Previous in Forum: Profibus   Next in Forum: Line Monitoring
Close
Close
Close
6 comments
Rate Comments: Nested
Anonymous Poster

PD Controller Use

06/02/2010 2:56 AM

We use the PD controller in temperature loop, not the pressure loop. Why?

Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.
Guru
United Kingdom - Member - Indeterminate Engineering Fields - Control Engineering - New Member

Join Date: Jan 2007
Location: In the bothy, 7 chains down the line from Dodman's Lane level crossing, in the nation formerly known as Great Britain. Kettle's on.
Posts: 32175
Good Answers: 839
#1

Re: controller

06/02/2010 3:06 AM

It really isn't possible to see why at this distance. How does the loop behave when the derivative is turned off? That may give some insight.

  • Proportional is about "how hard one presses the accelerator or the brake pedal".
  • Integral is about "for how long one presses the accelerator or brake pedal".
  • Derivative is about "how quickly one moves between accelerator and brake pedal".

Does that help?

__________________
"Did you get my e-mail?" - "The biggest problem in communication is the illusion that it has taken place" - George Bernard Shaw, 1856
Reply
Guru
Engineering Fields - Electrical Engineering - Analog and Digital Circuit Design Engineering Fields - Electromechanical Engineering - Transformers, Motors & Drives, EM Launchers Engineering Fields - Engineering Physics - Applied Electrical, Optical, and Mechanical

Join Date: Jan 2008
Location: NY
Posts: 1207
Good Answers: 119
#2

Re: PD Controller Use

06/02/2010 6:26 PM

Simplified explanation:

"P" Proportional system control gets you close to a setpoint, but dynamic performance and setpoint error may not be acceptable in certain systems.
"I" Integral part of control is used to minimize or eliminate setpoint error.
"D" Derivative part of control is used to predict how more complex systems will behave (dynamic performance) and attempts to prevent overshoots, undershoots, and/or ringing.

Thermal systems typically have a large time constant or time delay from when the heat is added to when the temperature probe registers a change. The "PD" control loop can greatly improve dynamic performance in thermal systems.

Pressure systems (the ones I'm familiar with) don't have the large time constant and generally don't need the "D" derivative control to reduce overshoot.

The forward system response and the desired loop response will dictate which type of control (P, PI, PD, PID) should be used.

Reply
Participant

Join Date: Jun 2010
Posts: 1
#3

Re: PD Controller Use

06/03/2010 5:56 PM

The first question you must answer is how does the PV respond to a change in controller Output for both variables? Once you know this, then the choice of which PID Modes to use is obvious. It's also important to understand which PID Form your control system provides: "Ideal" or "Interactive" as this affects the choice of tuning rules to apply when you must use Derivative.

Also be aware that Pressures (& Levels) are very different from Flows & Temperatures as they are "Integrating Processes". PID tuning for integrators is significantly different compared to flows & temperatures (even when the dynamics are similar). In general, PID tuning for Integrators looks much more aggressive compared to open-loop stable processes, but it is required in order to properly address the integrating nature of the process.

In general, Derivative is used ONLY when there is Dead Time in the process. If there is no Dead Time, then there should be absolutely no Derivative (even for a "long" Lag Time process). If you find yourself trying to use Derivative to counter overshoot, try reducing Gain and/or increasing the Integral time first. Remember: you can't change the physics of the process with tuning.

Reply
Anonymous Poster
#4
In reply to #3

Re: PD Controller Use

06/04/2010 1:24 AM

thanks sir

but one think i do not understand that why dead time come to picture in PD controller.

Reply Off Topic (Score 5)
Guru
Engineering Fields - Electrical Engineering - Analog and Digital Circuit Design Engineering Fields - Electromechanical Engineering - Transformers, Motors & Drives, EM Launchers Engineering Fields - Engineering Physics - Applied Electrical, Optical, and Mechanical

Join Date: Jan 2008
Location: NY
Posts: 1207
Good Answers: 119
#5

Re: PD Controller Use

06/04/2010 2:31 AM

While I wouldn't rely on wiki alone as learning reference, it is a convenient and usually accurate starting point for web discussions.

http://en.wikipedia.org/wiki/Control_theory

http://en.wikipedia.org/wiki/PID_controller

If you have trouble understanding these pages, don't feel bad. Many engineers have trouble also. There is a good reason auto-tune PID controllers are popular. Even when you are familiar with control theory and the specific process you want to control, manual tuning a PID controller can be a challenging task.

Reply
Anonymous Poster
#6
In reply to #5

Re: PD Controller Use

06/04/2010 2:46 AM

thanks sir

it is nice reply

Reply Off Topic (Score 5)
Reply to Forum Thread 6 comments
Copy to Clipboard

Users who posted comments:

Anonymous Poster (2); MichaelTaube (1); mjb1962853 (2); PWSlack (1)

Previous in Forum: Profibus   Next in Forum: Line Monitoring

Advertisement