BTP Report PDF

BTP REPORT
PID TUNING
USING EXTREMUM SEEKING CONTROL
by
VINEET MISHRA
B.Tech - Electrical Engineering
IIT PATNA
Supervised by
Dr. ALI
PID TUNING USING
EXTREMUM SEEKING CONTROL METHOD
THEORY OF EXTREMUM SEEKING CONTROL :

Extremum seeking control is a non-model based adaptive
control method ( it means that it is able to adapt itself to the
variations in cost function with time etc. ) , it iteratively
modifies the arguments of a cost function so that the output
of the cost function reaches a local minimum or a local
maximum .It is also effective in cases of dynamic objective
function or function with changing parameters .
COST FUNCTION AND PID CONTROLLER :
Extremum seeking is used to tune the parameters of a PID
controller so as to minimise the given cost function. We use
the integrated square error cost function .
1 𝑇
𝐽(𝜃 ) = ∫
𝑇−𝑡0 𝑡0
𝑒 2 (𝑡, 𝜃 )dt
where the error e(t,θ) = r(t) − y(t,θ) is the difference
between input reference signal and the output signal of the closed
loop system and 𝜃 = [K 𝑇𝑖 𝑇𝐷 ]𝑇 contains the PID parameters .
EXTREMUM SEEKING TUNING SCHEME :
ES seeks to tune the PID controller by finding a minimizer of
cost function . However , since ES is a gradient method , the
PID parameters found by ES are not necessarily a global
minimizer of cost function .
HOW EXTREMUM SEEKING WORKS :
For any given objective function there is a set of input

parameters which maximize the objective (u*) . This is the
value which the extremum seeking control tries to reach at .
Suppose we assumed u1 as our initial optimal guess point ,
at this point if we give a sine wave perturbation , we observe
another sine wave response from the given objective
function . Now if we multiply the sine wave perturbation with
the response obtained from the objective function and
remove its DC portion using a high pass filter, we obtain a
wave that is positive at all times now if we add this positive
value to the initial guess point we move closer to u* .
Suppose our initial guess point was u2 , to the right of u*. We
will observe that upon multiplication of the sine wave
perturbation to the response obtained from objective
function we will get a value that is negative , now if we add
this value to the initial guess point we will again move
towards u*.
Suppose after a lot of iterations we reach at u* now if at this
point we give a sine wave perturbation and multiply the
result obtained from the objective function with this sine
wave and use a high pass filter to remove the DC portion we
will get the wave which is positive as well as negative so its
overall value is zero . Hence if we will add this value to u* it
will not change .
COMPLETED TASK –
Here I have tried to show how extremum seeking is implemented by
assuming a predefined static objective function and tried to find its
maxima using extremum seeking method .
Assumed objective function is a static function with a single maxima
at u=5 .
FUNCTIONS OF DIFFERENT BLOCKS IN EXTREMUM

SEEKING METHOD
HIGH PASS FILTER – It is used to eliminate the DC content of the
perturbed wave at the point of inspection .
MULTIPILER – It is used to multiply the sine wave perturbation to the
objective function to give the amount by which the initial assumed
guess has to be changed .
INTEGRATOR – It is used to accumulate all the derived changes in the
initial assumed point of maxima to finally arrive at the local maxima
after repeated iterations .
GAIN – It decides how quickly we will reach the point of maxima . It’s
value has to be adjusted otherwise we will not get smooth path to the
point of maxima .
SUMMER – It is used to add the sine wave perturbation to the point
where we have reached in the iteration before , so as to check whether
we need to move forward or backward or we are at the point of
maxima .
RESULTS –
For gain=10
OUTPUT AT SCOPE_2—It gives the maximum value of the objective
function. The maximum value is 25
OUTPUT AT SCOPE_1 – It gives the point at which the maxima of the
objective function occurs. It occurs at u=5
PROPOSED TASK –
Now I am trying to replace this static objective function by a plant and
a controller which would give the a dynamic objective function whose
point of maxima and maximum value will change with time . So there
we will get real power of extremum seeking of finding the maxima of a
changing dynamic function .
REFERENCES –
[1] K.J. Åström and T. Hägglund, PID Controllers: Theory, Design and Tuning 2nd ed. Research
Triangle Park, NC: Instrum. Soc. Amer., 1995.
[2] K.J. Åström, T. Hägglund, C.C. Hang, and W.K. Ho, “Automatic tuning and adaptation for PID
controllers—A survey,” Contr. Eng. Pract., vol. 1, no. 4, pp. 699–714, 1993.
[3] K.J. Åström and T. Hägglund, “Automatic tuning of simple regulators with specifications on phase
and amplitude margins,” Automatica, vol. 20, no. 5, pp. 645–651, 1984.
[4] A. Leva, “PID autotuning algorithm based on relay feedback,” IEEE Proc. D Contr. Theory
Applicat., vol. 140, no. 5, pp. 328–338, 1993.
[5] A.A. Voda and I.D. Landau, “A method for the auto-calibration of PID controllers,” Automatica,
vol. 31, no. 1, pp. 41–53, 1995.
[6] M. Jun and M.G. Safonov “Automatic PID tuning: An application of unfalsified control,” in Proc.
IEEE Int. Symp. CACSD, Hawaii, 1999. pp. 328–333.
[7] M. Saeki, “Unfalsified control approach to parameter space design of PID controllers,” in Proc.
42nd IEEE Conf. Decision and Control, Maui, HI, 2003, pp. 786–791.
[8] M. Saeki, A. Takahashi, O. Hamada, and N. Wada, “Unfalsified parameter space design of PID
controllers for nonlinear plants,” in Proc. IEEE Int. Symp. CACSD, Taipei, Taiwan, 2004, pp. 1521–
1526.
[9] H. Hjalmarsson, M. Gevers, S. Gunnarsson, and O. Lequin, “Iterative feedback tuning: Theory and
applications,“ IEEE Contr. Syst. Mag., vol. 18, no. 4, pp. 26–41, 1998.
[10] O. Lequin, M. Gevers, and T. Triest, “Optimizing the settling time with iterative feedback
tuning,” in Proc. 14th IFAC World Congress, Beijing, P.R. China, 1999, pp. 433–437.
THANK YOU

BTP Report PDF

Uploaded by

Copyright:

Available Formats

BTP Report PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

BTP Report PDF

Uploaded by

Copyright:

Available Formats

BTP REPORT

B.Tech - Electrical Engineering

THEORY OF EXTREMUM SEEKING CONTROL :

For any given objective function there is a set of input

FUNCTIONS OF DIFFERENT BLOCKS IN EXTREMUM

You might also like