Design of decentralized proportional–integral–derivative controller based on decoupler matrix for two-input/two-output process with active disturbance rejection structure

Abstract

The framework of the active disturbance rejection internal model control is proposed to solve the problem of the model reduced-order error in the process of the controller design for two-input/two-output system with time delay. In the controller design process of the two-input/two-output system, the decoupler matrix method is used to decompose a multi-loop control system into a set of equivalent independent single loops. Then, a complex equivalent model is obtained, and its order should be reduced for each individual loop. Maclaurin series method is used to reduce the order of the decoupling model. After reducing model order, the proposed method is applied to lessen the effect of the reduced-order error and improve the anti-interference ability and robustness for the control system. Simulation results show that the proposed method possesses a good disturbance rejection performance.

Keywords

Decoupler matrix internal model control–proportional–integral–derivative active disturbance rejection internal model controller Maclaurin series

Introduction

Many chemical industrial processes can be described as the multivariable process. Due to the interactions between the input and output variables, the design process of the multivariable controller is complicated. To simplify the design procedure of the multivariable controller, a decoupler is designed to decompose a multi-loop control system into a set of equivalent independent single loops. Three types of basic decoupler design method are available: ideal, simplified, and inverted decoupling.

The two-input/two-output (TITO) process is the most common multivariable system in industry, such as continuous stirred tank reactor and chemical reactor. Many scholars concern on the design controller of the TITO with dead-time process.^1,2 Tong and Wang³ addressed the problem of fuzzy robust decentralized control for a class of fuzzy large-scale systems with time delay based on the Takagi–Sugeno (T-S) fuzzy model and developed both fuzzy state feedback decentralized controller and fuzzy observer-based decentralized controller. Liu et al.⁴ studied an adaptive neural output feedback tracking control of uncertain nonlinear multi-input–multi-output (MIMO) systems in the discrete-time form and developed an adaptive neural output feedback control for MIMO nonlinear discrete-time systems with the couplings in unknown functions. Li et al.⁵ investigated the adaptive neural network (NN) control for a class of block triangular MIMO nonlinear discrete-time systems with each subsystem in pure-feedback form with unknown control directions and designed adaptive output NN controls for the nonlinear MIMO system based on the prediction control. Liu et al.⁶ addressed the problems of stability and tracking control for a class of large-scale nonlinear systems with unmodeled dynamics by designing the decentralized adaptive fuzzy output feedback approach. Bhalodia and Weber⁷ described a model for a TITO interacting process which characterizes the four open-loop responses by six time constants and two interacting gain factors; and then, proportional–integral (PI) controllers were applied to two feedback control loops. Lawrence and Douglas⁸ detailed an applied investigation of pattern recognition based on adaptive control for TITO systems. Palmor et al.⁹ presented a new algorithm for automatic tuning of decentralized proportional–integral–derivative (PID) control for TITO plants that fully extends the single-loop relay auto-tuner to the multi-loops case. Liu et al.¹⁰ proposed a new analytical decoupling control scheme that was based on a standard internal model control (IMC)¹¹ structure, which was capable of absolute decoupling regulation for the nominal binary system output responses. Li et al.¹² designed a new identification method based on the decentralized step-test for TITO processes with significant interactions. Pontus and Tore¹³ treated controller design and tuning for systems with two input signals and two output signals in the process industry and proposed a controller structure which consisted of a decoupler and a diagonal PID controller. Rao et al.^14,15 proposed a decentralized controller design based on the direct synthesis method for TITO processes in the Smith predictor configuration.

Recently, many decoupling methods are proposed to deal with the loop interaction. Jevtović and Mataušek¹⁶ proposed a method for designing multivariable controller based on ideal decoupler and PID controller optimization under constraints on the robustness and sensitivity to measurement noise. Vu and Lee¹⁷ proposed a new analytical method based on the direct synthesis approach for the design of a multi-loop PI controller to achieve the desired closed-loop response for MIMO processes with multiple time delays. Nie et al.^18,19 presented a single-iteration strategy for the design of a multi-loop PI controller to achieve the desired gain and phase margins for TITO processes. Vargas et al.²⁰ concerned the stabilization of discrete-time TITO linear time-invariant (LTI) systems over communication channels.

This article aims to improve the disturbance rejection performance of decoupling control system for TITO processes with the dead-time process. A compound control structure is introduced into the controller design process of decoupler matrix. The proposed control scheme can handle the error of the reduced-order process. Meanwhile, the proposed method shows the trade-off between the performance and the robustness.

Diagonal decoupling control

The TITO system is one of the most prevalent categories of multivariable systems, because there are real processes of this nature or because a complex process has been decomposed into TITO process with non-negligible interactions between its inputs and outputs. Decentralized PID controllers with decoupler are most commonly preferred for TITO processes.²¹ The traditional decoupling control system is shown in Figure 1.

Decoupler matrix method²² is used in the design process of decentralized PID controllers for TITO plant. The procedure of the decoupler matrix method is given. Let the TITO plant be as follows

G (s) = [\begin{matrix} g_{11} (s) e^{- τ_{11} s} & g_{12} (s) e^{- τ_{12} s} \\ g_{21} (s) e^{- τ_{21} s} & g_{22} (s) e^{- τ_{22} s} \end{matrix}]

(1)

To design decoupler matrix, the following two cases are considered:

Case 1. The off-diagonal elements of G(s) have no right-half-plane (RHP) poles and diagonal elements of G(s) have no RHP zeros. The decoupler matrix is

D (s) = [\begin{matrix} v_{1} (s) & d_{12} (s) v_{2} (s) \\ d_{21} (s) v_{1} (s) & v_{2} (s) \end{matrix}]

(2)

where

v_{1} (s) = {\begin{matrix} 1 & τ_{21} \geq τ_{22} \\ e^{(τ_{21} - τ_{22}) s} & τ_{21} < τ_{22} \end{matrix}

