Nonlinear Control of Bioprocess Using Feedback Linearization, Backstepping, and Luenberger Observers ()
1. Introduction
In process control, a major difficulty is to provide direct real-time measurements of the state variables required to implement advanced monitoring and control methods on bioreactors [1] -[5] . Dissolved oxygen concentration in bioreactors, temperature in non-isothermal reactors and gaseous flow rates, are available for on-line measurement while the values of concentration of products, reactants and/or biomass are often available only via on-line analysis [2] - [4] , which means that these variables are not available for real-time feedback control. An alternative is to use state observers which, in conjunction with the process model and available measurements, can generate accurate estimates of the unmeasured and/or inaccessible states effectively. Exponential and asym- ptotic observers and their variants to estimate unmeasured states in bioprocess systems have appeared in [1] - [5] . In [6] , Dochain and Perrier applied backstepping [7] - [9] , techniques to the nonlinear control of microbial growth problem in a CSTR (continuously stirred tank reactor) and two controllers were proposed. The first one was a non-adaptive version, while the second one was an adaptive version in which the maximum specific growth rate was estimated on-line. However, backstepping-based observer design was not considered in [6] .
In this paper, a complete development of observer-based control is presented that includes feedback linearization [7] [8] [10] [11] , backstepping [7] - [9] , Luenberger observer [12] with feedback linearization, and Luenberger observer with backstepping.
The paper is organized as follows. Section 2 presents the bioprocess model for control design. Theoretical foundation of input-output feedback linearization (FBL) and controller design are outlined in Section 3 with simulation results. Section 4 addresses the formulation and application of backstepping (BS) control with simulation results. In Section 5, simulation results are compared for both approaches, i.e., FBL and BS. Section 6 addresses the design of Luenberger observers for FBL and BS controls with simulations. The conclusions are presented in Section 7.
2. Bioprocess Model
The model dynamics in a CSTR (continuous stirred tank reactor) with a simple microbial growth reaction, with one substrate
and biomass
, are given by the following equations [1] :
, (1)
, (2)
where
, represent the yield coefficient, specific growth rate (h−1), dilution rate (h−1), and substrate concentration (grams/lit) in the influent and reactor, respectively.
The biomass concentration
(grams/lit) is the variable which is to be controlled. Defining the parameter
as
and expressing specific growth rate
as
, the dynamical Equations (1) and (2) above can be written as [6]
(3)
where it is assumed that the biomass concentration
can be measured with a sensor, i.e., the output is given by
, while
denotes the substrate concentration and
is the control input. The bioprocess model given by (3) can be written compactly in an alternate state-space form as:
(4)
where
and
. Note that in (4),
has been written using “Monod form” for reac-
tion kinetics, which can expressed as
, (5)
. (6)
We will use (3) for back stepping control and observer design and (4) will be utilized for developing the control law and observer design using the feedback linearization approach. Typical values of the model parameters needed for the simulation studies are given in Table 1 [6] .
3. Feedback Linearization (FBL) Control Design
The main intent of this section is to investigate control design using the input-output feedback linearization (FBL) technique. Consider a general nonlinear control-affine SISO system described by [7] [8] [10] [11] ,
, (7)
, (8)
where
is the state vector,
are the control and output signals, respectively;
is a smooth function, and
are smooth vector fields on D, where D is an open set. Given the nonlinear system (7) and the measurement (8), our goal is to find a diffeomorphism or nonlinear transformation of the form
,
with
that transforms the nonlinear system in the
-coordinates to a linear system in the
-coordinates. Differentiating the output
with respect to t yields
, (9)
where
and
denote the Lie derivatives of
with respect to
and
, respectively. If
, then
is independent of
. Continuing successive differentiation
times until
appears explicitly, we obtain
, (10)
where
is a nonlinearity cancellation factor and
is a scalar function. The smallest integer
for which
appears is referred to as the relative degree, i.e., when
. The nonlinear system (7) - (8) is said to have a well-defined relative degree
in a region
if
(11)
for all
. Note that
. From (10), define
, (12)
where
is a one-dimensional transformed input created by the feedback linearization process. Equation (12) yields the linearizing feedback control law [7] [8] [10] [11] :
, (13)
provided
is invertible. If
, then
(14)
If ρ < n, the diffeomorphism Tfbl(x) comprises of both external and internal dynamics, i.e.,
, where
represents the external dynamics state vector and
the internal dynamics state vector, respectively; furthermore, the differential equation for
is linear, while that for
is typically nonlinear. For the bioprocess model given by (4),
, so the system is fully linearizable. We obtain, from (4), (12) and (14),
(15)
, (16)
, (17)
where
is a local diffeomorphism for the system. Using (17) with (12) for
, the original nonlinear system described by (3) and (4) is transformed into a linear system of the form
(18)
where
(19)
and
and
are, respectively, controllable and observable pairs. A suitable tracking control law for the transformed input
in the linear system (18) for
can be formulated as, with (17),
, (20)
where
,
is a bounded reference with bounded derivatives
and
,
,
and the constant feedback gain matrix
is determined such that
is Hurwitz. Furthermore, the gain matrix
can be determined by various design methods, such as pole placement (PP) and linear quadratic regulator (LQR). We shall focus on the PP design in this paper. Substituting (20) into (18) yields the closed-loop system
(21)
The linearizing feedback control law in the
-domain can be written by setting
in (13) as
, (22)
which yields the closed-loop system
(23)
The design of a PP control law (20) for the 2nd-order system (18) is achieved by choosing a damping ratio
that prohibits overshoot, and an undamped natural frequency
(rad/sec). The resulting closed-
loop poles are given by
where
and the resulting gain
is computed with Matlab’s ACKER command. Simulation studies for the closed-loop system with FBL control were conducted using (23). The controller performance was evaluated for a square-wave set-point reference
that alternates every 20 hours between 3 grams/lit and 4 grams/lit as shown in Figure 1 (dotted line). The initial conditions were chosen as
and
[6] . The simulation results are depicted in Figure 1 which shows that the responses are satisfactory.
