[PPT] - BSc Project What kinds of fault we may confront in a control loop? PowerPoint Presentation

SLIDE 1

13/12/2016 1

Shahriar Shahram Supervisor: Dr . Nobakhti Contributors : Milad Pooladsanj

BSc Project Fault Detection & Diagnosis in Control Valves

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

2

f

23

Sharif University of Technology

Content

What is fault? Why we detect fault in a control loop? What kinds of fault we may confront in a control loop? What is “stiction”? Comparing “stiction” with other factors Ways we can resolve “stiction”

Shahriar Shahram Fault Detection & Diagnosis in Control valve

3

f

23

Sharif University of Technology

Controller : poor tuning is not fault Actuator Fault: valve friction Plant Fault: leakage , human error Sensor Fault: calibration

What is fault?

Actuator Controller Plant Sensor

SP Output

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

4

f

23

Sharif University of Technology

Why we detect a fault?

A. Increasing product quality
B. Reducing the rate of rejection
C. Fault signals propagation in physical components
D. Minimizing risk of instability

SLIDE 2

13/12/2016 2

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

5

f

23

Sharif University of Technology

CLPM :Control Loop Performance Monitoring

An important prerequisite for assessing control loops which is based on routine data and closed loop. CLPM usually connects to PID controller, because we don’t want to specify additional parameters except the normal controller in order to have an automatic procedure. CLPM often used in chemical , petrochemical , pulp and paper industry.

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

6

f

23

Sharif University of Technology

CLPM :Control Loop Performance Monitoring

CLPM steps are:

1. Diagnosis
2. Fault identification
3. Isolation
4. Root cause

CLPM can be grouped according to three different criteria :

A. Mathematical Methods
B. Nature of Time Trends
C. Prevalence of Faults

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

7

f

23

Sharif University of Technology

Mathematical Method Results

Reference1 Shahriar Shahram Fault Detection & Diagnosis in Control Valve

8

f

23

Sharif University of Technology

Time Trend Properties

Saturation

Saturation manifests itself at the controller output. Saturation can result from a poorly dimensioned actuators, poor controller tuning or may be caused by integral wind-up of PID controllers after a large setpoint change occurred

Sinusoidal Oscillation

Generally, as oscillations become sinusoidal as they travel through a process via physical connections since the process acts as a low pass filter

Manual Control

If the controller is not acting satisfactory the operator or control engineer

ften switches it off. This is seen in the control system as a switch from

automatic to manual mode

SLIDE 3

13/12/2016 3

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

9

f

23

Sharif University of Technology

Time Trend Properties

Sluggish Behavior

This means that the process variable only slowly follows the setpoint. There is no overshoot after a setpoint change. The reason for sluggish behavior lies usually in the tuning settings.

Non-Linear Oscillation

Some control loops contain a nonlinear element such as a faulty valve. Nonlinear disturbance affects the process variable regularly and repetitively but contains higher frequency components and can therefore be distinguished from a sinusoidal oscillation

Quantisation of process variables

All signals are digital in amplitude. The resolution of the signal is determined by the sensor. The sensor may have a coarse resolution so that the time trend shows step-like features

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

10

f

23

Sharif University of Technology

Time Trend Results

Reference1 Shahriar Shahram Fault Detection & Diagnosis in Control Valve

11

f

23

Sharif University of Technology

Time Trend Results

Reference1 Shahriar Shahram Fault Detection & Diagnosis in Control Valve

12

f

23

Sharif University of Technology

Prevalence of Fault

Reference1

SLIDE 4

13/12/2016 4

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

13

f

23

Sharif University of Technology

Prevalence of Faults

Reference1 Shahriar Shahram Fault Detection & Diagnosis in Control Valve

14

f

23

Sharif University of Technology

Valve Stiction

What is stiction referred to?

According to the Instrument Society of America (ISA) (ISA Subcommittee

SP75.05, 1979)

‘‘stiction is the resistance to the start of motion, usually measured as the difference between the driving values required to overcome static friction upscale and downscale’’.

According to EnTech (1998)

‘‘stiction is a tendency to stick-slip due to high static friction. The phenomenon causes a limited resolution of the resulting control valve motion. ISA terminology has not settled on a suitable term yet. Stick-slip is the tendency of a control valve to stick while at rest, and to suddenly slip after force has been applied’’.

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

15

f

23

Sharif University of Technology

Valve Stiction

What is stiction referred to?

According to Horch (2000)

‘‘The control valve is stuck in a certain position due to high static friction. The (integrating) controller then increases the set point to the valve until the static friction can be overcome. Then the valve breaks off and moves to a new position (slip phase) where it sticks again. The new position is usually on the other side of the desired set point such that the process starts in the opposite direction again’’.

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

16

f

23

Sharif University of Technology

Valve Stiction

Why stiction is important ?

 Stiction as 1st frequent fault in control loops  Stiction increases the instability chance in whole system  Stiction reduces control valves lifetime  product quality will decrease

SLIDE 5

13/12/2016 5

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

17

f

23

Sharif University of Technology

Control Valves Types

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

18

f

23

Sharif University of Technology

Resolving Stiction

We have three ways to reduce the control valve stiction problem:

1. Test each valve individually

In this method we should disconnect the control valve and test it for a specific input .Then we measure the output and we can recognize if it has stiction or not.

Advantages

You will recognize all of the valves that have stiction

Disadvantages

The whole system must be shutdown
Loosing time

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

19

f

23

Sharif University of Technology

Resolving Stiction

2. Implementing sensors in valve

In this method we use several sensors so that we can analyze the output signal of sensors , monitoring them and check if there is stiction.

Advantages

You will find faulty valves without shutting down the plant
Time saving

Disadvantages

Expensive , because we use many sensors
The sensors may be faulty
(In Iran) even the main sensors of the system are not available!

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

20

f

23

Sharif University of Technology

Resolving Stiction

3. Proccesing output signal of the valve

In this method we can approximately recognize the faulty valves by doing some mathematical algorithms on different output signals in control loop.

Advantages

You will find faulty valves without shutting down the plant
Time saving
Low cost

Disadvantages

Some of the methods can be used only in certain conditions (such as availability of

some sensors)

Most of the methods are not applicable in industrial systems

SLIDE 6

13/12/2016 6

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

21

f

23

Sharif University of Technology

Conclusion

1. We can improve control loop performance by detecting the faulty valves and repair

them

2. Valve stiction is very important. The first reason is that valve stiction causes 30% of

loops work poorly. Another reason is when a valve has stiction , the risk of instability will rise.

3. By proccessing the output signals in control loop we can determine the faulty valve

with more advantages comparing to other methods.

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

22

f

23

Sharif University of Technology

Challenges

1. Derive a good modeling for valve stiction
2. Working on algorithms to process the output signal
3. Increasing the accuracy of results

Shahriar Shahram Fault Detection & Diagnosis in Control Valve

23

f

23

Sharif University of Technology

References

1) Margret Bauera, Alexander Horcha, Lei Xieb, Mohieddine Jelali , Nina Thornhilld . The current state of control loop performance monitoring – A survey of application in industry (2015) 2) T. Hagglund. Department of Automatic Control, Lund Institute of Technology, S-221 O0 Lund, Sweden. A CONTROL-LOOP PERFORMANCE MONITOR(1995) 3) M.A.A. Shoukat Choudhurya, N.F. Thornhillb, S.L. Shaha. Modelling valve stiction(2004)