(3a)

v_{2} (s) = {\begin{matrix} 1 & τ_{12} \geq τ_{11} \\ e^{(τ_{12} - τ_{11}) s} & τ_{12} < τ_{11} \end{matrix}

(3b)

d_{12} (s) = - \frac{g_{12} (s)}{g_{11} (s)} e^{- (τ_{12} - τ_{11}) s}

(3c)

d_{21} (s) = - \frac{g_{12} (s)}{g_{11} (s)} e^{- (τ_{12} - τ_{11}) s}

(3d)

Case 2. The diagonal elements have no RHP poles and off-diagonal elements have no RHP zeros of G(s). The decoupler matrix is

D (s) = [\begin{matrix} d_{11} (s) v_{1} (s) & v_{1} (s) \\ v_{2} (s) & d_{22} (s) v_{2} (s) \end{matrix}]

(4a)

where

v_{1} (s) = {\begin{matrix} 1 & τ_{22} \geq τ_{21} \\ e^{(τ_{22} - τ_{21}) s} & τ_{22} < τ_{21} \end{matrix}

(4b)

v_{2} (s) = {\begin{matrix} 1 & τ_{11} \geq τ_{12} \\ e^{(τ_{11} - τ_{12}) s} & τ_{11} < τ_{12} \end{matrix}

(4c)

d_{11} (s) = - \frac{g_{22} (s)}{g_{21} (s)} e^{- (τ_{22} - τ_{21}) s}

(4d)

d_{22} (s) = - \frac{g_{11} (s)}{g_{12} (s)} e^{- (τ_{11} - τ_{12}) s}

(4e)

As shown in equation (5), $\bar{G} (s)$ is a diagonal matrix

\bar{G} (s) = G (s) D (s) = diag ({\bar{g}}_{ii} (s)) = [\begin{matrix} {\bar{g}}_{11} (s) & 0 \\ 0 & {\bar{g}}_{22} (s) \end{matrix}]

(5)

The decoupled elements ${\bar{g}}_{ii} (s)$ , where i = 1,2, are to be controlled through decentralized PID controllers.

Model reduced-order method

The dynamics of the elements ${\bar{g}}_{ii} (s)$ of $\bar{G} (s)$ are quite complex. Due to the complexity of the decoupled model, it cannot be applied to design of controller using the traditional design method. So, the high-order model should be reduced in order to design the controller. The frequency of the reduced model approximates that of the original model. However, the reduced model order affects the performance of the control system. So, the proposed method is applied to lessen the effect of the reduced-order error and improve the anti-interference ability and robustness for the control system.

A first-order plus dead-time (FOPDT) model often reasonably describes the process gain, overall time constant, and effective dead time of higher order processes. Thus, the model order is reduced to simplify the derivation for higher order processes. Many methods can be used to reduce order, such as polynomial approximation,²³ the Gaussian frequency domain approach,²⁴ and least squares algorithm.²⁵ Model reduction can be described as searching a low-order model with a small time delay to approximate the full-order model. In this section, a simple Maclaurin series reduction-order method²⁶ is shown. A FOPDT model ${\bar{g}}_{ii}^{r} (s)$ of each element ${\bar{g}}_{ii} (s)$ of $\bar{G} (s)$ is obtained as follows

{\bar{g}}_{ii}^{r} (s) = \frac{K_{ii} e^{- θ_{ii} s}}{t_{ii} s + 1}

(6)

${\bar{g}}_{ii} (s)$ is expanded in a Maclaurin series. The result is given by

{\bar{g}}_{ii} (s) = a_{ii} + b_{ii} s + c_{ii} s^{2} + d_{ii} s^{3} + e_{ii} s^{4} + O (s^{5})

(7)

where the coefficients of equation (7) polynomial are

a_{ii} = {\bar{g}}_{ii} (0)

(8a)

b_{ii} = \frac{d {\bar{g}}_{ii} (s)}{ds} |_{s = 0}

(8b)

c_{ii} = \frac{1}{2} \frac{d^{2} {\bar{g}}_{ii} (s)}{d s^{2}} |_{s = 0}

(8c)

d_{ii} = \frac{1}{6} \frac{d^{3} {\bar{g}}_{ii} (s)}{d s^{3}} |_{s = 0}

(8d)

e_{ii} = \frac{1}{24} \frac{d^{4} {\bar{g}}_{ii} (s)}{d s^{4}} |_{s = 0}

(8e)

Expanding the reduced diagonal matrix, given by equation (6), into a Maclaurin series in s results in

{\bar{g}}_{i i}^{r} (s) = K_{i i} - K_{i i} (θ_{i i} + t_{i i}) s + K_{i i} [\frac{1}{2} θ_{i i}^{2} + (θ_{i i} + t_{i i}) t_{i i}] s^{2} + O_{i i} (s^{3})

(9)

where K, t, and $θ$ should be identified to approximate ${\bar{g}}_{ii} (s)$ as close as possible over relevant frequency ranges. Trying the following: comparing the first, the second, and the third terms in equation (9) with those in equation (7) leads to the following explicit equations for K, t, and $θ$

K_{ii} = a_{ii}

(10a)

t_{ii} = \sqrt{\frac{2 c_{ii}}{a_{ii}} - {(\frac{b_{ii}}{a_{ii}})}^{2}}

(10b)

θ_{ii} = - \frac{b_{ii}}{a_{ii}} - \sqrt{\frac{2 c_{ii}}{a_{ii}} - {(\frac{b_{ii}}{a_{ii}})}^{2}}

(10c)

To get feasible results for the FOPDT model, t and $θ$ should be real and positive. It is clear from equation (10) that the following condition should be satisfied for feasible t and $θ$ values

\sqrt{\frac{2 c_{ii}}{a_{ii}}} > (- \frac{b_{ii}}{a_{ii}}) > \sqrt{\frac{2 c_{ii}}{a_{ii}} - {(\frac{b_{ii}}{a_{ii}})}^{2}}

(11)

Remark 1

Any technique can be applied toward fitting the decoupled model into a low-order model. In this work, for the purpose of evaluating the proposed method, a simple model reduced technique was adopted based on the coefficient matching method. Maclaurin series method is based on the coefficient matching method without considering the frequency response fitting; Maclaurin series method makes the readers understood easily; the obtained reduced error fits to demonstrate the property of the proposed method. So, Maclaurin series is applied into model reduced order in this article.

