7

Fuzzyneural Model Predictive Control of

Multivariable Processes
Michail Petrov, Sevil Ahmed, Alexander Ichtev and Albena Taneva
Technical University Sofia, Branch Plovdiv/Control Systems Department
Bulgaria
1. Introduction
Predictive control is a model-based strategy used to calculate the optimal control action, by
solving an optimization problem at each sampling interval, in order to maintain the output
of the controlled plant close to the desired reference. Model predictive control (MPC) based
on linear models is an advanced control technique with many applications in the process
industry (Rossiter, 2003). The next natural step is to extend the MPC concept to work with
nonlinear models. The use of controllers that take into account the nonlinearities of the plant
implies an improvement in the performance of the plant by reducing the impact of the
disturbances and improving the tracking capabilities of the control system.
In this chapter, Nonlinear Model Predictive Control (NMPC) is studied as a more applicable
approach for optimal control of multivariable processes. In general, a wide range of
industrial processes are inherently nonlinear. For such nonlinear systems it is necessary to
apply NMPC. Recently, several researchers have developed NMPC algorithms (Martinsen et
al., 2004) that work with different types of nonlinear models. Some of these models use
empirical data, such as artificial neural networks and fuzzy logic models. The model
accuracy is very important in order to provide an efficient and adequate control action.
Accurate nonlinear models based on soft computing (fuzzy and neural) techniques, are
increasingly being used in model-based control (Mollov et al., 2004).
On the other hand, the mathematical model type, which the modelling algorithm relies on,
should be selected. State-space models are usually preferred to transfer functions, because
the number of coefficients is substantially reduced, which simplifies the computation;
systems instability can be handled; there is no truncation error. Multi-input multi-output
(MIMO) systems are modelled easily (Camacho et al., 2004) and numerical conditioning is
less important.
A state-space representation of a Takagi-Sugeno type fuzzy-neural model (Ahmed et al.,
2010; Petrov et al., 2008) is proposed in the Section 2. This type of models ensures easier
description and direct computation of the gradient control vector during the optimization
procedure. Identification procedure of the proposed model relies on a training algorithm,
which is well-known in the field of artificial neural networks.
Obtaining an accurate model is the first stage of the of the NMPC predictive control
strategy. The second stage involves the computation of a future control actions sequence. In
order to obtain the control actions, a previously defined optimization problem has to be
solved. Different types of objective and optimization algorithms (Fletcher, 2000) can be used

www.intechopen.com

126

Advanced Model Predictive Control

in the optimization procedure. Two different approaches for NMPC are proposed in Section
3. They consider the unconstrained and constrained model predictive control problem. Both
of the approaches use the proposed Takagi-Sugeno fuzzy-neural predictive model.
The proposed techniques of fuzzy-neural MPC are studied in Section 4, by experimental
simulations in Matlab environment in order to control the levels in a multi tank system
(Inteco, 2009). The case study is capable to show how the proposed NMPC algorithms
handle multivariable processes control problem.

2. Multivariable fuzzy-neural predictive model

The Takagi-Sugeno fuzzy-neural models are powerful modelling tools for a wide class of
nonlinear systems. Fuzzy reasoning is capable of handling uncertain and imprecise
information while neural networks can learn from samples. Fuzzy-neural networks combine
the advantages of both artificial intelligent techniques and incorporate them in adaptive
features. Those futures, based on a real time learning algorithm are the main advantage of
the fuzzy-neural models.
The importance of the used in MPC strategy models and their adaptive characteristics is
obvious. The accuracy of the model determines the accuracy of the control action. The
proposed fuzzy-neural model is implemented in a classical NMPC scheme (Fig. 1) as a
predictor (Camacho et al., 2004).

Fig. 1. Basic structure of the proposed Fuzzy-Neural NMPC

In this chapter a nonlinear discrete time state-space implementation is considered to
represent the system dynamic:
x(k + 1 ) = f x (x(k), u(k))
y(k) = f y (x(k), u(k))

(1)

where x(k) n , u(k) m and y(k) q are state, control and output variables of the
system, respectively. The unknown nonlinear functions fx and fy can be approximated by
Takagi-Sugeno type fuzzy rules in the next form:

www.intechopen.com

127

Fuzzyneural Model Predictive Control of Multivariable Processes

R l : if

z1 ( k ) is Ml 1

and and zi ( k ) is Mli

xl ( k + 1) = A l x( k ) + Bl u( k )
then
yl ( k ) = C l x( k ) + Dl u( k )

and zp ( k ) is Mlp
(2)

where Rl is the l-th rule of the rule base. Each rule is represented by an if-then conception.
The antecedent part of the rules has the following form zi(k) is Mli where zi(k) is an i-th
linguistic variable (i-th model input) and Mli is a membership function defined by a fuzzy set
of the universe of discourse of the input zi. Note that the input regression vector z(k) p in
this chapter contains the system states and inputs z(k)=[x(k) u(k)]T. The consequent part of
the rules is a mathematical function of the model inputs and states. A state-space
implementation is used in the consequent part of Rl, where Al nn , Bl nm , Cl qn
and Dl qm are the state-space matrices of the model (Ahmed et al., 2009).
The states in the next sampling time x ( k + 1) and the system output y ( k ) can be obtained by
taking the weighted sum of the activated fuzzy rules, using
x ( k + 1) = yl ( k )( Al x( k ) + Bl u( k ))
L

y ( k ) = yl ( k )(C l x ( k ) + Dl u( k ))
l =1

(3)

l =1

On the other hand the state-space matrices A, B, C, and D for the global state-space plant
model could be calculated as a weighted sum of the local matrices Al, Bl, Cl, and Dl from the
activated fuzzy rules (2):
A( k ) = Al yl ( k )
L

C ( k ) = C l yl ( k )
l =1
L

l =1

B( k ) = Bl yl ( k )
L

D( k ) = Dl yl ( k )
l =1
L

(4)

l =1

yl is the normalized value of the membership function degree

L

where yl = yl

yl

l =1

the l-th activated fuzzy rule and L is the number of the activated rules at the moment k.

