Professional Documents
Culture Documents
Design of ANFIS Based Estimation and Control For MIMO Systems
Design of ANFIS Based Estimation and Control For MIMO Systems
ABSTRACT
This work is an attempt to illustrate the usage and
effectiveness of soft computing techniques in the
estimation and control of multi input and multi output
systems. This paper focuses on neuro-fuzzy system
ANFIS (Adaptive Neuro Fuzzy Inference system). An
Adaptive Network based Fuzzy Interference System
architecture extended to cope with multivariable systems
has been used. The performance of the control strategy is
studied on the control of Distillation Column problem.
The results confirmed the control quality improvement
with MPC and PID controller.
Keywords - ANFIS, Distillation Column, PID controller,
MPC, Neural modeling.
1. INTRODUCTION
The controller always aims at the set point value of the given
process variable. This is the main task of the properly
designed controller. The controller should also provide some
flexibility in case of change in set point and disturbances.
Today there are many methods for designing intelligent
controllers, such as predictive controller, fuzzy control,
neural networks and expert systems. Various combinations of
these controllers give a number of design possibilities.
Artificial Neural Networks (ANNs) and Fuzzy Logic (FL)
have been increasingly in use in many engineering fields
since their introduction as mathematical aids [1-2]. Being
branches of Artificial Intelligence (AI), both emulate the
human way of using past experiences, adapting itself
accordingly and generalizing.
Obtaining a mathematical model for a system can be
rather complex and time consuming as it often requires some
assumptions such as defining an operating point and doing
linearization about that point and ignoring some system
parameters, etc. This fact has recently led the researchers to
exploit the neural and fuzzy techniques in modelling complex
systems utilizing solely the input-output data sets. Although
fuzzy logic allows one to model a system using human
knowledge and experience with IF- THEN rules, it is not
always adequate on its own. This is also true for ANNs,
which only deal with numbers rather than linguistic
expressions. This deficiency can be overcome by combining
the superior features of the two methods. To achieve the most
accurate set point, an appropriate extensions and
2. PID CONTROLLER
A ProportionalIntegralDerivative (PID) controller
is a generic control loop feedback mechanism widely used in
industrial control systems. A PID is the most commonly used
feedback controller. A PID controller calculates an "error"
value as the difference between a measured process variable
and a desired set point. The controller attempts to minimize
the error by adjusting the process control inputs.
The
PID controller calculation (algorithm) involves three separate
constant parameters, and is accordingly sometimes called
three-term control: the proportional, the integral and
derivative values, denoted P, I, and D. These values can be
interpreted in terms of time: P depends on the present error, I
on the accumulation of past errors, and D is a prediction of
future errors, based on current rate of change. The weighted
sum of these three actions is used to adjust the process via a
control element such as the position of a control valve, or the
power supplied to a heating element.
Tuning a control loop is the adjustment of its control
parameters (proportional band/gain, integral gain/reset,
derivative gain/rate) to the optimum values for the desired
control response. Stability (bounded oscillation) is a basic
requirement, but beyond that, different systems have different
behavior, different applications have different requirements,
and requirements may conflict with one another.
2803 | P a g e
y k si u k 1 s N u k N
(1)
i 1
where
12e s
18.6e3s
Y1 16.7 s 1 21.0s 1 u1
Y
7 s
19.8e3s u2
2 6.6e
10.9s 1 14.4s 1
(2)
2804 | P a g e
(4)
Ai or Bi 2 is a fuzzy
o3,i wi
wi
for i=1, 2
w1 w2
(6)
o4,i wi f i wi ( pi x qi y ri )
(7)
pi , qi , ri ) is the
5. ANFIS AS AN ESTIMATOR
ANFIS can be used for the estimation of some
dependent variables in chemical process. The designed
ANFIS estimator is used to infer the compositions of top and
bottom products. In estimator design process, different
ANFIS structure are constructed and trained to find the
architecture that gives the best performance as an estimator.
As a first step to design an estimator, training data sets
should be generated to train the estimator networks. These
data sets consist of estimator inputs and desired output
values. They are produced from the process input output data.