Active disturbance rejection control structure

A composite control structure is proposed to conquer the drawback of the reduced error and improve the system robustness. In other words, a design method of the active disturbance rejection internal model control (ADRIMC) is proposed. Then, the design procedures are shown.

Control structure

The system structure of ADRIMC is shown in Figure 2.

In Figure 2, r_i(s) and y_i(s) define the reference input and output, respectively; τ_disi(s) and ξ_i(s) define the internal disturbance and the external disturbance, respectively; G_ci(s) defines the controller of ADRIMC controller; u_i and c_i define the external control law and the internal control law, respectively; and ${\hat{τ}}_{disi}$ defines the estimation disturbance.

Let

{\bar{g}}_{ii}^{r} (s) = g_{ni} (s) e^{- θ_{ni} s}

(12)

Q(s) is low-pass filter (LPF), which is given by

Q (s) = \frac{1 + \sum_{k = 1}^{N - m} a_{k} {(λ_{1 i} s)}^{k}}{1 + \sum_{k = 1}^{N} a_{k} {(λ_{1 i} s)}^{k}}

(13)

where N is the order of Q(s), λ_1i is the filter time constant of the ith loop, and m is the relative degree of Q(s).²⁷

In Figure 2, the transfer function of the output response is given by

y_{i} = {\bar{g}}_{ii} (s) u_{i} + D_{i} (s)

(14)

where the equivalent disturbance input is given by

D_{i} (s) = ξ_{i} (s) + τ_{disi} (s)

(15)

Then, the estimate disturbance is given by

{\hat{τ}}_{disi} (s) = Q_{i} (s) g_{ni}^{- 1} (s) y_{i} - Q_{i} (s) u_{i} e^{- θ_{ni} s}

(16)

From equations (14)–(16), it is known that the estimate disturbance can be defined as

{\hat{τ}}_{disi} (s) = Q_{i} (s) g_{ni}^{- 1} (s) D_{i} (s)

(17)

In Figure 2, from the first summing point, we have

u_{i} = c_{i} - {\hat{τ}}_{disi} (s)

(18)

By substituting equation (17) into equation (18), the transfer function of the manipulate variable is obtained as

u_{i} = c_{i} - Q_{i} (s) g_{ni}^{- 1} (s) D_{i} (s)

(19)

Then, if the transfer function of the manipulate variable is known, according to equation (14), the transfer function of the output response is obtained as

y_{i} = c_{i} {\bar{g}}_{ii}^{r} (s) + [1 - Q_{i} (s) e^{- θ_{ni} s}] D_{i} (s)

(20)

Remark 2

Although the equivalent disturbance D_i(s) is unknown, the equivalent disturbance has a steady-state value, that is, satisfies $lim_{s \to 0} D (s) < \infty$ . The filter Q(s) in the ADRIMC satisfies $lim_{s \to 0} Q (s) = 1$ . So, the following equation is established: $lim_{s \to 0} [1 - Q_{i} (s) e^{- θ_{ni} s}] D_{i} (s) = 0$ . The final transfer function is $lim_{s \to 0} y_{i} = lim_{s \to 0} c_{i} {\bar{g}}_{ii}^{r} (s)$ .

Thus, when the equivalent disturbances have a steady-state value, the suitable $λ_{1 i}$ is selected to conquer the disturbance influence.

Note that

G_{pi} (s) = {\bar{g}}_{ii}^{r} (s)

(21)

G_{di} (s) = 1 - Q_{i} (s) e^{- θ_{ni} s}

(22)

then, the transfer function of the output response can be described as

y_{i} = G_{pi} (s) c_{i} + G_{di} (s) D_{i} (s)

(23)

where G_pi(s) is the modified controlled plant model and G_di(s) is the disturbance filter.

Based on the above discussion, the system structure of ADRIMC is shown in Figure 3. Compared with the traditional IMC structure, a disturbance filter G_di(s) is located at the disturbance input in the ADRIMC structure. The disturbance filter can ensure that the ADRIMC scheme realizes the effectiveness of the ADR for the error of the reduced order. The internal model structure can ensure that the ADRIMC scheme realizes the effectiveness of the deviation-integration in itself.

Tuning controller parameters

The parameter $λ_{1}$ is tuned by the balance tuning technique.²⁸ Assume that the controlled plant of the each loop is second-order plus dead-time (SOPDT) model, as shown in the following equation

G_{pi} (s) = \frac{k_{i}}{τ_{1 i} s^{2} + τ_{2 i} s + 1} e^{- θ_{i} s}, G_{p 0 i} (s) = \frac{k_{i}}{τ_{1 i} s^{2} + τ_{2 i} s + 1}

(24)

And then, the IMC controller is obtained by the following equation

G_{C i} (s) = \frac{τ_{1 i} s^{2} + τ_{2 i} s + 1}{k_{i}} \frac{β_{i} s + 1}{{(λ_{1 i} s + 1)}^{3}} = \frac{(β_{i} s + 1) (τ_{1 i} s^{2} + τ_{2 i} s + 1)}{k_{i} {(λ_{1 i} s + 1)}^{3}}

(25)

where $β_{i} = α_{i} λ_{1 i}$ , $α_{i}$ is the value of $β_{i} / λ_{1 i}$ and $α_{i}$ shows the filter characteristic.