Fig. 2. Gaussian membership functions of the i-th input

Fuzzy implication in the l-th rule (2) can be realized by means of a product composition

www.intechopen.com

upon

128

Advanced Model Predictive Control

p

yl = ij

(5)

i =1

where ij specifies the membership degree (Fig. 2) upon the activated j-th fuzzy set of the
corresponded i-th input signal and it is calculated according to the chosen here Gaussian
membership function (6) for the l-th activated rule:

ij ( zi ) = exp

( zi cGij ) 2
2 ij 2

(6)

where zi is the current input value of the i-th model input, cGij is the centre (position) and ij is
the standard deviation (wide) of the j-th membership function (j=1, 2, .., s) (Fig.2).
2.1 Identification procedure for the fuzzyneural model
The proposed identification procedure determines the unknown parameters in the TakagiSugeno fuzzy model, i.e. the parameters of membership functions, according to their shape
and the parameters of the functions fx and fy in the consequent part of the rules (2). It is
realised by a five-layer fuzzy-neural network (Fig. 3). Each of the layers performs typical
fuzzy logic strategy operations:

Fig. 3. The structure of the proposed fuzzy - neural model

Layer 1. The first layer represents the model inputs through its own input nodes Z1, Z2, ,
Zp. The network synaptic weights are set to one, so the model inputs are directly passed
through the nodes to the next layer. Neurons here are represented by the elements of the
regression vector z(k).

www.intechopen.com

129

Fuzzyneural Model Predictive Control of Multivariable Processes

Layer 2. The fuzzification procedure of the input variables is performed in the second layer.
The weights in this layer are the parameters of the chosen membership functions. Their
number depends on the type and the number of the applied functions. All these parameters
ij are adjustable and take part in the premise term of the Takagi-Sugeno type fuzzy rule
base (2). In that section the membership functions for each model input variable are
represented by Gaussian functions (Fig. 2). Hence, the adjustable parameters ij are the
centres cGij and standard deviations ij of the Gaussian functions (6). The nodes in the second
layer of the fuzzy-neural architecture represent the membership degrees ij(zi) of the
activated membership functions for each model input zi(k) according to (6). The number of
the neurons depends on the number of the model inputs p and the number of the
membership functions s in corresponding fuzzy sets. It is calculated as p s .
Layer 3. The third layer of the network interprets the fuzzy rule base (2). Each neuron in the
third layer has as many inputs as the input regression vector size p. They are the corresponding
membership degrees for the activated membership functions calculated in the previous layer.
Therefore, each node in the third layer represents a fuzzy rule Rl, defined by Takagi-Sugeno
fuzzy model. The outputs of the neurons are the results of the applied fuzzy rule base.
Layer 4. The fourth layer implements the fuzzy implication (5). Weights in this layer are set
to one, in case the rule Rl from the third layer is activated, otherwise weights are zeros.
Layer 5. The last layer (one node layer) represents the defuzzyfication procedure and forms
the output of the fuzzy-neural network (3). This layer also contains a set of adjustable
parameters l. These are the parameters in the consequent part of Takagi-Sugeno fuzzy
model (2). The single node in this layer computes the overall model output signal as the
summation of all signals coming from the previous layer.
I 5 = f yl yl or I 5 = f xl yl O 5 =
L

l =1

l =1

f yl yl
L

yl

l =1
L

f xl yl
L

or O 5 =

l =1

yl

l =1
L

(7)

l =1

where f xl = Al x( k ) + Bl u( k ) and f yl = C l x( k ) + Dl u( k ) .
2.2 Learning algorithm of the fuzzyneural model
Two-step gradient learning procedure is used as a learning algorithm of the internal fuzzyneural model. It is based on minimization of an instant error function EFNN. At time k the
function is obtained from the following equation

EFNN ( k ) = 2 ( k ) / 2

(8)

where the error (k) is calculated as a difference between the controlled process output y(k)
and the fuzzy-neural model output (k):

( k ) = y( k ) y ( k )

(9)

During step one of the procedure, the consequent parameters of Takagi-Sugeno fuzzy rules
are calculated according to summary expression (10) (Petrov et al., 2002).

EFNN

l ( k + 1) = l ( k ) +

www.intechopen.com

(10)

130

Advanced Model Predictive Control

where is a learning rate and l represents an adjustable coefficient aij, bij, cij, dij (11) for the
activated fuzzy rule Rl (2). The coefficients take part in the state matrix Al, control matrix Bl
and output matrices Cl and Dl of the l-th activated rule (Ahmed et al., 2009). The matrices
approximate the unknown nonlinear functions fx and fy according to defined fuzzy rule
model (2). The matrix dimensions are specified by the system parameters numbers of
inputs m, outputs q and states n of the system.
a11 a1n
Al =
an1 ann

b11 b1m

Bl =
bn 1 bnm

c11 c1n

Cl =
c

q 1 cqn

d11 d1m

(11)
Dl =
d

q 1 dqm

In order to find a weight correction for the parameters in the last layer of the proposed
EFNN
fuzzy-neural network the derivative
of the instant error should be determined.

Following the chain rule, the derivative is calculated considering the expressions (7) and (8)

EFNN EFNN y I 5
=

y I 5 l
l

(12)

After the calculation of the partial derivatives, the matrix elements for each matrix of the
state-space equations corresponding to the l-th activated rule (2) are obtained according to
the summary expression (12) (Petrov et al., 2002; Ahmed et al., 2010):
aij ( k + 1) = aij ( k ) + ( k ) yl ( k )xi ( k )

i = j =1n

bij ( k + 1) = bij ( k ) + ( k ) yl ( k )u j ( k )

i = 1 n, j = 1 m

cij ( k + 1) = cij ( k ) + ( k ) yl ( k )x j ( k )

i = 1 q, j = 1 n

dij ( k + 1) = dij ( k ) + ( k ) yl ( k )u j ( k )

i = 1 q, j = 1 m

(13)

The proposed fuzzy-neural architecture allows the use of the previously calculated output
error (8) in the next step of the parameters update procedure. The output error EFNN is
propagated back directly to the second layer, where the second group of adjustable
parameters are situated (Fig. 3). Depending on network architecture, the membership
degrees calculated in the fourth and the second network layer are related as yl ij.
Therefore, the learning rule for the second group adjustable parameters can be done in
similar expression as (10):
E
FNN
ij

ij ( k + 1) = ij ( k ) +

(14)

where the derivative of the output error EFNN is calculated by the separate partial
derivatives:

EFNN EFNN y ij
=

ij
y ij ij

www.intechopen.com

(15)

131

Fuzzyneural Model Predictive Control of Multivariable Processes

The adjustable premise parameters of the fuzzy-neural model are the centre cGij and the
deviation ij of the Gaussian membership function (6). They are combined in the
representative parameter ij, which corresponds to the i-th model input and its j-th activated
fuzzy set. Following the expressions (14) and (15) the parameters are calculated as follows
(Petrov et al., 2002; Ahmed et al., 2010):
[ zi ( k ) cGij ( k )]

cGij ( k + 1) = cGij ( k ) + ( k ) yl ( k )[ f yl y( k )]
ij2 ( k )

ij ( k + 1) = ij ( k ) + ( k ) yl ( k )[ f yl y( k )]

[ zi ( k ) cGij ( k )]2

ij3 ( k )

(16)

(17)

The proposed identification procedure for the fuzzy-neural model could be summarized in
the following steps (Table 1).
Initialize the membership functions number, shape, parameters;
Assign initial values for the network inputs;
Start the algorithm at the current moment k;
Fuzzify the network inputs and calculate the membership degrees upon the
activated fuzzy set of the membership functions according to (6);
Step 5. Perform fuzzy implication according to (5);
Step 6. Calculate the fuzzy-neural network output, which is represented by state-space
description of the modelled system (3) and (4);
Step 7. Calculate the instant error according to (8) and (9);
Step 8. Start training procedure for fuzzy-neural network;
Step 9. Adjust the consequent parameters according to (13);
Step 10. Adjust the premise parameters according to (16) and (17).
Repeat the algorithm from Step 3 for each sampling time.
Step 1.
Step 2.
Step 3.
Step 4.