Since, ANFIS is a data processing method, it is important
that the input-output data must be within the sufficient
operational range including the maximum and minimum
values for both input and output variables of the system. If
this is not provided, estimator performance cannot be
guaranteed and thus the designed estimator will not be
accurate. Having generated the training data, estimators that
have different architectures are trained with the obtained data
sets. Performances of the trained estimators are evaluated
with model simulations and best estimator architecture is
obtained. These simulations are made to verify and to
generalize the ANFIS structures. Verification is done to show
how good the estimator structure learned the given training
2805 | P a g e
OUTPUT RESPONSE
1.5
Training Data
ANFIS Output
1
0.5
-0.5
0
50
100
150
200
250
TIME
Training Data
ANFIS Output
0
-0.5
-1
-1.5
0
50
100
150
200
250
TIME
2806 | P a g e
0.8
0.6
0.4
0.2
0
-0.2
0
The best fit of ANFIS estimator data for the chosen MIMO
is used for control study. The ANFIS controller is designed
using MATLAB/SIMULINK software. The error and rate of
change of error is used for rule formation in ANFIS
controller. The Gaussian type input membership function is
used for control design. Fig 8, fig 9 presents the ANFIS
controller output.
The ANFIS controller response is compared with
conventional multi loop PID controller tuned by BLT method
and Model Predictive controller (MPC) tuned by Dynamic
Matrix Control. From the fig 10, fig 11 the response of
ANFIS controller is better than other two controllers. The
performance indices ISE and IAE of all three controllers are
listed in Table 1 and Table 2.
50
100
150
200
250
TIME
OUTPUT RESPONSE
Training Data
ANFIS Output
0
0.6
0.4
0.2
0
0
50
100
150
200
250
Time
-0.5
-1
-1.5
0
50
100
150
200
250
TIME
OUTPUT RESPONSE
0.5
0.8
0
-0.2
-0.4
-0.6
-0.8
-1
0
50
100
150
200
250
Time
2807 | P a g e
1
0.8
MPC
ANFIS
PID
0.6
0.4
0.2
0
0
50
100
150
200
250
Time
0.2
0
-0.2
-0.4
MPC
ANFIS
PID
-0.6
-0.8
-1
0
50
100
150
200
250
Time
7. CONCLUSION
The performance of the control strategy is studied on the control of distillation column problem. The results confirmed
the control quality improvement with MPC and multi-loop PID controller. Thus the optimization of the output is obtained and the
performance criteria are compared with the ISE and IAE values.
REFERENCES
[1]
E. H. Mamdani, S. Assilian, An Experiment in Linguistic Synthesis with a Logic Controller, International Journal of ManMachine Studies,8, 1975, 1-13.
[2]
S. Takagi, M. Sugeno, Fuzzy Identification of Fuzzy Systems and its Application to Modeling and Control, IEEE
Transaction on System, Man and Cybernetics, 15, 1985, 116-132.
[3]
R. Jang, ANFIS: Adaptive Network based Fuzzy Inference System, IEEE Transaction on System, Man and Cybernetics,
23(3), 1993, 665-684.
[4]
P. Tatjewski, M. L. Czuk, Soft Computing in Model based Predictive Control, International Journal of Applied Maths and
Computer Science, 16(1), 2006, 7-26.
2808 | P a g e
S. Skogestad, Dynamics and Control of Distillation Column: A Tutorial Introduction, Transaction of Institution of
Chemical Engineers, 75(A), 1997, 539-562.
[6]
J. Fernandez de Canete, An Adaptive Neuro-Fuzzy Approach to Control a Distillation Column, International Journal of
Neural Computing and Applications, 2000, 211-217.
[7]
R. Sivakumar, K. Suresh Manic, V. Nerthiga, R. Akila, K. Balu, Application of Fuzzy Model Predictive Control in
Multivariable Control of Distillation Column, International Journal of Chemical Engineering and Applications, 1(1), 2010,
39-42.
[8]
R. Sivakumar, K. Balu, ANFIS based Distillation Column Control, International Journal of Computer Applications
Special issue on Evolutionary Computation, (2), 2010, 67-73.
2809 | P a g e