The equivalent feedback controller $G_{0 i} (s)$ is given by

G_{0 i} (s) = \frac{G_{C i} (s)}{1 - G_{C i} (s) G_{i} (s)} = \frac{(β_{i} s + 1) (τ_{1 i} s^{2} + τ_{2 i} s + 1)}{k_{i} {(λ_{1 i} s + 1)}^{3} - k_{i} (β_{i} s + 1) e^{- θ_{i} s}}

(26)

When $r_{i} (s) = 1$ , $d_{i} (s) = 0$ , and the model is matched, the closed-loop response of the system is shown in equation (27)

y_{ri} (s) = \frac{G_{0 i} (s) G_{pi} (s)}{1 + G_{0 i} (s) G_{pi} (s)} r_{i} (s)

(27)

When $r_{i} (s) = 0$ , $d_{i} (s) = 1$ , and the model is matched, the closed-loop response of the system is shown in equation (28)

y_{di} (s) = \frac{G_{di} (s)}{1 + G_{0 i} (s) G_{pi} (s)} D_{i} (s)

(28)

where $y_{di} (s)$ is the disturbance output of ith loop and $G_{0 i} (s)$ is equivalent feedback controller of ith loop.

And then, the closed-loop response of the whole system is given by

y_{i} (s) = \frac{G_{0 i} (s) G_{pi} (s)}{1 + G_{0 i} (s) G_{pi} (s)} r_{i} (s) + \frac{G_{di} (s)}{1 + G_{0 i} (s) G_{pi} (s)} D_{i} (s)

(29)

The input error of the control system is given by

e_{ri} (s) = y_{ri} (s) - r_{i} (s) = \frac{G_{0 i} (s) G_{pi} (s)}{1 + G_{0 i} (s) G_{pi} (s)} r_{i} (s) - r_{i} (s)

(30)

The sensitivity function and the complementary sensitivity function are shown in equation (31)

S_{i} = \frac{1}{1 + G_{0 i} (s) G_{pi} (s)}, T_{i} = \frac{G_{0 i} (s) G_{pi} (s)}{1 + G_{0 i} (s) G_{pi} (s)}

(31)

where $S_{i} G_{Ci} (s) G_{i} (s) - 1 = T_{i} - 1 = - S_{i}$ .

Furthermore, the relationship between the input error and the sensitivity function is given by

e_{ri} (s) = - S_{i} r_{i} (s) = (T_{i} - 1) r_{i} (s)

(32)

According to equation (32), when the input sensitivity function is smaller and the complementary sensitivity function is larger, the system error is smaller. By substituting equation (27) into equation (31), the analytical expression of the sensitivity function is shown in equation (33)

S_{i} = \frac{1}{1 + G_{0 i} (s) G_{p i} (s)} = \frac{1}{1 + \frac{(β_{i} s + 1) (τ_{1 i} s^{2} + τ_{2 i} s + 1) / k_{i}}{{(λ_{1 i} s + 1)}^{3} - (β_{i} s + 1) e^{- θ_{i} s}} G_{p i} (s)}

(33)

By simplifying equation (33), equation (34) can be obtained

S_{i} = 1 - \frac{(β_{i} s + 1) e^{- θ_{i} s}}{{(λ_{1 i} s + 1)}^{3}}

(34)

First-order Taylor approximation method is adopted to deal with the time delay part

e^{- θ_{i} s} \approx 1 - θ_{i} s

(35)

By substituting equation (35) into equation (34), equation (36) can be obtained

S_{i} = \frac{λ_{1 i}^{3} s^{3} + (3 λ_{1 i}^{2} - β_{i} θ_{i}) s^{2} + (3 λ_{1 i} - β_{i} + θ_{i}) s}{λ_{1 i}^{3} s^{3} + 3 λ_{1 i}^{2} s^{2} + 3 λ_{1 i} s + 1}

(36)

Let

μ_{i} = \frac{λ_{1 i}}{θ_{i}}, γ_{i} = \frac{β_{i}}{θ_{i}}, s_{i} = \frac{η_{i}}{θ_{i}}

(37)

where $γ_{i} / μ_{i} = β_{i} / λ_{1 i} = α_{i}$ .

By substituting equation (37) into equation (36), the expression of the sensitivity function is obtained

S_{i} = \frac{{μ_{i}}^{3} {η_{i}}^{3} + (3 {μ_{i}}^{2} - γ_{i}) {η_{i}}^{2} + (3 μ_{i} - γ_{i} + 1) η_{i}}{{μ_{i}}^{3} {η_{i}}^{3} + 3 {μ_{i}}^{2} {η_{i}}^{2} + 3 μ_{i} η_{i} + 1}

(38)

The analytical expression of the maximum sensitivity function is shown in equation (39)

M_{si} = max_{0 < ω < \infty} | \frac{{μ_{i}}^{3} {η_{i}}^{3} + (3 {μ_{i}}^{2} - γ_{i}) {η_{i}}^{2} + (3 μ_{i} - γ_{i} + 1) η_{i}}{{μ_{i}}^{3} {η_{i}}^{3} + 3 {μ_{i}}^{2} {η_{i}}^{2} + 3 μ_{i} η_{i} + 1} |

(39)

Based on $β_{i} = α λ_{1 i}$ , the relationship of $M_{si} - μ_{i}$ can be found, as shown in Figure 4. Where M_si is the maximum sensitivity function, and then the typical values of M_si are in the range of [1.0–2.0].

Assuming that the system input of each loop is step response, integral absolute error (IAE) value from input to output is given by

IA E_{ri} = \int_{0}^{\infty} | y_{ri} (t) - r_{i} (t) | dt

(40a)

IA E_{ri} = \int_{0}^{\infty} | ζ^{- 1} (\frac{1}{s} (1 - \frac{(β_{i} s + 1) e^{- θ_{i} s}}{{(λ_{1 i} s + 1)}^{3}})) | dt

(40b)

The result is obtained by

IA E_{ri} = {\begin{matrix} t_{i} \\ 3 λ_{1 i} - β_{i} + θ_{i} \end{matrix} \begin{matrix} t_{i} \leq θ \\ t_{i} > θ \end{matrix}

(41)

IAE value is manipulated by the scale transformation, as shown in equation (42)

IA {E'}_{ri} = 3 μ_{i} - γ_{i} + 1

(42a)

IA {E'}_{ri} = (3 - α_{i}) μ_{i} + 1

(42b)

where $IA {E'}_{ri} = IA E_{i} / θ_{i}$ and $γ_{i} / μ_{i} = α_{i}$ .