Table 1. Fuzzy-neural model identification procedure

3. Optimization algorithm of multivariable model predictive control strategy

The model provided by the Takagi-Sugeno type fuzzy-neural network is used to formulate
the objective function for the optimization algorithm and to calculate the future control
actions. The second stage of the predictive control strategy includes an optimization
procedure. It utilizes the obtained results during the first (modelling) stage predictive model
of the system. Using the Takagi-Sugeno fuzzy-neural model (3), the optimization algorithm
computes the future control actions at each sampling period, by minimizing the typical for
MPC strategy (Generalized Predictive Control GPC) cost function (Akesson, 2006):

H p + H w 1

J( k ) =

i =Hw

2
y ( k + i ) r ( k + i ) Q +

Hu 1
i =0

u( k + i )

2
R

(18)

where (k), r(k) and u(k) are the predicted outputs, the reference trajectories, and the
predicted control increments at time k, respectively. The length of the prediction horizon is

www.intechopen.com

132

Advanced Model Predictive Control

Hp, and the first sample to be included in the horizon is Hw. The control horizon is given by
Hu. Q 0 and R >0 are weighting matrices representing the relative importance of each
controlled and manipulated variable and they are assumed to be constant over the Hp.
The cost function (18) may be rewritten in a matrix form as follows
2

J ( k ) = Y ( k ) T ( k ) Q + U ( k )

2
R

(19)

where Y(k), T(k), U(k), Q and R are predicted output, system reference, control variable
increment and weighting matrices, respectively,

u( k| k )
y ( k|k )
r ( k|k )

Y(k) =
, T(k) =
, U ( k ) =

y ( k + H - 1| k )
r ( k + H - 1| k )
u( k + H u - 1| k )
p
p

Q (1)
0

Q=

Q ( H p )
0

R (1)
0

R=
0
R ( H u )

The linear state-space model used for Takagi-Sugeno fuzzy rules (2) could be represented in
the following form:

x ( k + 1) = Ax( k ) + Bu( k 1) + Bu( k )

y ( k ) = Cx ( k ) + Du( k 1) + Du( k )

(20)

Based on the state-space matrices A, B, C and D (4), the future state variables are calculated
sequentially using the set of future control parameters:
x ( k + 1) = Ax( k ) + Bu( k 1) + Bu( k )
x ( k + 2) = A2 x( k ) + ( AB + B)u( k 1) + ( AB + B)u( k ) + Bu( k + 1)
x ( k + 3) = A3 x( k ) + ( A2 B + AB + B)u( k 1) + ( A2 B + AB + B)u( k ) + ( AB + B)u( k + 1) + Bu( k + 2)
.....................................................................................................................
x ( k + j ) = A j x( k ) + Ai Bu( k 1) + Ai B

j 1

j 1

j 1 i

i =0

i =0

m=0

u( k + m)

.....................................................................................................................
H
x ( k + H p ) = A p x( k ) +

H p 1
i =0

H p 1

Ai Bu( k 1) +

i =0

Hp 2

Ai Bu( k ) +

Ai Bu( k + 1) + + A

H p Hu

Bu( k + H u 1)

i =0

The predictions of the output y for j steps ahead could be calculated as follows
y ( k + 1) = Cx( k + 1) + Du( k + 1) = CAx( k ) + (CB + D)u( k 1) + (CB + D)u( k ) + Du( k + 1)
y ( k + 2) = CA2 x( k ) + (CAB + CB + D)u( k 1) + (CAB + CB + D)u( k ) + (CB + D)u( k + 1) + Du( k + 2)
........................................................................................................................................................................

www.intechopen.com

133

Fuzzyneural Model Predictive Control of Multivariable Processes

j 1

j 1

y ( k + j ) = CA j x( k ) + C Ai B + D u( k 1) + C Ai B + D u( k ) + (CB + D)u( k + j 1) + Du( k + j )

i =0

i =0

........................................................................................................................................................................
Hp 2

Hp 2

H 1
y ( k + H p 1) = CA p x( k ) + C Ai B + D u( k 1) + C Ai B + D u( k ) +
i =0

i =0

H
H
H
3
1

p u

+ C Ai B + D u( k + 1) + + C Ai B + D u( k + H u 1)
i =0

i =0

The recurrent equation for the output predictions y ( k + jp ) , where jp= 1, 2,..., Hp 1, is in the
next form:
j p jp 1

C A j B + D u( k + i ), jp < H u

i = 0 j = 0
jp 1

. y ( k+ j ) = CA jp x( k ) + C
.
Ai B + D u( k 1) +

p
j
i
1

i =0

Hu 1
p

j
C A B + D u( k + i ), j p > H u
i
j
0
0
=
=

(21)

The prediction model defined in (21) can be generalized by the following matrix equality
Y ( k ) = x( k ) + u( k - 1) + U ( k )

(22)

where
C
CA

= CA2

Hp 1
CA

CB + D

CAB + CB + D
=

Hp 2

C Ai B + D
i = 0

0
D

+
CB
D
D

CAB + CB + D CB + D

= Hu 2 i
C A B + D
i =0

Hp 2

i
C A B + D
i =0

H p Hu 1

i
C A B + D
i =0

All matrices, which take part in the equations above, are derived by the Takagi-Sugeno
fuzzy-neural predictive model (4).
It is also possible to define the vector
E( k ) = T ( k ) - u( k - 1) - U ( k )

(23)

This vector can be thought as a tracking error, in the sense that it is the difference between the
future target trajectory and the free response of the system, namely the response that would
occur over the prediction horizon if no input changes were made, i.e. U(k)=0. Hence, the
quantity of the so called free response F(k) is defined as follows
F( k ) = x( k ) + u( k - 1)

www.intechopen.com

(24)