4. Backstepping (BS) Control Design
We shall address the design of back stepping (BS) [7] -[9] control in this section, where the parameter
is assumed known. The objective here is to design a BS control law
such that the output
tracks the reference
. We will also compare the performance of the closed-loop bioprocess under FBL control
given by (22) and the BS control
to be developed below. The formulation presented here considers a general bounded differentiable reference signal
instead of the constant set-point regulation in [6] . Consider the nonlinear system in the form of (3) reproduced below for ease of reference:
(3)
where
is a known constant parameter. We treat
as the virtual control of the first subsystem
in (3) and let
be the stabilizing function such that
racks
. Define the tracking errors as
(24)
, (25)
![]()
Figure 1. Responses of closed-loop bioprocess (23) under FBL control
with PP design.
where
is the error between
and
. Taking the derivative of
yields, with (25),
. (26)
Consider the Lyapunov function candidate
, (27)
which yields the derivative, with (26),
. (28)
Choosing the stabilizing function
to make
in (28) yields
. (29)
Substituting (29) into (26) and (28) yields, respectively,
, (30)
(31)
From (31), if
, then
and the origin
is globally asymptotically stable, whereby achieving global tracking with
. The term
will be addressed in the next step.
The next step is to develop a BS control law for
. The derivative of
given by (25) satisfies, with (3) and (29),
, (32)
, (33)
where
, (34)
and
is given by (6).
To stabilize the
-system the Lyapunov function candidate as
. (35)
The derivative of
is given by, with (31) and (33),
. (36)
Defining
to be the BS control, and choosing
to make the term
in (36) yields
. (37)
Substituting (37) into (33) and (36), we obtain,
(38)
. (39)
Since
for all
, it follows that
is globally asymptotically stable. Additionally, the stability result can also be established by combing the error equations from (30) and (38) as
. (40)
Since
is a skew-symmetric Hurwitz matrix for all
and
, it follows that the equilibrium
is globally asymptotically stable. Moreover,
is an observable pair, where
(see (53)). Since (40) is in the form of a standard linear time-invariant (LTI) system, a Luenberger observer [12] for state estimation can be constructed for the system, and will be investigated in Section 6.
Meanwhile, the closed-loop bioprocess under BS control is given by,
(41)
where
is given by (37).
Simulation studies were conducted using (41) with the backstepping gains
. The reference signal
and the initial condition
were same as those used for the FBL control in Section 3. The simulation is depicted in Figure 2 which shows that the responses were satisfactory.
5. Comparison of FBL and BS Designs
The simulation results for the FBL versus BS designs using the gains reported in Sections 3 and 4 are shown in Figure 3 and Figure 4 for comparison purposes.
It can be seen that both
and
asymptotically with no overshoot. It can also be seen that the magnitudes of
are slightly larger than those of
. However, the reverse can also be obtained by tuning
and
.
6. Observer-Based FBL and BS Controls
As mentioned before that not all state variables are measured in the bioreactor systems; therefore, suitable observers are needed for realizing the full-state feedback control designs proposed in Sections 3 and 4. We shall
![]()
Figure 2. Responses of closed-loop bioprocess (41) under BS control
for
.