According to Figure 5, it is known that if $0 < α_{i} < 3$ , then $min (IA {E'}_{ri}) = 1$ ; if $α_{i} > 3$ , then $min (IA E'_{ri}) = 0$ and $μ_{i} = 1 / α_{i} - 3$ . According to Figure 5, it is known that when $α_{i} > 3$ , the system robustness is bad. So, $0 < α_{i} < 3$ is only considered, as shown in Figure 5.

It is known that balance tuning method is used for tuning the controller parameters $λ_{1 i}$ by considering the robustness and stability. Furthermore, PID parameters of the controller are shown by

G_{0 i} (s) = \frac{(β_{i} s + 1) (τ_{1 i} s^{2} + τ_{2 i} s + 1)}{k_{i} {(λ_{1 i} s + 1)}^{3} - k_{i} (β_{i} s + 1) e^{- θ_{i} s}} = \frac{1}{s} \bar{f_{i}} (s)

(43)

Then, expanding $C_{i} (s)$ in a Maclaurin series with s is given by

P_{i} (s) = \frac{\bar{f_{i}} (s)}{s} = \frac{1}{s} ({\bar{f}}_{i} (0) + {\bar{f_{i}}}^{'} (0) s + \frac{{\bar{f}}^{''}_{i} (0)}{2!} s^{2} + \dots)

(44)

The corresponding PID expressions are given by

C_{i} (s) = K_{mi} (1 + \frac{1}{T_{Ii} s} + T_{Di} s)

(45)

{\begin{matrix} K_{Ci} = \bar{f'_{i}} (0) \\ T_{Ii} = \frac{\bar{f'_{i}} (0)}{\bar{f_{i}} (0)} \\ T_{Di} = \frac{\bar{f_{i}^{″}} (0)}{2 \bar{f_{i}'} (0)} \end{matrix}

(46)

The results are shown in equation (47)

{\begin{matrix} K_{Ci} = \frac{(3 λ_{1 i} - β_{i} + θ_{i}) τ_{2 i} + 3 λ_{1 i}^{2} - \frac{1}{2} {θ_{i}}^{2} + β_{i} θ_{i}}{k_{i} {(3 λ_{1 i}^{2} - β_{i} + θ_{i})}^{2}} \\ T_{Ii} = \frac{(3 λ_{1 i} - β_{i} + θ_{i}) τ_{2 i} + 3 λ_{1 i}^{2} - \frac{1}{2} {θ_{i}}^{2} + β_{i} θ_{i}}{k_{i} (3 λ_{1 i}^{2} - β_{i} + θ_{i})} \\ T_{D} = \frac{k_{i} (τ_{2 i} - τ_{2 i}) (3 λ_{1 i} - β_{i} + θ_{i}) - k_{i} τ_{2 i} (3 λ_{1 i}^{2} - \frac{1}{2} {θ_{i}}^{2} + β_{i} θ_{i}) + k_{i} [λ_{1 i}^{2} - \frac{1}{6} (3 β_{i} {θ_{i}}^{2} - {θ_{i}}^{3})]}{- k_{i} (3 λ_{1 i} - β_{i} + θ_{i}) + k_{i} (3 λ_{1 i}^{2} - \frac{1}{2} {θ_{i}}^{2} + β_{i} θ_{i})} + T_{Ii} \end{matrix}

(47)

Similarly, the parameter $λ_{2 i}$ can be tuned. The error between the disturbance and the output is given by

e_{di} (s) = \frac{G_{di} (s)}{1 + G_{0 i} (s) G_{pi} (s)} D_{i} (s) - D_{i} (s)

(48)

When $lim_{s \to 0} G_{di} (s) = 0$ , the error between the disturbance and the output is maximum, and the disturbance rejection effective of the system is good.

And then, calculation of the IAE value from the disturbance to the output is obtained by

IA E_{di} = \int_{0}^{\infty} | y_{di} (t) - d_{i} (t) | dt

(49)

Assuming that the system input of each loop is step response

IA E_{di} = \int_{0}^{\infty} | ζ^{- 1} {\frac{1}{s} [\frac{G_{di} (s)}{1 + G_{0 i} (s) G_{pi} (s)} D_{i} (s) - D_{i} (s)]} | dt

(50)

Let

IA {E'}_{di} = \int_{0}^{\infty} | ζ^{- 1} [\frac{1}{s} (1 - Q_{i} (s) e^{- θ_{i} s})] | dt

(51)

It can be seen from equation (51) that if $IA {E'}_{di} \to 0$ , then $IA E_{di} \to$ maximum (Figure 6). Assuming that

Q_{i} (s) = \frac{ρ_{i} s + 1}{{(λ_{2 i} s + 1)}^{2}}

(52)

where $ρ_{i} = κ_{i} λ_{2 i}$ .

By substituting equation (52) into equation (51)

IA {E'}_{di} = \int_{0}^{\infty} | ζ^{- 1} [\frac{1}{s} (1 - \frac{ρ_{i} s + 1}{{(λ_{2 i} s + 1)}^{2}} e^{- θ_{i} s})] | dt

(53)

Equation (54) is obtained by calculating equation (53)

IA {E'}_{di} = {\begin{matrix} t_{i} \\ 2 λ_{2 i} + θ_{i} - ρ_{i} \end{matrix} \begin{matrix} t_{i} \leq θ_{i} \\ t_{i} > θ_{i} \end{matrix}

(54)

When $λ_{2 i} = - θ_{i} / 2$ and $ρ_{i} = 0$ , the error between the disturbance and the output is maximum. When $ρ_{i} = κ_{i} λ_{2 i}$ , the relationship of $IA {E'}_{di} - λ_{2 i}$ can be obtained by

\frac{IA {E'}_{di}}{θ_{i}} = {\begin{matrix} t_{i} \\ (2 - κ_{i}) ν_{i} + 1 \end{matrix} \begin{matrix} t_{i} \leq θ_{i} \\ t_{i} > θ_{i} \end{matrix}

(55)

where $λ_{2 i} / θ_{i} = ν_{i}$ .