134

Advanced Model Predictive Control

3.1 Unconstrained model predictive control

In this section, the study is focused on the optimization problem of the unconstrained
nonlinear predictive control with the quadratic cost function (18). The section presents an
approximate solution of the problem where the information given by the obtained fuzzyneural model is used to solve the problem.
The unconstrained optimization problem can be formulated in a matrix form. First, the
predictor can be constructed as follows
Y ( k ) = U ( k ) + F ( k )

(25)

Second, the cost function (19) can be rewritten as

J ( U ) = (T Y)T Q(T Y) + U T RU

(26)

Hence, substituting the predictive model (25) into expression (26), the cost function of the
model predictive optimization problem can be specified as follows:
J ( U ) = U T (T Q+R )U+ 2(FT )T QU + (T F )T Q(T F )

(27)

The minimum of the function J(U) can be obtained by calculating the input sequence U so
that J/U = 0:

U T (T Q+R )U + 2(FT )T QU + (F T )T Q( F T ) = 0
J ( U ) =

U
U

(28)

Then the optimal sequence U* is

U * = (T Q+R )1 T Q(TF )

(29)

The input applied to the controlled plant at time k is computed according to the receding
horizon principle, i.e. the first element from the control sequence u*(k) of the vector U* is
taken. Then, control signal is calculated from:
u( k ) = u( k - 1) + u * ( k )

(30)

It is evident that the expression given by the matrix equation (29) is the same as expression
obtained for the generalized predictive control. However, in the GPC formulation the
components involved in the calculation of the formula (29) are obtained from a linear model.
In the present case the components introduced in this expression are generated by the
designed nonlinear fuzzy-neural model. A more rigorous formulation of (29) will be
representation of the components as time-variant matrices, as they are shown in the
expression (22). In this case the matrix (k) and the vectors (k), T(k) are being reconstructed
at each sampling time. The vector (k) is obtained by simulating the fuzzy model with the
current input u(k); the matrix (k) is also rebuilt using a method described below.

U * ( k ) = T ( k )Q( k )+R T ( k )Q [T ( k )F( k )]

1

(31)

The proposed method solves the problem of unconstrained MPC. A system of equations is
solved at each sampling time k. The proposed approach decreases computational burden
avoiding the necessity to inverse the gain matrix in (31) at each sampling time k.

www.intechopen.com

Fuzzyneural Model Predictive Control of Multivariable Processes

135

Applying this method, minimization of the GPC criterion (18) is based on a calculation of
the gradient vector of the criterion cost function J at the moment k subject to the predicted
control actions:
T
J ( k )

J ( k )
J ( k )
J ( k ) =
,
, ,

u( k + H u 1)
u( k ) u( k + 1)

(32)

Each element of this gradient vector (32) can be calculated using the following derivative
matrix equation:
J ( k )
Y ( k )
U ( k )
T
= 2 [ T ( k ) Y ( k )] Q
+ 2 U ( k )T R

U ( k )
U ( k )
U ( k )

(33)

From the above expression (33) it can be seen that it is necessary to obtain two groups of
Y ( k )
U ( k )
partial derivatives. The first one is
, and the second one is
. The first
U
(
k
)

U ( k )
partial derivatives in (33) have the following matrix form:

y ( k + H w )
y ( k + H w )

u
k
u
k
H

(
)
(
1)
u

Y(k)

U ( k )
y ( k + H p + H w 1)
y ( k + H p + H w 1)

u( k )
u ( k + H u 1)

(34)

For computational simplicity assume that Hw=0 (18). Then each element of the matrix (34)
is calculated by the expressed equations according to the Takagi-Sugeno rules consequents (2).
For example the derivatives from first column of the matrix (34) have the following form:
L
y ( k )
= Dl yl ( k )
u( k ) l = 1

y ( k + 1) L
= (C l Bl + Dl ) yl ( k + 1)
u( k )
l =1
y ( k + 2) L
= (C l Al Bl + C l Bl + Dl ) yl ( k + 2)
u( k )
l =1

(35)

(36)

(37)

..
y ( k + H p 1)
u( k )

Hp 2
L

= C l Alj Bl + Dl yl ( k + H p 1)

l =1
j =0

The second group partial derivatives in (33) has the following matrix form:

www.intechopen.com

(38)

136

Advanced Model Predictive Control

u (k)
u (k)

....

u ( k )
u ( k + H u 1)

U ( k )

U ( k )

u ( k + H u 1)
u ( k + H u 1)
....

u ( k )
u ( k + H u 1)

(39)

Since u( k ) = u( k ) u( k 1) , the matrix (39) has the following form:

1 0
1 1
U ( k )
= 1
U ( k )

1
1 1

1
1

(40)

Following this procedure it is possible to calculate the rest column elements of the matrix
(34) which belongs to the next gradient vector elements (32). Finally, each element of the
gradient-vector (32) could be obtained by the following system of equations:
y ( k + H p )
y ( k + 1)
J ( k )
= 2 e( k + 1)Q (1)
+ ... + 2 e( k + H p )Q ( H p )
+
u( k + 1)
u( k )
u( k )
+2 R (1)u( k ) 2 R (2)u( k + 1) + ... 2 R ( H )u( k + H 1) = 0

(41)

y ( k + H p )
y ( k + 1)
J ( k )
= 2 e( k + 1)Q (1)
+ ... + 2 e( k + H p )Q ( H p )
+
u( k + 2)
u( k + 1)
u( k + 1)
+2 R (2)u( k + 1) 2 R (3)u( k + 2) + ... 2 R ( H )u( k + H 1) = 0

(42)

y ( k + H p )
y ( k + 1)
J ( k )
= 2 e( k + 1)Q (1)
+ ... + 2 e( k + H p )Q ( H p )
+
u( k + H u 2)
u( k + H u 2)
u( k + H u 2)
(43)

+2 R( H 1)u( k + H 2) 2 R( H )u( k + H 1) = 0
u

y ( k + H p )
y ( k + 1)
J ( k )
= 2 e( k + 1)Q (1)
+ ... + 2 e( k + H p )Q ( H p )
+
u( k + H u 1)
u( k + H u 1)
u( k + H u 1)
+2 R ( H )u( k + H 1) = 0
u

where e( k + j ) = r( k + j ) y ( k + j ),

(44)

j = 1, 2, , H p is the predicted system error.

