ti is the integral action time

Question 26

What happens if the integral time (Ti) is too small?

Answer 26

There is strong integral action and the offset is quickly removed but there is potentially unstable behaviour.

Question 27

What happens if the integral time (Ti) is too large?

Answer 27

There is weak integral action control, it takes longer for the offset to be removed but the response is more stable.

Question 28

What is the relationship between derivative action and error.

Answer 28

The derivative control action is proportional to the rate of change of the error

Question 29

Disadvantage of derivative only control?

Answer 29

There could be a large offset from the setpoint, but as long as there is no change in the behaviour then no action is taken. Derivative control can lead to the amplification of noise.

Question 30

Advantage of using derivative action with proportional-integral action?

Answer 30

It removes the overshoot

Question 31

The equation for PID control

Answer 31

V = K(epsilon + (1/Ti)integral of epsilon dt + Tdderivative epsilon/dt)

Question 32

When is PID not used?

Answer 32

When the measurement is noisy because then this leads to instability.

Question 33

Proportional, Integral, Derivative in terms of time frames

Answer 33

Proportional action controls the present
Integral action controls the past
Derivative action controls the future

Question 34

What happens if derivative action time, Td is large?

Answer 34

There is vigorous derivative control action but this leads to instability.

Question 35

Briefly describe feedback control

Answer 35

The measured variable is the controlled variable. The sensor measures the system variable and transforms it into an electrical signal. The signal is then passed on to the controller. The controller compares to setpoint and determines the action to be taken. The electrical signal is then passed on to the final control element and the appropriate action is taken

Question 36

Advantages of feedback control

Answer 36

Little knowledge is required for the process. No model is required for the system, the manipulated variable is adjusted according to the extent of the error. It is entirely empirical, as long as an adjustment is being made in the correct sense then the control system should remove disturbances

Question 37

Disadvantages of feedback control

Answer 37

• The error is required to pass through the system before action is taken
• Large magnitude disturbances can sometimes not be corrected for - use feedforward control This most commonly occurs because there is not a linear relationship between the disturbance and the controlled variable.
• Large time delay - then use a cascade control loop. This is especially true when only the feed flow rate can be varied.

Question 38

Briefly describe feedforward control

Answer 38

The sensor measures the disturbance which has the most dire effect on the controlled variables. The controller then predicts the effect that this disturbance would have on the controlled variable and computes how much action needs to be taken. The manipulated variable is adjusted accordingly using the final control element.

Question 39

Disadvantages of feedforward control

Answer 39

A lot of operator skill is required - there needs to a deep understanding of all the disturbances that have occurred on the process in the past. There is complex computation required, it involves determining exactly how much change is required for specific change in disturbance. There is no monitoring of controlled variable. Assume that all disturbance s are known in advances.

Question 40

Advantages of feedforward control

Answer 40

The disturbances have no effect on the overall process.

Question 41

Two causes of disturbances in a feedback control loop.

Answer 41

1. The system has been disturbance

2. The setpoint has been changed. In absence of any external disturbance, a change in setpoint will introduce error.

Question 42

Describe cascade control in terms of the manipulated and measured variables

Answer 42

There is one manipulated variable and more than one measured variable.

Question 43

Describe the inner and outer loop of cascade control

Answer 43

The outer loop - the input to the controller is the measured value of the variable to be controlled. The setpoint is supplied by the operator. It passes output signal to inner loop.
The inner loop - it measures a second variable whose value affects the controlled variable. (MEASURES A DISTURBANCE) Setpoint is supplied by output from outer loop. Its output signal is supplied to manipulated variable

Question 44

Advantage of using cascade control rather than feedback control

Answer 44

It gives a faster response. It is especially if inner loop is much faster than outer loop.

Question 45

Disadvantages of PID control

Answer 45

The computation is complex because 3 tuning parameters are required, also it can cause the amplification of noisy signals.

Question 46

Why is the valve always placed downstream of the indicator?

Answer 46

Because the valve alters the properties of the stream

Question 47

Why is the valve always downstream from a compressor?

Answer 47

Compressors have high energy requirements, making them expensive. It will be most cost effective to place the compressor at the front of the process and then vary the pressure by throttling the output downstream.

Question 48

What is the rise time in proportional control?

Answer 48

It is the time taken to go from 10 % to 90% of its steady state value.

Question 49

What is the settling time in proportional control?

Answer 49

It is the time taken to reach 95% of its steady state value.