According to equation (55), it is known that if $0 < κ_{i} < 2$ , then $min (IA {E'}_{ri}) = 1$ ; if $κ_{i} > 2$ , then $min (IA {E'}_{ri}) = 0$ and $ν_{i} = 1 / κ_{i} - 2$ .

Robustness analysis

A TITO multivariable control system is nominally stable if and only if²⁹

ρ = [\begin{matrix} 0 & \frac{G_{12} C_{11}}{1 + G_{11} C_{11}} \\ \frac{G_{21} C_{22}}{1 + G_{22} C_{22}} & 0 \end{matrix}] < 1 \forall ω

(56)

A practical way²⁹ to identify the closed-loop system robustness stability in the presence of process uncertainties is to lump multiple sources of uncertainty into a multiplicative form

ρ (K_{ii} (I + G^{- 1}) G Δ_{I}) < 1 \forall ω

(57)

ρ (G K_{ii} (I + G^{- 1}) Δ_{O}) < 1 \forall ω

(58)

where Δ_I and Δ_O are stable process multiplicative input and output uncertainties, respectively. Hence, for a specified bound of Δ_I or Δ_O, control system robustness stability is evaluated by observing the magnitude plots of the left sides of equations (57) and (58) with $ω \in [0, + \propto]$ , which should fall below unity.

Remark 3

From the aspect of computation amount, the proposed method is increased complexity as compared to standard decoupling structure available in the literature. However, the performance of the control system is improved. Thus, the proposed method shows a trade-off between the dynamic performance and system robustness.

Simulation example

We consider two simulation examples to show the simplicity and effectiveness of the proposed method. One is Wood and Berry (WB) column (methanol–water distillation column)³⁰ and other one is Alatiqi subsystem distillation column.³¹

Example 1

Taking WB column as an example, the matrix function expression of the plant model is shown in the following equation

G_{p} (s) = [\begin{matrix} \frac{12.8 e^{- s}}{16.7 s + 1} & \frac{- 18.9 e^{- 3 s}}{21 s + 1} \\ \frac{6.6 e^{- 7 s}}{10.9 s + 1} & \frac{- 19.4 e^{- 3 s}}{14.4 s + 1} \end{matrix}]

According to section “Diagonal decoupling control,” the decoupler of WB column can be obtained by

D (s) = [\begin{matrix} 1 & \frac{1.476 (16.7 s + 1) e^{- 2 s}}{21.3 s + 1} \\ \frac{0.33 (14.2 s + 1) e^{- 4 s}}{10.89 s + 1} & 1 \end{matrix}]

Then, according to equation (5), the diagonal matrix $\bar{G} (s)$ of WB column is given by

\begin{matrix} {\bar{g}}_{11} (s) = \frac{12.8 e^{- s}}{16.7 s + 1} - \frac{6.237 (14.36 s + 1)}{228.69 s^{2} + 31.89 s + 1} e^{- 7 s} \\ {\bar{g}}_{22} (s) = \frac{- 19.4 e^{- 3 s}}{14.4 s + 1} + \frac{9.745 (16.7 s + 1)}{228.69 s^{2} + 31.89 s + 1} e^{- 9 s} \end{matrix}

Then, according to section “Model reduced-order method,” the reduced-order model of $\bar{G} (s)$ can be obtained by

\begin{matrix} {\bar{g}}_{11}^{r} (s) = \frac{6.563}{10.529 s + 1} e^{- 0.309 s} \\ {\bar{g}}_{22}^{r} (s) = \frac{- 9.656}{6.273 s + 1} e^{- 4.266 s} \end{matrix}

The Bode plots of the first loop and the second loop are shown in Figure 7. Figure 7(a) shows the frequency characteristics comparison between the original model and the reduced-order model in the first loop. Figure 7(b) shows the frequency characteristics between the original model and the reduced-order model in the second loop.

Applying the HO-PI,³² BLT,³¹ and SAT³³ controllers to WB system output responses to a unit step function in the first or second input is shown in Figures 8 –11. The obtained results are compared with those of the proposed method. Figures 8 –11 show the track performance of the control system for WB control system. Figure 8 shows the set-point response and control law of the first loop for WB control system. Figure 9 shows the set-point response and control law of the second loop for WB system. In Figures 8 and 9, the obtained results show that proposed method possesses the favorable track performance.

Figure 10 shows the disturbance response and control law of the first loop for WB control system. Figure 8 shows the disturbance response and control law of the second loop for WB control system. In Figures 10 and 11, the obtained results show that proposed method possesses the favorable ability of the disturbance rejection. The performance index of the set-point response for WB system is shown in Table 1.

Table 1.

Performance indexes for WB column.

Method	Loop	K_mi	T_Ii	T_Di	λ _1i	λ _2i	M_si	IAE_I	IAE_d
HO-PI	1	0.57	0.027	–	–	–	–	15.44	19.84
HO-PI	2	−0.11	−0.0085	–	–	–	–	8.51	10.35
BLT	1	0.375	0.045	–	–	–	–	25.83	38.94
BLT	2	−0.075	−0.003	–	–	–	–	16.31	28.63
SAT	1	0.87	0.267	–	–	–	–	13.79	18.32
SAT	2	−0.09	−0.0086	–	–	–	–	6.68	9.07
Proposed	1	0.41	0.074	0.001	1.75	0.72	1.31	11.34	15.91
Proposed	2	−0.12	−0.024	−0.023	2.84	4.91	1.53	4.69	7.48

IAE: integral absolute error.

To further investigate and evaluate the control performance at frequency, we impose the frequency responses on the control system. The impulse response diagram is shown in Figure 12.