The obtained system of equations (41)-(44) can be solved very easily, starting from the last
equation (44) and calculating the last control action u(k+Hu-1) first. Then, the procedure can
continue with finding the previous control action u(k+Hu-2) from (43). The calculations
continue until the whole number of the control actions over the horizon Hu is obtained. The

www.intechopen.com

Fuzzyneural Model Predictive Control of Multivariable Processes

137

calculation order of the control actions is very important, since the calculations should
contain only known quantities. After that, only the first control action u(k) (30) will be used
at the moment k as an input to the controlled process. The software implementation of the
proposed algorithm is realized easily according to the following equations:
u( k + H u 1) = R ( H u )

y ( k + H p )
y ( k + 1)
+ + e( k + H p )Q ( H p )
e( k + 1)Q (1)

u( k + H u 1)
u( k + H u 1)

(45)

u( k + H u 2) = u( k + H u 1) +
+ R ( H u 1)

y ( k + H p ) (46)
y ( k + 1)
+ + e( k + H p )Q ( H p )
e( k + 1)Q (1)

u( k + H u 2)
u( k + H u 2)

y ( k + H p )
y ( k + 1)
+ + e( k + H p )Q ( H p )
e( k + 1)Q (1)

u( k + 1)
u( k + 1)

(47)

y ( k + H p )

y ( k + 1)
u( k ) = u( k + 1) + R (1)1 e( k + 1)Q (1)
+ + e( k + H p )Q ( H p )

u( k )
u( k )

(48)

u( k + 1) = u( k + 2) + R (2)

The proposed unconstrained predictive control algorithm could be summarized in the

following steps (Table 2).
Step 1.
Step 2.
Step 3.
Step 4.
Step 5.
Step 6.
Step 7.

Initial identification of the Takagi-Sugeno fuzzy-neural predictive model;

Start the algorithm at the sample k with the initial parameters;
Calculate the predicted model output (k+j) using the tuned fuzzy-neural model (2);
Calculate the derivatives for the matrix (34) according to the equations (35)-(38);
Calculate predicted control actions according to (45)-(48) and update the sequence;
Apply the first optimal control action u(k);
Modify the model parameters into the rule (3) and update them for the next step 3
for the next sample k

Table 2. Basic fuzzy-neural model unconstrained predictive control algorithm

3.2 Constrained model predictive control
The constrained nonlinear predictive control problem can be described as a problem of
finding the optimal input sequence to move a dynamic system to a desired state, taking
into account the constraints on the inputs and the outputs of the control systems. This
section reveals the formulation of the constrained control problem for MPC uses.
Essentially, the problem becomes a quadratic programming problem with linear inequality
constraints (LICQP). It follows by the nature of the operational constraints, which are
usually described by linear inequalities of the control and plant variables.
The problem of nonlinear constrained predictive control is formulated as a nonlinear
quadratic optimization problem. By means of local linearization (20) the problem can be
solved using QP. That way the solution to the linear constrained predictive control problem
is obtained. At each sampling time the LICQP is solved with new parameters, which are
obtained by the Takagi-Sugeno fuzzy-neural model. An active set method is used for
solving the constructed quadratic programming problem.

www.intechopen.com

138

Advanced Model Predictive Control

3.2.1 Constraint types in model predictive control

The operational constraints may be classified in three major types according to the type of
the system variables, which they are imposed on. The first two types of constraints deal with
the control variable incremental variation u(k) and control variable u(k). The third type is
concerned with output y(k) or state variable x(k) constraints.
Related to the origin model predictive control problem, the constraints are expressed in a set
of linear equations. All types of constraints are taken into consideration for each moving
horizon window.

Umin ( k ) U ( k ) Umax ( k )
U min ( k ) U ( k ) Umax ( k )

(49)

Ymin ( k ) Y ( k ) Ymax ( k )
Where
umax ( k )

u ( k + 1)
max

U max ( k ) =

umax ( k + N u 1)

umax ( k )

u ( k + 1)
max

U max ( k ) =

umax ( k + N u 1)

umin ( k )

u ( k + 1)
min

Umin ( k ) =

umin ( k + N u 1)

umin ( k )

u ( k + 1)
min

Umin ( k ) =

umin ( k + N u 1)

ymax ( k )

y ( k + 1)
max

Ymax ( k ) =

ymax ( k + N p 1)
ymin ( k )

y ( k + 1)
min

Ymin ( k ) =

ymin ( k + N p 1)

Therefore, for multi-input case the number of the constraints for the change of the control
variable u(k) is mNu. Similarly, the number of the constraints for the control variable
amplitude is also mNu and for the output constraints it is qNp.
3.2.2 Quadratic programming in use of constrained MPC
Since the cost function J(k) (19) is quadratic and the constraints are linear inequalities, the
problem of finding an optimal predictive control becomes one of finding an optimal solution
to a standard quadratic programming problem with linear inequality constraints

www.intechopen.com

Fuzzyneural Model Predictive Control of Multivariable Processes

1 T
x Hx + f T x
2
subject to Ax b
minJ ( x ) =

139

(50)

where H and f are the Hessian and the gradient of the Lagrange function, x is the decision
variable. Constraints on the QP problem (50) are specified by Ax b according to (49).
The Lagrange function is defined as follows
L( x , ) = J ( x ) + i ai , i = 1, 2,N ,
N

(51)

i =1

where i are the Lagrange multipliers, ai are the constraints on the decision variable x, N is
the number of the constraints considered in the optimization problem.
Several algorithms for constrained optimization are described in (Fletcher, 2000). In this
chapter a primal active set method is used. The idea of active set method is to define a set S
of constraints at each step of algorithm. The constraints in this active set are regarded as
equalities whilst the rest are temporarily disregarded and the method adjusts the set in
order to identify the correct active constraints on the solution to (52)
1 T
x Hx + f T x
2
subject to ai x = bi
ai x bi

min J ( x ) =

(52)

At iteration k a feasible point x(k) is known which satisfies the active constraints as
equalities. Each iteration attempts to locate a solution to an equality problem (EP) in which
only the active constraints occur. This is most conveniently performed by shifting the origin
to x(k) and looking for a correction (k) which solves
1

min T Hx + f T
2

subject to ai = 0 ai S

(53)

where f(k) is defined by f(k) =f + Hx(k) and is J ( x ( k )) for the function defined by (52). If
(k) is feasible with regard to the constraints not included in S, then the feasible point in
next iteration is taken as x(k+ 1) = x(k) + (k). If not, a line search is made in the direction of
(k) to find the best feasible point. A constraint is active if the Lagrange multipliers i 0, i.e.
it is at the boundary of the feasible region defined by the constraints. On the other hand, if
there exist i < 0, the constraint is not active. In this case the constraint is relaxed from the
active constraints set S and the algorithm continues as before by solving the resulting
equality constraint problem (53). If there is more than one constraint with corresponding
i < 0, then the min i ( k ) is selected (Fletcher, 2000).
iS