![]()
Figure 3. Comparing FBL responses in Figure 1 and BS responses in Figure 2.
investigate the constructions of Luenberger observers for the FBL-based and BS-based control approaches in this section
6.1. Luenberger Observer for FBL Control
Since only
is measured in (4), a Luenberger observer [12] can be constructed for full-state estimation needed for full-state control of the bioprocess system. Using (21), a full-state observer can be constructed as
(42)
where
is the observer system matrix and
the observer gain matrix to be determined such that
is Hurwitz, provided that
is an observable pair (which is the case in the present problem). The gain matrix L in (42) can be computed using a Luenberger observer [12] with pole placement (PP) and/or Kalman-Bucy filter [13] design techniques. We shall focus on the PP design method; henceforth L and
in (42) will be denoted by
and
, respectively. It should be noted that for a general LTI system characterized by
, where
is an
observable pair, the pair
may not be observable, because full-state feedback can destroy ob-
servability; furthermore,
may be unstable even though
is designed to be stable [14] [15] .
Now using (21) and (42), it can readily be shown that the estimation error
satisfies,
, (43)
where the initial condition
is arbitrary. Since
is Hurwitz, it follows that
(44)
for all
.
Using the transformation defined by (17), the observer described by (42) in the
-coordinates can be transformed back to the
-coordinates as
, (45)
where
and
is the Jacobian matrix associated with (17) given by
. (46)
In summary, the observer-based control system with feedback linearization for the bioprocess under consideration has the form
(47)
, (48)
where
is the initial estimate of
and
, (49)
, (50)
, (51)
(52)
6.2. Luenberger Observer for BS Control
In this section we pursue our final objective, i.e., to design a Luenberger observer based on the BS formulation using the error Equation (40). To construct an observer for (40), we need an output equation which can be defined as,
, (53)
where
is known and
is an observable pair.
We present the following proposition.
Proposition 1
Consider the bioprocess system described by (3) and (4). A Luenberger observer for the associated error system (40) with measurement given by (53) can be constructed as
(54)
where
and
is the observer gain matrix to be determined by, for example poleplace- ment, such that the observer matrix
is Hurwitz. Since
is already Hurwitz and
is an observable pair,
can be determined such that the real parts of the eigenvalues of
lie on the left-side of those of
on the open left-half plane, if desirable. Furthermore, (54) can be expressed in the x-coordinates as
(55)
Proof: First, we need to show that the estimate
converges to its true value
. Define the estimation error as
. From (40) and (54), it follows that
satisfies,
(56)
where
is the initial condition. Since
is Hurwitz, it follows that
, (57)
for arbitrary
Next, using, (24), (25) and (29), the coordinates transformation for the error-system can be obtained as
. (58)
Equation (58) yields the Jacobian matrix
, (59)
where
is nonsingular for
so that
is a local diffeomorphism for (3). Equa-
tions (58) and (59) yield
and, from (53) and (54),
, (60)
where
is the inverse of
, and
, (61)
(62)
which complete the proof. ![]()
The observer design technique developed here is interesting and attractive and is different from the two-filter approach in [9] . The technique can be applied to a wide class of BS-based error systems.
In summary, the observer-based control system with the BS formulation for the bioprocess is described by
, (63)
(64)
where
is the initial estimate of
.
Simulation studies for the proposed observer-based FBL and BS controls were conducted and compared. The initial conditions were chosen as
and
. The set-point reference
was the same as before. The model parameters were given in Table 1 and Table 2.
In Figure 5, results for observer-based FBL control scheme described by (47) and (48) are shown. It can be seen that the estimates
and
converged to the true states around
h. In Figure 6, results for the observer-based BS control scheme are presented. Convergence of the estimated states to the actual states can also be seen from this figure, and are similar to those presented in Figure 2.
In Figure 7, the behavior of the error variables
and
defined by (24) and (25) which satisfy (40) in the backstepping scheme is shown. It is evident that
and
smoothly after the transients are over around
h.
7. Conclusion
Observers are critical to control system analysis and designs that employ full-state feedback, where not all the state variables are accessible for on-line, real-time measurements, and/or where the measurements are corrupted by noise. Indeed, the design of suitable linear or nonlinear observers or filters leading to observer-based control technology is an integral part of real world control system applications. In this paper, observer-based control strategies were developed for a nonlinear bioprocess system using feedback linearization and backstepping control techniques; in particular, a Luenberger observer for backstepping control was formulated using the error equation resulted from the backstepping design procedure. The observer design technique developed here is interesting and attractive and is different from the two-filter approach known in the literature. Simulation results with and without observers for both the FBL and BS schemes are presented and compared. The results were excellent and demonstrated the feasibility and effectiveness of the proposed approaches.
![]()
Table 2. Controllers and observer gains.
![]()
Figure 5. Responses of observer-based FBL control scheme (47) and (48):
and
smoothly.
![]()
Figure 6. Responses of observer-based BS control scheme (63) and (64):
and
smoothly.
![]()
Figure 7. Evolution of the backstepping error variables:
and
smoothly.
Acknowledgements
The authors would like to thank all the reviewers for their feedbacks and constructive criticisms.