The impulse response of control system provides information about the output of WB control system at different frequencies. From impulse response, it can be seen that the magnitude of the proposed method is smaller than that of other three methods for both loops. For the stability being evaluated in the frequency domain, the Nichols charts of each control system are drawn in Figure 13. In Figure 13, the gain margin of the proposed method is 9.3 dB and the phase margin of the proposed method is 35° for the first loop; and then, for the second loop, the gain margin of the proposed method is 6.5 dB and the phase margin of the proposed method is 60°. The results show that the proposed method possesses the favorable stability and robustness.

Example 2

A modified Alatiqi subsystem distillation column is shown in the following equation

G (s) = [\begin{matrix} \frac{- 0.51 e^{- 7.5 s}}{{(32 s + 1)}^{2} (2 s + 1)} & \frac{1.68 e^{- 2 s}}{{(28 s + 1)}^{2} (2 s + 1)} \\ \frac{- 1.25 e^{- 2.8 s}}{(43.6 s + 1) (9 s + 1)} & \frac{4.78 e^{- 1.15 s}}{(48 s + 1) (5 s + 1)} \end{matrix}]

According to sections “Diagonal decoupling control” and “Model reduced-order method,” the decoupler matrix of Alatiqi subsystem can be given by

\begin{matrix} D (s) = [\begin{matrix} 1 & \frac{3.294 {(32 s + 1)}^{2}}{28 s + 1} \\ \frac{0.262 (48 s + 1) (5 s + 1) e^{- 1.65 s}}{(43.6 s + 1) (9 s + 1)} & e^{- 5.5 s} \end{matrix}] \\ {\bar{g}}_{11} (s) = \frac{- 0.51 e^{- 7.5 s}}{(32 s + 1) (2 s + 1)} + \frac{0.439 (48 s + 1) (5 s + 1) e^{- 3.65 s}}{{(28 s + 1)}^{2} (43.6 s + 1) (9 s + 1) (2 s + 1)} \\ {\bar{g}}_{22} (s) = \frac{4.78 e^{- 6.65 s}}{(48 s + 1) (5 s + 1)} - \frac{4.118 (32 s + 1) e^{- 2.8 s}}{{(28 s + 1)}^{2} (43.6 s + 1) (9 s + 1)} \end{matrix}

Then, the reduced-order model of $\bar{G} (s)$ can be obtained by

\begin{matrix} {\bar{g}}_{11}^{r} (s) = \frac{- 0.071}{58.763 s + 1} e^{- 77.231 s} \\ {\bar{g}}_{22}^{r} (s) = \frac{- 0.662}{32.141 s + 1} e^{- 61.976 s} \end{matrix}

The Bode plots of the first loop and the second loop are shown in Figure 14. Figure 14(a) shows the frequency characteristics between the original model and the reduced-order model in the first loop. Figure 14(b) shows the frequency characteristics between the original model and the reduced-order model in the second loop.

Applying the MV,³⁴ NDT,²² and Wang³⁵ controllers to the modified Alatiqi subsystem output responses to a unit step function in the first or second input is shown in Figures 12 –15. The results are compared with those of the proposed method. Figures 15 –18 show the track performance of the control system for the modified Alatiqi subsystem.

Figure 15 shows the set-point response and control law of the first loop for the modified Alatiqi subsystem. Figure 16 shows the set-point response and control law of the second loop for the modified Alatiqi subsystem. In Figures 15 and 16, the obtained results show that proposed method shows the trade-off between the dynamic performance and robustness. Figure 17 shows the disturbance response and control law of the first loop for the modified Alatiqi subsystem. Figure 18 shows the disturbance response and control law of the second loop for the modified Alatiqi subsystem.

In Figures 17 and 18, the obtained results show that proposed method possesses the favorable ability of the disturbance rejection. The performance index of the set-point response for the modified Alatiqi control system is shown in Table 2.

Table 2.

Performance indexes for Alatiqi subsystem.

Method	Loop	K_mi	T_Ii	T_Di	λ _1i	λ _2i	M_si	IAE_I	IAE_d
MV	1	−2.12	−0.03	–	–	–	–	22.36	31.64
	1	−0.126	0.013	0.467	–	–	–	22.36	31.64
	2	−0.218	−0.009	0.624	–	–	–	25.04	36.19
	2	0.049	0.004	–	–	–	–	25.04	36.19
NDT	1	−6.393	−0.092	–	–	–	–	18.03	23.42
NDT	2	0.499	0.012	–	–	–	–	21.72	30.37
Wang	1	−2.176	−0.059	0.807	–	–	–	15.94	21.79
Wang	2	0.154	0.005	−0.117	–	–	–	19.08	24.04
Proposed	1	−2.14	−0.046	0.913	5.965	134.298	1.39	17.16	22.95
Proposed	2	0.504	0.0089	−0.091	8.378	65.850	1.63	20.67	28.47

IAE: integral absolute error.

To further investigate and evaluate the control performance at frequency, we impose the frequency responses on the control system. The impulse response diagram is shown in Figure 19. For the stability being evaluated in the frequency domain, the Bode diagrams of each control system are shown in Figure 20.

According to the simulation results of Examples 1 and 2, it is observed that the proposed method can improve the anti-interference ability and realize the effectiveness of the ADR. However, the improvement of the anti-interference ability is at the cost of losing the partial dynamic performance. Hence, the proposed method shows a trade-off between the dynamic performance and system robustness.

Conclusion

In this article, a compound control structure of ADRIMC is proposed and applied to TITO system with time delay process for improving the system robustness. The decoupler matrix method is applied to decoupled TITO system, and then, Maclaurin series reduced-order method is used to find an equivalent FOPDT for each loop. The controller parameters are tuned by IMC-PID technique. Using this design method, the system robustness is improved on the structure uncertainty aspect of the control system. Further research extending the presented methodology to the design of multivariable controllers for $n \times n$ dimension system and non-square systems with large model errors is currently ongoing.

Footnotes

Acknowledgements

The authors are grateful to the anonymous reviewers for their valuable recommendations.