The QP, described in that way, is used to provide numerical solutions in constrained MPC
problem.

www.intechopen.com

140

Advanced Model Predictive Control

3.2.3 Design the constrained model predictive problem

The fuzzy-neural identification procedure from the Section 2 provides the state-space matrices,
which are needed to construct the constrained model predictive control optimization problem.
Similarly to the unconstrained model predictive control approach, the cost function (18) can
be specified by the prediction expressions (22) and (23).
T

J(k) = [ x(k) + u(k -1) + U(k)-T(k)] Q [ x(k) + u(k -1) + U(k)-T(k)] + U T (k)R U(k)=
T

= [ U(k)-E(k)] Q [ U(k)-E(k)] + U T (k)R U(k)=

= U T (k) T Q+R U(k) + E T (k)QE(k) - 2 U T (k) T QE(k)

Assuming that
H = T Q + R and = 2T QE( k ),

(54)

the cost function for the model predictive optimization problem can be specified as follow
J ( k ) = U T ( k ) U ( k ) - U T ( k ) + ET ( k )QE( k )

(55)

The problem of minimizing the cost function (55) is a quadratic programming problem. If
the Hessian matrix H is positive definite, the problem is convex (Fletcher, 2000). Then the
solution is given by the closed form
U =

1 1
H
2

(56)

The constraints (49) on the cost function may be rewritten in terms of U(k).
U min ( k ) I uu( k - 1) + I u U ( k ) Umax ( k )
U min ( k ) U ( k ) U max ( k )

(57)

Ymin ( k ) x( k ) + u( k - 1) + U ( k ) Ymax ( k )

where Im mm

Im
Im
I
I
is an identity matrix, I u = m mNu m , I u = m

I
m
Im

0
Im

Im

0
0
mNu mN u .

Im

All types of constraints are combined in one expression as follows

U min + I uu( k 1)
I u

(
1)
U
I
u
k

u
max
u

U min

U max
I

Y + ( x( k ) + u( k 1))

min

Ymax ( x( k ) + u( k 1))

www.intechopen.com

(58)

Fuzzyneural Model Predictive Control of Multivariable Processes

141

where I mN u mN u is an identity matrix.

Finally, following the definition of the LIQP (50), the model predictive control in presence of
constraints is proposed as finding the parameter vector U that minimizes (55) subject to the
inequality constraints (58).
min J ( k ) = U T H U - U T + ET QE
subject to U

(59)

In (59) the constraints expression (58) has been denoted by U , where is a matrix
with number of rows equal to the dimension of and number of columns equal to the
dimension of U. In case that the constraints are fully imposed, the dimension of is equal
to 4mNu + 2qNp, where m is the number of system inputs and q is the number of
outputs. In general, the total number of constraints is greater than the dimension of the U.
The dimension of represents the number of constraints.
The proposed model predictive control algorithm can be summarized in the following steps
(Table 3).
At each sampling time:
Step 1. Read the current states, inputs and outputs of the system;
Step 2. Start identification of the fuzzy-neural predictive model following Algorithm 1;
Step 3. With A(k), B(k), C(k), D(k) from Step 2 calculate the predicted output Y(k) according
to (17);
Step 4. Obtain the prediction error E(k) according to (23);
Step 5. Construct the cost function (55) and the constraints (58) of the QP problem;
Step 6. Solve the QP problem according to (59);
Step 7. Apply only the first control action u(k).
Table 3. State-space implementation of fuzzy-neural model predictive control strategy
At each sampling time, LIQP (59) is solved with new parameters. The Hessian and the
Lagrangian are constructed by the state-space matrices A(k), B(k), C(k) and D(k) (4) obtained
during the identification procedure (Table 1). The problem of nonlinear constrained
predictive control is formulated as a nonlinear quadratic optimization problem. By means of
local linearization a relaxation can be obtained and the problem can be solved using
quadratic programming. This is the solution of the linear constrained predictive control
problem (Espinosa et al., 2005).

4. Fuzzy-neural model predictive control of a multi tank system. Case study

The case study is implemented in MATLAB/Simulink environment with Inteco Multi
tank system. The Inteco Multi tank System (Fig. 4) comprises from three separate tanks
fitted with drain valves (Inteco, 2009). The additional tank mounted in the base of the set-up
acts as a water reservoir for the system. The top (first) tank has a constant cross section,
while others are conical or spherical, so they are with variable cross sections. This causes the
main nonlinearities in the system. A variable speed pump is used to fill the upper tank. The
liquid outflows the tanks by the gravity. The tank valves act as flow resistors C1, C2, C3. The
area ratio of the valves is controlled and can be used to vary the outflow characteristic. Each
tank is equipped with a level sensor PS1, PS2, PS3 based on hydraulic pressure measurement.

www.intechopen.com

142

Advanced Model Predictive Control

Fig. 4. Controlled laboratory multi tank system

The linearized dynamical model of the triple tank system could be described by the linear
state-space equations (2) where the matrices A, B, C and D are as follow (Petrov et al., 2009):

awH 111

1
A=
1
w ( c + b H 1 H 1 max ) H 1 1

B=

1
aw

awH 11 1

www.intechopen.com

0
2
w ( c + b H 2 H 2 max ) H 212

2
w R ( H 3 max H 3 ) 2 H 212
2

1 3
2
2
w R ( H 3 max H 3 ) H 3
0

w ( c + b H 2 H 2 max ) H 21 2
0

1
w R ( H 3 max H 3 )2 H 31 3
2

Fuzzyneural Model Predictive Control of Multivariable Processes

143

1 0 0
C = 0 1 0
0 0 1

(60)

0 0 0 0

D = 0 0 0 0
0 0 0 0
The parameters 1, 2 and 3 are flow coefficients for each tank of the model. The described
linearized state-space model is used as an initial model for the training process of the fuzzyneural model during the experiments.

4.1 Description of the multi tank system as a multivariable controlled process

Liquid levels 1, 2, 3 in the tanks are the state variables of the system (Fig. 4). The Inteco
Multi Tank system has four controlled inputs: liquid inflow q and valves settings C1, C2, C3.
Therefore, several models of the tanks system can be analyzed (Fig. 5), classified as pumpcontrolled system, valve-controlled system and pump/valve controlled system (Inteco, 2009).

