-1
\$\begingroup\$

Sir my question is how integral component eliminates the steady state error and physical significance of integral and derivative component in a PID controller.

\$\endgroup\$
2
  • \$\begingroup\$ If you want a bit on the maths of integral component eliminating steady state error for a step signal, read about Final Value Theorem and apply your closed loop transfer function to it. \$\endgroup\$ – Vicente Cunha Mar 9 '16 at 11:50
  • \$\begingroup\$ I explain the basics of what a PID controller is at electronics.stackexchange.com/a/27861/4512. That's not the main point of that answer, but take a look at paragraphs 4-8. \$\endgroup\$ – Olin Lathrop Mar 9 '16 at 12:27
1
\$\begingroup\$

An integrator has infinite gain for steady signal therefore any sudden error (due to a step change in demand for instance) will progressively become amplified until the action of feedback minimizes that error to zero.

Why not just have proportional gain at infinity? This would likely cause severe instability problems - a control system would hunt about the set point in many, many situations. Integration provides a "calmer" and more gradual way of dispensing infinite gain.

Derivative is a means of slowing down the action of a control system when faced with a step change in demand - this naturally means that mechanical overshoots are not as great and therefore some systems can "come into lock" quicker.

\$\endgroup\$
2
  • \$\begingroup\$ Thank you sir I am confused about PID controller action..which variable(P,I,D)will take in action at first.I just want to know about the sequence of action. \$\endgroup\$ – Kofil Uddin Mar 9 '16 at 12:04
  • \$\begingroup\$ Derivative and proportional act first. Integral acts last in simple terms. \$\endgroup\$ – Andy aka Mar 9 '16 at 12:32

Not the answer you're looking for? Browse other questions tagged or ask your own question.