Academic Editor: Liyuan Sheng

Author’s Note

Author Liye Liu is now affiliated to Department of Electrical and Electronic Engineering, ShiJiaZhuang Vocational Technology Institute, Shijiazhuang City, P.R. China.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported by the National Natural Foundation of China (61273132) and higher school specialized research fund for the doctoral program (20110010010) and partly by the Automation Institute Beijing University of Chemical Technology.

References

Bakule

Decentralized control: an overview. Annu Rev Control 2008; 32: 87–98.

Tong

Chen

GR.

Adaptive fuzzy decentralized control for a class of large-scale nonlinear systems. IEEE T Syst Man Cy B 2004; 34: 770–775.

Tong

Wang

Decentralized robust control for uncertain TS fuzzy large-scale systems with time-delay. Int J Innov Comput I 2007; 3: 657–672.

Liu

Chen

CLP

Wen

. Adaptive neural output feedback tracking control for a class of uncertain discrete-time nonlinear systems. IEEE T Neural Networ 2011; 22: 1162–1167.

Yang

. Adaptive output feedback NN control of a class of discrete-time MIMO nonlinear systems with unknown control directions. IEEE T Syst Man Cy B 2011; 41: 507–517.

Liu

Tong

Chen

CLP

. Adaptive fuzzy control via observer design for uncertain nonlinear systems with unmodeled dynamics. IEEE T Fuzzy Syst 2013; 21: 275–288.

Bhalodia

Weber

TW.

Feedback control of a two-input, two-output interacting process. Ind Eng Chem Proc DD 1979; 4: 599–607.

Lawrence

Douglas

JC.

Pattern recognition based adaptive control of two-input/two-output systems using ART2-A neural networks. Ind Eng Chem Proc DD 1994; 33: 1510–1519.

Palmor

Halevi

Krasney

Automatic tuning of decentralized PID controllers for TITO processes. Automatica 1995; 7: 1001–1010.

10.

Liu

Zhang

DY.

Analytical design of decoupling internal model control (IMC) scheme for two-input-two-output (TITO) processes with time delays. Ind Eng Chem Res 2006; 45: 3149–3160.

11.

Garcia

Morari

Internal model control: a unifying review and some new results. Ind Eng Chem Proc DD 1982; 21: 308–323.

12.

Cai

Mei

. Effective decentralized TITO process identification from closed-loop step responses. Asian J Control 2005; 2: 154–162.

13.

Pontus

Tore

Decoupler and PID controller design of TITO systems. J Process Contr 2006; 16: 923–936.

14.

Rao

VSR

Chidambaram

Delay-compensated controllers for two-input/two-output (TITO) multivariable processes. Asia-Pac J Chem Eng 2007; 2: 510–516.

15.

Rao

VSR

Chidambaram

. Direct synthesis-based controller design for integrating processes with time delay. J Fran Ins 2009; 346: 38–56.

16.

Jevtović

Mataušek

MR.

PID controller design of TITO system based on ideal decoupler. J Process Contr 2010; 20: 869–876.

17.

TNL

Lee

. Multi-loop PI controller design based on the direct synthesis for interacting multi-time delay processes. ISA T 2010; 49: 79–86.

18.

Nie

Wang

. Tuning of multi-loop PI controllers based on gain and phase margin specifications. J Process Contr 2011; 21: 1287–1295.

19.

Nie

Wang

. A novel computational method for loop gain and phase margins of TITO systems. J Frankl Inst 2013; 350: 503–520.

20.

Vargas

Silva

Chen

Stabilization of two-input two-output systems over SNR-constrained channels. Automatica 2013; 49: 3133–3140.

21.

Wang

Ning

Yang

Multivariable PID decoupling control method of electroslag remelting process based on improved particle swarm optimization (PSO) algorithm. Information 2014; 5: 120–133.

22.

Tavakoli

Griffin

Fleming

PJ.

Tuning of decentralized PI (PID) controllers for TITO processes. Control Eng Pract 2006; 14: 1069–1080.

23.

Gawthrop

Nihtila

MT.

Identification of time-delays using a polynomial identification method. Syst Control Lett 1985; 5: 267–271.

24.

Pintelon

Biesen

LV.

Identification of transfer functions with time delay and its application to cable fault location. IEEE T Instrum Meas 1990; 39: 479–484.

25.

Strejc

Least squares parameter estimation. Automatica 1980; 16: 535–550.

26.

TNL

Lee

. Independent design of multi-loop PI/PID controllers for interacting multivariable processes. J Process Contr 2010; 20: 922–933.

27.

Umeno

Kaneko

Hori

Robust servo system design with two degrees of freedom and its application to novel motion control of robot manipulators. IEEE T Ind Electron 1993; 40: 473–485.

28.

Zhao

Liu

Zhang

IMC-PID tuning method based on sensitivity specification for process with time-delay. J Cent South Univ T 2011; 18: 1153–1160.

29.

Mokadam

Patre

Maghade

DK.

Tuning of multivariable PI/PID controllers for TITO processes using dominant pole placement approach. Int J Control Autom 2013; 7: 21–41.

30.

Wood

Berry

MW.

Terminal composition control of binary distillation column. Chem Eng Sci 1973; 28: 1707–1717.

31.

Luyben

WL.

Simple method for tuning SISO controllers in multivariable system. Ind Eng Chem Proc DD 1986; 25: 654–660.

32.

Lee

Gan

OP.

Tuning of multi-loop PID controllers based on gain and phase margin specifications. Ind Eng Chem Res 1997; 6: 2231–2238.

33.

Loh

Hang

Quek

CK.

Auto-tuning of multi-loop proportional integral controllers using relay feedback. Ind Eng Chem Res 1993; 6: 1102–1107.

34.

Wang

Zou

Lee

. Auto-tuning of multi-variable PID controllers from decentralized relay feedback. Automatica 1997; 33: 319–330.

35.

Wang

Huang

Guo

Auto-tuning of TITO decoupling controllers from step tests. ISA T 2000; 39: 407–418.