Fig. 5. Model of the Multi Tank system as a pump and valve-controlled system
In this case study a multi-input multi-output (MIMO) configuration of the Inteco Multi Tank
system is used (Fig. 5). This corresponds to the linearized state-space model (60). Several
issues have been recognized as causes of additional nonlinearities in plant dynamics:

nonlinearities (smooth and nonsmooth) caused by shapes of tanks;

saturation-type nonlinearities, introduced by maximum or minimum level allowed in

tanks;

nonlinearities introduced by valve geometry and flow dynamics;

nonlinearities introduced by pump and valves input/output characteristic curve.

The simulation results have been obtained with random generated set points and following
initial conditions (Table 4):
Model predictive
controller parameters
Inteco Multi tank
system parameters
Operational constraints
on the system
Simulation parameters

Prediction horizon Hp=10

First included sample of the prediction horizon Hw=1
Control horizon Hu=3
Flow coefficients for each tank
1=0.29; 2=0.2256; 3=0.2487
Constraints on valve cross section ratio 0 Ci 2e-04, i=1,2,3
Constraint on liquid inflow 0 q 1e-04 m3/s
Constraints on liquid level in each tank 0 Hi 0.35 m, i=1,2,3
Time of simulation 600 s
Sample time Ts=1 s

Table 4. Simulation parameters for unconstrained and constrained fuzzy-neural MPC

www.intechopen.com

144

Advanced Model Predictive Control

Figures below show typical results for level control problem. The reference value for each
tank is changed consequently in different time. The proposed fuzzy-neural identification
procedure ensures the matrices for the optimization problem of model predictive control at
each sampling time Ts. The plant modelling process during the unconstrained and
constrained MPC experiments are shown in Fig. 6 and Fig. 9, respectively.
4.2 Experimental results with unconstrained model predictive control
The proposed unconstrained model predictive control algorithm (Table 2) with the TakagiSugeno fuzzy-neural model as a predictor has been applied to the level control problem. The
experiments have been implemented with the parameters in Table 4. The weighting
matrices are specified as follow: Q = 0.01 * diag(1, 1, 1) and R = 10 e 4 * diag(1, 1, 1, 1) . Note

identification H3, m

identification H2, m

identification H1, m

that the weighting matrix R is constant over all prediction horizon, which allows to avoid
matrix inversion at each sampling time with one calculation of R 1 at time k=0.
plant output
model output

0.3
0.2
0.1
0
-0.1

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

0.4
0.2
0

0.4

0.2

Fig. 6. Fuzzy-neural model identification procedure of the multi tank system

unconstrained NMPC
The next two figures - Fig. 7 and Fig. 8, show typical results regarding level control, where
the references for H1, H2 and H3 are changed consequently in different time. The change of
every level reference behaves as a system disturbance for the other system outputs (levels).
It is evident that the applied model predictive controller is capable to compensate these
disturbances.

www.intechopen.com

145

Fuzzyneural Model Predictive Control of Multivariable Processes

0.3

reference
FNN MPC

H1, m

0.2
0.1
0
0

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

0.3

H2, m

0.2
0.1
0

0.3

H3, m

0.2
0.1
0

Fig. 7. Transient responses of multi tank system outputs unconstrained NMPC

pump flow, m3/s

-4

x 10

1
0

100

200

300
time,sec

400

500

600

0
-4
x 10

100

200

300
time,sec

400

500

600

0
-4
x 10

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

-4

C1

2
1
0

C2

2
1
0
2
C3

x 10

1
0

Fig. 8. Transient responses of multi tank system inputs unconstrained NMPC

www.intechopen.com

146

Advanced Model Predictive Control

identification H3, m

identification H2, m

identification H1, m

4.3 Experimental results with fuzzy-neural constrained predictive control

The experiments with the proposed constrained model predictive control algorithm (Table
3) have been made with level references close to the system outputs constraints. The
weighting matrices in GPC cost function (19) are specified as Q = diag(1, 1, 1) and
R = 15 e 4 * diag(1, 1, 1, 1) . System identification during the experiment is shown on Fig. 9.
The proposed identification procedure uses the linearized model (60) of the Multi tank
system as an initial condition.

0.4
plant output
model output

0.3
0.2
0.1
0

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

0.4
0.2
0
-0.2

0.4
0.2
0
-0.2

Fig. 9. Fuzzy-neural model identification procedure of the multi tank system

constrained NMPC
The proposed constrained fuzzy-neural model predictive control algorithm provides an
adequate system response as it can be seen on Fig. 10 and Fig. 11. The references are achieved

www.intechopen.com

147

Fuzzyneural Model Predictive Control of Multivariable Processes

0.4
reference
liquid level

H1, m

0.3
0.2
0.1
0

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

0.4

H2, m

0.3
0.2
0.1
0

0.4

H3, m

0.2
0
-0.2

Fig. 10. Transient responses of the multi tank system outputs constrained NMPC

pump flow, m3/s

-5

20

x 10

10
0
0
-4
x 10

100

200

300
time,sec

400

500

600

0
-4
x 10

100

200

300
time,sec

400

500

600

0
-4
x 10

100

200

300
time,sec

400

500

600

100

200

300
time,sec

400

500

600

C1

2
1
0
-1

C2

2
1
0
-1

C3

2
1
0
-1

Fig. 11. Transient responses of the multi tank system inputs constrained NMPC

www.intechopen.com

148

Advanced Model Predictive Control

without violating the operational constraints specified in Table 4. Similarly to the

unconstrained case, the Takagi-Sugeno type fuzzy-neural model provides the state-space
matrices A, B and C (the system is strictly proper, i.e. D=0) for the optimization procedure of
the model predictive control approach. Therefore, the LIQP problem is constructed with
fresh parameters at each sampling time and improves the adaptive features of the applied
model predictive controller. It can be seen on the next figures that the disturbances, which
are consequences of a sudden change of the level references, are compensated in short time
without violating the proper system work.

5. Conclusions
This chapter has presented an effective approach to fuzzy model-based control. The
effective modelling and identification techniques, based on fuzzy structures, combined with
model predictive control strategy result in effective control for nonlinear MIMO plants. The
goal was to design a new control strategy - simple in realization for designer and simple in
implementation for the end user of the control systems.
The idea of using fuzzy-neural models for nonlinear system identification is not new,
although more applications are necessary to demonstrate its capabilities in nonlinear
identification and prediction. By implementing this idea to state-space representation of
control systems, it is possible to achieve a powerful model of nonlinear plants or processes.
Such models can be embedded into a predictive control scheme. State-space model of the
system allows constructing the optimization problem, as a quadratic programming problem.
It is important to note that the model predictive control approach has one major advantage
the ability to solve the control problem taking into consideration the operational constraints
on the system.
This chapter includes two simple control algorithms with their respective derivations. They
represent control strategies, based on the estimated fuzzy-neural predictive model. The twostage learning gradient procedure is the main advantage of the proposed identification
procedure. It is capable to model nonlinearities in real-time and provides an accurate model
for MPC optimization procedure at each sampling time.
The proposed consequent solution for unconstrained MPC problem is the main contribution
for the predictive optimization task. On the other hand, extraction of a local linear model,
obtained from the inference process of a TakagiSugeno fuzzy model allows treating the
nonlinear optimization problem in presence of constraints as an LIQP.
The model predictive control scheme is employed to reduce structural response of the
laboratory system - multi tank system. The inherent instability of the system makes it
difficult for modelling and control. Model predictive control is successfully applied to the
studied multi tank system, which represents a multivariable controlled process. Adaptation
of the applied fuzzy-neural internal model is the most common way of dealing with plants
nonlinearities. The results show that the controlled levels have a good performance,
following closely the references and compensating the disturbances.
The contribution of the proposed approach using TakagiSugeno fuzzy model is the
capacity to exploit the information given directly by the TakagiSugeno fuzzy model. This
approach is very attractive for systems from high order, as no simulation is needed to obtain
the parameters for solving the optimization task. The models state-space matrices can be

www.intechopen.com

Fuzzyneural Model Predictive Control of Multivariable Processes

149

generated directly from the inference of the fuzzy system. The use of this approach is very
attractive to the industry for practical reasons related with the capacity of this model
structure to combine local models identified in experiments around the different operating
points.

6. Acknowledgment
The authors would like to acknowledge the Ministry of Education and Science of Bulgaria,
Research Fund project BY-TH-108/2005.

7. References
Ahmed S., M. Petrov, A. Ichtev (July 2010). Fuzzy Model-Based Predictive Control
Applied to Multivariable Level Control of Multi Tank System. Proceedings of 2010
IEEE International Conference on Intelligent Systems (IS 2010), London, UK. pp. 456
- 461.
Ahmed S., M. Petrov, A. Ichtev, Model predictive control of a laboratory model coupled
water tanks, in Proceedings of International Conference Automatics and Informatics09,
October 14, 2009, Sofia, Bulgaria. pp. VI-33 - VI-35.
kesson Johan. MPCtools 1.0Reference Manual. Technical report ISRN LUTFD2/TFRT-7613--SE, Department of Automatic Control, Lund Institute of Technology,
Sweden, January 2006.
Camacho E. F., C. Bordons (2004). Model Predictive Control (Advanced Textbooks in
Control and Signal Processing). Springer-Verlag London, 2004.
Espinosa J., J. Vandewalle and V. Wertz. Fuzzy Logic, Identification and Predictive Control.
(Advances in industrial control). Springer-Verlag London Limited, 2005.
Fletcher R. (2000). Practical Methods of Optimization. 2nd.ed., Wiley, 2000.
Inteco
Ltd.
(2009).
Multitank
System
Users
Manual.
Inteco
Ltd.,
http://www.inteco.com.pl.
Lee, J.H.; Morari, M. & Garcia, C.E. (1994). State-space interpretation of model predictive
control, Automatica, 30(4), pp. 707-717.
Maciejowski J. M. (2002). Predictive Control with Constraints. Prentice Hall Inc., NY, USA,
2002.
Martinsen F., Lorenz T. Biegler, Bjarne A. Foss (2004). A new optimization algorithm with
application to nonlinear MPC, Journal of Process Control, vol.14, pp 853865, 2004.
Mendona L.F., J.M. Sousa J.M.G. S da Costa (2004). Optimization Problems in
Multivariable Fuzzy Predictive Control, International Journal of Approximate
Reasoning, vol. 36, pp. 199221, 2004 .
Mollov S, R. Babuska, J. Abonyi, and H. Verbruggen (October 2004). Effective Optimization
for Fuzzy Model Predictive Control. IEEE Transactions on Fuzzy Systems, Vol. 12,
No. 5, pp. 661 675.
Petrov M., A. Taneva, T. Puleva, S. Ahmed (September, 2008). Parallel Distributed NeuroFuzzy Model Predictive Controller Applied to a Hydro Turbine Generator.
Proceedings of the Forth International IEEE Conference on "Intelligent Systems",

www.intechopen.com

150

Advanced Model Predictive Control

Golden Sands resort, Varna, Bulgaria. ISBN 978-1-4244-1740-7, Vol. I, pp. 9-20 - 925.
Petrov M., I. Ganchev, A. Taneva (November 2002). Fuzzy model predictive control of
nonlinear processes. Preprints of the International Conference on "Automation and
Informatics 2002", Sofia, Bulgaria, 2002. ISBN 954-9641-30-9, pp. 77-80.
Rossiter J.A. (2003). Model based predictive control A practical Approach. CRC Press, 2003.

www.intechopen.com

Advanced Model Predictive Control

Edited by Dr. Tao ZHENG

ISBN 978-953-307-298-2
Hard cover, 418 pages
Publisher InTech

Published online 24, June, 2011

Published in print edition June, 2011

Model Predictive Control (MPC) refers to a class of control algorithms in which a dynamic process model is
used to predict and optimize process performance. From lower request of modeling accuracy and robustness
to complicated process plants, MPC has been widely accepted in many practical fields. As the guide for
researchers and engineers all over the world concerned with the latest developments of MPC, the purpose of
"Advanced Model Predictive Control" is to show the readers the recent achievements in this area. The first part
of this exciting book will help you comprehend the frontiers in theoretical research of MPC, such as Fast MPC,
Nonlinear MPC, Distributed MPC, Multi-Dimensional MPC and Fuzzy-Neural MPC. In the second part, several
excellent applications of MPC in modern industry are proposed and efficient commercial software for MPC is
introduced. Because of its special industrial origin, we believe that MPC will remain energetic in the future.

How to reference

In order to correctly reference this scholarly work, feel free to copy and paste the following:
Michail Petrov, Sevil Ahmed, Alexander Ichtev and Albena Taneva (2011). Fuzzyneural Model Predictive
Control of Multivariable Processes, Advanced Model Predictive Control, Dr. Tao ZHENG (Ed.), ISBN: 978-953307-298-2, InTech, Available from: http://www.intechopen.com/books/advanced-model-predictivecontrol/fuzzy-neural-model-predictive-control-of-multivariable-processes

InTech Europe

University Campus STeP Ri

Slavka Krautzeka 83/A
51000 Rijeka, Croatia
Phone: +385 (51) 770 447
Fax: +385 (51) 686 166
www.intechopen.com

InTech China

Unit 405, Office Block, Hotel Equatorial Shanghai

No.65, Yan An Road (West), Shanghai, 200040, China
Phone: +86-21-62489820
Fax: +86-21-62489821